American Journal of Critical Care - ajcc.aacnjournals.orgajcc.aacnjournals.org/content/27/3/local/complete-issue.pdf · ISI Alerting Services, Current Contents/Clinical Medi- cine,

Integrative Nursing Intervention to Reduce Patients’ Pain

Barriers to Nurse- Facilitated Patient Mobility in the ICU

Mobilization Therapy in the PICU

Early Blood Transfusions in Sepsis

Critical Care Nurses’ Experiences With Spiritual Care

Survival of Patients With Severe ARDS Treated Without ECMO

Fluid Response to Passive Leg Raising

Early Warning Score Communication Bundle

Effect of Dynamic Light on Nurses

May 2018 • Volume 27, Number 3

American Journal ofCritical Care

LEADER IN NASAL DECOLONIZATION™

LEADER IN NASAL DECOLONIZATION™

Nozin.com 877-669-4648

Ask a Nozin advisor for a savings analysis and professional samples.

• Safely Reduce MRSA Contact Precautions

• 30 Bed ICU Could Save $600,000 Per Year*

• Improve Patient Care and Satisfaction

1. Huang SS et al. Targeted versus universal. N Engl J Med, 2013: 368(24): 2255-65. Mupirocin and CHG used in study. 2. Huang SS et al. Cost Savings of Universal Decolonization

to Prevent Intensive Care Unit Infection. ICHE, 2014: 35 (S3): S23-S31. 3. Steed L, et al. Reduction of nasal Staphylococcus aureus carriage. AJIC, 2014: 42(8): 841-846. ©2018 Global

Life Technologies Corp. All rights reserved. Made in USA. Nozin®, Nasal Sanitizer®, Leader in Nasal Decolonization™ are trademarks of Global Life Technologies Corp. Nozin® Nasal

Sanitizer® antiseptic is an OTC topical drug. No claim is made that it has an effect on any specific disease. * Savings estimates are for example only. Any actual savings may vary.

Nasal Decolonization is The Key to Reduce CP

Contact isolation precautions (CP) can adversely affect patient care, utilization of staff and costs. Research shows that universal skin and nasal decolonization can help.1, 2

Nozin ICU programs use universal patient decolonization to safely reduce screening and CP for MRSA colonization.

With the Nozin program, healthcare facilities are able to lessen CP burdens on patients, address staff isolation fatigue, and support antibiotic stewardship.

Nozin® Nasal Sanitizer® antiseptic is clinically proven to decolonize nasal S. aureus day one without antibiotics.3 Learn how a Nozin program can be the key to safely reducing CP in critical care at your facility.

NTI 2018 - Visit Nozin Booth 2973 - Earn ExpoEd CERPs

www.ajcconline.org

Evidence-based interdisciplinary knowledge for high acuity and critical care

AMERICAN ASSOCIATION OF CRITICAL-CARE NURSESPresident, CHRISTINE SCHULMAN, RN, MS, CNS, CCRN-K; President-elect, LISA RIGGS, RN, MSN, APRN-BC, CCRN-K; Secre-tary, MICHELLE KIDD, RN, MS, ACNS-BC, CCRN-K; Trea surer, LOUISE SALADINO, RN, DNP, MHA, CCRN-K; Directors, ELIZABETH BRIDGES, RN, PhD, CCNS; KIMBERLY CURTIN, RN, DNP, APRN, ACNS-BC, CCRN, CEN, CNL; JUSTIN DiLIBERO, RN,

DNP, CCRN, CCNS, ACCNS-AG; NIKKI DOTSON-LORELLO, RN, BSN,

CCRN, CPTC; WENDI FROEDGE, RN-BC, MSN, CCRN-K; DEBORAH JONES, RN, MS, PhD; MARY BETH FLYNN MAKIC, RN, PhD, CNS,

CCNS, CCRN-K; ROSEMARY TIMMERMAN, RN, DNP, CCNS,

CCRN-CSC-CMC; BETH WATHEN, RN, MSN, APRN, CCRN; Chief Executive Officer, DANA WOODS, MBA

EDITORIAL OFFICEAmerican Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. (800) 899-1712, (949) 362-2000. E-mail address: [email protected]. Web address: www.ajcconline.org Publishing Manager, MICHAEL MUSCAT; Managing Editor, KATIE L. SPILLER, MS; Art and Production Director, LeROY HINTON; Copy Editors, JANE CALAYAG, BA; BARBARA HALLIBURTON, PhD; JULIE HENDERSON, RN, MS, ELS; LAURIE ANNE WALDEN, DVM,

ELS; Graph ics Specialist, MATT EDENS; Peer-Review Coor-dinator, DENISE GOTTWALD; Publishing Assistant, SAM MARSELLA

ADVERTISING SALES OFFICESLACK Incorporated, 6900 Grove Rd, Thorofare, NJ 08086. (800) 257-8290, (856) 848-1000. National Account Manager, NICOLE RUTTER; Recruit ment Sales Representative BERNADETTE HAMILTON; Admin-is trator, JOANN CAMPISI

AMERICAN JOURNAL OF CRITICAL CARE® Telephone: (949) 362-2000. Fax: (949) 362-2049. Copyright 2018 by AACN. All rights reserved. The AMERICAN JOURNAL OF CRITICAL CARE is an official peer-reviewed publication of AACN. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mech an ical, including photo copying, recording or by any information storage retrieval system, without per mission of AACN. For all permission requests, please contact the Copyright Clearance Center, Customer Ser-vice, 222 Rosewood Dr, Danvers, MA 01923. (978) 750-8400. Prices on bulk reprints of articles available on request from AACN at (800) 899-1712. Printed on acid-free paper. The AMERICAN JOURNAL OF CRITI-CAL CARE is indexed in MEDLINE/PubMed, Psy-cINFO, Cum u lative Index to Nursing & Allied Health Literature (CINAHL), Science Citation Index Expanded, ISI Alerting Services, Current Contents/Clinical Medi-cine, EBSCO collections, Ovid collections, and Scopus.

Annual individual subscriptions: US and possessions, $59; Canada and Mexico, $120 (US); all other countries, $120 (US). Institutional rates: US, print and online, $640; print only, $465; online only, $440. Institutional rates for all countries outside the US: print and online, $780; print only, $605; online only, $440. Single copies: US, $50; all other countries, $60 (US). For subscription questions: AACN members, please call toll-free (800) 899-2226 or (949) 362-2000; nonmembers, please call toll-free (800) 336-6348 or (818) 487-2075. Change of ad dress: Send notices to AMERICAN JOURNAL OF CRITICAL CARE, Subscrip-tion Service Depart ment, 101 Columbia, Aliso Viejo, CA 92656. Allow 4 to 6 weeks for change to take effect.

The statements and opinions contained in the articles in the AMERICAN JOURNAL OF CRITICAL CARE are solely those of the individual contributors and not of the editors or the American Association of Critical-Care Nurses. The edi-tors and the American Association of Critical-Care Nurses assume that articles emanating from a particular institu-tion are submitted with the approval of the requisite authority, including all matters pertaining to human stud-ies and patient privacy requirements. Advertisements in this journal are not a warranty, endorsement, or approval of the products by the editors of this journal or the Amer-ican Association of Critical-Care Nurses, who disclaim all responsibility for any injury to persons or property result-ing from any ideas or products referred to in the articles or advertisements.

Editors in Chief

CINDY L. MUNRO, RN, PhD, ANP

Dean and Professor, School of Nursing and Health Studies, University of Miami, Coral Gables, Florida

RICHARD H. SAVEL, MD

Adjunct Professor of Clinical Medicine and Neurology, SUNY Downstate College of Medicine, New York, New York

Clinical Advisers

LINDA BELL, RN, MSN

American Association of Critical-Care Nurses Aliso Viejo, California

Founding Coeditors

CHRISTOPHER W. BRYAN-BROWN, MD, and KATHLEEN DRACUP, RN, DNSc

Editorial Board

SARAH A. DELGADO, RN, MSN, ACNP-BC

American Association of Critical-Care NursesAliso Viejo, California

MICHAEL H. ACKERMAN, RN, DNS

Rochester, New York

THOMAS AHRENS, RN, DNS, CCRN

St Louis, Missouri

JOANN GRIF ALSPACH, RN, MSN, EdD

Annapolis, Maryland

JUDY L. BEZANSON, RN, DSN

Dallas, Texas

STIJN I. BLOT, RN, PhD

Ghent, Belgium

ELIZABETH J. BRIDGES, RN, PhD, CCNS, CCRN

Seattle, Washington

TIMOTHY G. BUCHMAN, PhD, MD, MCCM

Atlanta, Georgia

LINDA L. CHLAN, RN, PhD

Rochester, Minnesota

MARIANNE CHULAY, RN, DNSc

Southern Pines, North Carolina

MARTHA A. Q. CURLEY, RN, PhD

Boston, Massachusetts

RHONDA D’AGOSTINO, ACNP-BC

New York, New York

LYNN DOERING, RN, DNSc Los Angeles, California

BARBARA DREW, RN, PhD San Francisco, California

LEWIS A. EISEN, MD

Bronx, New York

DOUG ELLIOTT, RN, PhD

Sydney, New South Wales, Australia

SUSAN K. FRAZIER, RN, PhD Lexington, Kentucky

DORRIE K. FONTAINE, RN, DNSc

Charlottesville, Virginia

MARJORIE FUNK, RN, PhD

New Haven, Connecticut

MICHAEL A. GROPPER, MD, PhD

San Francisco, California

SANDRA HANNEMAN, RN, PhD

Houston, Texas

KATHRYN HAUGH, RN, PhD

Charlottesville, Virginia

STEVEN HOLLENBERG, MD

Camden, New Jersey

CONNIE JASTREMSKI, RN, MS, CNAA

Syracuse, New York

RUTH KLEINPELL, RN, PhD

Chicago, Illinois

CONSTANTINE MANTHOUS, MD

Bridgeport, Connecticut

PETER E. MORRIS, MD

Winston Salem, North Carolina

DEBRA K. MOSER, RN, DNSc Lexington, Kentucky

JANET D. PIERCE, DSN, ARNP Kansas City, Kansas

KATHLEEN PUNTILLO, RN, PhD

San Francisco, California

MARY LOU SOLE, RN, PhD

Orlando, Florida

THEODORE A. STERN, MD

Boston, Massachusetts

M. CHRISTINE STOCK, MD

Chicago, Illinois

KATHLEEN M. VOLLMAN, RN, MSN, CCNS, CCRN

Detroit, Michigan

DOUGLAS WHITE, MD, MAS

Pittsburgh, Pennsylvania

SUSAN WOODS, RN, PhD

Seattle, Washington

Printed in the USA.

www.ajcconline.org AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2018, Volume 27, No. 3 165

AMERICAN JOURNAL OF CRITICAL CARE® (Print ISSN 1062-3264, Online ISSN 1937-710X) is published bi monthly (January, March, May, July, September, Nov ember) by the American Association of Critical-Care Nurses (AACN), 101 Columbia, Aliso Viejo, CA 92656. Periodicals postage paid at Laguna Beach, CA, and additional mailing office(s). Postmaster: Send address changes to the AMER ICAN JOURNAL OF CRITICAL CARE, Subscription Service Depart ment, 101 Columbia, Aliso Viejo, CA 92656.

Coming in July …Distinguished Research Lecturer Margaret L.

Campbell addresses the role of critical care

nurses in ensuring breathing comfort in pa-

tients at the end of life.

On the Cover

Detail from “Yeon-gyel (coupling) 1507”

Jeong Han Yun & Choon-Hyang Yun

16.5'' x 16.5''

Mixed media

2015

To view other works by

Jeong Han & Choon-Hyang Yun,

visit their website at

www.jeonghan.net

Critical Care Evaluation

Early Mobility in Critical Care

Critical Care Management

172 Effects of an Integrative Nursing Intervention on Pain in Critically Ill Patients: A Pilot Clinical TrialElizabeth D. E. Papathanassoglou, Maria Hadjibalassi,

Panagiota Miltiadous, Ekaterini Lambrinou, Evridiki Papastavrou,

Lefkios Paikousis, and Theodoros Kyprianou

186 Identifying Barriers to Nurse-Facilitated Patient Mobility in the Intensive Care UnitDaniel L. Young, Jason Seltzer, Mary Glover, Caroline Outten, Annette Lavezza,

Earl Mantheiy, Ann M. Parker, and Dale M. Needham

194 Mobilization Therapy in the Pediatric Intensive Care Unit: A Multidisciplinary Quality Improvement Initiative Blair R. L. Colwell, Cydni N. Williams, Serena P. Kelly, and Laura M. Ibsen

205 Early Blood Transfusions in Sepsis: Unchanged Survival and Increased CostsKarthik Raghunathan, Mandeep Singh, Brian H. Nathanson, Elliott Bennett-Guerrero,

and Peter K. Lindenauer

166 AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2018, Volume 27, No. 3 www.ajcconline.org

252 Abstracts of articles available exclusively online at www.ajcconline.org

e1 2018 National Teaching Institute Research Abstracts

May 2018, Volume 27, No. 3

212 Critical Care Nurses' Experiences With Spiritual Care: The SPIRIT StudyNigel Bone, Marilyn Swinton, Neala Hoad, Feli Toledo, and Deborah Cook

220 Survival of Patients With Severe Acute Respiratory Distress Syndrome Treated Without Extracorporeal Membrane OxygenationSarina K. Sahetya, Roy G. Brower, and R. Scott Stephens

228 Noninvasive Blood Pressure Monitoring and Prediction of Fluid Responsiveness to Passive Leg RaisingJoya D. Pickett, Elizabeth Bridges, Patricia A. Kritek, and JoAnne D. Whitney

238 Early Warning Score Communication Bundle: A Pilot StudyCheryl Gagne and Susan Fetzer

245 Effect of Dynamic Light Application on Cognitive Performance and Well-being of Intensive Care NursesKoen S. Simons, Enzio R. K. Boeijen, Marlies C. Mertens, Paul Rood, Cornelis

P.C. de Jager, and Mark van den Boogaard

End-of-Life Care

Pulmonary Critical Care

Cardiovascular Critical Care

Brief Report

168 Editorial Celebrating May—and Nursing!

Cindy L. Munro and

Richard H. Savel

170 Clinical Pearls Rhonda Board

171 Distinguished Research Lecture Abstract

Ensuring Breathing Comfort at

the End of Life: The Integral

Role of the Critical Care Nurse

Margaret L. Campbell

204 Patient Care Page No Time for Early Mobility?

Cindy Cain

243 Evidence-Based Review and Discussion Points

Ronald L. Hickman

249 ECG Puzzler The Value of Lead aVR:

A Frequently Neglected Lead

Salah S. Al-Zaiti, Teri M. Kozik,

Michele M. Pelter, and Mary G. Carey

252 Education Directory

Visit AJCC’s website, www.ajcconline.org, to submit a manuscript or for author guidelines, full text of selected articles, OnlineNOW articles, digital edition access, links to AACN’s online continuing education tests, and more.

An Official Publication of the American Association of Critical-Care Nurses



CELEBRATING MAY—AND NURSING! By Cindy L. Munro, RN, PhD, ANP, and Richard H. Savel, MD

Editorial

Comedian Robin Williams said, “Spring is

nature’s way of saying, ‘Let’s party!’”1 May is

the last month of spring, and it is a month

of celebration. Holidays include May Day (May 1st),

Cinco de Mayo (May 5th), Mother’s Day (the second

Sunday in May), and Memorial Day (the last Monday

in May). May is also a special time of celebration for

nursing. National Nurses Week traditionally begins

on May 6th with National Nurses Day and concludes

on May 12th with International Nurses Day, which is

also Florence Nightingale’s birthday. National Student

Nurses Day is May 8th, and School Nurses Day falls

on the Wednesday of Nurses Week. The American

Association of Critical-Care Nurses (AACN) National

Teaching Institute (NTI)—the premier gathering of

critical care nurses—is held in May of every year.

Recognition and celebration of the contributions of

nurses are appropriate all year long, but the special

emphasis in May is appreciated.

A new global initiative to recognize and celebrate

nursing was launched in February 2018. Nursing

Now is a 3-year campaign focused on acknowledg-

ing and expanding the worldwide involvement of

nursing in health.2 Key components of the campaign

are improving public perceptions of nurses, enhanc-

ing the infl uence of nurses, and maximizing nursing’s

contributions to health and access to health care.

Nursing Now is a collaborative effort with the Inter-

national Council of Nurses and the World Health

Organization. Nursing Now already has star power

on board; Her Royal Highness Catherine The Duch-

ess of Cambridge (née Kate Middleton) is the cam-

paign’s patron. The campaign will conclude in 2020,

coinciding with the 200th anniversary of Nightin-

gale’s birth.

The Nursing Now website ambitiously states,

“We work to empower nurses to take their place at

the heart of tackling 21st Century health challenges.”2

This statement affi rms that nurses are central to improv-

ing the health care system. The active involvement

of nurses in advancing health care in all settings is

crucial; nurses have both the capacity and the num-

bers to make a real difference.

Nurses’ capacity to drive change is centered in

extraordinary knowledge and patient care skills. Crit-

ical care nurses are grounded by their initial nursing

education, which enables them to care for patients

and families throughout the life span and in a vari-

ety of settings. Additional specialty education is built

upon this foundation, giving nurses who work in

critical care the tools they need to care for the highly

vulnerable patients and families entrusted to their care.

Some nurses will pursue graduate nursing education

at the master’s and doctoral level to expand their exper-

tise and to contribute to the scientifi c base of critical

care nursing. Many nurses pursue and achieve certifi -

cation in critical care, as a visible recognition of their

commitment to excellence. AACN offers certifi cation

in multiple specialties and subspecialties relevant to

critical care practice for bedside nurses, nurse manag-

ers, educators, and advanced practice nurses.

Nurses are well positioned to advance change

in the health care system because they are the largest ©2018American Association of Critical-Care Nursesdoi:https://doi.org/10.4037/ajcc2018206


group of health care clinicians worldwide, and they

are integral to every aspect of health care. The United

States has nearly 3 million registered nurses,3 and

registered nurses are the largest group of health care

providers. Registered nurses are also the largest com-

ponent of the critical care workforce. In the United

States alone, more than 500 000 critical care regis-

tered nurses work together with their colleagues

(15 000 acute care nurse practitioners, 10 000 physi-

cian intensivists, and many other members of the

critical care team) to meet the needs of critically

ill patients and their families.4

Leveraging the power of nursing’s capacity and

numbers should be an overarching goal of any cam-

paign to improve health. Particularly in critical care,

nursing exemplifies both leadership and collabora-

tive effort. Nursing Now will provide important

opportunities to advance the role of nursing world-

wide. We encourage nurses and other health profes-

sionals to follow the activities of Nursing Now as

the campaign matures and to contribute to advanc-

ing its important goals.

Every May, AACN’s NTI “offers learning, inspi-

ration and celebration for high-acuity and critical

care nurses.”5 This year’s conference in Boston will

bring an estimated 7000 critical care nurses and

their colleagues together for 4 days of celebration,

mentoring, education, and companionship. NTI

sessions offer up-to-date information about new

research and evidence-based practice that invigo-

rates patient care. Supersessions and interactions

with colleagues provide inspiration. Recognition

and celebration are central to NTI. Visionary

Leadership Awards are presented to nurses and

other critical care leaders for lifetime achievements.

Individual and AACN Chapter Circle of Excellence

Awards are celebrated. The AACN Distinguished

Research Lecture is a celebration of outstanding

critical care research. Units that have received Gold,

Silver, and Bronze Beacon Awards along a journey

for excellence are recognized during NTI. Attendees

leave NTI energized and empowered to improve

their units and their care of patients and families.

Of course, nurses are not the only critical care

professionals who deserve recognition and celebra-

tion! Meaningful recognition is one of 6 essential

standards underpinning the AACN Healthy Work

Environment Initiative. That standard is based on

the idea that, “Nurses must be recognized and must

recognize others for the value each brings to the work

of the organization.”6(p29) A sampling of upcoming

events relevant to recognizing the contributions of

critical care team members includes

National Physician Assistants Day: October 6,

2018

National Physical Therapy Month: October 2018

National Radiologic Technology Week: Novem-

ber 4-10, 2018

National Pharmacists Day: January 12, 2019

National Women Physicians Day: February 3,

2019

Certified Nurses Day: March 19, 2019

National Doctors Day: March 30, 2019

Recognition is not a zero-sum game, where a

win for one party can come only at the expense

of a loss for others. Rather, recognition should be a

“win-win.” The celebrations of nursing that occur

in May are a spring party that we can all enjoy!

The statements and opinions contained in this editorial are solely those of the coeditors in chief.

FINANCIAL DISCLOSURESNone reported.

REFERENCES1. Brainy Quotes. Robin Williams. https://www.brainyquote

.com/quotes/robin_williams_107638. Accessed March 3, 2018.2. Nursing Now. http://www.nursingnow.org/. Accessed

March 3, 2018.3. Bureau of Labor Statistics, US Department of Labor. Occu-

pational Outlook Handbook, Registered Nurses. https://www.bls.gov/ooh/healthcare/registered-nurses.htm. Accessed March 4, 2018.

4. Society of Critical Care Medicine. Critical Care Statistics, Staffing/Salary. http://www.sccm.org/Communications/Pages /CriticalCareStats.aspx. Accessed March 4, 2018.

5. American Association of Critical-Care Nurses. NTI 2018. https://www.aacn.org/conferences-and-events/nti. Accessed March 8, 2018.

6. American Association of Critical-Care Nurses. 2016. AACN Standards for Establishing and Sustaining Health Work Envi-ronments: A Journey to Excellence. 2nd ed. Aliso Viejo, CA: AACN; 2016.

To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, [email protected].

Nurses must be recognized and must recognize others for the value each brings to the work of the organization.

About the AuthorsCindy L. Munro is coeditor in chief of the American Journal of Critical Care. She is dean and professor, School of Nursing and Health Studies, University of Miami, Coral Gables, Florida. Richard H. Savel is coeditor in chief of the American Journal of Critical Care. He is director, Adult Critical Care Services, Maimonides Medical Center and adjunct professor of clinical medicine and neurology, SUNY Down state College of Medicine, both in New York City.

Clinical Pearls Rhonda Board, RN, PhD, CCRN, Section Editor

Clinical Pearls is designed to help implement evidence-based care at the bedside by summarizing some of the most clinically useful material from select articles in each issue. Readers are encouraged to photocopy this ready-to-post page and share it with colleagues. Please be advised, however, that any substantive change in patient care protocols should be carefully reviewed and approved by the policy-setting authorities at your institution.

Nurse-Facilitated Patient Mobility

Early movement can improve the muscle weakness commonly

experienced by patients in the intensive care unit and prevent potential long-term impairments. Nurse-facilitated mobility improves patient out-comes and decreases length of hospital stay. However, a commonly named barrier to nurse-facilitated mobilization is nurses’ lack of time. To understand time-related barriers, Young and colleagues designed a multidisciplinary team process to directly observe the work performed by nurses and clinical care technicians. They found the following: • Four categories of nurse work: patient care (47%), provider com-munication (25%), documentation (18%), and down time (10%). • The best times for potential mobility events occurred during direct patient care or down time. • Nursing team members noted that when mobility could be possible, it was not necessarily thought about. Direct observation of nursing care activities is a process that could be replicated by other institutions to provide insight in identifying missed opportunities for nurse-facilitated patient mobility.

See Article, pp 186-193

Patient Early Warning Scores

About half of adults admitted to intensive care units (ICUs) are patients whose condition has deteriorated while on a medical-surgical

unit. Patients’ early warning scores (EWSs) are based on physiological measures and were developed as a decision tool to help bedside nurses identify and take action when a patient decompensates. Fetzer and colleagues created an EWS bundle of interventions, including alerting an experienced ICU nurse, to improve communication and patient outcomes related to patient deterioration. After testing the bundle, they found the following: • The number of medical-surgical transfers to the ICU decreased. • The percentage of patients admitted to the ICU after a rapid response team (RRT) call decreased. • RRT calls increased in general but decreased for patients with an EWS greater than 4 (indicat-ing clinical deterioration), suggesting earlier identification and intervention occurred with deteriorating patients. Findings suggest that use of an electronically embedded EWS and a communication bundle with experienced ICU nurse collaboration can improve patient care and preserve health care costs.


©2018 American Association of Critical-Care Nurses, doi:https://doi.org/10.4037/ajcc2018970

Nurses’ Experiences With Spiritual Care

Patients and families in the critical care setting often experience spiritual distress. Although most nurses do not receive education in how to provide spiritual

care to patients, most recognize it as part of holistic nurs-ing practice. Bone and colleagues interviewed nurses in an intensive care unit (ICU) to understand their experiences when making a referral for spiritual care for a dying patient and/or the patient’s family. They found the following 3 categories related to spiritual presence: 1. The value and role of chaplains: Chaplains were con-sidered an essential part of the ICU team and provided sup-port to both families and nurses. 2. Nurses’ experiences with chaplains: Nurses appreciated sharing care with chaplains and made referrals to them throughout a patient’s stay. 3. How ICU nurses provide spiritual care: Although nurses stated that compassion came naturally to them, providing spir-itual care was not always intentional or recognized as such. Nurses considered spiritual care important and valued chaplain support in the holistic care they provide to patients and their families.



Putting Evidence-Based

Care in Your Hands

Effects of an Integrative Nursing Intervention on Pain in ICU Patients

Pain is a common symptom for many patients in the inten-sive care unit (ICU). Unrelieved pain can contribute to physical and psychological complications such as hemo-

dynamic instability, infections, anxiety, delirium, and post-ICU syndrome. The complex nature of pain management requires both pharmacological and nonpharmacological interventions. Papathanassoglou and colleagues examined the effects of a multimodal integrative intervention that included relax-ation and guided imagery, moderate pressure massage, and music listening. They found that patients receiving the inter-vention had • Decreased incidence of pain • Lower systolic blood pressure • Reduced fear • Decreased anxiety levels • Improved quality of sleep Although many of the patients had low acuity of illness, the authors suggest use of a multimodal daily intervention to reduce pain and improve pain-related outcomes in critically ill adults.


Distinguished Research Lecture AbstractPresented May 21, 2018, at the AACN National Teaching Institute in Boston, Massachusetts

©2018 American Association of Critical-Care Nursesdoi:https://doi.org/10.4037/ajcc2018487

About the AuthorMargaret L. Campbell is a professor in the College of Nursing at Wayne State University, Detroit, Michigan.

Dyspnea is one of the worst symptoms experi-

enced by patients in the intensive care unit

and patients approaching the end of life.

The patients in the intensive care unit who are at the

highest risk include those with underlying cardiopul-

monary conditions and those with respiratory failure.

Critical care nurses are integral to assessing and treat-

ing dyspnea during the trajectory of critical care illness,

especially when a patient is not expected to survive

and care goals are shifted to focus on comfort. Para-

doxically, cognitive impairment develops along with

worsening dyspnea in dying patients, preventing

patients from reporting their distress while they

may still be able to experience it. Inability to report

distressing symptoms can lead to undertreatment or

overtreatment. The Respiratory Distress Observation

Scale (RDOS), developed by the author, is the only

known valid, reliable tool for assessing respiratory

distress when the patient cannot self- report dyspnea,

as typifies many critically ill patients. An evidence-

based approach to dyspnea assessment by patient

report and RDOS and treatment is addressed in this

lecture. Interventions are categorized into those that

are effective, interventions with limited effectiveness,

and interventions whose effectiveness has not been

established. In addition, a nurse-led, respiratory

therapist–supported ventilator withdrawal algorithm

guided by the RDOS is introduced.

Margaret L. Campbell’s presentation will be published in its

entirety in the July 2018 issue of AJCC.

ENSURING BREATHING COMFORT AT THE END OF LIFE: THE INTEGRAL ROLE OF THE CRITICAL CARE NURSEBy Margaret L. Campbell, RN, PhD


Critical Care Evaluation


Background Pain, a persistent problem in critically ill patients, adversely affects outcomes. Despite recom-mendations, no evidence-based nonpharmacological approaches for pain treatment in critically ill patients have been developed.Objectives To investigate the effects of a multimodal integrative intervention on the incidence of pain and on secondary outcomes: intensity of pain, hemodynamic indices (systolic and mean arterial pressure, heart rate), anxiety, fear, relaxation, optimism, and sleep quality.Methods A randomized, controlled, double-blinded repeated-measures trial with predetermined eligibility criteria was conducted. The intervention included relax-ation, guided imagery, moderate pressure massage, and listening to music. The primary outcome was incidence of pain (score on Critical Care Pain Observation Tool > 2). Other outcomes included pain ratings, hemodynamic measurements, self-reported psychological outcomes, and quality of sleep. Repeated-measures models with adjustments (baseline levels, confounders) were used.Results Among the 60 randomized critically ill adults in the sample, the intervention group experienced signifi-cant decreases in the incidence (P = .003) and ratings of pain (P < .001). Adjusted models revealed a significant trend for lower incidence (P = .002) and ratings (P < .001) of pain, systolic arterial pressure (P < .001), anxiety (P = .01), and improved quality of sleep (P = .02).Conclusion A multimodal integrative intervention may be effective in decreasing pain and improving pain- related outcomes in critically ill patients. (American Journal of Critical Care. 2018; 27:172-185)

EFFECTS OF AN INTEGRATIVE NURSING INTERVENTION ON PAIN IN CRITICALLY ILL PATIENTS: A PILOT CLINICAL TRIAL By Elizabeth D. E. Papathanassoglou, RN, MSc, PhD, Maria Hadjibalassi, RN, MSc,

PhD, Panagiota Miltiadous, PhD, Ekaterini Lambrinou, RN, MSc, PhD, Evridiki Papastavrou, RN, MSc, PhD, Lefkios Paikousis, and Theodoros Kyprianou, PhD, MD

1.0 HourC EThis article has been designated for CE contact

hour(s). See more CE information at the end of

this article.


Intensive care unit (ICU) patients experience pain at rest and during procedures.1 Unre-lieved pain is common among ICU patients and may compromise outcomes, by contrib-uting to unstable hemodynamic parameters, hypercatabolism, hyperglycemia, infections, delirium, and posttraumatic stress.2 Pain in ICU patients may also be part of a vicious circle implicating anxiety3 and insomnia.4 Moreover, pain has been linked to the post-ICU

syndrome5 and may become chronic in survivors, a situation associated with poor quality of life and poor psychological and physiological outcomes.6

Management of pain in patients with a critical

illness is challenging.7 Opioids, the drug class of

choice, are associated with marked side effects, includ-

ing respiratory depression, hypotension, decreased

gastrointestinal motility, delirium, and higher costs

due to increased use of resources and prolonged stay

in the ICU.8 Nociception, the perception of painful

stimuli, entails a complex interaction among sensory,

affective, and social components.9 Moreover, anxiety,

fear, and negative expectations are common in criti-

cally ill patients and may contribute to heightened

perception of pain.10 Thus, the multifactorial nature

of ICU pain calls for approaches that address both

physiological and psychosocial responses to pain.

Current guidelines highlight the need to test and

implement nonpharmacological strategies for pain

treatment in critically ill patients.2 Interventions

that elicit a relaxation response, via parasympathetic

activation, appear to influence a some patients’

outcomes.11 For example, the authors of a recent

review12 concluded that guided imagery can decrease

pain and anxiety in critically ill patients. However,

studies on nonpharmacological interventions for

ICU pain and related outcomes are scarce. The

physiological pathway involved in decreasing pain

via relaxation-inducing interventions is unclear.13

Multiple synergistic mechanisms may be involved,

including distraction of attention, diminished

transmission of nociceptive signals due to descending

impulses from the brain during processing of relax-

ation cues, downregulation of the affective nociceptive

pathway, and perception of social connectedness.14-18

Objectives Our aim was to investigate measures of the effect

of a multimodal integrative intervention on the inci-

dence of pain (primary outcome) and several second-

ary outcomes: intensity of pain; hemodynamic indices

(systolic arterial pressure [SAP], mean arterial pressure

[MAP]), and heart rate); psychological outcomes (anxi-

ety, fear, feeling of relaxation, optimism); quality of

sleep; patient outcomes (complications, mortality);

and daily analgesic doses. Measures of effect were tested

before and after the intervention and longitudinally.

Literature Review and Definitions Relaxation, guided imagery, and music therapy

are categorized as mind-body interventions, whereas

touch and massage are con-

sidered body-based prac-

tices.19 Relaxation promotes a

sense of calmness often asso-

ciated with parasympathetic

activation.20 Guided imagery

involves focusing one’s imag-

ination on pleasurable cir-

cumstances in a way that

elicits emotion.19 Interper-

sonal touch or massage is a powerful means of mod-

ulating emotion, triggering neuroendocrine and

immune effects, vagal stimulation, and a reduction

in stress, pain, and depression.21

Recent evidence22 suggests that nonpharmaco-

logical interventions for pain, such as hand massage,

may be feasible and acceptable in critical care settings.

Moreover, both families of ICU patients and nurses

seem to regard nonpharmacological interventions

as relevant and feasible approaches for relief of ICU

pain.23 However, despite reports of increased effec-

tiveness of multimodal integrative interventions,24

few studies have addressed the impact of combining

About the AuthorsElizabeth D. E. Papathanassoglou is an associate profes-sor, Faculty of Nursing, University of Alberta, Edmonton, Alberta, Canada. Maria Hadjibalassi is an assistant pro-fessor Panagiota Miltiadous is special teaching staff, and Ekaterini Lambrinou and Evridiki Papastavrou are associate professors, Department of Nursing, Cyprus University of Technology, Limassol, Cyprus. Lefkios Paikousis is an analyst, Improvast Analytical Services Company, Nicosia, Cyprus. Theodoros Kyprianou is an associate professor, St Georges University of London Medical Program, Uni-versity of Nicosia Medical School, Nicosia, Cyprus.

Corresponding author: Elizabeth D. E. Papathanassoglou, RN, MSc, PhD, Faculty of Nursing, University of Alberta, Edmonton, Alberta, Canada, 5-262 Edmonton Clinic Health Academy, 11405-87th Ave, Edmonton, Alberta, Canada T6G 1C9 (email:[email protected]).


Perception of pain entails a complex interaction among sensory, affective, and social components.


integrative approaches in critically ill patients. In

most instances, relaxation was combined with guided

imagery25-27; in 1 study,28 relaxation and guided imag-

ery were combined with gentle massage. Anxiety, pain,

and sleep were the most common outcomes addressed,

and the results varied. However, little attention was

given to confounders and the effect and interactions

with time in previous studies.12

Methods The study has been registered at ClinicalTrials.gov

(Identifier: NCT02423252).

Ethical approval was obtained from the Cyprus

National Bioethics Committee, Republic of Cyprus.

Written informed consent was obtained from all

patients or their surrogates before recruitment. Each

participant’s assent was acquired when the partici-

pant regained capacity.

DesignWe conducted a randomized, controlled, double-

blinded (clinicians, outcome assessors) repeated-

measures pilot trial with 2 parallel groups (intervention

and standard care groups; Figure 1). The sample con-

sisted of patients admitted to a 17-bed academic

teaching general systems ICU in Cyprus. Patients

were eligible for the study if they were more than 18

years old, understood Greek, had a score of -2 to +2

on the Richmond Agitation-Sedation Scale, had a score

greater than 9 on the Glasgow Coma Scale (GCS) at

Figure 1 Schematic of study design.

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; CPOT, Critical Care Pain Observation Tool; ICU, intensive care unit; MODS, multiple organ dysfunction syndrome; NRS, Numeric Rating Scale; RASS, Richmond Agitation-Sedation Scale; SOFA, Sequen-tial Organ Failure Assessment.

Before enrollment

Visit 1: time point 1 (day 1)

Visit 2-5: time point 2-5 (day 2-5) or until transfer

At ICU discharge

60 participants: Obtain informed consent, screen potential participants by inclusion and exclusion criteria, obtain history, document

Arm 1 (n = 30), study groupStandard care + intervention

Perform once-daily measurementsSleep NRS, APACHE II, SOFA, and MODS scores, daily dose of opioid and nonopioid analgesic agents

Perform preintervention measurementsVital signs; sleep NRS, APACHE II, SOFA, MODS, CPOT, pain NRS, anxiety NRS, relaxation NRS,

fear level NRS, and optimism NRS scoresRepeat study intervention

Perform postintervention assessmentsVital signs; CPOT, pain NRS, anxiety NRS, relaxation NRS, fear level NRS, and optimism NRS scores

Perform baseline assessmentsAge, sex, admission diagnosis, history of alcohol use, depression, baseline clinical data, intravenous sedation, anal-

gesia, vasoactive medication dose, and baseline RASS, CPOT (or pain NRS), APACHE II, MODS, and SOFA scoresAdminister study intervention

Randomize

Final follow-upsRate of complications: hospital-acquired infections, thromboembolism,

stress-related gastrointestinal bleeding, deliriumICU length of stay

Survival

Arm 2 (n = 30), control groupStandard care only


the time of inclusion, and had an arterial catheter

in place.

Patients were excluded if they had an expected

ICU length of stay less than 48 hours, had a current

history of severe mental health problems or demen-

tia, had a hearing impairment or conditions that

did not permit the use of headphones, were receiv-

ing neuromuscular blockers, were confused at the

time of screening (according to the assessment of

an expert research nurse; no formal tool for confu-

sion was used in the unit at the time), or required,

at the time of screening, special contact or isolation

precautions for any reason.

The size of the sample was predefined as 30

patients per arm. This pilot study was not powered to

determine a difference in a primary outcome, because

our aim was to assess estimates of effect and to

attain high probability of equivalence at baseline.29

Consecutive patients were screened daily (March

2013-March 2015) and were recruited for the study

by a research nurse (Figure 2). For patients who were

uncommunicative or incapable for any reason of

providing informed consent, the patients’ families

were approached to obtain written informed consent.

Participants were randomized (http://www

.randomization.com/) to an intervention (n = 30)

or a control (n = 30) group. Patients in the interven-

tion group received, in addition to standard care,

the daily 55-min intervention. Randomization blocks

of 4 allocations were based on participants’ age (≤ 45

years, > 45 years), sex (men, women), and systemic

inflammatory response syndrome (SIRS) status (SIRS,

no SIRS). Concealment during the intervention was

maintained by drawing the curtains around a partic-

ipant’s bed and by the presence of an intervention

nurse at the bedside for the set amount of time in

both the intervention and control groups. Before

the intervention, the intervention nurse negotiated a

time at which the participant (in either group: inter-

vention or control) could remain uninterrupted for

55 minutes with the bedside nurse and the partici-

pant’s family. However, if a clinical need arose, clini-

cians were free to enter the room. Clinicians and

all study personnel had no knowledge of the group

allocations. Outcomes were assessed by persons not

involved in patient care or in other aspects of the

trial and with no knowledge of the study. Allocation

was disclosed to the intervention nurse only. Partici-

pants were not blinded to the allocation.

InterventionThe intervention was delivered once daily

(between 9:30 AM and 11:30 AM) by a trained

intervention nurse not involved in patient care, for

up to 5 days during the ICU stay, starting the day after

ICU admission. The multimodal intervention, with a

duration of approximately 55 minutes, was based on

a literature review; recommendations of the American

Holistic Nurses Association30; recommendations of a

group of experts, including academics and clinicians

(n = 5); results of a small feasibility pilot test (n = 10);

and feedback of patients and patients’ families. The

selection of music (Mozart piano sonata KV 283)

was based on previous evidence of physiological

effects in the critically ill.31 The intervention included

relaxation and guided imagery (40 minutes) and

moderate-pressure massage (15 minutes). The seg-

ment of relaxation and guided imagery included

guided relaxation, a use of a structured guided

imagery script, and listening to music through head-

phones for 15 minutes. Moderate-pressure, low-

velocity (4 N, 1-5 cm/s) massage consisted of

squeezing movements with a wide area of contact

over the head, neck, trapezius muscles, and fore-

arms. Moderate-pressure massage was used because

Figure 2 Flowchart of enrollment of participants in the study.

Excluded (n = 552) Did not meet inclusion criteria (n = 337) Met exclusion criteria (n = 215)

Refused consent (n = 13)No timely access to family for consent (n = 14)

Patients screened for eligibility (n = 639)

Randomized (n = 60)

Allocated to standard-care group (n = 30)

Analyzed (n = 30)

Allocated to intervention group (n = 30)

Analyzed (n = 30)

Eligible for participation (n = 87)


After the intervention, participants in the intervention group

had 44% less chance of having indications

of pain than did the control participants.

it elicits parasympathetic activation, which light-

pressure massage does not.32 The efficacy of multi-

modal integrative interventions is superior to that of

1-dimensional ones.24

OutcomesThe primary outcome was incidence of pain as

indicated by a score greater than 2 on the Critical

Care Pain Observation Tool (CPOT). Secondary

outcomes were the CPOT score, self-reported pain

intensity according to a numeric rating scale (NRS),

observer-reported pain (NRS), systolic arterial pres-

sure (SAP), heart rate, respiratory rate, anxiety,

fear, relaxation, optimism, quality of sleep (self-

reported NRS scales), length of ICU stay (hours),

ICU survival, score on the

Sequential Organ Failure

Assessment (SOFA), multiple-

organ dysfunction syndrome

(MODS) score, and daily doses

of opioid (morphine equiva-

lents)33 and nonopioid (mg/

kg per 24 hour) analgesics.34

The aim of the interven-

tion was to elicit a relaxation

response, which, despite lack

of any pertinent evidence,

could in theory cause

increased parasympathetic tone and a decrease in

mean arterial pressure (MAP). Thus, MAP measure-

ments were used as a safety outcome for

intervention-induced impairment of tissue perfusion

(lower cutoff: 65 mm Hg).35 Physiological and

behavioral alterations and incidence of complica-

tions were reported (infections, thromboembolism,

organ dysfunction, delirium). Adverse events, irre-

spective of causal relationship, were noted for all

participants.

Data Collected. All clinical assessment scales

we used are routinely used in clinical practice with

established psychometrics, with the exception of

the NRS scale for the assessment of psychological

responses, which had been tested during the pre-

liminary pilot phase during which we pilot tested

use of the instruments. Measurements were col-

lected by trained data collectors with no knowledge

of the study. Interrater reliability was established

during the pilot phase ( > 0.80).

Three pain assessment scales were used: CPOT,

the 0 to 10 NRS, and the 0 to 10 observer-rated NRS.

The Greek CPOT has reliability and validity similar

to those of the original version.36,37 Although a CPOT

score greater than 2 indicates presence of pain, the

CPOT value has also been used as a pain score.38 We

used CPOT to assess both presence and intensity of

pain, because of the lack of tools for uncommunicative

patients.39 Communicative patients also indicated their

pain on the NRS. Following the recommendation of

the panel of experts, we used an observer NRS as an

indicator of nurses’ assessment of patients’ pain that

can be a basis for clinical judgments. Observers’ NRS

ratings correlate highly with patients’ NRS values, but

observers tend to underestimate pain when a patient’s

NRS score is greater than 4.40 We also collected val-

ues for clinical variables to use as control variables in

the analysis: age, sex, vital signs, SOFA score, MODS

score, presence of SIRS, score on the Acute Physiology

and Chronic Health Evaluation (APACHE) II, use of

analgesics, and use of vasoactive medications.

Data AnalysisWe checked variables for normality and used

transformations as needed. We did baseline and

cross-sectional comparisons between the interven-

tion and control groups by using an independent t

test or the Fisher exact test. We analyzed the pri-

mary outcome by using a logistic regression model

and the binary logistic link function41 based on

generalized estimating equations with autoregres-

sive first-order correlation structure. We used the

quasi-likelihood under independence model crite-

rion to compare model fit across covariance struc-

tures. To assess the effect of the intervention, we

used the adjusted estimated marginal means of

the proportions at each time measurement. These

proportions were adjusted for within participant

“time,” taking into account the correlation of pain

incidents between time points. The odds ratio (OR)

was calculated at each time point.

For effects on continuous variables, we used a

linear mixed models approach with autoregressive

or unstructured covariance structure for parsimony.

The best model fit was selected on the basis of the

Akaike Information Criterion. We calculated effect

sizes by using estimated marginal means and calcu-

lating the Cohen d. For effects at discrete time points,

analysis of covariance (ANCOVA) was performed to

control for pretreatment measurements and a num-

ber of confounders. The Cohen f was calculated as a

measure of effect size.42

Not all participants were able to self-report

NRS scores, especially after day 2, when most

communicative participants would be discharged.

The effect of the intervention on self-reported NRS

scores was assessed by using a linear mixed model

for the first 2 days only and ANCOVA for the first


Within the first 2 days of interven-tion, anxiety levels in the intervention group decreased, whereas those in the control group remained stable.

day of intervention. Linear mixed models for self-

reported psychological responses were fit for the

first 2 days under the unstructured covariance matrix,

with 1 covariate at a time, because adding more

covariates resulted in nonconvergence. ORs and

95% CIs were calculated for rates of complications.

We used SPSS, version 21, software (IBM SPSS) for

analysis; with an of .05.

Results Participants and Baseline Characteristics

In the final sample of 60 participants, the 2

groups had no statistically significant differences

(Table 1). No losses in the number of participants

occurred after randomization, and no participants

skipped a session. Only a small proportion of partici-

pants completed 5 days of intervention (6 in the

intervention group and 2 in the control group)

because of earlier discharge from the unit. No

adverse events were reported.

Primary Outcome: Presence of PainIncidence of pain (CPOT > 2) had a downward

trend in both groups, with a clear trend for postin-

tervention decreases in the intervention group (Table

2A). Analysis with generalized estimating equations

revealed that the trend for decreased pain incidence

was significantly greater in the intervention group

than in the control group: Wald 21 = 18.0; P = .003;

(Table 2B). After adjustment (sex and age), the

effect of the intervention remained significant

(P = .002).

At the first preintervention measurement, 43%

of participants in the control group and 57% in the

intervention group had indications of pain (CPOT > 2;

OR = 1.76). After the intervention, participants in

the intervention group had 44% less chance of

having indications of pain than did participants

in the control group (OR = 0.42; Table 3C).

Secondary OutcomesCPOT Scores. Mean CPOT scores on day 1 before

the intervention were equivalent (P = .52) in the con-

trol (2.7; SD, 1.2) and the intervention (2.5; SD, 1.1)

groups. On day 1 after the intervention, mean CPOT

scores were 1.44 (SD, 1.26) in the intervention

group and 2.5 (SD, 1.29) in the control group, a

finding that suggests a large effect size (P = .004;

Cohen d = 0.83; Table 3A). Over time, mean CPOT

scores in both groups showed a downward trend,

with a consistent trend for decreased CPOT values

after the intervention in the intervention group.

Adjusted linear mixed models (age, sex) indicated

a significant interaction effect of the intervention

group (P < .001). When only postintervention mea-

surements were used, the intervention effect over

time remained significant, with a large effect size

(P < .001; Cohen d = 0.77-1.10; Table 3A).

An ANCOVA for CPOT scores 1 day after the

intervention (adjustments: preintervention scores

for pain, SOFA, MODS, SIRS, and APACHE II; doses

of analgesics), showed a significant difference in

postintervention pain in the intervention group

(P < .001). The study group explained 30% of the

variation of postintervention CPOT scores (partial 2 = 0.3; Table 2C).

Self-reported Pain NRS. Preintervention self-

reported pain did not differ between the 2 groups

(P = .30, Table 1). On day 1 after the intervention,

pain NRS scores were lower in the intervention

group than in the control group, with a large effect

size (P < .001; Cohen d = 1.21; Table 3B).

The effect of the intervention over time was sig-

nificant, with a large effect size (adjustments for age,

sex; linear mixed model; P < .01; Tables 3B and 4B).

An ANCOVA of self-reported pain

NRS scores 1 day after the interven-

tion (adjustments for preinterven-

tion scores of self-reported pain

NRS, SOFA, SIRS, and APACHE II

and doses of analgesics), indicated

a significant decrease in pain in the

intervention group (P < .001; partial 2 = 0.353).

Hemodynamic Measurements. On

day 1 after the intervention, partici-

pants in the intervention group had

a decrease in SAP, with a moderate

to large effect size (P = .02; Cohen

d = 0.63; Table 3G). The interven-

tion effect over time was significant, even after

adjustment for confounders, with a moderate

effect size (P = .008; Cohen d = 0.63-0.89; Table 4).

An ANCOVA for mean SAP on day 1 after the

intervention (adjustments: preintervention scores

for SAP, SOFA, SIRS, APACHE II; doses of vasoactive

medication) indicated a significant decrease in SAP

in the intervention group (P < .04).

Before the intervention, MAP levels did not

differ significantly (P = .81) between the 2 groups or

in linear mixed model and ANCOVA analyses. No

statistically significant differences in heart rate and

respiratory rate were evident (Table 1). Both adjusted

and unadjusted linear models and ANCOVA analyses

indicated no significant differences (P > .06; Cohen

d = 0.21-0.52; relative risk = 0.59-0.71).


Variable

Variable

Variable

Variable

Table 1Baseline clinical data of the intervention and control groups

Demographic and clinical data Female sex, No. (%) Age, mean (SD), y Body mass index,c mean (SD) SIRS, No. (%) Mechanical ventilation (intubated), No. (%)

Clinical assessment score, mean (SD) RASS GCS APACHE II SOFA MODS

Vital signs, mean (SD) SAP DAP MAP Heart rate Respiratory rate Body temperature

Pain score, mean (SD) CPOT NRS self-reported NRS observer

Subjective baseline assessment score, mean (SD) Anxiety Fear Relaxation Optimism Sleep quality

Diagnostic categories, No. (%) Neurosurgical Heart surgery Burns Medical Coronary Surgical Pulmonary

Comorbid conditions, No. (%) CAD Hyperlipidemia Hypertension Diabetes COPD/respiratory failure Renal failure

Vasoactive medication, day 1, No. (%) Any type Dobutamine Dopamine Norepinephrine Nitroglycerin Esmolol

11 (37) 63.9 (12.7) 30.7 (7.1) 7 (23) 4 (13)

-0.1 (0.5) 14.1 (3.0) 17.0 (5.9) 4.2 (1.2) 3.4 (1.8)

125.7 (18.6) 62.4 (15.1) 82.1 (14.0) 84.8 (15.5) 21.1 (5.9) 36.7 (0.6)

2.7 (1.2) 3.7 (2.1) 4.9 (1.5)

4.0 (1.7) 3.2 (2.2) 5.5 (1.6) 5.8 (1.5) 4.2 (2.1)

2 (7) 25 (83) 0 (0) 0 (0) 0 (0) 1 (3) 2 (7)

12 (40) 3 (10) 22 (73) 11 (37) 3 (10) 1 (3)

15 (50) 4 (13) 12 (40) 1 (3) 3 (10) 0 (0)

9 (30) 62.4 (12.9) 28.7 (5.7) 12 (40) 3 (10)

-0.5 (0.5) 13.4 (2.6) 15.1 (6.2) 4.7 (2.0) 3.3 (1.6)

128.9 (22.2) 60.1 (10.5) 81.3 (12.1) 90.8 (15.6) 23.0 (4.4) 36.6 (0.6)

2.5 (1.1) 4.4 (2.4) 5.0 (1.6)

4.7 (2.4) 4.4 (3.1) 5.0 (1.4) 6.1 (1.9) 4.4 (2.4)

5 (17) 15 (50) 1 (3) 2 (7) 2 (7) 2 (7) 3 (10)

13 (43) 0 (0) 19 (63) 11 (37) 3 (10) 1 (3)

8 (27) 1 (3) 6 (20) 0 (0) 2 (7) 1 (3)

.39a

.66b

.23b

.27a

>.99a

.11 .35 .23 .25 .87

.55 .48 .81 .14 .16 .82

.52 .30 .74

.23 .23 .18 .69 .79

.16

>.99 .24 .58 .79>.99>.99

.06 .35 .16>.99>.99>.99

-1.667-0.942-1.207 1.169-0.160

0.600-0.706-0.241 1.488 1.440-0.234

-0.640 1.042 0.338

1.212 1.228-1.344 0.403 0.274

7 (12)40 (67) 1 (2) 2 (3) 2 (3) 3 (5) 5 (8)

25 (42) 3 (5)41 (68)20 (33) 6 (10) 2 (3)

23 (38) 5 (8)18 (30) 1 (2) 5 (8) 1 (2)

Intervention (n = 30)

Intervention (n = 30)

Control (n = 30)

Control (n = 30)

TotalControl (n = 30)Intervention (n = 30)

TotalControl (n = 30)Intervention (n = 30)

P

Pb

Pa

Pd

t

Continued

Variable

Table 1Continued

Analgesics, day 1, No. (%) Any type Paracetamol Morphine ASA Fentanyl Pethidine

26 (87)18 (60) 6 (20)13 (43) 5 (17) 0 (0)

20 (67)17 (57) 4 (13) 5 (17) 5 (17) 2 (7)

.12>.99 .73 .05>.99 .49

46 (77)35 (58)10 (17)18 (50)10 (17) 2 (3)

Intervention (n = 30) Control (n = 30) Total Pd

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; ASA, acetylsalicylic acid; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; CPOT, Critical Care Pain Observation Tool; DAP, diastolic arterial blood pressure; GCS, Glasgow Coma Scale; ICU, intensive care unit; MAP, mean arterial blood pressure; MODS, multiple organ dysfunction syndrome; NRS, Numeric Rating Scale; RASS, Richmond Agitation-Sedation Scale; RGI, relaxation and guided imagery; SAP, systolic arterial blood pressure; SIRS, systemic inflammatory response syndrome; SOFA: Sequential Organ Failure Assessment.

a From 2 test. b From t test. c Calculated as weight in kilograms divided by height in meters squared.d Fisher exact test.

Day

Source

Source

Table 2Effects of the intervention on the presence of pain (CPOT score > 2) and CPOT pain ratings

First Before intervention After interventionSecond Before intervention After interventionThird Before intervention After intervention

Corrected modelInterceptInterventionSexPain at baselineAgeAnalgesicsAPACHE II scoreSIRS scoreSOFA scoreMODS score

Intercept Group Time Group x Time, QIC = 233. AR(1) covariance structure.

57 (9)24 (8)

38 (14) 9 (9)

23 (16)12 (12)

5.084 0.10611.798 0.81119.360 1.153 0.249 0.118<0.001 1.650 1.551

15.6 0.024.718.0

43 (9)43 (9)

18 (7)13 (7)

23 (13)18 (12)

0.6240.0040.3000.0290.4130.0400.0090.0040.0000.0570.053

6.834 0.14215.861 1.09026.026 1.550 0.335 0.159 0.000 2.218 2.085

91111111111

1155

1.76 0.42

2.790.66

1.000.62

<.001.71

<.001.30

<.001.22.57.69

>.99.14.16

<.001.94

<.001 .003

Intervention

Mean square F df

Wald 2

A. Adjusted prevalence of paina

Mean percentage (SE)

C. ANCOVA (dependent variable: first day postintervention CPOT scores)c

B. GEE for repeated measurements (dependent variable: CPOT score > 2)b

Control

Partial 2

df

Odds ratio (intervention vs control)

P

P

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; AR (1), autoregressive first order; CPOT, Critical Care Pain Observation Tool; MODS, multiple organ dysfunction syndrome; QIC, quasi-information criterion; SIRS, systemic inflammatory response syndrome; SOFA: Sequential Organ Failure Assessment. a Adjusted (for the within-subject correlation across time) prevalence of pain (CPOT > 2) during the first 3 days of observation and related odds ratios for pain.b Generalized estimating equations (GEE) accounting for repeated measurements.c Analysis of covariance (ANCOVA) for first day postintervention CPOT scores.

Continued

Content

Table 3Point estimates of repeated measurements and differences in pain, psychological outcomes, and sleep quality from before to after the intervention for the 5 days of the interventiona

A. CPOTIntervention N Mean SDControl N Mean SDAdjusted Cohen d

B. Self-reported NRS pain ratingsIntervention N Mean SDControl N Mean SDAdjusted Cohen d

C. AnxietyIntervention N Mean SDControl N Mean SDAdjusted Cohen d

D. FearIntervention N Mean SDControl N Mean SDAdjusted Cohen d

E. Relaxation Intervention N Mean SDControl N Mean SDAdjusted Cohen d

F. OptimismIntervention N Mean SDControl N Mean SDAdjusted Cohen d

29 2.50 1.14

30 2.70 1.24 0.16

24 3.71 2.14

22 4.41 2.42 0.01

264.01.7

22 4.7 2.4 0.42

183.22.2

114.43.1

0.43

265.51.6

225.01.4

0.35

11 6.090.50

18 5.91 0.39 0.11

25 1.44 1.26

28 2.50 1.29 0.88

20 2.35 1.63

20 4.55 1.99 1.37

22 2.9 1.7

16 4.3 2.5 1.05

131.91.9

83.93.41.14

22 6.6 2.3

16 4.8 1.6 1.21

95.740.52

146.920.400.79

12 2.00 1.13

12 1.67 1.61 0.03

82.252.55

63.001.790.82

82.62.3

74.02.60.78

72.02.2

53.02.80.74

86.41.8

75.41.30.66

46.530.68

76.940.520.29

11 1.18 0.98

11 1.73 1.68 0.77

82.382.07

53.201.921.83

62.31.9

54.22.61.25

52.22.0

33.02.61.12

66.22.6

55.00.70.80

36.860.83

56.970.620.16

81.751.04

91.781.300.44

42.001.63

43.252.221.41

51.02.2

53.82.21.62

50.81.8

33.02.62.20

58.01.0

55.40.91.16

36.860.80

56.860.680.01

81.500.76

71.571.400.60

40.250.50

24.500.712.26

41.82.1

33.02.61.18

41.82.1

22.02.81.44

47.81.7

35.71.21.22

28.340.87

46.810.650.20

61.170.41

52.201.641.10

30.6671.154

15.000.00

42.02.4

22.53.5

41.82.1

22.53.5

48.30.5

26.01.4

60.830.41

41.500.581.50

200

15.000.00

32.32.1

10.00.0

32.32.1

10.00.0

38.70.6

17.00.0

41.250.50

21.000.00

32.3302.516

31.32.3

10.00.0

30.71.2

10.00.0

38.30.6

16.00.0

41.000.82

21.000.00

20.5000.707

31.01.7

10.00.0

30.71.2

10.00.0

38.31.2

16.00.0

BeforeBeforeBeforeBeforeBefore AfterAfterAfterAfterAfter

First day Second day Third day Fourth day Fifth day


Psychological Responses. Within the first 2 days,

anxiety levels in the intervention group decreased,

whereas those in the control group were stable

(Table 3C). The intervention had a significant effect

over time, even after adjustments for heart rate or

SAP (Cohen d = 1.04-1.24; Table 4). An adjusted

ANCOVA for anxiety 1 day after the intervention

indicated a significant decrease in the intervention

group (P = .01; f = 0.506).

Fear was moderate with a general downward

trend in both groups (Table 3D). Analysis with lin-

ear models indicated a nonstatistically significant

effect of the intervention (over a 2-day period;

P = .05; Cohen d = 1.13-1.44; Table 4). An adjusted

ANCOVA for fear on day 1 after the intervention

decreased significantly in the intervention group

(P = .04; f = 0.562).

Relaxation NRS ratings indicated a trend for

improvement in both groups (Table 3E). Linear

mixed models showed a significant effect of the

intervention in a 2-day observation period (P < .001;

Cohen d = 0.79-1.22; Table 4). An unadjusted

ANCOVA for ratings obtained 1 day after the inter-

vention indicated a significant increase in relaxation

in the intervention group (P = .004; f = 0.517); how-

ever, the adjusted model showed no statistical signif-

icance (P = .07; f = 0.352).

Optimism ratings fluctuated and had a moder-

ate upward trend in both groups (Table 3F). Linear

mixed models indicated that the effect of the inter-

vention during a 3-day observation period was sig-

nificant (P = .01; Cohen d = 0.76-1.97; Table 4). An

unadjusted ANCOVA showed a significant increase

in optimism in the intervention group (P = .02;

f = 0.574); however, after adjustments, the increase

was not significant (P = .16; f = 0.348).

Quality of Sleep. Self-reported quality of sleep

showed a trend toward gradual improvement in

both groups (Table 3G). During the first 2 days,

changes in sleep quality did not differ significantly

between the 2 groups (P = .98). Nevertheless, during

a 4-day observation period, sleep quality improved

significantly in the intervention group (P = .02,) with

a progressively increasing effect size (Cohen d = 0.1-

3.5). The effect of the intervention over time was sig-

nificant (linear models P = .02; Table 4).

Participants’ Outcomes. Incidence of respiratory

complications (atelectasis and pneumonia) was

higher in the control group (7) than in intervention

group (3): 2 = 1.9; P = .16; OR = 0.36; 95% CI = 0.08-

1.57). The incidence of delirium was also higher in

the control group: ( 2 = 3.2; P = .08). Renal failure

developed in 1 participant in the intervention group

and in none of the participants in the control group.

Content

Table 3Continued

G. Sleep qualityIntervention N Mean SDControl N Mean SDAdjusted Cohen d

H. SAPIntervention N Mean SDControl N Mean SDAdjusted Cohen d

254.22.1

214.62.40.22

30125.7 18.6

30128.9 22.20.16

30120.8 17.5

30133.3 21.50.63

75.33.4

66.31.20.06

13133.4 22.9

14130.5 19.30.03

13127.2 19.8

14138.6 24.10.80

66.81.7

55.01.21.09

9122.7 16.1

11131.9 20.00.87

9122.2 19.2

10133.6 20.70.88

48.01.2

25.50.71.60

7124.7 22.2

7141.4 16.91.39

7125.9 21.1

5134.8 25.10.89

38.71.5

16.00.0

6126.5 12.0

2118.5 21.9

5121.2 14.5

2108.5 12.0

Before BeforeBeforeBeforeBeforeAfter AfterAfterAfterAfter

First day Second day Third day Fourth day Fifth day

Abbreviations: CPOT, Critical Care Pain Observation Tool; N, number of participants; NRS, Numeric Rating Scale; SAP, systolic arterial blood pressure.

a Adjusted Cohen d effect sizes are based on estimated marginal means and are shown for those days for which linear mixed models could converge.


Mortality was 0% in both groups. The interven-

tion had no effect on the discharge SOFA score

(F = 0.021; P = .85). ICU length of stay did not differ

significantly (t = 0.469; P = .64) between the 2

groups, even though the mean length of stay was

almost 2 days shorter in the intervention group

Dependent variable

Table 4Linear mixed models for intervention effects accounting for repeated measurements

CPOT

Self-reported pain (NRS)

Anxiety (NRS)

Fear

Relaxation (NRS)

Optimism (NRS)

Sleep quality (NRS)

SAP

Opioid analgesic dose

Nonopioid analgesic dose

<.001<.001 .78<.001 .008 .003

.002 .02 .04 .01 .004 .11

<.001 .04 .02 .001 .04 .04

<.001 .14 .03 .05

<.001 .10<.001<.001

<.001 .98 .26 .01

<.001 .15 .07 .02

<.001 .41 .24 .007 .56 .009 .01 .004

.008 .37 .33 .24

<.001 .29 .04 .13

184.526 4.066 0.080 3.943 7.438 2.956

11.581 3.808 4.271 3.948 9.129 2.643

75.446 4.660 4.913 12.349 5.319 4.970

46.528 2.335 5.675 4.268

347.081 3.136 82.645 70.123

446.787 0.001 1.433 3.894

268.266 2.340 9.717 32.952

27.930 0.678 1.465 5.368 0.336 7.286 7.228 8.624

7.524 0.800 1.190 1.434

35.320 1.189 3.487 2.239

210.86129.99 82.86131.90 68.04 27.15

38.97 49.82 48.32 49.36 38.64 38.91

21.30 37.73 11.55 11.05 9.90 9.90

24.32 24.32 7.08 7.08

14.95 14.95 14.59 14.59

28.61 28.61 18.45 18.45

11.81 11.81 2.45 2.45

70.80 56.72 54.62 55.10 55.04 55.58 54.37120.25

77.88 77.88 34.62 34.62

16.51 6.51 1.01 11.01

1 9 1 9 131

1 3 1 3 1 1

1 1 3 3 1

1 1 3 3

1 1 3 3

1 1 5 5

1 1 3 3

1 1 2 2 1 1 1 1

1 1 4 4

1 1 4 4

InterceptTimeInterventionIntervention x time SexAge

InterceptTimeInterventionTime x interventionSexAge

InterceptInterventionTimeIntervention x timeHeart rateSAP

InterceptInterventionTimeIntervention x time




InterceptInterventionTimeIntervention x timeSexAgeVasoactive medicationRespiratory rate



PFDenominator dfNumerator dfSource

Abbreviations: CPOT, Critical Care Pain Observation Tool; NRS, Numeric Rating Scale; SAP, systolic arterial blood pressure.


(mean, 7.6; SD, 13.8) than in the control group

(mean, 9.4; SD, 14).

Use of Analgesics. The intervention had no sig-

nificant effect on the use of opioid (P = .24) or

nonopioid (P = .13) analgesics (Table 4), despite a

nonstatistically significant trend for decreased use

of nonopioids in the intervention group. Over time,

use of opioid analgesics remained approximately

constant in both groups.

Discussion We used a multimodal intervention and com-

prehensively assessed its effects on participants’

pain via subjective and behavioral assessments,

analgesic use, and related physiological and psy-

chological measures. We included a longer term

intervention, and, for the first time, we used

repeated-measures modeling. Our main findings

included a moderate to large effect size in decreas-

ing pain incidence and subjective, objective, and

behavioral pain ratings; a moderate effect size in

lowering SAP; a moderate to progressively large

effect size on the quality of sleep; and a moderate

effect size on anxiety, fear, and relaxation.

The intervention-associated decrease in pain is

consistent with previous reports28,43-45 of relaxation-

inducing interventions in critical and noncritical care

settings. The decrease in pain did not correspond to

use of analgesics, which remained almost stable, in

line with previous results.46 This finding may be under-

standable on the basis of clinical practices, because at

the time of the study, pain assessment tools were not

used in the unit, and, therefore, prescription of

analgesics was not based on valid assessments.

The multifactorial physiology of nociception

may favor multimodal rather than unimodal inter-

ventions, and this physiological context may be a

reason for the large effect sizes we observed.

Although the difference in complication rates may

reflect the bias of using a small sample, increased

atelectasis or pneumonia and delirium in the con-

trol group may be commensurate with higher pain.

The decrease in SAP is in accordance with

reports47-53 of interventions consisting of music or

massage. The intervention did not evoke unfavorable

hemodynamic effects; SAP and MAP were maintained

within the normal reference range. Although the tim-

ing of the effect may indicate parasympathetic trig-

gering, this notion is not commensurate with the

lack of effect on heart rate. Future research needs

to address the underlying mechanisms.

Anxiety and fear interfere with perception of

pain.9 The intervention-associated decrease in

anxiety is in line with previous reports on relax-

ation,28,44-46,54 music,48,49,55 and touch-massage.21,55,56

We also found some effect of the intervention on

patients’ fear, which is an important outcome

because fear is involved in pain perception.57 No

previous studies have addressed fear responses.

The effect of the intervention on relaxation agrees

with findings of a small pretest-posttest trial58 and is

important because anxiety can initiate a neuroendo-

crine cascade that interferes with recovery.59,60 The

moderate effect on optimism is noteworthy, because

optimism may modulate anxiety responses.61 The

improvement in self-reported quality of sleep is an

important finding, because lack of sleep may elicit

feedback to anxiety and pain.62 Previous research

with unimodal interventions did not show signifi-

cant effects on sleep54; multimodal interventions can

be more effective.25,26 Although testing the mediating

effect of sedation on relaxation and anxiety outcomes

would be interesting, only a few patients in our

study (1 participant per arm) continued to receive

sedation after day 1. Therefore such an investiga-

tion was not possible.

LimitationsLimitations include the small number of partic-

ipants and the progressive loss of participants, and

thus the loss of statistical power, as ICU patients were

discharged. Although we used block randomization

and adjustment, a larger study is needed to confirm

our results. On the basis of the eligibility criteria,

participants had overall low acuity and sedation

levels; therefore our conclusions may not be extrap-

olated to more severely ill patients. In the future, a

thorough assessment of the effect of sedation on

the effectiveness of the intervention is warranted.

The progressive loss of participants as they got better

and were discharged might have introduced type II

error and bias (either overestimation or underesti-

mation of the effects of the intervention) because

only the most severely ill participants remained in

the study. Moreover, the self-reporting nature of

many of the outcome variables contributed to miss-

ing values if participants were unable to self-report.

To mitigate the loss of statistical power as ICU patients

were discharged, we confirmed analyses by applying

longer and shorter term follow-up periods. However,

our results need to be tested in a larger sample with

inclusion criteria that would allow for longer follow-up.

Another limitation might be due to incom-

plete concealment; clinicians might have entered a

participant’s space at the time of intervention. We

tried to mitigate this risk by having independent


assessors, not employed in the unit, perform assess-

ments before and after the intervention. Although evi-

dence63 indicates the merits of allowing patients to

self-select music, for this pilot study we decided to

keep all aspects of the intervention constant. How-

ever, not allowing patients to self-select music might

have resulted in underestimation of the usefulness

of the intervention, especially in participants who

did not enjoy classical music.

Conclusions A multimodal integrative intervention delivered

once daily may be effective and safe in decreasing

pain and in improving pain-related outcomes in

critically ill patients. Cost-effectiveness associated

with costs for use of resources by personnel must be

addressed in the future. Moreover, in future attempts

to replicate our results with similar eligibility crite-

ria, a 2-day intervention would be more appropriate

to avoid postintervention attrition.

Financial DisclosuresThis research was supported by a Cyprus University of Technology faculty grant and a University of Alberta establishment grant to Dr Papathanassoglou.

REFERENCES1. Puntillo KA, Max A, Timsit JF, et al. Determinants of proce-

dural pain intensity in the intensive care unit: the Europain study. Am J Respir Crit Care Med. 2014;189(1):39-47.

2. Barr J, Fraser GL, Puntillo K et al; American College of Criti-cal Care Medicine. Clinical practice guidelines for the man-agement of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013;70(1):263-306.

3. Oh J, Sohn JH, Shin CS, et al. Mutual relationship between anxiety and pain in the intensive care unit and its effect on medications. J Crit Care. 2015;30(5):1043-1048.

4. Jaiswal SJ, Malhotra A, Owens RL. Sleep and critical illness: bridging the two pillars at the ATS 2016. J Thorac Dis. 2016; 8(suppl 7):S536-S538.

5. Davidson JE, Harvey MA, Bemis-Dougherty A, Smith JM, Hopkins RO. Implementation of the pain, agitation, and delirium clinical practice guidelines and promoting patient mobility to prevent post-intensive care syndrome. Crit Care Med. 2013;41(9)(suppl 1):136-145.

6. Battle CE, Lovett S, Hutchings H. Chronic pain in survivors of critical illness: a retrospective analysis of incidence and risk factors. Crit Care. 2013;17(1): R101.

7. Georgiou E, Hadjibalassi M, Lambrinou E, Andreou P, Papa-thanassoglou ED. The impact of pain assessment on criti-cally ill patients’ outcomes: a systematic review. Biomed Res Int. 2015;2015:503830. doi:10.1155/2015/503830.

8. Hughes CG, McGrane S, Pandharipande PP. Sedation in the intensive care setting. Clin Pharmacol. 2012;4:53-63.

9. Papathanassoglou E. Recent advances in understanding pain: what lies ahead for critical care? Nurs Crit Care. 2014; 19(3):110-113. doi: 10.1111/nicc.12097

10. Peris A, Bonizzoli M, Iozzelli D, et al. Early intra-intensive care unit psychological intervention promotes recovery from post traumatic stress disorders, anxiety and depression symptoms in critically ill patients. Crit Care. 2011; 15(1):R41.

11. Oke SL, Tracey KJ. The inflammatory reflex and the role of complementary and alternative medical therapies. Ann N Y Acad Sci. 2009;1172:172-180.

12. Hadjibalassi M, Lambrinou E, Papastavrou E, Papathanas-soglou E. The effect of guided imagery on physiological and psychological outcomes of adult ICU patients: a sys-tematic literature review and methodological implica-tions [published online March 29, 2017]. Aust Crit Care. doi:10.1016/j.aucc.2017.03.001.

13. Papathanassoglou E, Park T. To put the patient in the best condition: integrating integrative therapies in critical care. Nurs Crit Care. 2016;21(3):123-126.

14. Umezawa S, Higurashi T, Uchiyama S, et al. Visual distraction alone for the improvement of colonoscopy-related pain and satisfaction. World J Gastroenterol. 2015; 21(15):4707-4714.

15. Volz MS, Suarez-Contreras V, Portilla AL, Fregni F. Mental imagery-induced attention modulates pain perception and cortical excitability. BMC Neurosci. 2015;16:15.

16. Bonapace J, Chaillet N, Gaumond I, Paul-Savoie E, March-and S. Evaluation of the Bonapace Method: a specific edu-cational intervention to reduce pain during childbirth. J Pain Res. 2013;6:653-661.

17. Eisenberger NI. The neural bases of social pain: evidence for shared representations with physical pain. Psychosom Med. 2012;74(2):126-135.

18. Kross E, Berman MG, Mischel W, Smith EE, Wager TD. Social rejection shares somatosensory representations with phys-ical pain. Proc Natl Acad Sci U S A. 2011;108(15):6270-6275.

19. Complementary, alternative, or integrative health: what’s in a name? National Center for Complementary and Integra-tive Health. NCCIH publication D347. https://nccih.nih.gov /health/integrative-health. Updated June 2016. Accessed January 22, 2017.

20. Sanders S, reviewer. Am J Clin Hypn. 1981;24(1):65-66. Review of: Edmonston WE Jr. Hypnosis and Relaxation: Modern Verification of an Old Equation.

21. Papathanassoglou ED, Mpouzika MD. Interpersonal touch: physiological effects in critical care. Biol Res Nurs. 2012; 14(4):431-443.

22. Martorella G, Boitor M, Michaud C, Gélinas C. Feasibility and acceptability of hand massage therapy for pain man-agement of postoperative cardiac surgery patients in the intensive care unit. Heart Lung. 2014;43(5):437-444.

23. Gélinas C, Arbour C, Michaud C, Robar L, Côté J. Patients and ICU nurses’ perspectives of non-pharmacological inter-ventions for pain management. Nurs Crit Care. 2013; 18(6):307-318.

24. Hermann C. Psychological interventions for chronic pediat-ric pain: state of the art, current developments and open questions. Pain Manag. 2011;1(5):473-483.

25. Richards KC. Effect of a back massage and relaxation inter-vention on sleep in critically ill patients. Am J Crit Care. 1998;7(4):288-299.

26. Richardson S. Effects of relaxation and imagery on the sleep of critically ill adults. Dimens Crit Care Nurs. 2003; 22(4): 182-190.

27. Stein T, Olivo E, Grand S, Namerow P, Costa J, Oz M. A pilot study to assess the effects of a guided imagery audiotape intervention on psychological outcomes in patients under-going coronary artery bypass graft surgery. Holist Nurs Pract. 2010;24(4):213-222.

28. Kshettry V, Carole L, Henly S, Sendelbach S, Kummer B. Complementary alternative medical therapies for heart surgery patients: feasibility, safety, and impact. Ann Thorac Surg. 2006;81(1):201-205.

29. Moore CG, Carter RE, Nietert PJ, Stewart PW. Recommen-dations for planning pilot studies in clinical and translational research. Clin Transl Sci. 2011;4(5):332-337.

30. Dossey BM, Keegan L. Holistic Nursing: A Handbook for Practice. 5th ed. Boston MA: Jones & Bartlett Learning; 2009.

31. Conrad C, Niess H, Jauch KW, Bruns CJ, Hartl W, Welker L. Overture for growth hormone: requiem for interleukin-6? Crit Care Med. 2007;35(12):2709-2713.

32. Diego MA, Field T. Moderate pressure massage elicits a parasympathetic nervous system response. Int J Neurosci. 2009;119(5):630-638.

33. Faculty of Medicine and Dentistry, University of Alberta. Opioid conversion guide. https://www.ualberta.ca/medicine /institutes-centres-groups/multidisciplinary-pain-clinic/for -healthcare-professionals/opioid-conversion-guide. Accessed January 22, 2018.

34. Koh W, Nguyen KP, Jahr JS. Intravenous non-opioid analge-sia for peri- and postoperative pain management: a scientific review of intravenous acetaminophen and ibuprofen. Korean J Anesthesiol. 2015;68(1):3-12.

35. Callaway CW, Donnino MW, Fink EL, et al. Part 8: Post-cardiac arrest care: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care [published correction appears in Circulation. 2017;136(10):e197]. Circulation. 2015; 132(18)(suppl 2):S465-S482.


36. Mamoura K. Validation of the Greek Version of the Critical Care Pain Observation Tool [master’s thesis]. Athens, Greece: National and Kapodistrian University of Athens; 2017.

37. Gélinas C, Fillion L, Puntillo KA, Viens C, Fortier M. Valida-tion of the Critical-Care Pain Observation Tool in adult patients. Am J Crit Care. 2006;15(4):420-427.

38. Li Q, Wan X, Gu C, et al. Pain assessment using the Critical-Care Pain Observation Tool in Chinese critically ill ventilated adults. J Pain Symptom Manage. 2014;8(5):975-982.

39. Chanques G, Viel E, Constantin JM, et al. The measurement of pain in intensive care unit: comparison of 5 self-report intensity scales. Pain. 2010;151(3):711-721.

40. Ahlers SJ, van Gulik L, van der Veen AM, et al. Comparison of different pain scoring systems in critically ill patients in a general ICU. Crit Care. 2008;12(1):R15.

41. Twisk J. Applied Longitudinal Data Analysis for Epidemiology. 2nd ed. Cambridge, UK: Cambridge University Press; 2013.

42. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.

43. Deisch P, Soukup SM, Adams P, Wild MC. Guided imagery: replication study using coronary artery bypass graft patients. Nurs Clin North Am. 2000;35(2):417-425.

44. Tusek D L, Cwynar R, Cosgrove DM. Effect of guided imag-ery on length of stay, pain and anxiety in cardiac surgery patients. J Cardiovasc Manag. 1999;10(2):22-28.

45. Kuo B, Bhasin M, Jacquart J, et al. Genomic and clinical effects associated with a relaxation response mind-body interven-tion in patients with irritable bowel syndrome and inflam-matory bowel disease [published correction appears in PLos One. 2017;12(2):e0172872]. PLoS One. 2015; 10(4): e0123861.

46. Ashton C Jr, Whitworth GC, Seldomridge JA, et al. Self-hypnosis reduces anxiety following coronary artery bypass surgery: a prospective, randomized trial. J Cardiovasc Surg (Torino). 1997;38(1):69-75.

47. Mofredj A, Alaya S, Tassaioust K, Bahloul H, Mrabet A. Music therapy, a review of the potential therapeutic benefits for the critically ill. J Crit Care. 2016;35:195-199.

48. Wong HL, Lopez-Nahas V, Molassiotis A. Effects of music therapy on anxiety in ventilator-dependent patients. Heart Lung. 2001;30(5):376-387.

49. Han L, Li JP, Sit JW, Chung L, Jiao ZY, Ma WG. Effects of music intervention on physiological stress response and anxiety level of mechanically ventilated patients in China: a ran-domised controlled trial. J Clin Nurs. 2010;19(7-8):978-987.

50. Dijkstra BM, Gamel C, van der Bijl JJ, Bots ML, Kesecioglu J. The effects of music on physiological responses and sedation scores in sedated, mechanically ventilated patients. J Clin Nurs. 2010;19(7-8):1030-1039.

51. Albert NM, Gillinov AM, Lytle BW, Feng J, Cwynar R, Black-stone EH. A randomized trial of massage therapy after heart surgery. Heart Lung. 2009;38(6):480-490.

52. Vahedian-Azimi A, Ebadi A, Asghari Jafarabadi M, Saadat S,

Ahmadi F. Effect of massage therapy on vital signs and GCS scores of ICU patients: a randomized controlled clinical trial. Trauma Mon. 2014;19(3):e17031.

53. Adib-Hajbaghery M, Abasi A, Rajabi-Beheshtabad R. Whole body massage for reducing anxiety and stabilizing vital signs of patients in cardiac care unit. Med J Islam Repub Iran. 2014;28:47.

54. Casida JM, Yaremchuk KL, Shpakoff L, Marrocco A, Babicz G, Yarandi H. The effects of guided imagery on sleep and inflam-matory response in cardiac surgery: a pilot randomized con-trolled trial. J Cardiovasc Surg (Torino). 2013;54(2):269-279.

55. Lee OK, Chung YF, Chan MF, Chan WM. Music and its effect on the physiological responses and anxiety levels of patients receiving mechanical ventilation: a pilot study. J Clin Nurs. 2005;14(5):609-620.

56. Henricson M, Ersson A, Määttä S, Segesten K, Berglund AL. The outcome of tactile touch on stress parameters in inten-sive care: a randomized controlled trial. Complement Ther Clin Pract. 2008;14(4):244-254.

57. Palstam A, Larsson A, Löfgren M, et al. Decrease of fear avoid-ance beliefs following person-centered progressive resistance exercise contributes to reduced pain disability in women with fibromyalgia: secondary exploratory analyses from a random-ized controlled trial. Arthritis Res Ther. 2016; 18(1): 116.

58. Hattan J, King L, Griffiths P. The impact of foot massage and guided relaxation following cardiac surgery: a randomized controlled trial. J Adv Nurs. 2002;37(2):199-207.

59. Lusk B, Lash AA. The stress response, psychoneuroimmunol-ogy, and stress among ICU patients. Dimens Crit Care Nurs. 2005;24(1):25-31.

60. Papathanassoglou ED, Giannakopoulou M, Mpouzika M, Bozas E, Karabinis A. Potential effects of stress in critical illness through the role of stress neuropeptides. Nurs Crit Care. 2010;15(4):204-216.

61. Castillo MI, Cooke M, Macfarlane B, Aitken LM. Factors asso-ciated with anxiety in critically ill patients: a prospective observational cohort study. Int J Nurs Stud. 2016;60:225-233.

62. Stewart JA, Green C, Stewart J, Tiruvoipati R. Factors influencing quality of sleep among non-mechanically ventilated patients in the intensive care unit. Aust Crit Care. 2017; 30(2):85-90.

63. Heiderscheit A, Chlan L, Donley K. Instituting a music listen-ing intervention for critically ill patients receiving mechani-cal ventilation: exemplars from two patient cases. Music Med. 2011;3(4):239-246.

To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; email, [email protected].

1.0 Hour Category BC ENotice to CE enrollees:

This article has been designated for CE contact hour(s). The evaluation demonstrates your knowledge of the

following objectives:

1. Identify factors implicated in the perception of pain in the critically ill.

2. Describe the role of nonpharmacological therapies in the treatment of critical illness pain.

3. Discuss the research design issues for studies examining the effectiveness of integrative interventions in criti-

cal care.

To complete the evaluation for CE contact hour(s) for this article #A1827033, visit www.ajcconline.org and

click the “CE Articles” button. No CE evaluation fee for AACN members. This expires on May 1, 2021.

The American Association of Critical-Care Nurses is an accredited provider of continuing nursing education by the American Nurses Credentialing Center’s Commission on Accreditation. AACN has been approved as a provider of continuing education in nursing by the State Boards of Registered Nursing of California (#01036) and Louisiana (#LSBN12).



Background Nurse-facilitated mobility of patients in the intensive care unit can improve outcomes. However, a gap exists between research findings and their implemen-tation as part of routine clinical practice. Such a gap is often attributed, in part, to the barrier of lack of time. The Translating Evidence Into Practice model provides a framework for research implementation, including rec-ommendations for identifying barriers to implementa-tion via direct observation of clinical care.Objectives To report on design, implementation, and outcomes of an approach to identify and understand lack of time as a barrier to nurse-facilitated mobility in the intensive care unit.Methods An interprofessional team designed the obser-vational process and evaluated the resulting data by using qualitative content analysis. Results During three 4-hour observations of 2 nurses and 1 nursing technician, 194 distinct tasks were per-formed (ie, events). A total of 4 categories of nurses’ work were identified: patient care (47% of observation time), provider communication (25%), documentation (18%), and down time (10%). In addition, 3 types of potential mobility events were identified: in bed, edge of bed, and out of bed. The 194 observed events included 34 instances (18%) of potential mobility events that could be implemented: in bed (53%), edge of bed (6%), and out of bed (41%). Conclusions Nurses have limited time for additional clinical activities but may miss potentially important opportunities for facilitating patient mobility during existing patient care. The proposed method is feasible and helpful in empirically investigating barriers to nurse-facilitated patient mobility in the intensive care unit. (American Journal of Critical Care. 2018; 27:186-193)

IDENTIFYING BARRIERS TO NURSE-FACILITATED PATIENT MOBILITY IN THE INTENSIVE CARE UNITBy Daniel L. Young, PT, DPT, PhD, Jason Seltzer, PT, DPT, Mary Glover, RN, Caroline Outten, RN, BSN, CCRN, Annette Lavezza, OTR/L, Earl Mantheiy, BA, Ann M. Parker, MD, and Dale M. Needham, MD, PhD, FCPA


This article is followed by an AJCC Patient Care Page on page 204.

The muscle weakness commonly experienced by survivors of critical illness is a potentially modifiable risk factor for long-term functional impairment, mortality, and use of health care resources.1-12 Facilitating early mobility of patients in the intensive care unit (ICU) can reduce such muscle weakness13,14 and can be done safely and feasibly, even in patients receiving mechanical ventilation who have an

endotracheal tube in place.15-21 Many ICU mobilization efforts have engaged physical and occupational therapists (PTs and OTs), but successfully changing an ICU culture of bed rest requires the involvement of all clinicians.13,19,22

Engaging ICU nurses to mobilize patients can

have positive outcomes.23-25 In an international, mul-

ticentered, blinded randomized controlled trial,8,26

a multidisciplinary team set a daily mobility goal for

each patient, and ICU nurses worked with the patient

to meet that goal. The results indicated that such

nurse-facilitated mobility improved patients’ mobil-

ity status and decreased length of stay.8 However, a

gap exists between research evidence and its imple-

mentation as part of routine clinical practice in the

ICU.17,27,28 Structured quality improvement projects

play a critical role in bridging that gap.29,30

One framework for designing and executing

structured quality improvement projects is the

Translating Evidence Into Practice (TRIP) model.31

The model has 4 stages: summarize the evidence,

identify local barriers, measure performance, and

ensure patients receive the intervention via an itera-

tive process of engaging, educating, executing, and

evaluating. Within the TRIP model, stage 2 high-

lights the need to identify local barriers to imple-

mentation and recommends direct observation of

clinicians as a successful approach to inform quality

improvement efforts. For nurse-facilitated mobiliza-

tion in the ICU, a commonly cited barrier is lack of

time.17,20,32,33 However, to our knowledge, design

and implementation of a method to understand

this specific barrier have not been undertaken. Hence,

our objective was to report on the design, imple-

mentation, and outcomes of an approach to identi-

fying and understanding the barrier of lack of time

for nurse-facilitated mobility in the ICU.

Methods Background

An early rehabilitation quality improvement

project with the TRIP model was previously con-

ducted in the Johns Hopkins Hospital medical ICU

(MICU).29,34 This project increased early rehabilita-

tion, with interventions primarily

performed by PTs and OTs. However,

lighter sedation35 and a growing cul-

ture of mobility clearly indicated

that MICU patients have the poten-

tial for mobility and activity via inter-

ventions facilitated by nurses and

clinical technicians in addition to

interventions performed by PTs and

OTs. To understand time-related bar-

riers to performing such interventions in our MICU,

we directly observed the work carried out by nurses

and a clinical technician and measured the time

required for that work in order to understand the

feasibility of integrating mobility interventions into

the daily activities of nurses and technicians.

Interprofessional Generation of IdeasThe first step in this project involved interpro-

fessional discussions and planning by members of

the quality improvement team, including physi-

cians (D.M.N., A.M.P.), nurses (M.G., C.O.), physical

About the AuthorsDaniel L. Young is an associate professor, Department of Physical Therapy, University of Nevada Las Vegas, Las Vegas, Nevada, and a visiting scientist, Department of Physical Medicine and Rehabilitation, and Outcomes After Critical Illness and Surgery (OACIS) Group, Johns Hopkins University, Baltimore, Maryland. Jason Seltzer is intensive care unit rehabilitation team coordinator, Department of Physical Medicine and Rehabilitation, and OACIS Group, Johns Hopkins Hospital, Baltimore, Maryland. Annette Lavezza is therapy manager, Depart-ment of Physical Medicine and Rehabilitation, and OACIS Group, Johns Hopkins Hospital. Mary Glover is a nurse clinician, medical intensive care unit, Johns Hopkins Hospital. Caroline Outten is a nurse clinician, Department of Medicine, Johns Hopkins Hospital. Earl Mantheiy is senior clinical coordinator, Division of Pulmonary and Critical Care Medicine, and OACIS Group, Johns Hopkins University. Ann M. Parker is an assistant professor, Divi-sion of Pulmonary and Critical Care Medicine, and OACIS Group, Johns Hopkins University. Dale M. Needham is a professor, Division of Pulmonary and Critical Care Medi-cine, Department of Physical Medicine and Rehabilitation, and OACIS Group, Johns Hopkins University.

Corresponding author: Dale M. Needham, MD, 5th Floor, 1830 E Monument St, Baltimore, MD 21205 (e-mail: dale .needham @jhmi.edu).


Understanding of barriers to nurse-facilitated patient mobility needs to improve.


We directly observed nurses and measured

the time required for their work.

therapists (J.S., A.T.), an occupational therapist (A.L.),

and a senior clinical coordinator for the ICU clinical

rehabilitation program (E.M.). In these meetings,

team members discussed the idea of directly observ-

ing and recording work-related activities of nurses

and clinical technicians. Including clinical techni-

cians was suggested because they are delegated

certain tasks to perform by MICU nurses, including

mobility-related interventions. The quality improve-

ment team decided that direct observation would be

a valuable way to “walk the process” of nurses’ work

activities and clarify the lack-of-time barrier. The Johns

Hopkins University institutional review board deemed

the project quality improvement, under the US Office

for Human Research Protections guidance, with no

need for consent of patients or clinicians.

Planning the ObservationsWithin the framework of typical case sampling,36

a type of purposive sampling, representative days,

times of day, and personnel for the observations

were determined via discussions with nurses on

the quality improvement team. The quality improve-

ment team wanted to observe times that included

typical responsibilities of nurses, such as morning

report, initial patient assess-

ment, routine patient assess-

ments, and MICU team rounds,

because these routine nursing

events were generalizable to

most patients. The quality

improvement team wanted to

avoid observations of times when

nonroutine events occur more

frequently or when mobility may be less feasible. In

our MICU, such times include 11:30 AM to 3:30 PM,

when tests and procedures unique to individual

patients occur more frequently, and during the

night shift. Hence, observations were planned for 3

consecutive days: day 1, from 3 PM to 7:30 PM; day 2,

from 7 AM to 10 AM; and day 3, from 7 AM to 11 AM.

The nurse or technician to be observed was selected

to achieve variability in staff experience and

patient characteristics. In the MICU, 1 nurse with

more than 5 years of experience and 1 with less

than 2 years of experience were chosen. One nurse

had a 2-patient assignment; the other, a 1-patient

assignment. The third observation was designated

for a clinical technician, as previously explained.

The quality improvement team thought that this

mix of days, times, and personnel was representa-

tive and feasible for this project.

Executing Direct ObservationsA physical therapist (J.S.), the rehabilitation

team coordinator for the hospital’s adult ICUs,

performed all observations, for a total of 10 hours.

During these times, the observer’s goal was to watch

the designated nurse or clinical technician and record

all of their activities and the time spent on each activ-

ity. Using a paper-based logbook, the observer chrono-

logically recorded a description of each task performed

by the nurse or technician (eg, assessments, admin-

istering medications) with the associated starting and

ending time for each event. In order to reduce the

potential for modifying staff attitudes or behaviors,

the specific purpose of the observation was described

as better understanding nurses’ workflow, without

mentioning the specific focus on understanding

barriers to patient mobilization.

Review of the Observation Logbook Four members of the project team, 2 nurses

(M.G., C.O.) and 2 physical therapists (J.S., D.L.Y),

met in person to receive a brief orientation (by

J.S.) to the observation logbook, including defini-

tion of abbreviations, and to receive a copy of the

logbook for their individual reviews. These 4 team

members used qualitative content analysis37 to

evaluate the logbooks. First, each of the 4 mem-

bers independently read the logbook to become

familiar with its content, identify “meaning units,”

and categorize those meaning units into codes.37

For example, 1 activity logbook entry reads, “Addi-

tional family member calls to talk to RN. On

phone discusses social aspects of pt care, too many

family members involved asking for updates. RN

arranges specific family members who are identi-

fied by current family member (P.O.A.) who can

receive info.” The meaning unit here would be

family member calls to talk to RN, and the code

created was family-centered care. D.L.Y. then reviewed

all 4 coded logbooks and identified 4 categories37

into which all codes fit: patient care (eg, measur-

ing blood pressure, administering medication,

talking to a patient’s family), documentation (eg,

keying in the electronic medical record), provider

communication (eg, talking with the patient’s

physician, taking report from another nurse),

and down time (eg, time during which the nurse

or technician had no active demands). The catego-

ries assigned to each event were then reviewed by

the other 3 members of the research team in a

group meeting, and discrepancies were resolved

by consensus.


More mobility opportunities were identified for in-bed activities than out-of-bed or edge-of-bed activities.

Review of Potential for Mobility Using the analyses of the logbooks, the 4 mem-

bers of the project team (M.G., C.O., J.S., D.L.Y.)

met and collectively discussed potential for nurse-

facilitated mobility events, what those activities might

have been, and when the activities could have

occurred. For example, when the logbook stated,

“Bed laid flat, bed elevated to begin clean up. Chang-

ing chucks [disposable bed pads], linens that are

soiled, places new brief down,” the 4 members

thought that the nurse could have encouraged the

patient to “roll and bridge actively” (1 example of

a code from this process) as the linens, chuck, and

brief were changed. Another example that was noted

several times in the logbook was meals being served

in bed, and the 4 members identified the opportu-

nity for the nurse to help the patient sit at the edge

of bed, or get out of bed to a chair, to eat the meal

(“out of bed for meals” was the code used). When

the logbook indicated that time was spent in work

away from the patient (eg, documentation), the 4

members thought that mobility was not possible.

On the basis of these group discussions, 4 categories

of potential mobility events emerged: not possible,

in bed, edge of bed, and out of bed. These categories

for potential mobility events, along with the codes

from which they were created, were added to the

logbook. As a result of this process, each event in

the logbook was assigned 2 categories, 1 for general

workload (patient care, documentation, provider

communication, down time) and 1 for potential

mobility (not determined, in bed, edge of bed, out

of bed).

Results Design

The first step in the design process was assem-

bling the interdisciplinary quality improvement

team. This process was an iterative one, conducted

during 3 MICU quality improvement group meet-

ings, devoted to identifying appropriate personnel

needed for this specific project. Once the team

members to execute this specific project were identi-

fied, two 1-hour meetings were needed to outline

the plan for the observations and discuss questions

about those specific methods. One of the more

time-consuming items on the agenda of those 2

meetings was discussing who and when to observe.

Selecting the right person to perform the observa-

tions was also discussed in detail. The final issue

was creating the logbook. Attention was given to

create adequate space for detailed descriptions of

desired elements for the observed work and for

notation of potential mobility activities. The origi-

nal logbooks were created and implemented in

paper form, but for analysis, they were transcribed

into an electronic document.

ImplementationAs planned, the observations took approximately

11 hours (655 minutes). We found no indication that

the observations created disruption or concern for the

clinicians being observed, the

patient, or anyone else coming

into a patient’s room. Transcrip-

tion of the observation log into

an electronic document took 1.5

hours. The qualitative analysis

involved three 1-hour meetings

with 4 of the team members

(M.G., C.O., J.S., D.L.Y.), 1 hour

for each person to identify the

meaning units and codes inde-

pendently, and another 2 hours

for D.L.Y. to apply categories to the logs. Total time

for this part of the project was approximately 11 hours

for observation and 20 person-hours for analysis.

OutcomesThe results of the qualitative content analysis

of activities and time revealed 194 distinct events

observed during a period of 655 total minutes (Table

1). As expected, patient care accounted for the larg-

est part of observed time (47%) among the 4 defined

categories. The second largest time commitment

(25%) was communication with other health care

providers. Documentation accounted for 18% and

down time for 10%. Table 2 provides examples of

logbook entries and the codes and categories that

were assigned to those entries during the qualitative

content analysis.

The final step in completing the qualitative

content analysis of the logbook data was identifying

potential mobility events. The number of logbook

entries during which mobility could have been

Activity observed

Table 1Direct observation of nurse and clinical technician activity, by category

Patient care

Provider communication

Documentation

Down time

Total

305 (47)

166 (25)

116 (18)

68 (10)

655 (100)

115 (59)

47 (24)

27 (14)

5 (3)

194 (100)

Time required, min (%)

Distinct tasks observed, No. (%)


facilitated is reported in Table 3. As expected, all

potential mobility events were identified during

times of direct patient care or down time. The not

possible category for mobility events had the most

events (82%) and time (79% of observed minutes)

and included all time spent in communication

with other providers and in documentation. More

opportunities were identified for in-bed activities

(9%) and out-of-bed activities (7%) than for edge-

of-bed activities (1%). However, in-bed activities

were the third largest category in terms of minutes

(6%), after not possible (79%) and out of bed

(15%). Table 4 provides examples of logbook

entries, codes, and categories that were assigned to

those entries during the qualitative content analysis.

Discussion We devised a process to investigate the lack-of-

time barrier that nurses often express when asked

about facilitating mobility of their patients. Using

the TRIP model31 as a guide, we explored this barrier

by using a direct observation approach. Herein we

have provided the design, implementation, and out-

comes of the approach. This approach was feasible

and provided valuable insight into how nurses and

clinical technicians spend their time and where

potential mobility could occur. We observed that

nurses and clinical technicians working with nurses

have relatively little time not already filled with

patient care and related tasks. However, coincid-

ing with direct patient care, we detected important

opportunities for patient mobility that could be

facilitated by a nurse or a technician.

Our observations confirm that nurses are

busy.38-41 The nurses and clinician technician were

completely free from work tasks for only 10% of the

observed time, and most of that time was attributed

to the clinical technician, rather than nurses. This

Logbook observation

Table 2Examples of logbook observations of clinician activities with assigned codes and categories

Performing assessment

Face-to-face communication

Comfort and safety

Facilitating ADLs/IADLs

NA

Communication with patient

Getting supplies/medications

Documentation

Facilitating ADLs/IADLsPerforming assessment

Documentation

Patient safety and comfort


Enters room 58, checks blood pressure to assess vitals prior to giving BP meds

Approached by buddy to take report on 2 patients (room 59 and…) while buddy is on lunch break

Pillows replaced under both arms, pt assists with placement of pillows, demonstrating shoulder AROM

Prepare to enter room 60 where pt is delirious, asking to go to bathroom. Pt is put on bedpan

Done charting

Pt spontaneously moving, RN addressing pt directly at eye level to attempt to calm down

Gathers supplies from supply room for 58: chucks, briefs, wipes, linens

Starts filing out rounding template (update to provide to team since last report, issues, concerns, drips pt has been started on)

Breakfast arrives, sets pt up to eat in bed and takes dexi

Updates and rereads notes from multi D rounds for better understanding of pt social issues

Checks on pt to see if he wants or needs anything

Reviews CT scan with RN for pt who was scheduled for level 1 to OR

Patient care


Patient care

Patient care

Down time

Patient care

Patient care

Documentation

Patient care

Documentation

Patient care


CategoryCode

Abbreviations: ADL, activities of daily living; AROM, active range of motion; BP, blood pressure; chucks, disposable bed pads; CT, computed tomography; dexi, Dextrostix (reagent strip for estimating blood sugar levels); drips, infusions; D rounds, multidisciplinary rounds; IADL, instrumental activities of daily liv-ing; meds, medications; NA, not applicable; OR, operating room; pt, patient; RN, nurse.

Table 3Potential mobility events during observed clinician activities

a Because of rounding, percentages do not total 100.

Mobility event category

Not possible

In bed

Out of bed

Edge of bed

Total

518 (79)

39 (6)

95 (15)

3 (< 1)

655 (100)

160 (82)

18 (9)

14 (7)

2 (1)

194 (100)

Minutes (%)aNo. (%)a


Coinciding with direct patient care are opportunities for nurse-facilitated patient mobility.

down time may be a target for increasing patient

mobility. Documentation was a major time require-

ment; nearly 20% of all time observed was devoted

to this task. Other reports of the proportion of nurses’

time spent documenting range from 7%38 to 35%.42

In 2 studies of critical care nurses, values were rela-

tively similar to ours: 24%39 and 30%.40 These results

support efforts to streamline documentation to allow

more time for direct patient care.43,44 Despite the

relative lack of down time among the nurses in our

project, we observed opportunities to facilitate

patient mobility concurrently with other tasks of

direct patient care.

During the logbook review, when searching

for potential mobility opportunities, the nursing

members of the project team repeatedly commented

that, in retrospect, mobility was possible, but not

performed, because nurses “just don’t think to do

it.” In order for nurses to think more about mobil-

ity and to facilitate mobility, a shift in culture and

behavior must occur.19,29,45-51 In a systematic review,

Coquhoun et al52 highlighted 4 key components:

identifying barriers, selecting intervention compo-

nents, using theory, and engaging end users. Direct

observation is a potentially important component

of identifying barriers and engaging end users. The

results from direct observation could be effectively

used as a starting point for discussions about the

what and when of nurse-facilitated patient mobility.

Another strategy to help detect missed mobility

opportunities could be setting specific mobility

goals for a nursing shift, facilitated by using a sim-

ple and clear mobility scale, as successfully done in

a recent randomized controlled trial.8,53,54

Our project has potential limitations. First, this

project was conducted in a single MICU in which 2

nurses and 1 clinical technician were observed by

only 1 physical therapist. Observing other providers,

at different times, in different

units or hospitals, could change

the findings. The observer also may

not have been completely objective

regarding documentation of obser-

vations in the logbook, and a dif-

ferent physical therapist or a nurse

might have differed in documen-

tation. Although qualitative evalua-

tions have this type of limitation,

they offer an important framework

for other researchers to replicate and obtain results

specific to other institutions.55 Finally, because the

potential mobility events were not actually

observed, but inferred from post hoc review of log-

books, the number of those potential mobility events

and the time required may contain measurement

error.

Conclusion We have described a process for better under-

standing the lack-of-time barrier to nurse-facilitated

Logbook observation

Table 4Examples of logbook observations of potential mobility events with assigned codes and categories

Ask pt to sit at side of bed for assessment

In-bed exercise

Have patient get out of bed to chair for oral medication

Have patient get out of bed to chair for meals


Ambulate with patient

Help patient stand at sink in room

In-bed exercise: rolling, bridging, scooting

In-bed exercise: scooting, bridging

Assessment: lung/abdominal sounds, pain, pulses, strength by assessing resisted DF/hip flexion via heel slide, orientation questions mixed in throughout; asks pt what he would like for breakfast

Gets up to check on rm 58 who is agitated, vent alarming

Prepares to give meds to pt. Pt is mobile in bed, moves into long sitting in bed to take oral meds, RN raises HOB to support pt while pt is sitting upright. Pt returns to resting against support of bed

Breakfast arrives, sets pt up to eat in bed and takes dexi

Talks to providers about labs requested

Walk past rm 56 and pt now in chair at bedside, with assistance of son

Gives pt his toothbrush while pt seated in chair. Tech gets TB from table on opposite side of room, prepares it with water cup, gives in to pt while pt is seated in chair

BP did not record, needs retake—asks pt how he does moving around in bed

Pt asks to be boosted/moved to HOB

Edge of bed

In bed

Out of bed

Out of bed

Not possible

Out of bed

Out of bed

In bed

In bed

CategoryCode

Abbreviations: BP, blood pressure; dexi, Dextrostix (reagent strip for estimating blood sugar levels); DF, dorsiflexion; HOB, head of bed; labs, laboratory tests; meds, medications; pt, patient; rm, room; RN, nurse; TB, toothbrush; tech, technologist; vent, ventilator.


patient mobility. The design and implementation

of this process were feasible and provided valuable

insights for our quality improvement project. We

observed that nurses have relatively little available

time for additional clinical activities, but they may

miss opportunities to facilitate patient mobility as

part of existing patient care activities. Using our

observational process, other researchers can begin

to better understand barriers to nurse-facilitated

patient mobility in other clinical settings.

ACKNOWLEDGMENTThis work was performed at Johns Hopkins Hospital.

FINANCIAL DISCLOSURESDr Young was supported by the Foundation for Physical Therapy’s Center of Excellence in Physical Therapy Health Services and Health Policy Research and Training Grant. Dr Parker was supported by award 1K23HL138206-01 from the National Institutes of Health and received an honorarium for a presentation for Vizient.

SEE ALSO For more about patient mobility, visit the Critical Care Nurse website, www.ccnonline.org, and read the article by Messer et al, “Implementation of a Progressive Mobilization Program in a Medical-Surgical Intensive Care Unit” (October 2015).

REFERENCES 1. Herridge MS, Cheung AM, Tansey CM, et al; Canadian Critical

Care Trials Group. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med. 2003; 348(8):683-693.

2. Herridge MS, Tansey CM, Matté A, et al; Canadian Critical Care Trials Group. Functional disability 5 years after acute respira-tory distress syndrome. N Engl J Med. 2011; 364(14): 1293-1304.

3. Mueller N, Murthy S, Tainter CR, et al. Can sarcopenia quan-tified by ultrasound of the rectus femoris muscle predict adverse outcome of surgical intensive care unit patients as well as frailty? A prospective, observational cohort study. Ann Surg. 2016;264(6):1116-1124.

4. Heyland DK, Stapleton RD, Mourtzakis M, et al. Combining nutrition and exercise to optimize survival and recovery from critical illness: conceptual and methodological issues. Clin Nutr. 2016;35(5):1196-1206.

5. Fan E, Dowdy DW, Colantuoni E, et al. Physical complications in acute lung injury survivors: a two-year longitudinal pro-spective study. Crit Care Med. 2014;42(4):849-859.

6. Dinglas VD, Aronson Friedman L, Colantuoni E, et al. Muscle weakness and 5-year survival in acute respiratory distress syndrome survivors. Crit Care Med. 2017;45(3):446-453.

7. Schweickert WD, Pohlman MC, Pohlman AS, et al. Early physical and occupational therapy in mechanically venti-lated, critically ill patients: a randomised controlled trial. Lancet. 2009;373(9678):1874-1882.

8. Schaller SJS, Anstey M, Blobner M, et al; International Early SOMS-guided Mobilization Research Initiative. Early, goal-directed mobilisation in the surgical intensive care unit: a randomised controlled trial. Lancet. 2016;388(10052):1377-1388.

9. Ruhl AP, Huang M, Colantuoni E, et al; with the National Insti-tutes of Health, National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network. Healthcare utilization and costs in ARDS survivors: a 1-year longitudinal national US multicenter study. Intensive Care Med. 2017;43(7):980-991.

10. Ruhl AP, Lord RK, Panek JA, et al. Health care resource use and costs of two-year survivors of acute lung injury: an observational cohort study. Ann Am Thorac Soc. 2015; 12(3): 392-401.

11. Ruhl AP, Huang M, Colantuoni E, et al. Healthcare resource use and costs in long-term survivors of acute respiratory distress syndrome: a 5-year longitudinal cohort study. Crit Care Med. 2017;45(2):196-204.

12. Hermans G, Van Mechelen H, Clerckx B, et al. Acute outcomes and 1-year mortality of intensive care unit-acquired weakness. A cohort study and propensity-matched analysis. Am J Respir Crit Care Med. 2014;190(4):410-420.

13. Hashem MD, Parker AM, Needham DM. Early mobilization and rehabilitation of patients who are critically ill. Chest. 2016; 150(3):722-731.

14. Tipping CJ, Harrold M, Holland A, Romero L, Nisbet T, Hodgson CL. The effects of active mobilisation and rehabilitation in ICU on mortality and function: a system-atic review. Intensive Care Med. 2017;43(2):171-183.

15. Wahab R, Yip NH, Chandra S, et al. The implementation of an early rehabilitation program is associated with reduced length of stay: a multi-ICU study. J Intensive Care Soc. 2016;17(1):2-11.

16. Perme C, Nalty T, Winkelman C, Kenji Nawa R, Masud F. Safety and efficacy of mobility interventions in patients with femoral catheters in the ICU: a prospective observa-tional study. Cardiopulm Phys Ther J. 2013;24(2):12-17.

17. Nydahl P, Sricharoenchai T, Chandra S, et al. Safety of patient mobilization and rehabilitation in the intensive care unit: systematic review with meta-analysis. Ann Am Thorac Soc. 2017;14(5):766-777.

18. Bailey P, Thomsen GE, Spuhler VJ, et al. Early activity is feasible and safe in respiratory failure patients. Crit Care Med. 2007;35(1):139-145.

19. Dubb R, Nydahl P, Hermes C, et al. Barriers and strategies for early mobilization of patients in intensive care units. Ann Am Thorac Soc. 2016;13(5):724-730.

20. Parry SM, Knight LD, Connolly B, et al. Factors influencing physical activity and rehabilitation in survivors of critical ill-ness: a systematic review of quantitative and qualitative studies. Intensive Care Med. 2017;43(4):531-542.

21. Sricharoenchai T, Parker AM, Zanni JM, Nelliot A, Dinglas VD, Needham DM. Safety of physical therapy interventions in critically ill patients: a single-center prospective evaluation of 1110 intensive care unit admissions. J Crit Care. 2014;29(3): 395-400.

22. Parry SM, Nydahl P, Needham DM. Implementing early physical rehabilitation and mobilisation in the ICU: institu-tional, clinician, and patient considerations [published online ahead of print August 2017]. Intensive Care Med. doi:10.1007/s00134-017-4908-8.

23. Titsworth WL, Hester J, Correia T, et al. The effect of increased mobility on morbidity in the neurointensive care unit. J Neu-rosurg. 2012;116(6):1379-1388.

24. Piva S, Dora G, Minelli C, et al. The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian pop-ulation of medical, surgical, and neurologic intensive care unit patients. J Crit Care. 2015;30(6):1251-1257.

25. Dong ZH, Yu BX, Sun YB, Fang W, Li L. Effects of early rehabili-tation therapy on patients with mechanical ventilation. World J Emerg Med. 2014;5(1):48-52.

26. Meyer MJ, Stanislaus AB, Lee J, et al. Surgical Intensive Care Unit Optimal Mobilisation Score (SOMS) trial: a proto-col for an international, multicentre, randomised controlled trial focused on goal-directed early mobilisation of surgical ICU patients. BMJ Open. 2013;3(8):e003262.

27. Sibilla A, Nydahl P, Greco N, et al. Mobilization of mechan-ically ventilated patients in Switzerland [published online ahead of print August 2017]. J Intensive Care Med. doi: 10.1177/0885066617728486.

28. Jolley SE, Moss M, Needham DM, et al. Point prevalence study of mobilization practices for acute respiratory failure patients in the United States. Crit Care Med. 2017;45(2):205-215. doi:10.1097/CCM.0000000000002058.

29. Needham DM, Korupolu R. Rehabilitation quality improvement in an intensive care unit setting: implementation of a quality improvement model. Top Stroke Rehabil. 2010;17(4):271-281.

30. Hashem MD, Nelliot A, Needham DM. Early mobilization and rehabilitation in the ICU: moving back to the future. Respir Care. 2016;61(7):971-979.

31. Pronovost PJ, Berenholtz SM, Needham DM. Translating evidence into practice: a model for large scale knowledge translation. BMJ. 2008;337(6):963-965.

32. Zanni JM, Korupolu R, Fan E, et al. Rehabilitation therapy and outcomes in acute respiratory failure: an observational pilot project. J Crit Care. 2010;25(2):254-262.

33. Anekwe DE, Koo KK-Y, de Marchie M, Goldberg P, Jayara-man D, Spahija J. Interprofessional survey of perceived barriers and facilitators to early mobilization of critically ill patients in Montreal, Canada [published online ahead

of print January 1, 2017]. J Intensive Care Med. doi: 10.1177/0885066617696846.

34. Needham DM, Korupolu R, Zanni JM, et al. Early physical medicine and rehabilitation for patients with acute respira-tory failure: a quality improvement project. Arch Phys Med Rehabil. 2010;91(4):536-542.

35. Hager DN, Dinglas VD, Subhas S, et al. Reducing deep seda-tion and delirium in acute lung injury patients: a quality improvement project. Crit Care Med. 2013;41(6):1435-1442.

36. Suri H. Purposeful sampling in qualitative research synthe-sis. Qual Res J. 2011;11(2):63-75.

37. Erlingsson C, Brysiewicz P. A hands-on guide to doingcontent analysis. Afr J Emerg Med. 2017;7(3):93-99.doi:10.1016/j.afjem.2017.08.001.

38. Westbrook JI, Duffield C, Li L, Creswick NJ. How much time do nurses have for patients? A longitudinal study quantify-ing hospital nurses’ patterns of task time distribution and interactions with health professionals. BMC Health Serv Res. 2011;11:319.

39. Douglas S, Cartmill R, Brown R, et al. The work of adultand pediatric intensive care unit nurses. Nurs Res. 2013; 62(1): 50-58.

40. Ballermann MA, Shaw NT, Mayes DC, Gibney RT, West-brook JI. Validation of the Work Observation Method By Activity Timing (WOMBAT) method of conducting time-motion observations in critical care settings: an observa-tional study. BMC Med Inform Decis Mak. 2011;11(1):32.

41. Farquharson B, Bell C, Johnston D, et al. Frequency of nursing tasks in medical and surgical wards. J Nurs Manag. 2013;21(6):860-866.

42. Hendrich A, Chow MP, Skierczynski BA, Lu Z. A 36-hospital time and motion study: how do medical-surgical nurses spend their time? Perm J. 2008;12(3):25-34. http://www.ncbi.nlm.nih.gov/pubmed/21331207. Accessed February 6, 2018.

43. Freeman R, Maley K. Mobilization of intensive care cardiac surgery patients on mechanical circulatory support. Crit CareNurs Q. 2013;36(1):73-88.

44. Cheevakasemsook A, Chapman Y, Francis K, Davies C. The study of nursing documentation complexities. Int J Nurs Pract. 2006;12(6):366-374.

45. Parker AM, Sricharoenchai T, Needham DM. Early rehabilita-tion in the intensive care unit: preventing impairment of physical and mental health. Curr Phys Med Rehabil Rep.2013;1(4):307-314.

46. Drolet A, DeJuilio P, Harkless S, et al. Move to improve: the feasibility of using an early mobility protocol to increase ambulation in the intensive and intermediate care settings. Phys Ther. 2013;93(2):197-207.

47. Engel HJ, Needham DM, Morris PE, Gropper MA. ICU early mobilization: from recommendation to implementation at three medical centers. Crit Care Med. 2013;41(9)(suppl 1): S69-S80.

48. Hopkins RO, Spuhler VJ, Thomsen GE. Transforming ICU cul-ture to facilitate early mobility. Crit Care Clin. 2007; 23(1): 81-96.

49. Cameron S, Ball I, Cepinskas G, et al. Early mobilization in the critical care unit: a review of adult and pediatric litera-ture. J Crit Care. 2015;30(4):664-672.

50. Hester JM, Guin PR, Danek GD, et al. The economic and clini-cal impact of sustained use of a progressive mobility program in a neuro-ICU. Crit Care Med. 2017;45(6):1037-1044.

51. Messer A, Comer L, Forst S. Implementation of a progres-sive mobilization program in a medical-surgical intensive care unit. Crit Care Nurse. 2015;35(5):28-42.

52. Colquhoun HL, Squires JE, Kolehmainen N, Fraser C, Grim-shaw JM. Methods for designing interventions to change healthcare professionals’ behaviour: a systematic review. Implement Sci. 2017;12(1):30.

53. Hodgson CL, Bailey M, Bellomo R, et al; Trial of Early Activ-ity and Mobilization Study Investigators. A binational multi-center pilot feasibility randomized controlled trial of early goal-directed mobilization in the ICU. Crit Care Med. 2016; 44(6):1145-1152.

54. Tipping CJ, Bailey MJ, Bellomo R, et al. The ICU MobilityScale has construct and predictive validity and is respon-sive: a multicenter observational study. Ann Am Thorac Soc. 2016;13(6):887-893.

55. Portela MC, Pronovost PJ, Woodcock T, Carter P, Dixon-Woods M. How to study improvement interventions: a brief overview of possible study types. BMJ Qual Saf. 2015; 24(5):325-336.


TemporalScanner™

The Exergen TemporalScanner Temporal Artery Thermometer

More than 70 published studies - supporting accuracy from preemies to geriatrics in all areas of care.

• Patients love the TemporalScanner!

• Cost savings of 90% over other thermometry methods

• Lifetime Warranty – unique to thermometry

• Chemical resistant materials stand up to harsh disinfectants

• On-demand, innovative, inservicing results in successful usage for all levels of nursing skills

To evaluate, email: [email protected]

For general information: www.exergen.com

For clinical information, visit: www.TAThermometry.org

For educational videos, clinical studies, and manuals: www.exergen.com/ww

Exergen Corporation 400 Pleasant Street Watertown MA 02472 Tel 617-923-9900 Fax 617-923-9911

www.exergen.com/s

Invented, designed, assembled, tested, & packaged in the U.S.A. by Exergen



Background Mobilization is safe and associated with improved outcomes in critically ill adults, but little is known about mobilization of critically ill children. Objective To implement a standardized mobilization therapy protocol in a pediatric intensive care unit and improve mobilization of patients.Methods A goal-directed mobilization protocol was instituted as a quality improvement project in a 20-bed cardiac and medical-surgical pediatric intensive care unit within an academic tertiary care center. The mobilization goal was based on age and severity of illness. Data on severity of illness, ordered activity limitations, baseline functioning, mobilization level, complications of mobili-zation, and mobilization barriers were collected. Goal mobilization was defined as a ratio of mobilization level to severity of illness of 1 or greater.Results In 9 months, 567 patient encounters were ana-lyzed, 294 (52%) of which achieved goal mobilization. The mean ratio of mobilization level to severity of illness improved slightly but nonsignificantly. Encounters that met mobilization goals were in younger (P = .04) and more ill (P < .001) patients and were less likely to have barriers (P < .001) than encounters not meeting the goals. Compli-cation rate was 2.5%, with no difference between groups (P = .18). No serious adverse events occurred.Conclusions A multidisciplinary, multiprofessional, goal-directed mobilization protocol achieved goal mobilization in more than 50% of patients in this pediatric intensive care unit. Undermobilized patients were older, less ill, and more likely to have mobilization barriers at the patient and provider level. (American Journal of Critical Care. 2018; 27:194-203)

MOBILIZATION THERAPY IN THE PEDIATRIC

INTENSIVE CARE UNIT: A MULTIDISCIPLINARY QUALITY

IMPROVEMENT INITIATIVEBy Blair R. L. Colwell, MD, Cydni N. Williams, MD, Serena P. Kelly, CPNP-AC, and Laura M. Ibsen, MD



this article.

This article is followed by an AJCC Patient Care Page on page 204.


Mobilization therapy is safe and beneficial in critically ill adults and is associated with decreased length of stay, improved physical function at discharge, and reduced hospital costs,1,2 but little is known about mobilization therapies in the pediatric intensive care unit (PICU). The definition of mobilization therapy and the timing of mobilization vary widely in the literature.3-5 Mobilization

therapy in the intensive care unit (ICU) is rehabilitating physical activity that escalates from passive range of motion exercises to standing and walking.

A variety of adverse effects, including ICU-

acquired weakness, delirium, and poor quality of

life and functional outcomes, have been associated

with adult ICU admission6 and are supported in

limited pediatric studies.7-11 Functional impairment

may persist in children after PICU admission, with

up to one-third of children displaying impairment

upon discharge and up to 13% showing

impairment at 2-year follow-up.12 Studies in

adults have shown that mobilization therapy may

mitigate some of these long-term adverse effects of

ICU admission. A recent prospective quality

improvement (QI) study in a tertiary PICU by

Wieczorek et al13 demonstrated that implementa-

tion of a mobilization bundle in children admitted for

more than 3 days doubled the frequency of mobili-

zation events and was safe; the main barrier to

mobilization was a lack of appropriate equipment.

Another study showed that about half of children

admitted to a PICU in Canada received some type of

physical therapy.14 A survey of pediatric critical care

physicians and physical therapists in Canada

showed that most clinicians believe mobilization is

important but that a major barrier to early mobili-

zation is lack of established guidelines, in addi-

tion to safety concerns and lack of physician orders

for mobilization.4 After the implementation of a

nurse-led, progressive early mobilization protocol (in

conjunction with a delirium bundle and sedation

protocol) in a mixed PICU, the prevalence of delirium

dropped from 19% to 12%.15 Because of improvements

in outcomes in adult ICU patients who receive

mobilization therapy, we developed a mobilization

protocol designed to improve PICU patient care

through nurse-driven mobilization interventions.

We implemented a unique goal-directed multi-

disciplinary and multiprofessional mobilization

protocol for critically ill infants and children admit-

ted to the PICU. Our pro-

tocol defined a minimum

goal for mobility, with

adjusted goals based on

patients' severity of illness

(SOI) and age. We tailored

mobilization activities on

the basis of minimum

goals to be completed

primarily by the bedside

nurse. We tracked proto-

col adherence, adverse events, and perceived barriers

to goal mobilization. In this article, we report our

preliminary experience with a focus on safety and

identifying areas to target for continued improve-

ment in mobilizing critically ill children.

Methods Ethical Issues

We implemented this project as a QI protocol

that applied to all PICU patients and designed it to

improve patient care by standardization of mobility

practices without comparison to other interventions.

It was therefore exempt from institutional review

board approval. Dissemination of results was also

deemed exempt. We tracked adverse events closely

to ensure that no additional patient risk occurred as

a result of practice change.

SettingOur hospital is an urban, academic tertiary care

children’s hospital and a referral center for several

states. The PICU has 20 beds and is a combined

About the AuthorsBlair R. L. Colwell is a pediatric critical care physician at University of California Davis, Sacramento, California. Cydni N. Williams is an assistant professor of pediatrics at Oregon Health and Science University, Portland, Oregon. Serena P. Kelly is an assistant professor of pediatrics at Oregon Health and Science University. Laura M. Ibsen is a professor of pediatrics and anesthesiology at Oregon Health and Science University.

Corresponding author: Blair R. L. Colwell, MD, University of California Davis, Department of Pediatrics, Division of Critical Care, 2516 Stockton Blvd, Sacramento, CA 95817 (email: [email protected]).


A goal-directed multidis-ciplinary mobilization protocol for critically ill infants and children was implemented in a PICU.


cardiac and medical-surgical unit with a typical

nurse to patient ratio of 1 to 1 or 1 to 2. There are

approximately 1300 admissions and approximately

200 cardiac surgery cases per year. The PICU uses

nurse-driven sedation protocols for most intubated

patients, as published in the RESTORE (Randomized

Evaluation of Sedation Titration for Respiratory Fail-

ure) trial.16 Physical and occupational therapists are

available for consultation in our PICU 8 hours per

day (9 AM to 5 PM), 6 days per week. Before implemen-

tation of the goal-directed mobilization protocol,

no guidelines were in place for initiation of rehabili-

tation or physical or occupational therapy in the

PICU, and mobilization practices depended on care

providers and bedside nurses.

Planning the InterventionWe first conducted a survey of nursing staff in

the PICU, focusing on perceptions of and barriers

to mobilization. The most significant preimple-

mentation barriers were perception of physiologic

instability and lack of additional staff to help with

mobilization activities. We formed a multidisci-

plinary, multiprofessional group including bedside

nurses, a PICU nurse practitioner, nursing leaders,

PICU physicians, and physical, occupational, and

respiratory therapists, and we created a goal-directed

mobilization protocol. Goal minimum mobilization

was based on age and SOI (Table 1). The protocol

applied to all PICU patients upon admission, includ-

ing patients supported by vasoactive infusions, inva-

sive respiratory support, and extracorporeal membrane

oxygenation, but it did not supersede practitioner-

ordered activity limitations. To simplify the protocol,

we did not specifically address patients with primary

neurologic disease (eg, elevated intracranial pressure).

However, these patients have activity orders in place

to limit elevations in intracranial pressure. Addition-

ally, children with developmental delays are addressed

in the special considerations section of the protocol

(see Appendix).

The protocol was embedded into the electronic

medical record's admission order sets and was also

available as a stand-alone order, allowing the

Description

Age 0-12 months

1-4 years

>4 years

Mobilization level

No invasive respi-ratory or inotro-pic support

Add:Up to chair, seat,

or swing 2-3 times per day; floor time

Add:Sit in chair; stand

side of bed; ambulate 3 times per day

Add:Sit in chair; stand

side of bed; ambulate 3 times per day

4

PEEP < 10 cm H2O, FIO2 < 60%, stable dopamine < 10 μg/kg/min

Add:Parent hold in chair; active

positioning; up to chair, seat, or swing

Add:Sit on edge of bed, in

chair, or in pediatric positioning chair; twice daily ambulation; con-sider physical and/or occupational therapy consultation

Add:Sit on edge of bed, in

large pediatric position-ing chair, or cardiac chair; active assist with transfer; twice daily ambulation; consider physical and/or occupational therapy consultation

3

PEEP 10 cm H2O, dopamine 10 μg/kg/min, any epinephrine or norepinephrine

Add:Active range of motion; use of

bumper or nesting for supported flexion, supine, side-lying, and prone positioning; use of towel roll to support hip flexion

Add:Active range of motion; supine,

side-lying, and prone positioning; use of towel roll to support hip flexion; resting hand/foot splints if prolonged intubation

Add:Active range of motion; supine,

side-lying, and prone positioning; use of towel roll to support hip flexion; resting hand/foot splints if prolonged intubation

2

Vital sign lability requiring titration of interventions

Turn; gentle passive range of motion of all 4 extrem-ities and neck; alternate position of ventilator tubing

Turn; gentle passive range of motion of all 4 extrem-ities and neck; alternate position of ventilator tub-ing; float heels

Turn; gentle passive range of motion of all 4 extrem-ities and neck; alternate position of ventilator tub-ing; float heels

1

4: Most stable3: Stable2: Somewhat unstable1: Very unstable

Abbreviations: FIO2, fraction of inspired oxygen; PEEP, positive end-expiratory pressure. a For long-stay, chronically critically ill patients, consider more active exercise when still in acute phase of illness, as tolerated (see Appendix).

Severity of illness

Feature

Table 1Goal-directed mobilization protocola


Goal minimum mobiliza-tion was based on age and severity of illness for all PICU patients.

protocol to be initiated on the first hospital day

and continue throughout PICU hospitalization.

Mobilization activities could be initiated immedi-

ately after arrival in the PICU because most activi-

ties were structured to be performed by the bedside

nurse with or without additional support staff. A

laminated copy of the protocol was placed at each

patient’s bedside for review by staff as necessary.

An additional order for supplemental support from

a respiratory therapist could also be placed to allow

for extra support personnel when mobilizing intu-

bated patients.

We educated all PICU staff regarding the safety

and importance of mobilization therapy by presen-

tations on educational days, access to references,

and email updates. Attending physicians on service

in the PICU conducted biweekly rounds with bed-

side nurses to collect data on all patients and solicit

perceived barriers to mobilization (as reported by

bedside nurses). A multiprofessional team conducted

additional “mobilization rounds” on weekday

mornings. During these rounds, the charge nurse,

PICU nurse practitioner, and physical, occupational,

and respiratory therapists discussed each patient.

Discussion during mobilization rounds allowed

appropriate allocation of staff and facilitated earlier

mobilization by identifying patients who might

benefit from the assistance of additional support

staff, such as physical, occupational, or respiratory

therapists or a more experienced additional bed-

side nurse, to achieve mobilization goals.

Planning the Study of the InterventionWe chose protocol adherence as the primary

outcome because there is no gold standard for type

or timing of mobilization therapy in pediatric patients.

We quantified protocol adherence as a mobilization

ratio (level of mobilization to SOI) using the defini-

tions contained in Table 1. A ratio of 1 or greater

defined protocol adherence. Secondary outcome

measures included adverse events and barriers to

mobilization. We grouped perceived barriers to

mobilization by category. We defined serious adverse

events as desaturation requiring escalation of ther-

apy, unplanned extubation, removal of other medi-

cal equipment (eg, arterial catheter, central venous

catheter), and falls. We included all children admit-

ted to the PICU, and we excluded adult patients who

were boarding in the PICU solely for management

of extracorporeal membrane oxygenation because

these patients were primarily cared for by other

services. We ensured data accuracy by having only

PICU attending physicians collect data.

Methods of EvaluationData collection began in August 2014 and

included patients' age, SOI level from 1 to 4 (with

1 indicating the most ill patients), ordered activity

limitations, preadmission pediatric overall perfor-

mance category, mobilization level from 1 to 4 (see

Table 1), and complications of and barriers to mobi-

lization. Attending physicians collected data in real

time during biweekly rounds with bedside nurses.

We tracked the mean mobilization ratio every 2

weeks from all included encounters (defined as

admissions during time points at which data were

collected) and reviewed progress at monthly QI

meetings. We also reviewed adverse events, the data

collection process, and barriers to mobilization at

monthly meetings. We distributed summaries from

monthly meetings to all PICU staff to relay the cur-

rent status of the mobilization project and to pro-

mote continued improvement.

AnalysisThe first 3 months of data defined the implemen-

tation period. We used standard statistical process

control charts to plot mean mobilization ratio over

time and the proportion of patients not adherent to

the protocol (mobilization ratio < 1) over time to

account for skew related to

mobilization ratios greater

than 1 when calculating

means. Three limits were

used to set the upper and

lower control limits. Some

patients were present during

multiple data collection

encounters, but because the

primary outcome was adherence to the mobiliza-

tion protocol within the PICU, we analyzed data by

encounter and not by individual patient. Continuous

variables were reported as medians with interquartile

ranges; categorical variables, as counts with percent-

ages. We used the Mann-Whitney U (continuous)

and 2 (categorical) tests to compare variables

between encounters achieving goal mobilization

(mobilization ratio ≥1) and encounters not achiev-

ing goal mobilization (mobilization ratio <1). We

analyzed the data by using statistics software (SPSS

version 22, IBM, and Minitab 16.2.4, Minitab Inc).

Results During the first 9 months, we analyzed 567

encounters. We designated the initial 3 months as

the implementation period, during which the roll-

out of the protocol occurred with extensive staff


education. We designated the following 6 months

as the postintervention period. The implementation

and postintervention period populations are com-

pared in Table 2. Patients in the implementation

period were less ill (SOI, 3-4) than patients in the

postintervention period (P = .009) but were otherwise

similar. Notably, 18% of patients in the postinter-

vention period were classified as SOI 1 to 2 (intu-

bated with positive end-expiratory pressure ≥ 10 cm

H2O and/or receiving moderate amounts of vasoac-

tive infusions).

Of the 567 encounters, 294 (52%) achieved

the goal mobilization ratio (level of mobilization

to SOI ratio of ≥ 1). Figure 1 displays the study’s

mean mobilization ratio over time. The mean mobi-

lization ratio was slightly higher in the postinterven-

tion period (0.86) than in the implementation period

(0.84), but this difference was not statistically signif-

icant. The proportion of encounters meeting mobili-

zation goals remained stable over time (Figure 2).

In the postintervention period, some patients were

able to mobilize well beyond their minimum goals,

skewing some data points toward a higher mean

mobilization ratio. For instance, some patients ven-

tilated on high positive end-expiratory pressure

(> 10 cm H2O) were able to walk. This variability

accounts for the small increase in the postintervention

mean mobilization ratio among a stable proportion

of patients meeting mobilization goals.

Comparing goal-mobilized versus undermobi-

lized encounters revealed improved adherence to

mobilization goals in patients who were younger

(P = .04), more ill (P < .001), and less likely to have

barriers (P < .001) than in patients not meeting mobi-

lization goals, as displayed in Table 3. The compli-

cation rate was 2.5% (n = 14), with no difference

between groups (P = .18, Table 3). No complications

prevented mobility activities. Transient desatura-

tion was the most common complication (Table 4).

Characteristic of patients

Table 2Implementation versus postintervention period

Age, median (IQR), y

Pediatric overall performance category, No. (%)

Normal/mild disability Moderate disability or worse

Severity of illness, No. (%) Most ill (level 1-2) Less ill (level 3-4)

.94

.35

.009

1.3 (0.3-9.4)

291 (67)143 (33)

77 (18)357 (82)

2.5 (0.4-9.1)

83 (62)50 (38)

11 (8)122 (92)

P valuePostintervention period (n = 434)

Implementation period (n = 133)

Abbreviation: IQR, interquartile range.

Figure 1 Mean mobilization ratios (level of mobilization to severity of illness) of all encounters during the 3-month implementation period and the 6-month postintervention period. A mean value of 1 or greater on the y-axis describes a patient meeting the minimum mobilization goal. The mean mobilization ratio over time was 0.86 (represented by the horizontal green line). Data points represent 2-week intervals. The process remained in control during the study period.

Abbreviations: LCL, lower control limit; UCL, upper control limit.

Implementation

Implementation

1.0

0.9

0.8

0.7

0.6

0.20

0.15

0.10

0.00

0.05

1

1

4

4

7

7

10

10

13

13

16

16

19

19

22

22

25

25

28

28

LCL = 0.7279

LCL = 0

UCL = 1.0016

UCL = 0.1681

Mean = 0.8648

Mean = 0.0514

Ind

ivid

ual

val

ue

Mo

vin

g r

ang

e

Observation number

Observation number

Postintervention

Postintervention


No serious adverse events, such as hemodynamic

instability or unintended extubation, occurred. Bed-

side nursing staff identified barriers in 104 encoun-

ters not achieving goal mobilization. These barriers

are displayed in Table 5.

Discussion We implemented a unique multidisciplinary

and multiprofessional goal-directed mobilization

protocol based on SOI and age in our PICU as a

QI initiative, achieving goal mobilization in more

than 50% of encounters. The mean mobilization

ratio improved over time as staff seemed to become

more comfortable mobilizing the most critically ill

patients, although this improvement was not statis-

tically significant. However, we continue to have a

proportion of patients, mostly older and less ill,

who do not achieve the minimum mobilization

goals, most likely because of the increasing mobili-

zation requirements needed to achieve goal levels

in these children. The protocol was safe during this

period of preliminary implementation, with no seri-

ous adverse events even in the most severely ill

patients. We identified potentially modifiable bar-

riers to mobilization, including parent and patient

refusal. In addition, we found that staff sometimes

perceived patients to be too physiologically unsta-

ble for mobilization and that patients' families per-

ceived medical equipment to be a barrier. This study

was a collaborative effort among multidisciplinary

Figure 2 Proportion of encounters not achieving minimum mobility goals (on y-axis) during the 3-month implementa-tion period and the 6-month postintervention period. The mean proportion of patients not achieving the minimum mobility goal was 0.52 (represented by the horizontal green line). Data points represent 2-week intervals. The process remained in control during the study period.

Abbreviations: LCL, lower control limit; UCL, upper control limit.

Implementation

1 4 7 10 13 16 19 22 25 28

LCL = 0.358

UCL = 0.691

Mean = 0.525

Observation number

Postintervention

1.0

0.6

0.8

0.4

0.2

0.0

Pro

po

rtio

n

Characteristic of patients

Table 3Goal-mobilized versus undermobilized encounters

Age, median (IQR), y

Pediatric overall performance category, No. (%)

Normal/mild disability Moderate disability or worse

Severity of illness, No. (%) Most ill (level 1-2) Less ill (level 3-4)

Complications, No. (%)

Perceived barriers, No. (%)

.04

.86

< .001

.18

< .001

2.6 (0.5-9.1)

179 (66) 94 (34)

19 (7)254 (93)

4 (1)

139 (51)

1.1 (0.4-9.6)

195 (66) 99 (34)

69 (23)225 (77)

10 (3)

85 (29)

P valueUndermobilized

(n = 273)Goal-mobilizeda

(n = 294)

Abbreviation: IQR, interquartile range.a Mobilization ratio 1.


Goal mobilization was achieved in

52% of encounters.

providers in the PICU and resulted in a novel

goal-directed mobilization protocol.

PICU mobilization protocols are innovative;

however, few studies on this topic are available.

A recent report showed that implementation of a

mobilization bundle for children in a tertiary PICU

led to a significant increase in mobilization events,

although the study included only 100 postimple-

mentation patients admitted to the PICU for more

than 3 days.13 Our study was larger and included all

admissions but similarly showed improvement in

mobilization activities and no serious adverse events.

Choong et al14 noted that early mobilization within

48 hours of admission occurred in only 9.5% of

patients and was more common

in older patients. In contrast, our

results demonstrate a goal mobiliza-

tion of 52% of patients and easier

mobilization of younger children,

possibly because of our specific

age-based protocol goals and a

primarily nurse-driven protocol. Choong et al did

not use a mobilization protocol, which most likely

accounts for the higher proportion of patients

mobilized in our study. We chose to describe our

protocol as a goal-directed protocol rather than as

an early mobilization protocol, although the dis-

tinction is unclear in the literature, and timing of

mobilization activities in the 2 studies may account

for some differences.

The incidence of adverse events during early

mobilization in adult patients varies from 0% to

4% of all mobilization sessions.3,17-20 Studies in

children also support a very low rate of adverse

events,13,21,22 consistent with the 2.5% rate of adverse

events in our study. We were able to track adverse

events in a prospective manner, which most likely

allowed us to accurately estimate the number of

adverse events because mild events may be missed

during retrospective chart review. The most common

complications in our study were transient desaturation

and tachypnea (93% of all adverse events), similar

to studies of adults in which approximately half of

all adverse events were respiratory complications.23

Studies in adults have shown that barriers to

mobilization therapy are multifaceted and include

unit culture, perceived lack of resources, and factors

intrinsic to patients.24 Pediatric barriers to mobiliza-

tion therapy may include institutional barriers, such

as the absence of practice guidelines or an order for

mobilization, in addition to unit culture, safety, and

resource concerns.4,13,22 Our data indicate that disease

severity (or the perception thereof) represents a sub-

stantial barrier to mobilization, as this accounted

for almost half of identified barriers in our study.

In addition, a substantial number of patients or

their families refused to participate, providing a new

target to improve mobilization in our unit. Interest-

ingly, nurse refusal was not identified as a barrier in

our study, in contrast to a study by Cui et al22 in

which nurses deferred physical or occupational ther-

apy sessions in 50% of cases. This finding suggests

that the mobilization protocol was well received

by the bedside nursing staff in our study. Medical

equipment barriers, such as access to wheelchairs at

home, were also identified as potentially modifiable

barriers to mobilization. Notably, lack of an order to

proceed with mobilization and lack of practice

guidelines were not barriers in our study because

strategies to overcome institutional barriers were

built into this QI project. Because we collected our

data in real time, our study most likely reveals more

barriers to mobilization than did the study by

Wieczorek et al,13 which identified barriers only

through a retrospective chart review.

Our study has several limitations, including

implementation in a single setting and lack of

assessment of outcome measures because the pri-

mary goals were protocol adherence and safety

monitoring. The preliminary data are underpow-

ered to detect changes in important outcomes such

as incidence of delirium, duration of mechanical

respiratory support, and length of stay. We adjusted

the mobilization protocol on the basis of SOI pri-

marily according to patients’ respiratory and cardio-

vascular requirements by administering invasive

Complication

Table 4Complications among all encounters during mobilization therapy

Transient desaturation

Tachypnea

Emesis

11 (79)

2 (14)

1 (7)

No. (%) of 14 complications

Barrier

Table 5Barriers among patients not achieving minimum mobilization goals

Diagnosis or severity-of-illness concerns

Medical equipment–related concerns

Parent’s refusal

Patient’s refusal

Timing of admission

48 (47)

31 (30)

10 (10)

3 (3)

11 (11)

No. (%) of 103 encounters


mechanical ventilation and vasoactive infusions as

needed. However, the protocol did not specifically

address patients with primary neurological disease.

Although the protocol applied to all patients with

neurological disease, orders for management of ele-

vated intracranial pressure and spinal precautions

superseded the protocol. The omission of neuro-

logical criteria was intended to simplify the proto-

col and is a limitation of the study. Patients with

primary neurological disease can be challenging to

mobilize, which could have contributed to the

inability of some patients to adhere to the mobiliza-

tion protocol and meet mobilization goals, thereby

decreasing the mean mobilization ratio. We plan to

modify the protocol for this population of patients

by including active management of elevated intra-

cranial pressure in SOI category 1.

We chose to analyze data by the mean mobili-

zation ratio instead of analyzing the number of

mobilization activities per patient to set a minimum

mobilization goal for our patients. By choosing this

strategy, we aimed to have each patient meet a spe-

cific mobilization goal and to identify barriers to

early mobilization at the individual patient level.

We analyzed data by encounter, not by individual

patient. Therefore, some patients were counted mul-

tiple times in the data set, potentially causing P val-

ues to be falsely elevated or skewing the mean

mobilization ratio over time.

As this QI project continues, we are targeting

patients who are older and less critically ill for

improved adherence to the mobilization protocol.

We are also making efforts to ensure the availabil-

ity of equipment needed to mobilize patients,

such as encouraging parents to bring wheelchairs

from home upon hospital admission. We are cur-

rently surveying families to better understand

their response to mobilization, hoping to learn

how we can decrease parents' and patients' refusal

of mobilization therapies. In the future, we intend

to collect data on outcome measures such as length

of stay, duration of mechanical ventilation, and

incidence of delirium.

Conclusions Implementation of a novel multidisciplinary

and multiprofessional goal-directed mobilization

protocol adjusted for patients' age and SOI as a QI

initiative has improved patient mobility in children

admitted to our PICU. We have observed no serious

adverse events related to our protocol. During this

preliminary period, we identified barriers at both

the patient and staff levels to target for future

improvements, and we seek to improve protocol

adherence with ongoing educational interventions.

Appendix: Special Considerations of Goal-Directed Mobilization ProtocolFractures

Pelvic fractures • Range of motion as tolerated to 90° of hip flexionAcetabular fractures • Range of motion through functional mobility, no

repetitive range of motionIntramedullary nailing of lower extremity fractures • Range of motion as toleratedUpper extremity fractures • Require orthopedic clarification for range of motion

restrictions

BracesCervical collar • Follow cervical spine precautions Cervical thoracic orthosis • Follow cervical and thoracic spine precautionsThoracolumbosacral orthosis • Follow lumbar spine precautions Knee immobilizer • Ankle and hip range of motion only Hinged knee brace • Range of motion within locked parameters of brace

Spines Not ClearedCollar or braces must be worn at all times unless

otherwise ordered.Cervical spine precautions • Logroll to sitting position: Do not let them sit

straight up • Avoid neck rotation • No shoulder range of motion greater than 90° • No lifting elbow above shoulders • No lifting, pushing, pulling, or carrying more

than 5 pounds (2.25 kg) • Do not use walker or crutches to assist with

ambulation or transfersThoracic spine precautions • Logroll: No twisting or bending • No lifting, pushing, pulling, or carrying more

than 5 pounds (2.25 kg) • No head-of-bed elevation greater than 30°

(unless brace is on or otherwise indicated) • If upper thoracic injury, avoid repetitive overhead

activitiesLumbar spine precautions • Logroll: No twisting or bending • No excessive hip flexion

Spines ClearedAll activities acceptable unless otherwise specifically

orderedSpinal Fusion

Logroll, side-lie, sitNo lifting, pushing, pulling, or carrying more than 5

pounds (2.25 kg)No twisting or bendingAvoid lengthy overhead activity

CraniotomyNo head below heartAvoid anything that will increase intracranial pressure,

for example


• Stifling a sneeze • Plugging nose to “pop” ears • Holding breath • Lifting anything heavier than 10 pounds (4.5 kg)

Bone Flap OutHelmet must be worn with • Edge-of-bed or out-of-bed activity • Head-of-bed elevation greater than 30° • During transportNo head below heart No lifting, pushing, pulling, or carrying more than 5

pounds (2.25 kg)Sternal Precautions

Logroll, side-lie, sitNo lifting, pushing, pulling, or carrying more than 5

pounds (2.25 kg)No pushing up with arms to get out of bed/chairNo twisting or bendingNo shoulder range of motion greater than 90°No lifting from under shouldersScoop patient

Cardiac PrecautionsWarm up and cool down: • For at least 2 minutes • Before and after any change of position • Ankle pumps/hand squeezesNo head below heart No bending down

Abdominal Surgery PrecautionsLogroll, side-lie, sitNo lifting, pushing, pulling, or carrying more than 5

pounds (2.25 kg)Avoid Valsalva maneuver and bearing down

Developmental DelayHave parents bring patient’s personal equipment, for

example, wheelchair, chair, etcIncorporate home occupational/physical therapy

regimen if possibleChronically Critically Ill Patients (Length of Stay > 7 Days)

Consider more active exercise program to avoid critical illness myopathy

Use sedation/muscle relaxation holds to encourage activity as tolerated

Exercise upper and/or lower extremities as catheters,

wires, tubes, etc allow

ACKNOWLEDGMENTSWe sincerely thank all the physicians, nurses, and respir a-tory, physical, and occupational therapists for their help.

FINANCIAL DISCLOSURESDr Williams is supported by grant number K12HS022981 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

SEE ALSO For more about mobilization of patients, visit the Critical Care Nurse website, www.ccnonline.org, and read the article by Messer et al, “Implementation of a Progressive Mobilization Program in a Medical-Surgical Intensive Care Unit” (October 2015).

REFERENCES

1. Adler J, Malone D. Early mobilization in the intensive care unit: a systematic review. Cardiopulm Phys Ther J. 2012; 23(1):5-13.

2. Cameron S, Ball I, Cepinskas G, et al. Early mobilization in the critical care unit: a review of adult and pediatric literature. J Crit Care. 2015;30(4):664-672.

3. Bailey P, Thomsen GE, Spuhler VJ, et al. Early activity is feasible and safe in respiratory failure patients. Crit Care Med. 2007;35(1):139-145.

4. Choong K, Koo KK, Clark H, et al. Early mobilization in critically ill children: a survey of Canadian practice. Crit Care Med. 2013;41(7):1745-1753.

5. Kayambu G, Boots R, Paratz J. Physical therapy for the critically ill in the ICU: a systematic review and meta-analysis. Crit Care Med. 2013;41(6):1543-1554.

6. Herridge MS, Cheung AM, Tansey CM, et al; Canadian Critical Care Trials Group. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med. 2003;3 48(8):683-693.

7. Choong K, Al-Harbi S, Siu K, et al; Canadian Critical Care Trials Group. Functional recovery following critical illness in children: the “wee-cover” pilot study. Pediatr Crit Care Med. 2015;16(4):310-318.

8. Daoud A, Duff JP, Joffe AR; Alberta Sepsis Network. Diagnostic accuracy of delirium diagnosis in pediatric intensive care: a systematic review. Crit Care. 2014;18(5):489.

9. Field-Ridley A, Dharmar M, Steinhorn D, McDonald C, Marcin JP. ICU-acquired weakness is associated with differences in clinical outcomes in critically ill children. Pediatr Crit Care Med. 2016;17(1):53-57.

10. Silver G, Traube C, Kearney J, et al. Detecting pediatric delirium: development of a rapid observational assessment tool. Intensive Care Med. 2012;38(6):1025-1031.

11. Wieczorek B, Burke C, Al-Harbi A, Kudchadkar SR. Early mobilization in the pediatric intensive care unit: a systematic review. J Pediatr Intensive Care. 2015;2015:129-170.

12. Ong C, Lee JH, Leow MK, Puthucheary ZA. Functional outcomes and physical impairments in pediatric critical care survivors: a scoping review. Pediatr Crit Care Med. 2016;17(5):e247-259.

13. Wieczorek B, Ascenzi J, Kim Y, et al. PICU up!: impact of a quality improvement intervention to promote early mobi-lization in critically ill children. Pediatr Crit Care Med. 2016; 17(12):e559-566.

14. Choong K, Foster G, Fraser DD, et al; Canadian Critical Care Trials Group. Acute rehabilitation practices in critically ill children: a multicenter study. Pediatr Crit Care Med. 2014; 15(6):e270-279.

15. Simone S, Edwards S, Lardieri A, et al. Implementation of an ICU bundle: an interprofessional quality improvement project to enhance delirium management and monitor delirium prevalence in a single PICU. Pediatr Crit Care Med. 2017;18(6):531-540.

16. Curley MA, Wypij D, Watson RS, et al; RESTORE Study Investigators and the Pediatric Acute Lung Injury and Sep-sis Investigators Network. Protocolized sedation vs usual care in pediatric patients mechanically ventilated for acute respiratory failure: a randomized clinical trial. JAMA. 2015;313(4):379-389.

17. Bourdin G, Barbier J, Burle JF, et al. The feasibility of early physical activity in intensive care unit patients: a prospec-tive observational one-center study. Respir Care. 2010; 55(4):400-407.

18. Sricharoenchai T, Parker AM, Zanni JM, Nelliot A, Dinglas VD, Needham DM. Safety of physical therapy interventions in critically ill patients: a single-center prospective evalua-tion of 1110 intensive care unit admissions. J Crit Care. 2014; 29(3):395-400.

19. Burtin C, Clerckx B, Robbeets C, et al. Early exercise in criti-cally ill patients enhances short-term functional recovery. Crit Care Med. 2009;37(9):2499-2505.

20. Schweickert WD, Pohlman MC, Pohlman AS, et al. Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial. Lancet. 2009;373(9678):1874-1882.

21. Abdulsatar F, Walker RG, Timmons BW, Choong K. “Wii-hab” in critically ill children: a pilot trial. J Pediatr Rehabil Med. 2013;6(4):193-204.

22. Cui LR, LaPorte M, Civitello M, et al. Physical and occupational therapy utilization in a pediatric intensive care unit. J Crit Care. 2017;40:15-20.

23. Nydahl P, Ewers A, Brodda D. Complications related to early mobilization of mechanically ventilated patients on intensive care units [published online November 7, 2014]. Nurs Crit Care. 2016;21(6):323-333. doi:10.1111/nicc.12134.

1.0 Hour Category AC E Notice to CE enrollees:



1. Collaborate to implement a multiprofessional and multidisciplinary quality improvement project.

2. Design a protocol for early mobilization of PICU patients.

3. Identify barriers to early mobilization in critically ill pediatric patients.




24. Dubb R, Nydahl P, Hermes C, et al. Barriers and strategies for early mobilization of patients in intensive care units. Ann Am Thorac Soc. 2016;13(5):724-730.


THE NEXT STEP IN PROVIDING EXCELLENT CARE to your critical care patients in any practice settingNew to the 2nd Edition• Updated information throughout reflects the latest evidence-based knowledgeplus national and international treatment guidelines• Streamlined content places a greater focus on need-to-know information for nurses in today’s high acuity, progressive, and critical care settings• Expanded coverage of the most topical emerging and infectious diseases including Zika virus and multidrug-resistant infections• Additional content on alternative settings including the eICU and remotemonitoringThe book is organized by body systems with synthesis chapters addressing patient conditions involving multiple body systemsEndorsed by the American Association of Critical-Care Nurses, this comprehensive, nursing-focused text underscores clinical reasoning skills as it helpsreaders comprehend, analyze, synthesize, and apply advanced critical care knowledgeand conceptsAdvanced Critical Care Nursing, 2nd edition, is a must-have resource dedicated to helping you oversee or care for critical care patients in any practice setting

Edited by Vicki S. Good and Peggy L. KirkwoodMarch 2017, 912 pagesISBN: 978-1455758753 VISIT aacn.org/store to get your copy today!

NEW EDITION!

NO TIME FOR EARLY MOBILITY?By Cindy Cain, RN, DNP, CNS, CCRN

AJCC Patient Care Page The AJCC Patient Care Page is a service of the American Journal of Critical Care and the American Association of Critical-Care Nurses.

Designed to elaborate on AACN practice guidelines based on content in select articles, this page may be photocopied noncommer-

cially for use by readers in their workplace, in continuing education programs, or for distribution to colleagues, patients, or

patients’ families. To purchase bulk reprints, call (800) 899-1712.

Based on material from and published as a supplement to the article by Young et al, “Identifying Barriers

to Nurse-Facilitated Patient Mobility in the Intensive Care Unit” (American Journal of Critical Care.

2018;27:186-193) and the article by Colwell et al, “Mobilization Therapy in the Pediatric Intensive

Care Unit: A Multidisciplinary Quality Improvement Initiative” (American Journal of Critical Care.

2018;27:194-203).

We know that early mobility has many

benefits. Nurse-facilitated early mobility

reduces incidence of delirium, improves

muscle strength, increases independent

functional status after discharge, and

improves overall quality of life.1 Early mobility also decreases

health care costs with fewer days of mechanical ventilation

and shorter stays in the intensive care unit.1 The question

then becomes, how do I fit one more task into my shift?

That is the question that Young et al and Colwell et al

seek to answer in their respective studies “Identifying Barri-

ers to Nurse-Facilitated Patient Mobility in the Intensive Care

Unit” and “Mobilization Therapy in the Pediatric Intensive

Care Unit: A Multidisciplinary Quality Improvement Initia-

tive.” Although early mobility is an interprofessional respon-

sibility, nurses have the unique distinction of being present

with the patient more than any other member of the health

care team. Thus nurses have the opportunity to assist patients

with passive range of motion exercises, dangling legs at the

bedside, or other types of early mobility at times that are

convenient to both the patient and the nurse.

Here’s what you can do:• Use the minimum required sedation for your patients

receiving mechanical ventilation.

• Consider using an early progressive mobility tool to

determine when it is safe for a patient to participate in early

mobility and the type of early mobility that is appropriate

for your patient.

• Look for opportunities to incorporate patients’ mobil-

ity activities such as passive range of motion exercises, dan-

gling legs at the bedside, or sitting/standing at the bedside

during times of routine nursing care.

• Assess your patient daily for readiness to mobilize or

advancement to the next level of a progressive mobility plan.

Other helpful resources:• View the AACN webinar “Executing Evidence-Based

Progressive Mobility in the ICU” and download the accom-

panying Tools and Tactics resources, https://www.aacn.org

/education/webinar-series/wb0007/executing-evidencebased

-progressive-mobility-in-the-icu. Share the tools with your team.

• Review CSI Mobility Projects: “Walk This Way: Early

Progressive Mobility in the ICU,” “Don't Be a Bedhead: Help

Your Head, Get Out of Bed,” and “Keep the Beat, Move Your

Feet” at https://www.aacn.org/nursing-excellence/csi-academy

/projects?page=1&topic=Pulmonary.

• Review early mobility and exercise information from

the ICU Delirium and Cognitive Impairment Study Group

at http://www.icudelirium.org/earlymobility.html.

• Review the Society of Critical Care Medicine’s ICU

Liberation web page on early mobility at http://www

.iculiberation.org /Mobility/Pages/early-mobility.aspx.

REFERENCE1. Parker A, Needham DM. The importance of early rehabilitation and

mobility in the ICU. Crit Connect. Society of Critical Care Medicine web-site. Published August 4, 2013. http://www.sccm.org/Communications /Critical-Connections/Archives/Pages/Importance-Early-Rehabilitation -Mobility-ICU.aspx. Accessed February 26, 2018.



Illustration by Steve Oh

Critical Care Management


Background Early red blood cell transfusions are a common treatment for adults hospitalized for sepsis without shock. However, their utility and association with mortality and costs have not been well studied.Objectives To examine early transfusion rates for patients with sepsis treated outside intensive care units, and to find a correlation between transfusion rates and survival rates and costs. Methods Data were obtained from hospital members of the Premier Healthcare Alliance that admitted at least 50 adults with sepsis between January 1, 2006, and December 31, 2010. Early transfusion rates at each hos-pital were calculated as the observed incidence of allo-geneic red blood cells administered by hospital day 2. A multivariable logistic regression model was constructed to estimate the expected or risk-adjusted transfusion rates, mortality rates, and costs.Results A total of 256 396 adults were hospitalized with sepsis without major bleeding or surgery at 364 US hospitals. Approximately 84% of all patients admitted with sepsis, without vasopressor therapy, were treated outside the intensive care unit (by day 2). The mean insti-tutional early transfusion rate was 6.9%. After risk stan-dardization, the median (interquartile range) transfusion rate was 6.7% (5.8%-7.6%), mortality rate was 15.5% (13.1%-18.1%), and costs were $13 333 ($11 939-$14 986). Early transfusion rates were not correlated with mortal-ity but were modestly positively correlated with costs (Spearman = 0.157; P = .003). Conclusions Early transfusion rates during hospitaliza-tion for sepsis without shock varied widely across the hospitals. Transfusion rates were associated with increased costs but not with mortality rates. (American Journal of Critical Care. 2018; 27:205-211)

EARLY BLOOD TRANSFUSIONS IN SEPSIS:UNCHANGED SURVIVAL AND INCREASED COSTSBy Karthik Raghunathan, MD, MPH, Mandeep Singh, MD, Brian H. Nathanson, PhD, DSHS, Elliott Bennett-Guerrero, MD, and Peter K. Lindenauer, MD, MSc


Between 1997 and 2011, the rate of hospitalization for sepsis increased more rapidly than that for any other condition in adults older than 45 years in the United States.1,2 During such hospitalization, allogeneic red blood cell (RBC) transfusions are often ordered by internists in the general medicine service.3,4 Although the Surviving Sep-sis Campaign recommends a generally restrictive approach to RBC transfusions, it

also endorses early goal-directed therapy for patients with tissue hypoperfusion, including a hemoglobin level lower than 10 g/dL (to convert hemoglobin to g/L, multiply by 10.0).5,6

Efforts to reduce medical resource

waste are increas-ingly relevant.


The benefit of such transfusions in septic shock

with central venous desaturation was recently called

into question by 4 large randomized controlled tri-

als.7-10 However, unlike septic shock, sepsis is treated

mostly outside of intensive care units (ICUs) by

internists.2 The utility of RBC transfusion in patients

with sepsis and the association of such transfusion

with survival and costs remain unknown. Efforts to

reduce medical resource waste (eg, the Choosing

Wisely campaign) may be especially

relevant because of the large num-

ber of adults with sepsis admitted

to US hospitals each year.11 There-

fore, in this study, we aimed to eval-

uate the institutional variations in

transfusion practices and their asso-

ciation with the institutional varia-

tions in mortality and costs. We hypothesized that

early RBC transfusions are a common treatment for

sepsis without shock and do not improve survival.

Methods After obtaining approval for this study from

the institutional review board at Baystate Medical

Center, we collected retrospective data on a cohort

of adult patients with sepsis admitted between Janu-

ary 1, 2006, and December 31, 2010, to hospitals

that are members of the Premier Healthcare Alliance.

Most hospital members of Premier are small (< 200

beds) to medium-sized (200-399 beds) institutions

in urban locations and without teaching affiliations.

The details of this voluntary, fee-supported, admin-

istrative and financial data set are described in previous

studies on sepsis and other hospital-based condi-

tions.12,13 In addition to the International Classifica-

tion of Diseases, Ninth Revision, Clinical Modification

(ICD-9-CM) diagnosis and procedure codes, item-

ized charges for diagnostic and therapeutic interven-

tions (eg, laboratory, radiology, and microbiology

tests; transfusions; mechanical ventilation; dialysis;

medications) as well as for other billable services for

each patient at each hospital are included in the data

set in a standardized format for each hospital day.

Study PopulationOur study focused on early RBC transfusions,

defined as the administration of allogeneic RBCs on

the day of hospital admission (day 1) or on the next

day (day 2), because such therapy was likely to be

associated with the admitting diagnosis. The patients

in our study population had a principal diagnosis

of sepsis14 (with ICD-9-CM codes including 038 and

995.91; see Supplement 1, available online only at

www.ajcconline.org, for a list of the ICD-9-CM codes

included) and had charges for blood cultures and

intravenous antibiotics initiated by day 2 and con-

tinued for at least 3 consecutive days. We excluded

patients who were discharged alive or died by day

2. To focus on transfusions for sepsis, we excluded

patients with charges for gastrointestinal bleeding

therapy (eg, colonoscopy, endoscopy, infusions of

proton pump inhibitors of histamine-2 receptor

blockers) or with a diagnosis or treatment for gas-

trointestinal hemorrhage (ICD-9-CM code 578) at

any time during hospitalization as well as patients

who had any all-patient refined diagnosis-related

group (APR-DRG) codes signifying major surgical

About the AuthorsKarthik Raghunathan is an associate professor, Department of Anesthesiology, Duke University Medical Center and Durham Veterans Affairs Medical Center, Durham, North Carolina. Mandeep Singh is an assistant professor, Depart-ment of Anesthesiology, University of Southern Califor-nia, Los Angeles, California. Brian H. Nathanson is chief executive officer, OptiStatim, LLC, Longmeadow, Massa-chusetts. Elliott Bennett-Guerrero is a professor, Depart-ment of Anesthesiology, Stony Brook University Medical Center, Stony Brook, New York. Peter K. Lindenauer is director, Institute for Healthcare Delivery and Population Science, and professor of medicine, University of Mas-sachusetts–Baystate; professor of quantitative health sciences, University of Massachusetts Medical School, Worcester; and an adjunct professor, Tufts University School of Medicine, and Tufts Clinical and Translational Science Institute, Boston, Massachusetts.

Corresponding author: Karthik Raghunathan, MD, MPH, Department of Anesthesiology, Duke University Medical Center/Durham VAMC, DUMC 3094, Durham, NC 27710 (email: [email protected]).


A multivariable logistic regression model predicting early RBC transfusions on the basis of patient charac-teristics was constructed and applied at each hospital.

procedures (APR-DRG codes 005, 444, and 710).

In addition, we excluded patients who were likely

to have ongoing blood loss (ie, those with charges

for more than 3 units of RBCs on any single hospital

day, for blood products other than RBCs at any time

during hospitalization, or for more than 2 labora-

tory tests for hematocrit or hemoglobin levels on

any single hospital day). We also excluded patients

who were transferred between hospitals, as they were

not tracked within Premier and thus might have had

missing early exposure and late outcomes data. See

Supplement 1 (available online only) for the list of

inclusion and exclusion criteria.

Exposures, Outcomes, and Analyses To avoid bias inherent in patient-level analyses,

we conducted all analyses at the hospital level.15 At

each hospital, exposure was defined as the observed

rate of early transfusions or the proportion of patients

who received RBC transfusions by day 2, whereas the

expected rate was the proportion of patients predicted

to receive RBC transfusions by day 2. Mortality out-

come was defined as the proportion of patients who

died in the hospital after day 2. Costs were defined

as the mean of total charges for each patient.

To estimate the expected rate of early transfu-

sions, we constructed a multivariable logistic regres-

sion model that predicted early transfusions solely

on the basis of patient characteristics, and then

we applied the model at each hospital. This model

included the following predictors: patient demograph-

ics; chronic comorbidities, grouped using Healthcare

Cost and Utilization Project software (Agency for

Healthcare Research and Quality [AHRQ]); and

interventions received by day 2. Using date-stamped

charge codes (see Supplement 2, available online

only), we identified the early interventions admin-

istered such as intravenous fluid, vasopressor drug,

inotropic agent, central venous catheterization,

mechanical ventilation, and dialysis. Because large

volumes of crystalloid solution may result in

hemodilution and consequent RBC transfusion,

we accounted for the proportion of patients who

had received 5 L or more of crystalloid solution

by day 2.16 We ensured the accuracy of the model

by using the area under the receiver operating char-

acteristic curve, Hosmer-Lemeshow statistic, and

adjusted Brier score17 (see Supplement 3, available

online only).

We calculated risk-standardized rates at each

institution using ratios of observed and expected

transfusion rates, and we examined institutional

variations in transfusion practices (independent of

the underlying differences in patient populations)

by comparing these

rates. Using similar

methods, we com-

puted standardized

mortality ratios and

adjusted cost ratios at

the institutional level.

Then, we correlated

risk-standardized early

transfusion rates with

the adjusted mortality

rates and costs using Spearman rank correlations.

We repeated all analyses to conduct sensitivity analy-

ses for our subset of patients with septic shock.

This subset was defined by the patients’ admission

to an intensive care unit (ICU) and receipt of vaso-

pressors by hospital day 2 (see Supplements 1 and

4, available online only).

Results The study cohort consisted of 256 396 adults

with sepsis admitted to 364 US hospitals (Figure 1)

Figure 1 The study cohort.

n = 36 235 did NOT receive any

red blood cells

n = 4840received at least 1

unit of red blood cells

n = 238 581 did NOT receive any red blood

cells early

n = 17 815 received red blood cells early

Transfusion by day 2 Transfusion by day 2No transfusion by day 2 No transfusion by day 2

Excluded patients with any possibility of active bleeding, any major surgery, incomplete data or admission to low-volume hospitals (< 50 cases)

(n = 320 437)

Adults with valid sepsis codes, charges for intravenous antibiotics and blood cultures by day 2, and

a known mortality outcome (n = 576 833)

Study cohort of adults with sepsis

(n = 256 396 across 364 hospitals)

Septic shock subsetPatients who were admitted to an intensive care unit and receiving

vasopressors by day 2(n = 41 075 across 254 intensive

care units)


between January 1, 2006, and December 31, 2010.

The overall rate of early RBC transfusions was 6.9%,

representing 17 815 patients who received 34 898

units of RBCs by day 2 at a median dose of 2 units.

These early transfusions represented 42.0% (17 815)

of all patients receiving a transfusion (42 390) and

32.7% (34 898 of 106 783) of all units transfused

during hospitalization for sepsis. When compared

with patients who did not receive early transfusions,

those who did were more likely to have a history

of anemia, malignant neoplasms, and weight loss.

Conversely, early transfusions were less likely in

patients with chronic pulmonary disease, obesity,

hypertension, and neurological comorbidities (Table

1 and Supplement 2). Patients who received early

transfusions, compared with those who did not,

had higher mortality rates (22.7% [4038] vs 14.9%

[35 548])and greater hospital costs (US $18 030 vs

$13 268; Table 1). At the hospital level, the median

(interquartile range [IQR]) early transfusion rate was

6.4% (4.5%-8.4%), the median (IQR) mortality rate

was 15.3% (12.6%-18.0%), and the median (IQR)

costs were $12 783 ($11 151-$15 194).

The multivariable model used for risk standard-

ization of early transfusion practices had an area under

the receiver operating characteristic curve of 0.78

(Supplement 3). Applying this model, we computed

and compared the risk-standardized rates across 364

hospitals. The median (IQR) risk-standardized early

transfusion rate was 6.7% (5.8%-7.6%) (Figure 2).

Rates varied widely across hospitals. Thirty-eight of

the 364 hospitals (10.4%) had early transfusion rates

greater than 10. The median (IQR) risk-standardized

mortality rate was 15.5% (13.1%-18.1%), and the

median (IQR) adjusted costs were $13 333 ($11 939-

$14 986). We found a modest positive correlation

between transfusion rates and costs (Spearman

= 0.157; P = .003; Figure 3). We did not find any

Variable

Table 1Selected characteristics and outcomes of adults with sepsis admitted to the hospitala

Admission via emergency department

Comorbid condition Smoking (current) Hypertension Obesity Weight loss Other neurological disorders Chronic blood loss anemia Deficiency anemias

Treatments by day 2 Vasopressors 0 1 2 3

Admission to intensive care unit Mechanical ventilation Sodium bicarbonate Hydrocortisone Central venous pressure monitoring Arterial catheterization Supplemental iron administration Ferritin test Colloids Volume of crystalloid solution, mean (SD), mL 5 L of crystalloid solution

Outcomes Mortality Costs, $

0.103 0.149 0.144-0.1940.141-0.130-0.494

0.280-0.169-0.141-0.121

-0.340-0.203-0.212-0.192-0.258-0.137-0.303-0.284-0.202-0.158-0.134

-0.185-0.299

72.9

2.830.6 6.819.613.6 2.855.7

68.518.5 8.4 4.7

48.819.016.512.126.4 6.014.812.510.9341826.4

22.718 030

76.0

4.537.510.411.918.4 0.731.1

81.511.9 4.5 2.1

31.811.0 8.7 5.815.0 2.7 4.1 3.1 4.6

2599 (2939)20.5

14.913 268

Standardized differencec

Patients who received early transfusionb

(n = 17 815)

Patients who did not receive early transfusionb

(n = 238 581)

a See Supplement 1 (available online only, at www.ajcconline.org) for the complete list of variables compared. b Numbers in this column are percentage of total number of patients in the group unless otherwise indicated in the first column.c Standardized differences are shown, rather than P values, because standardized differences measure the differences between groups independent of sam-

ple size. Although there is no accepted threshold to determine the significance of standardized differences, we elected to display characteristics with a standardized difference greater than 0.1 or less than -0.1.


correlation between transfusion and mortality rates

(Spearman = 0.078; P = .14; Figure 3). After risk

standardization, mortality rates were statistically

similar at hospitals in the highest and the lowest

transfusion quintiles. Hospitals with 400 or more

beds and hospitals with teaching affi liations had

signifi cantly higher transfusion rates (Table 2). We

did not observe any geographic differences.

In sensitivity analyses, which were restricted to

41 075 patients with septic shock who were admitted

to the ICU and received vasopressors by day 2, risk-

standardized rates of early transfusion varied between

0% and 30.4% with a median rate of 10.5% (Sup-

plement 4 and Figure 2). Analyses at the ICU level

were consistent with hospital-level analyses, as trans-

fusion rates were associated with costs but not with

mortality rates. The ICUs in larger hospitals (≥400

beds) with teaching affi liations and in urban loca-

tions had higher transfusion rates.

Discussion Our main fi nding was that early RBC transfusions

for sepsis—the single most expensive reason for hos-

pitalization1,2—were associated with increased costs

but not with survival. This fi nding is an extension

of or similar to fi ndings in studies on patients with

septic shock.7-10 We examined the institutional rates

of early transfusion in adults with sepsis without

Figure 2 Variations in risk-standardized rates of red blood cell transfusion across 364 hospitals (left) and intensive care units in 254 hospitals (right).

Percentage of patients with sepsis receiving at least 1 unit of red blood cells by day 2 at hospital level

Percentage of patients with septic shock receiving at least 1 unit of red blood cells by day 2 at hospital level

1515

1010

55

000 105 2010 3015

0

Perc

enta

ge

of

ho

spit

als

(n =

364

)

Perc

enta

ge

of

ho

spit

als

(n =

254

)

Figure 3 At the hospital level, early red blood cell (RBC) transfusions were not correlated with mortality (left panel) but were positively correlated with costs (right panel).

Standardized transfusion ratio

Ris

k-st

and

ard

ized

mo

rtal

ity

rate

, %

Ad

just

ed c

ost

s, $

Standardized transfusion ratio

2.5 2.52.0 2.01.5 1.51.0 1.00.5 0.50

0

10

20

30

40

50

60

0

5 000

10 000

15 000

20 000

25 000

30 000

0

Restrictive RBC transfusions

Low

mo

rtal

ity

Bel

ow

-ave

rag

e co

sts

Ab

ove

-ave

rag

e co

sts

Hig

h m

ort

alit

y

Restrictive RBC transfusionsLiberal RBC transfusions Liberal RBC transfusions


Early RBC transfu-sions, common in

patients without sep-tic shock and usually

given outside the ICU, are associated with increased costs but

no better survival.

major bleeding or surgery. After adjusting for inter-

hospital differences in patient characteristics, we

found that early transfusion rates varied significantly

among the 256 396 adults treated at 364 hospitals.

By hospital day 2, approximately 84% of patients

with sepsis, without shock (no vasopressor therapy),

had been treated outside the ICU, with approximately

1 in 15 (6.7%) receiving RBC

transfusions. Such early trans-

fusion rates were not associ-

ated with mortality rates,

but higher transfusion rates

meant greater costs.

Our finding that transfu-

sions were a common treat-

ment for sepsis was confirmed

in an AHRQ report18 indicat-

ing that the RBC transfusion

rate during hospitalization

for sepsis increased from

9.3% in 2000 to 13.7% in

2013.Our cohort of patients

represented the largest number of adults at risk for

receiving transfusions in US hospitals.1,2

We observed a lack of association between

hospital-level transfusion rates and mortality rates,

and this finding is consistent with the results of sev-

eral patient-level randomized controlled trials but

has not been described previously.7-10,19 Similar

variations in transfusion practices were found in

patients who were admitted to ICUs or undergoing

surgical procedures.19,20 Patients in our cohort were

under the care of internists in the general medicine

service, who order more RBC transfusions than do

physicians in any other hospital service. Therefore,

efforts to improve cost-effectiveness should also be

focused on these clinicians.3,4 Direct and indirect

costs of RBC transfusion have been estimated at $761

per unit administered,21 and the number of patients

receiving this costly and potentially unnecessary treat-

ment is large. Internists may help reduce such waste

by ordering fewer transfusions and routine laboratory

tests of hemoglobin level as well as by promoting the

use of pediatric blood-collection tubes (which mini-

mize phlebotomy-related blood loss).20,21

Strengths and LimitationsThe strengths of this study include the involve-

ment of a large sample of hospitals, which represented

approximately 15% to 20% of all hospitalizations

for sepsis in the United States during the study

period, and the use of reliable definitions of sepsis,

exposure to RBCs, and the mortality and cost out-

comes. The Premier data set was well suited for

examining costs as it contained detailed billing

records, and clerical errors in the attribution of

charges were less likely because allogeneic RBC

transfusions were carefully audited for ABO and Rh

compatibility. By conducting analyses at the hos-

pital level, we minimized bias that may threaten the

validity of observational patient-level studies.15,17

We confirmed our hospital-level results in ICU-level

sensitivity analyses.

Our study has several limitations, the most sig-

nificant of which is the lack of data on hemoglobin

levels. Hence we were unable to define an optimal

threshold for clinically beneficial RBC transfusion.

Although we examined a homogeneous patient

cohort, we still observed variations in transfusion

practices, but such variations might have been due

to hospital-level differences in hemoglobin levels.

We accounted for interhospital differences in patient

mix, including differences in rates of diagnoses of

chronic blood loss anemia and deficiency anemias

as well as in rates of other treatments for anemia

(eg, supplemental iron), further minimizing the

alternative explanations for the observed institu-

tional variation in practice.

Another limitation is our use of data from 2006

through 2010. However, our findings were confirmed

by a report18 that RBC transfusions during hospital-

ization for sepsis increased from 2000 to 2013. Last,

our hospital-level analyses are subject to the ecological

Hospital characteristic

Table 2Variation in standardized transfusion ratios across 364 hospitals

.04

.02

.99

.13

0.87 (0.60-1.15)0.97 (0.71-1.16)0.94 (0.77-1.15)1.04 (0.78-1.28)1.08 (0.92-1.30)

1.00 (0.81-1.26)0.94 (0.67-1.16)

0.98 (0.70-1.17)0.96 (0.73-1.23)

1.01 (0.79-1.27)0.92 (0.66-1.18)0.93 (0.68-1.13)0.99 (0.75-1.19)

115 (31.6) 79 (21.7) 71 (19.5) 45 (12.4) 54 (14.8)

101 (27.7)263 (72.3)

296 (81.3) 68 (18.7)

82 (22.5) 60 (16.5)153 (42.0) 69 (19.0)

No. of beds < 200 200-299 300-399 400-499 500

Teaching status Teaching Nonteaching

Location Urban Rural

Region Midwest Northeast South West

P valuebStandardized

transfusion ratioaNo. (%) of hospitals

a Median (interquartile range: 25th-75th percentile).b From Kruskal-Wallis test.


fallacy, wherein correlations at an aggregate level

may not exist at the patient level and residual con-

founding due to unmeasured factors is possible.

Conclusion We studied early RBC transfusions during hos-

pitalization for sepsis and found that approximately

1 in every 15 adults with sepsis without major bleed-

ing or surgery received this therapy during the epi-

sode of care. Transfusion rates varied widely across

the hospitals in our study, and a comparable group

of patients with sepsis received transfusions as fre-

quently as 1 in every 10 patients or as rarely as 1 in

every 25 patients without any differences in mortal-

ity rates. Such variation was associated with increased

costs. A practical strategy for reducing costs without

decreasing the quality of care may be for internists

to limit early RBC transfusions following admission

for sepsis.22-25

FINANCIAL DISCLOSUREDr Nathanson reported receiving a consulting fee for statistical analysis from Duke University School of Medicine. No other disclosures were reported.

SEE ALSO For more about sepsis, visit the Critical Care Nurse web-site, www.ccnonline.org, and read the article by Droege et al, “Application of Antibiotic Pharmacodynamics and Dosing Principles in Patients With Sepsis” (April 2016).

REFERENCES1. Pfuntner A, Wier LM, Stocks C. Most frequent conditions in

U.S. hospitals, 2011. http://www.hcup-us.ahrq.gov/reports /statbriefs/sb162.jsp. Healthcare Cost and Utilization Project Statistical Brief number 162. Published September 2013. Accessed January 26, 2016.

2. Torio CM, Moore BJ. National inpatient hospital costs: the most expensive conditions by payer, 2013. http://www.hcup -us.ahrq.gov/reports/statbriefs/sb204-Most-Expensive -Hospital-Conditions.jsp. Healthcare Cost and Utilization Project Statistical Brief number 204. Published May 2016. Accessed September 14, 2016.

3. Whitaker BI, Hinkins S, US Department of Health and Human Services. The 2011 National Blood Collection and Utilization Survey report. http://www.hhs.gov/ash/bloodsafety/2011 -nbcus.pdf. Accessed January 26, 2016.

4. US Department of Health and Human Services Advisory Committee for Blood Safety and Availability. 40th meeting minutes. Advisory Committee for Blood Safety and Avail-ability; June 8, 2011; Baltimore, MD. http://nih.granicus.com /DocumentViewer.php?file=nih_279c20e5-c8ef-4e28-b457 -d86995ff40fa.pdf. Accessed January 26, 2016.

5. Surviving Sepsis Campaign. Recommendations: other sup-portive therapy of severe sepsis. http://survivingsepsis.org /Guidelines/Documents/Other%20supportive%20therapy .pdf. Accessed January 26, 2016.

6. Dellinger RP, Levy MM, Rhodes A, et al; Surviving Sepsis Campaign Guidelines Committee including the Pediatric Subgroup. Surviving Sepsis Campaign: international guide-lines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580-637.

7. Holst LB, Haase N, Wetterslev J, et al; TRISS Trial Group; Scandinavian Critical Care Trials Group. Lower versus higher haemoglobin threshold for transfusion in septic shock. N Engl J Med. 2014;371:1381-1391.

8. ProCESS Investigators, Yealy DM, Kellum JA, et al. A ran-domized trial of protocol-based care for early septic shock. N Engl J Med. 2014;370(18):1683-1693.

9. ARISE Investigators; ANZICS Clinical Trials Group, Peake SL, et al. Goal-directed resuscitation for patients with early septic shock. N Engl J Med. 2014;371(16):1496-1506.

10. Mouncey PR, Osborn TM, Power GS, et al.; ProMISe Trial Investigators. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med. 2015;372(14):1301-1311.

11. Bulger J, Nickel W, Messler J, et al; Choosing wisely in adult hospital medicine: five opportunities for improved healthcare value. J Hosp Med. 2013;8(9):486-492.

12. Raghunathan K, Shaw A, Nathanson B, et al. Association between the choice of IV crystalloid and in-hospital mortal-ity among critically ill adults with sepsis. Crit Care Med. 2014;42(7):1585-1591.

13. Lindenauer PK, Pekow PS, Lahti MC, Lee Y, Benjamin EM, Rothberg MB. Association of corticosteroid dose and route of administration with risk of treatment failure in acute exacerbation of chronic obstructive pulmonary disease. JAMA. 2010;303(23):2359-2367.

14. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiol-ogy of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003;348(16):1546-1554.

15. Sjoding MW, Luo K, Miller MA, Iwashyna TJ. When do con-founding by indication and inadequate risk adjustment bias critical care studies? A simulation study. Crit Care. 2015; 19(1):195.

16. Tsui AK, Dattani ND, Marsden PA, et al. Reassessing the risk of hemodilutional anemia: some new pieces to an old puz-zle. Can J Anaesth. 2010;57(8):779-791.

17. Reade MC, Delaney A, Bailey MJ, Angus DC. Bench-to-bedside review: avoiding pitfalls in critical care meta-analysis—funnel plots, risk estimates, types of heterogeneity, baseline risk and the ecologic fallacy. Crit Care. 2008;12(4):220.

18. West KA, Barrett ML, Moore BJ, Miller JL, Steiner CA. Trends in hospitalizations with a red blood cell transfusion, 2000-2013. http://www.hcup-us.ahrq.gov/reports/statbriefs /sb215-Red-Blood-Cell-Transfusions.pdf. Healthcare Cost and Utilization Project Statistical Brief number 215. Pub-lished December 2016. Accessed January 14, 2017.

19. Salpeter SR, Buckley JS, Chatterjee S. Impact of more restrictive blood transfusion strategies on clinical out-comes: a meta-analysis and systematic review. Am J Med. 2014;127(2): 124-131.

20. Bennett-Guerrero E, Zhao Y, O’Brien SM, et al. Variation in use of blood transfusion in coronary artery bypass graft surgery. JAMA. 2010;304(14):1568-1575.

21. Shander A, Hofmann A, Ozawa S, Theusinger OM, Gombotz H, Spahn DR. Activity-based costs of blood transfusions in surgi-cal patients at four hospitals. Transfusion. 2010;50(4): 753-765.

22. Scott BH, Seifert FC, Grimson R. Blood transfusion is associ-ated with increased resource utilisation, morbidity and mor-tality in cardiac surgery. Ann Card Anaesth. 2008; 11(1): 15-19.

23. Rogers MA, Blumberg N, Saint S, Langa KM, Nallamothu BK. Hospital variation in transfusion and infection after car-diac surgery: a cohort study. BMC Med. 2009;7:37.

24. Lagu T, Rothberg MB, Nathanson BH, Pekow PS, Steingrub JS, Lindenauer PK. Variation in the care of septic shock: the impact of patient and hospital characteristics. J Crit Care. 2012;27(4):329-336.

25. Abdelsattar ZM, Hendren S, Wong SL, Campbell DA Jr, Henke P. Variation in transfusion practices and the effect on out-comes after noncardiac surgery. Ann Surg. 2015;262(1):1-6.


End-of-Life Care


Background Little is known about the effect of chaplains on critical care nurses who are caring for critically ill patients and their families.Objective To understand nurses’ experiences when they make a referral to the Spiritual Care Department for a patient or the family of a patient who is dying or deceased. Specific aims were to explore spiritual care’s effect on nurses and how nurses understand the role of spiritual care in practice.Methods A qualitative descriptive study using in-person, semistructured interviews in a 21-bed medical-surgical intensive care unit in a teaching hospital. Purposeful sampling identified nurses who had at least 5 years of experience and had cared for at least 5 patients who died on their shift and at least 5 patients for whom they initiated a spiritual care referral. Interviews were digitally recorded and anonymized; conventional content analysis was used to analyze transcripts. Three investigators inde-pendently coded 5 transcripts and developed the prelimi-nary coding list. As analysis proceeded, investigators organized codes into categories and themes.Results A total of 25 nurses were interviewed. The central theme that emerged was presence, described through 3 main categories: the value of having chap-lains present in the intensive care unit and their role, nurses’ experiences working with chaplains, and nurses’ experiences providing spiritual care.Conclusion Nurses considered spiritual care essential to holistic care and valued the support chaplains provide to patients, families, and staff in today’s spiritually diverse soci-ety. (American Journal of Critical Care. 2018; 27:212-219)

CRITICAL CARE NURSES’ EXPERIENCES WITH SPIRITUAL CARE:THE SPIRIT STUDYBy Nigel Bone, MDiv, RP, Marilyn Swinton, MSc, Neala Hoad, RN, Feli Toledo, MDiv, RP, and Deborah Cook, MD



this article.


Although providing spiritual care has important meaning for nurses and enhances professional satisfaction,1 a “crisis of spirituality” has been identified within the nursing profession related to nurses’ preparedness to identify, assess, and provide spiritual care.1-4 This crisis may have evolved as modern nursing practice has been distanced from its original spiritual tradition through a more task-focused, problem-

solving approach to care.4 Another contributing factor may be technologic advances, which may have “led to a disconnect between caring for the body and caring for the soul.”5

Nurses report that attending to the spiritual care

of their patients is part of their scope of practice and

is rooted in holistic care.2-10 In 1988, the North Amer-

ican Nursing Diagnostic Association’s inclusion of

spiritual distress as a nursing diagnostic category

officially recognized the role of spiritual care in

nursing practice.11 However, according to several

studies,2,5,7-9,12 nurses do not receive education about

how to provide spiritual care.Some curricula “virtu-

ally ignore spiritual distress”9 despite recognition

that “spirituality is a way of being and experiencing

that shapes and impacts nursing presence.”13

According to results of a nursing survey, percep-

tions about the need for spiritual care differ depend-

ing on the care area; for example, in the operating

room, patients are usually unconscious and families

are not present, possibly attenuating the need for

spiritual care.12 In the intensive care unit (ICU), spir-

itual distress frequently is experienced by critically

ill patients requiring life support, their family mem-

bers, and staff.14 Referrals to professionals with

specialized knowledge and skills in spiritual care

are often made in this setting,15 because nurses may

lack the training to provide spiritual care and may be

uncomfortable with this aspect of practice.6-9,11,16-18

The role of spirituality in health care may be

assigned higher importance today than in the past,19

perhaps reflecting the growing societal interest in

spirituality.20 In a review of spirituality across disci-

plines, Swinton21 concluded that “it is clear that

people are trying to name and draw attention to

something that is missing from current ways of prac-

ticing.” The objective of our

study was to understand the

experiences of ICU nurses

when they make a referral to

spiritual care services for a

patient or the family of a

patient who is dying or is

deceased. Specific aims were to explore the effect of

spiritual care on nurses and how nurses understand

the role of spiritual care in the ICU.

Methods We used purposive sampling to identify ICU

nurses who had at least 5 years of experience, cared

for at least 5 patients who died on their shift, and

cared for at least 5 patients for whom they initiated

a referral to a chaplain (in this article, spiritual care

clinicians are referred to as chaplains). Nurses were

recruited through an email invitation.

Data were collected through semistructured

qualitative interviews. To frame a context for the

interview, we provided the following definition of

spirituality to participants:

Spirituality can mean different things

to different people and can be thought

of in terms of the ways that individuals

seek and express meaning and purpose,

and how they experience connections

with the moment, with themselves, with

others, with nature and with other things

that to them are significant or sacred. Reli-

gion is one way of expressing spirituality.15

Interviews were digitally recorded, transcribed verba-

tim, and anonymized.

AnalysisWe used qualitative description to produce a

descriptive summary of the findings.22 Conventional

About the AuthorsNigel Bone is a fellow in spiritual care and a registered psychotherapist, Spiritual Care Department, St Joseph’s Healthcare, Hamilton, Ontario, Canada. Marilyn Swinton is a research coordinator, Department of Health Research Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, Canada. Neala Hoad is a registered nurse and research coordinator, Department of Critical Care, St Joseph’s Healthcare. Feli Toledo is a chaplain and registered psychotherapist, Spiritual Care Department, St Joseph’s Healthcare. Deborah Cook is a professor and intensivist, Department of Health Research Methods, Evidence and Impact and Department of Medicine, McMaster University; and Department of Critical Care, St Joseph’s Healthcare.

Corresponding author: Nigel Bone, MDiv, RP, St Joseph’s Healthcare Hamilton, 50 Charlton Ave, E, Hamilton, ON, Canada L8N 4A6 (e-mail: [email protected]).


“Most nursing pro-grams virtually ignore spiritual distress.”9


content analysis was used to analyze transcripts.

This is a coding approach whereby codes are derived

directly from the data without imposing precon-

ceived categories or theoretical perspectives.23 Three

investigators independently completed line-by-line

coding of 5 interview transcripts and, through a

consensus, developed the preliminary coding list.

The remaining transcripts were coded by 1 investiga-

tor, who recorded the evolution of the coding list

in an audit trail.24 The research team reviewed cod-

ing reports and organized the codes into categories.

We held a member checking event with 6 nurses to

assess the extent to which our findings resonated with

their experiences. N’Vivo (version 11.0; QSR Interna-

tional) was used for data management. Quantita-

tive data were summarized using descriptive statistics.

Results A total of 25 ICU nurses (22 women, 3 men) were

interviewed (Table 1). The following 3 main categories

emerged from the data, all related to the central theme

of presence: (1) the value of having chaplains present

in the ICU and their role, (2) nurses’ experiences work-

ing with chaplains in the ICU, and (3) ICU nurses’

experiences of providing spiritual care through their

own practice (Table 2).

Characteristica

Table 1Demographic characteristics of the 25 nurse participants

Age, y

Years worked as a nurse anywhere

Years worked as a nurse in an ICU

Years worked as a nurse in this hospital’s ICU

No. of patients in past year that nurses cared for who died during their shift in this hospital’s ICU

No. of referrals nurses made to spiritual care services for patients who died in the past year in this hospital’s ICU

43.7 (11.5)

21.0 (11.2)

17.5 (10.1)

17.1 (9.8)

14.6 (13.4)

8.6 (4.7)

Mean (SD)

Abbreviation: ICU, intensive care unit.

The value and role of having chaplains present in the ICU

Table 2Examples of 3 main themes found

“I can provide physical care but [spir-itual care] . . . it’s almost like round-ing out the care that we provide and it’s . . . that’s a good feeling for me. It’s almost a feeling of completeness of care, like, holistic. We’ve kind of come full circle and, you know, we’ve . . . we’ve addressed everything.”

“Spiritual care isn’t just for the patient . . . spiritual care, I think, pro-vides support to clinicians who are going through the process of, you know, losing a patient.”

“I’m there for the patient and the fam-ily, but often in situations, my respon-sibility goes to the patient first and foremost, so it’s nice to have some-body that’s there to take care of the family—the family is their primary focus, so that we work together, but I know that the family is looked after.”

“I find spiritual care clinicians not only deal with the spiritual aspect of things, like they have been kind of good at sorting out, like, even fam-ily dynamics and kind of calming the waters between family members that [are] at odds.”

“We had one instance, actually, if I can share that. It happened last fall . . . [a] very sick patient going to the OR. The surgeon said, ‘We’re not sure if he’s going to make it back.’ And spiritual care couldn’t make it here in time and I could feel from the family that they wanted prayers said before he went to the OR, so I prayed with them.”

“If the patient has died, I make sure their hands are out so that the family can hold their hands . . . I just allow them the opportunity to do what-ever they feel is right in the moment. If they just want to cry for an hour, I’ll just get boxes of [tis-sues] and make sure [there are] chairs around and [a] glass of water—just to make them more comfortable.”

“I remember him [patient’s family member] com-ing and me just sitting with him and chatting about how he was feeling and what was going on and, you know, kind of reassuring him, too, like, we’re looking after her and, you know, now you can do a little bit for yourself, too, which I think he [found] kind of like, eye opening. Kind of like, you know, ‘I can actually do something for myself right now.’”

“Hear the stories . . . because often they’ll speak of the patient, and the stories and, ‘Remember Dad did this,’ and ‘Remember Mum did that.’ And it’s nice, we’ll often, you know, go in and look at the photos that they have up.”

“A lot of times what we do is we brief the spiritual care personnel overall, [about] what’s going on with the indi-vidual, with the patient and with the families, because sometimes families are really stressed and they have difficulty dealing with diagnosis and prognosis.”

“When spiritual care is there with me in the room . . . it makes me feel less guilty that I have these tasks to do when some-one is dying in front of me and their family is there watching them pass away.”

“The more recent referrals I’ve made are not about dying. They’ve been more about comfort for patients or family . . . somebody just needs someone who has 30 minutes of their undivided attention to do nothing but listen to them.”

“It’s a shared experience and kind of unspoken support. It’s a bit intangible at times, but there’s comfort in seeing spiri-tual care staff . . . in some unspoken way, I’m feeling supported. . . . I’m not going through this alone. . . . There’s a sense of comfort because that person is there to experience it with me.”

How ICU nurses provide spiritual care through their practice

Nurses’ experiences working with chaplains on the unit

Abbreviation: ICU, intensive care unit; OR, operating room.


Value of Having Chaplains Present in the ICU and Their Role

Based on interview findings, we determined the

nurses viewed chaplains as an essential part of the

multidisciplinary team who were valued for their

supportive presence in 3 contexts: with patients and

families, with ICU staff, and in the ICU in general.

Chaplain’s Presence Supports Patients and Families.

We found that chaplains are considered by nurses to

have the optimal language and approach to support

patients and their families, regardless of the patient’s

or family’s faith or belief system:

Talking about the patient and who they

are, and acknowledging them as a per-

son—just reflecting on their life [is] some-

thing I’m pretty comfortable doing with

families. But there’s something about

the spiritual care person [who] has the

right language and has the ability to . . .

elicit that kind of reflection.

Nurses affirmed that chaplains can ease the pain

associated with the dying experience for patients and

families. They attributed this to the presence and indi-

vidualized approach of the chaplain:

When pastoral care is involved . . .it makes

the death more like it should be. It’s an

important event and it should be personal.

Chaplains’ Presence Supports Staff. In addition

to the support for patients and their families, nurses

shared how they often feel a sense of relief after call-

ing chaplains, knowing they will jointly help support

the patient and the patient’s family:

Relieved would be a good word. Just

that there is 1 more member of the

team that can help support the family.

I feel relieved that perhaps they already

have a relationship with the spiritual

care worker, so it’s another familiar

face, someone who knows their jour-

ney, someone who knows the patient

and the family and so there’s some-

thing familiar . . . It’s comforting for the

family and I feel comforted that they’re

feeling . . . better about the situation.

Nurses described how the presence of a chap-

lain personally supported nurses caring for dying

patients. They valued the opportunity to debrief,

which usually was done informally:

Sometimes they come around after an

event . . .and say “How are you doing?

That was a difficult situation.” . . . Then

we might debrief a bit . . . Even though

we go through a really difficult or tragic

experience, sometimes it’s over and then

we move on to the next patient. So I

always appreciate it when someone

comes along and says, “So, how are

things going? How did you find that?”

Chaplains Educate Through Their Presence in the

ICU. The important role for chaplains educating

nurses was underscored during the interviews, as

indicated by the following:

There’s great knowledge and experience

that could be transferred to some of our

more inexperienced nurses from spiri-

tual care. It would be great if that rela-

tionship was nurtured from the very

beginning and just became . . . part of

our culture here.

I think we learned from them . . . the

compassion part . . . You watch how they

work and we learn ourselves.

Nurses’ Experiences Working With Chaplains in the Unit

Nurses make referrals to chaplains through

all phases of a patient’s critical illness, sometimes

immediately after ICU admission. Through early

engagement of chaplains, patients and their families

can develop relationships that facilitate access to

spiritual support during the patient’s stay. Nurses

described a shared-care model when working with

chaplains, which was articulated in the following

3 ways: (1) why nurses call spiritual chaplains, (2)

how nurses introduce the idea of spiritual care to

patients and families, and (3) the shared experi-

ence of nurses working with chaplains.

Why Nurses Call Chaplains. Nurses reflected

on how the presence of chaplains help provide an

essential aspect of holistic care at the end of life:

It’s the more holistic, humanistic

approach to dying than what we deal

with, which is more of the medical . . .

messy kind of things.

I always want to try and make sure the

family feels supported and that, if they are

religious or spiritual, they feel like we took

care of them from that perspective as well.

Nurses believed that the support delivered by

chaplains was a reflection of their own caring. Even

though they wanted to care for patients and their

families in every domain, nurses were sometimes

unable to do so because they were busy attending

to technical aspects of practice:


Sometimes we can’t spend as much

time at the bedside as we want to. And

so our spiritual care team just steps in

and they . . . just carry on that caring

that we always want the family to know

[is there] even though we can’t be there

all the time.

Some nurses were uncomfortable identifying

or addressing spiritual needs, preferring to have

chaplains involved:

I don’t always have the ability to find

comforting words. . . . I’m too intimi-

dated that I might say the wrong

thing . . . so I just find [the chaplain],

when they’re there, they deal with that

aspect, which is relieving [for] me.

How Nurses Introduce the Idea of Spiritual Care

Support to Patients and Families. Nurses talked about

different ways of introducing the service of spiri-

tual care to patients and families, depending on

the context:

Someone to be there if you need someone

to talk to because this is a hard time. That’s

pretty much how I say it.

Nurses sometimes guide hesitant families to

consider the potential role for chaplains by affirm-

ing their frequent involvement:

I also tell them that we use them often. I

do make sure that they know that because

I don’t want them to think that someone

will come into the room who really doesn’t

have much experience with that situation

or with someone becoming palliative or

someone dying.

Nurses’ Shared Experience of Working With Chap-

lains. Nurses appreciated sharing the caring experi-

ence with chaplains. They talked about wanting to

have chaplains present more often, not only at the

end of life:

I think we work really cooperatively. I

really appreciate and trust their ability

and their gifts. . . . Personally, I like to

have the presence of the spiritual care

team in the unit even when we don’t have

a dying patient at that very moment.

Dialogue between nurses and chaplains was

identified as helping to share information about the

patient, the patient’s family, and their circumstances,

particularly when people are experiencing difficulty

accepting a patient’s prognosis. Nurses described

that chaplains often forge an intimate relationship

with the patient and family, which facilitates sharing

important aspects of personhood with the interpro-

fessional team:

Spiritual care will come to us to ask us

about the family and we’ll ask them

what things they found out, because

sometimes the family will talk differ-

ently . . . to someone like [the chaplain,

saying some different than] what they

would tell the nurse or the doctor.

How ICU Nurses Provide Spiritual Care Through Their Practice

All of the nurses that we interviewed shared

examples of providing spiritual care to patients and

families through their presence, but not all of them

cited this presence as being part of providing spiri-

tual care. We learned about the intentional provision

of spiritual care from some nurses, and the unarticu-

lated, if not unrecognized, provision of spiritual care

by others. From these data, 3 ways that nurses pro-

vide spiritual care through presence in their practice

were identified: (1) intention, (2) being with the

patient and/or family, and (3) compassion.

Intention. One of the most common motivators

for nurses to provide the best end-of-life experience

possible was intention:

I don’t want to miss an opportunity when

somebody’s going to pass away that they

would have liked some spiritual advice,

prayers, or calling your own pastor.

Yeah, it’s just 1 of those things. It’s . . .

someone’s last moment and . . you

have to make it the best for them.

Being With the Patient and/or Family. Given the

large amount of time that nurses spend at the bed-

side with their patients, their presence was consid-

ered to be a manifestation of their provision of

spiritual care:

We’re not necessarily invited, but I auto-

matically go in and shut the door, and

I like to be part of it because I think it’s

nice for the families to see that and it

makes them feel like their loved one . .

they’re not just a patient.

Nurses often disclosed being unaware of when

they were providing spiritual care. One nurse said,

“I think we all do it; we just don’t realize that we’re

doing it.” Nurses gave examples, such as the follow-

ing, of unknowingly providing spiritual care through

their presence:


Nurses valued informal, immediate debriefing with chaplains after challenging clinical situations.

We had a young lady in our unit . . . the

parents were in the room and I wasn’t

even aware I was doing it, and the mom

said to the daughter, who was in the

bed, “This must be so reassuring for

you because this is what your Grandma

would do. She would hum for you.” . . .

So I do things like that without even

being aware, and it brings them some

peace and comfort.

Compassion. Nurses claimed compassion as

something that comes naturally to them, citing it

as a core component of nursing practice that posi-

tively affects nurses themselves:

I think I’d only been with the patient

maybe a few hours that day. And the

family member pulled me aside and

said, “I can see that you really, really

love your work by your actions.”

It’s not always what you say . . . but your

actions . . . you know . . . whether it be a

touch of their forehead as you are talking

to them or holding their hand or . . . get-

ting a glass of water.

One of the things that is most powerful

for me is when you have a patient [who]

doesn’t have any family and they don’t

really have anyone there when they’re

passing away. It’s kind of a peaceful

thing to go be with a patient and even

just hold their hand while they’re dying.

I think that’s kind of a spiritual thing.

Discussion We identified the presence of spiritual care to be

the central theme when nurses refer to chaplains for

dying or deceased patients in their care. Hailed as

“the most essential element of spiritual care,”25

“based on a healing relationship,”13 elements of

presence include a reciprocal relationship to the

whole person extending beyond the technical and

attending to their needs.26-29 Like spirituality, pres-

ence takes many forms and is challenging to define.

Spiritual care presence has been described as

being accompanying and comforting30—elements

that were identified by nurses in this study. Nurs-

ing presence has been described as reflecting

“uniqueness, connecting with the patient’s experi-

ence, sensing, going beyond the scientific data, know-

ing what will work and when to act, and being with

the patient.”27 The uniqueness of each nurse’s spiri-

tuality gives nursing presence its unique style.31,32

We identified a need for more guided discus-

sions by chaplains for nurses to learn how to pro-

vide spiritual care and make appropriate referrals.

This is an actionable item that could be imple-

mented through the integration of spiritual care

education into ICU orientation for new learners

and could lead to an overall decrease in the amount

of spiritual distress experienced by patients, fami-

lies, and unit staff. Lack of nursing preparation may

lead to hesitation inquiring about the spiritual needs

of patients and their families.1 Experienced nurses

suggested that better introduction to the hospital’s

spiritual care department and its roles would help

new graduates, who often have limited exposure to

spiritual care when starting their career. With guided,

repeated exposure to patients in crisis, nursing stu-

dents can recognize their nursing presence at work.13

In addition to periodic, scheduled, case-based

rounds after a death,33 nurses valued informal,

immediate debriefing with chaplains after chal-

lenging clinical situations; this finding aligned

with those of other reports.34

Strengths of this study include the descriptive

summary of the findings developed by a multidisci-

plinary team, with minimal theoretical interpreta-

tion. The interviews provided nurses an opportunity

to talk about and reflect on spirituality and their

practice. Results were presented in the words of

the research partici-

pants,21 and member

checking affirmed that

the findings resonated

with nurses. Limita-

tions of this study

include the single-

center design and the

focus on dying patients.

The generalizability of these findings should take

the setting into account: This study was conducted

in a faith-based hospital with a designated ICU

chaplain, 24-hour on-call chaplain coverage, a

chaplain to bed ratio of 0.70 to 21, periodic visiting

community clergy, and where a collaborative end-of-

life program was in place.35

Chaplains synergistically add a key dimension

to the care of the patient that no other member of

the health care team can provide,8 because spiritual

care is what chaplains do, rather than being a part

of what they do.11,36 The role of chaplains is crucial;

however, clearly they are not the only ones who pro-

vide spiritual care in the hospital.8,11,16 Ultimately,

nursing care focuses on wholeness, including spiri-

tuality,37 and meeting a patient’s spiritual needs is


not only consistent with best practice but may posi-

tively affect nurses themselves. Chaplains have the

skills and knowledge to help nurses identify when

they are engaging in spiritual care. Nurses in our study

recognized, with relative ease, acts of compassion

in their practice, manifested not just in words but in

actions. However, some nurses were unaware of

when, and how often, they provided spiritual care.38

In summary, we found that ICU nurses consid-

ered spiritual care to be an essential aspect of car-

ing holistically for critically ill patients. Nurses we

interviewed value the support chaplains provide to

patients, families, and clinicians, particularly, but

not only, when patients are dying. Nurses found

making referrals to the chaplains to be a positive

experience, contributing importantly to patient-

and family-centered care at the end of life in today’s

spiritually diverse society.

ACKNOWLEDGMENTSWe thank the nurses of the ICU at St Joseph’s Health-care Hamilton who participated in the SPIRIT Study. We appreciate the assistance of Diana Clancy with the tran-scription and of Nicole Zytaruk with the SPIRIT logo. We thank Gary Payne, manager of the Spiritual Care Depart-ment, for his encouragement with this project and for establishing a fellowship in spiritual care with a research component. We thank Lily Waugh, ICU manager, for her support of this study. We are grateful for Tammy French, RN, who helped to develop the project and provided clin-ical coverage for participants when they were being interviewed. This work was inspired by the work of the Sisters of St Joseph in Hamilton.

FINANCIAL DISCLOSURESThis study was supported by the St Joseph’s Healthcare Hamilton Spiritual Care Department and Academic Critical Care Office, and by peer-review funds from the St Joseph’s Healthcare Hamilton Professional Advisory Committee Research Award Program, the Canadian Foundation for Spiritual Care, and a grant from the Canadian Institutes of Health Research.

SEE ALSO For more about family support, visit the Critical Care Nurse website, www.ccnonline.org, and read the arti-cle by Mureau-Haines et al, “Family Support During Resuscitation: A Quality Improvement Initiative” (Decem-ber 2017).

REFERENCES

1. Canfield C, Taylor D, Nagy K, et al. Critical care nurses’ per-ceived need for guidance in addressing spirituality in criti-cally ill patients. Am J Crit Care. 2016;25(3);206-211.

2. Culliford L. Spirituality and clinical care: spiritual values and skills are increasingly recognized as necessary aspects of clinical care. BMJ. 2002;25(3):1434-1435.

3. Dickinson C. The search for spiritual meaning. Am J Nurs. 1975;75(10):1789-1794.

4. Hewa S, Hetherington RW. Specialists without spirit: crisis in the nursing profession. J Med Ethics. 1990;16(4):179-184.

5. Koenig HD. Spirituality in Patient Care: Why, How, When, and What? 2nd ed. West Conshohocken, PA: Templeton Press; 2007.

6. Davidhizar R, Bechtel GA, Cosey EJ. The spiritual needs of hospitalized patients. Am J Nurs. 2000;100(7):24C-24D.

7. Dickinson C. Reviewed Works: Spiritual Care; The Nurse’s Role by Sharon Fish, Judith Allen Shelly; Spiritual Care Workbook by Judith Allen Shelly. Am J Nurs. 1978;78(12): 2149.

8. Dimeo E. Nursing: significance for spiritual care. J Relig Health. 1980;19(3):240-245.

9. Murray CK. Addressing your patient’s spiritual needs. Am J Nurs. 1995;95(11);16N-16P.

10. Taylor EJ, Marnier I, Ricci-Allegra P, Foith J. Self-reported frequency of nurse-provided spiritual care. Appl Nurs Res. 2017;35:30-35.

11. Mohrmann ME. Ethical grounding for a profession of hospital chaplaincy. Hastings Cent Rep. 2008;38(6):18-23.

12. Grant D, O’Neil K, Stephens L. Spirituality in the workplace: new empirical directions in the study of the sacred. Sociol Relig. 2004;65(3):265-283.

13. Rankin, EA, Delashmutt MB. Finding spirituality and nursing presence the student’s challenge. J Holist Nurs. 2006; 24(4): 282-288.

14. Ho JQ, Nguyen CD, Lopes R, Ezeji-Okoye SC, Kuschner WG. Spiritual care in the intensive care unit: a narrative review [published online January 1, 2017]. J Intensive Care Med. doi:10.1177/0885066617712677.

15. Puchalski C, Ferrell B, Virani R, et al. Improving the quality of spiritual care as a dimension of palliative care: the report of the Consensus Conference. J Palliat Med. 2009; 12(10): 885-904.

16. Landry VL. Pastoral care in a trauma center. J Relig Health. 1996;35(3):211-214.

17. Sumner CH. CE credit: recognizing and responding to spiritual distress. Am J Nurs. 1998;98(1):26-31.

18. Hughes BP, DeGregory C, Elk R, Graham D, Hall EJ, Ressallat J. Spiritual Care and Nursing: A Nurse’s Contribution and Prac-tice. New York, NY: HealthCare Chaplaincy Network; 2017.

19. Pembroke NF, Pembroke JJ. The spirituality of presence in midwifery care. Midwifery. 2006:24(3);321-327.

20. Pesut B, Reimer-Kirkham S, Sawatzky R, Woodland G, Pever-all P. Hospitable hospitals in a diverse society: from chap-lains to spiritual care providers. J Relig Health. 2012; 51(3): 825–836.

21. Swinton J. What is missing from our practice? Spirituality as presence and absence. J Study Spiritual. 2011;1(1):13-16.

22. Sandelowski M. Whatever happened to qualitative descrip-tion? Res Nurs Health. 2000;23(1):334-340.

23. Hsieh HF, Shannon SE. Three approaches to qualitative con-tent analysis. Qual Health Res. 2005;15(9):1277-1288.

24. Rodgers BL, Cowles KV. The qualitative research audit trail: a complex collection of documentation. Res Nurs Health. 1993;16(3):219-226.

25. Burckhardt M., Nagai-Jacobson M. Spirituality: Living Our Connectedness. Albany, NY: Delmar; 2002.

26. Chase SK. Response to “The concept of nursing presence: state of the science.” Sch Inq Nurs Pract. 2001;15(4):323-327.

27. Doona ME, Chase SK, Haggerty LA. Nursing presence: as real as the Milky Way Bar. J Holist Nurs. 1999;17(1):54-70.

28. Duis-Nittsche ER. A Study of Nursing Presence [disserta-tion]. Galveston, TX: University of Texas at Galveston, 2002. Dissertation Abstracts International, 62, 10B.

29. Smith TD. The concept of nursing presence: state of the sci-ence. Sch Inq Nurs Pract. 2001;15(4):299-323.

30. Nolan S. Spiritual Care at the End of Life: The Chaplain as a “Hopeful Presence.” London, UK: Jessica Kingsley Pub-lishers; 2012.

31. DeLashmutt M. Spiritual Needs of Mothers Raising Children While Homeless [dissertation]. Fairfax, VA: George Mason University; 2000. Dissertation Abstracts International, 61, 03B.

32. Godkin J. Healing presence. J Holist Nurs. 2001;19(1): 5-21.33. Hough CL, Hudson LD, Salud A, et al. Death rounds: end-of-

life discussions among medical residents in the intensive care unit. J Crit Care. 2005;20(1):20-25.

34. Canadian Association for Spiritual Care. Scope of practice. http://www.spiritualcare.ca/profession/scope-of-practice/. Accessed January 25, 2018.

35. Cook D, Swinton M, Toledo F, et al. Personalizing death in the intensive care unit: the 3 Wishes Project: a mixed-methods study. Ann Intern Med. 2015;163(4):271-279.

36. Choi PJ, Curlin FA, Cox CE. “The patient is dying, please call the chaplain”: the activities of chaplains in one medical center’s intensive care units. J Pain Symptom Manage. 2015;50(4):501-506.

37. Canadian Nurses Association. Framework for the Practice of Registered Nurses in Canada. 2nd ed. Ottawa, Ontario,

Canada: Canadian Nurses Association; 2015. https://cna-aiic .ca/~/media/cna/page-content/pdf-en/framework-for-the-pracice-of-registered-nurses-in-canada.pdf?la=en. Accessed January 25, 2018.

38. Goldberg B. Connection: an exploration of spirituality in nursing care. J Adv Nurs. 1998;27:836-842.

1.0 Hour Category BC E Notice to CE enrollees:



1. Identify the value of having chaplains present in the intensive care unit for patients and family members as

well as the interprofessional team.

2. Describe the relationship between nurses and chaplains within the context of critical care.

3. Analyze the provision of spiritual care by critical care nurses and identify reasons they may be unaware of

providing it.





Pulmonary Critical CarePulmonary Critical Care


Background Case series have reported favorable out-comes with extracorporeal membrane oxygenation (ECMO) in patients with severe acute respiratory distress syndrome. However, those patients were generally young, with few comorbid conditions.Objective To characterize the clinical features and sur-vival rates of patients with severe acute respiratory distress syndrome who met criteria for ECMO but were managed without it.Methods Patients who met the study criteria were iden-tified prospectively. Inclusion criteria for ECMO included severe hypoxemia, uncompensated hypercapnia, or elevated end-inspiratory plateau pressures despite low tidal volume ventilation. Predicted survival rates with ECMO were calculated using the Respiratory ECMO Survival Prediction score.Results Of the 46 patients who met the criteria for severe acute respiratory distress syndrome and ECMO consid-eration, 5 received ECMO and 16 patients had at least 1 contraindication to it. The remaining 25 patients met ECMO criteria but did not receive the treatment. The patients’ mean age was 53.5 (SD, 14.3) years; 84% had at least 1 major comorbid condition. The median pre-dicted survival rate with ECMO was 57%. The actual hospi-tal discharge survival rate without ECMO was 56%.Conclusions The general medical intensive care patient population with severe acute respiratory distress syn-drome is older and sicker than patients reported in prior case series in which patients were treated with ECMO. In this study, the survival rate without ECMO was similar to predicted survival rates with ECMO. (American Journal of Critical Care. 2018; 27:220-227)

SURVIVAL OF PATIENTS WITH

SEVERE ACUTE RESPIRATORY

DISTRESS SYNDROME TREATED

WITHOUT EXTRACORPOREAL MEMBRANE OXYGENATIONBy Sarina K. Sahetya, MD, Roy G. Brower, MD, and R. Scott Stephens, MD


Most patients with acute respiratory distress syndrome (ARDS) require mechani-cal ventilation to facilitate gas exchange while allowing time for supportive care and healing to occur. However, in some patients, mechanical ventilation cannot maintain gas exchange goals and may cause ventilator-induced lung injury, exacerbating the underlying inciting injury. In recent years, extracorpo-

real membrane oxygenation (ECMO) has been used increasingly for the treatment of patients with severe ARDS. This support can allow mechanical ventilation to be modified or removed, lowering the risks of ventilator-induced lung injury and oxygen toxicity. Yet, ECMO itself con-fers risks, including bleeding, vascular access complications, thrombosis, and infection.1-3

Despite its increasing use, the benefits of

ECMO for the treatment of ARDS have not been

clearly established. Early randomized controlled

trials in 1979 and 1994 did not show benefit with

ECMO in ARDS.4,5 However, ECMO technology

has advanced significantly with improved pumps,

oxygenators, and biocompatible circuits since that

time.6 In a recent controlled clinical trial, patients

with ARDS were randomly assigned to either usual

care at their local hospitals or transfer to a single,

ECMO-capable, high-volume tertiary hospital. Clini-

cal outcomes were better in the patients transferred

to the ECMO-capable hospital than in patients who

remained at their local hospitals. However, only

75% of the intervention group actually received

ECMO at the tertiary hospital.7 Moreover, medical

care in the control group was not standardized, and

fewer patients in the control group received lung-

protective ventilation. As a result, it is unclear if ECMO

or better care at a high-volume tertiary center was

the variable most responsible for the reduction in

mortality between the 2 groups.6,8

Adoption of ECMO increased further after

researchers reported findings of several uncon-

trolled case series indicating low mortality rates in

patients with ARDS associated with H1N1 influ-

enza who received ECMO.9-12 However, a significant

limitation of these encouraging case series is that the

patients who received ECMO were young (median

age, 36 years) and had mostly single-organ disease

(median Sequential Organ Failure Assessment score,

9),13 in contrast to the typical ARDS population, which

is substantially older and has more comorbid condi-

tions.14,15 Moreover, in case series of comparably ill

patients with H1N1 influenza who were treated

without ECMO, similar low mortality rates were

reported.16,17 Because these were uncontrolled case

series, and owing to the young age of patients reported

therein, it is not clear whether ECMO actually provided

a survival advantage in the setting of severe ARDS.

Because of the lack of strong evidence support-

ing the use of ECMO in severe ARDS, most patients

with severe ARDS at our institution, a high-volume

tertiary center, are treated without ECMO. In this

prospective cohort study, we report the demographic

and clinical features and clinical outcomes of patients

with severe ARDS who were treated without ECMO

despite meeting previously published criteria for

receiving it.6

Methods This prospective, observational study was con-

ducted at a single tertiary academic center. This study

was approved by the institutional review board at

Johns Hopkins University. We screened for eligible

patients each day by reviewing electronic medical

records of patients admitted to

the medical intensive care unit

from February 2014 to June 2015.

Patients were identified on the

basis of the Berlin definition of

severe ARDS.18 Patients with severe

ARDS were included in our cohort

if they met 1 of the following crite-

ria for ECMO consideration, which

were modified from Brodie and

Bacchetta6: (1) ratio of PaO2 to fraction of inspired

oxygen (FIO2) was less than 100 mm Hg despite

levels of positive end-expiratory pressure of at least

10 cm H2O for at least 6 hours; (2) uncompensated

About the AuthorsSarina K. Sahetya, Roy G. Brower, and R. Scott Stephens are physicians, Division of Pulmonary and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland.

Corresponding author: R. Scott Stephens, MD, Division of Pulmonary and Critical Care Medicine, Johns Hopkins University School of Medicine, 1830 Building, 5th Floor–Pulmonary, Baltimore, MD 21287 (e-mail: rsteph13 @jhmi.edu).


Despite its increasing use, the benefits of ECMO in ARDS have not been clearly established.


hypercapnia with acidemia (PCO2 > 50 mm Hg with

arterial pH < 7.15 or venous pH < 7.12); or (3) end-

inspiratory plateau pressure greater than 35 cm H2O

with either tidal volume no greater than 6 mL/kg

predicted body weight or with arterial pH less than

7.15.6,18 Eligibility criteria were confirmed by 2 of

the physician investigators. Time of enrollment

was defined as the time and date when criteria for

severe ARDS and ECMO eligibility first were met.

We additionally noted whether a contraindication to

ECMO was present, such as inability to receive anti-

coagulation (eg, because of intracranial hemorrhage,

major internal or gastrointestinal bleeding requiring

red blood cell transfusion within 7 days); high pres-

sure ventilation (plateau pressure > 30 cm H2O) for

longer than 7 days; high FIO2 (ie, FIO

2 > 0.80)

requirements for longer than 7 days; or any condi-

tion that would limit the likelihood of overall bene-

fit from ECMO (eg, severe neurologic dysfunction,

diffuse anoxic brain injury, end-stage lung disease

and not a transplant candidate, untreatable meta-

static cancer).6

For patients who met the study’s inclusion

criteria, we recorded the following data from

inclusion until time of discharge: ventilator set-

tings, laboratory results, length of stay, vital status

at hospital discharge, and use of adjunct therapies

such as prone positioning and neuromuscular

blockade. Comorbid conditions were defined as

the presence of 1 or more of the following chronic

medical conditions in the past year:

• Chronic lung disease (eg, use of home oxygen,

chronic hypercapnia [PaCO2 ≥ 50 mm Hg] severe pul-

monary hypertension [mean positive airway pres-

sure ≥ 40 mm Hg or New York Heart Association

class IV symptoms], chronic ventilator dependence);

• Chronic liver disease (ie, biopsy-proven cirrho-

sis, portal hypertension, hepatic failure based on

Child-Pugh score ≥ 10);

• Chronic renal disease requiring dialysis;

• Vasculitis;

• Sickle cell disease;

• malignant neoplasia (eg, leukemia, lym-

phoma, known metastatic cancer, or currently

receiving cancer treatment);

• AIDS;

• Receiving immunosuppressive therapy (eg,

treatment in the past year with agents such as aza-

thioprine, mycophenolate, tacrolimus, sirolimus,

cyclophosphamide, or corticosteroids [≥ 15 mg of

prednisone daily or equivalent for ≥ 20 days]); or

• Other conditions, such as Parkinson disease.

Severity of illness was assessed by using Acute Physi-

ology and Chronic Health Evaluation (APACHE) II

and Sequential Organ Failure Assessment scores.

We calculated the Respiratory ECMO Survival

Prediction (RESP) score19 for each patient to esti-

mate the probability of survival while receiving

ECMO had ECMO been used. The RESP score is a

validated model for predicting hospital survival for

respiratory failure based on patient and treatment

variables before ECMO is initiated.19

Descriptive statistics are reported as frequency

(percentage) for categorical variables and mean (stan-

dard deviation) or median (interquartile range [IQR])

for continuous variables. Statistical analyses were

conducted by using Stata, version 14 (StataCorp).

Results At the end of the 16-month screening period, 46

patients met criteria for severe ARDS and ECMO con-

sideration. Of these 46 patients, 5 (11%) actually

received rescue ECMO. These patients were younger,

more hypoxemic, and had fewer comorbid conditions

than the remaining 41 patients who did not receive

Table 1Demographics, clinical features, and outcomes of the 5 patients receiving ECMO

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; ECMO, extracorporeal membrane oxygenation; FIO2, fraction of inspired oxygen; ICU, intensive care unit; IQR, interquartile range; PEEP, positive end-expiratory pressure; Pplat, plateau pressure; PBW, predicted body weight; RESP, Respiratory Extracor-poreal Membrane Oxygenation Survival Prediction; SOFA, Sequential Organ Fail-ure Assessment.

Characteristic

Age, mean (SD), y

APACHE II score, mean (SD)

SOFA score, mean (SD)

RESP score, median (IQR)

Pneumonia, No. (%) of patients

Lowest PaO2/FIO2, median (IQR)

Highest FIO2, mean (SD)

Highest PEEP, median (IQR), cm H2O

Highest Pplat, median (IQR), cm H2O

Highest PCO2, mean (SD), mm Hg

Lowest pH, mean (SD)

Lowest tidal volume, mL/kg PBW

Mechanical ventilation before ECMO, mean (SD), d

Adjunct therapies, No. (%) of patients Neuromuscular blockade Prone positioning

ICU length of stay, median (IQR), d

Hospital length of stay, median (IQR), d

Mortality, No. (%) of patients

29 (7)

21.6 (2.3)

12.6 (3.2)

0 (-1 to 1)

4 (80)

49 (41-50)

100 (0)

18 (16-20)

44 (41-46)

102.2 (29.3)

7.15 (0.10)

3.76 (0.70)

6.4 (2.9)

5 (100)4 (80)3 (60)

8 (5-10)

30 (16-33)

4 (80)

Patients receiving ECMO


ECMO (Table 1). The hospital discharge survival rate

of these 5 patients who received ECMO was 20%.

Of the 41 patients treated without ECMO, the

mean age was 52.6 (SD, 16.0) years, 19 patients

(46%) were female, and 37 patients (90%) had at

least 1 comorbidity. At day 1 of admission to the

intensive care unit, the mean APACHE II score was

28.3 (SD, 7.5), and the mean Sequential Organ Fail-

ure Assessment score was 12.6 (SD, 4.0) (Table 2).

Of the 41 patients, 16 patients had at least 1 contra-

indication to ECMO, which included an inability to

receive anticoagulation (7 patients) or a nonrecover-

able comorbidity (9 patients: 3 had anoxic brain

injury, 2 had metastatic cancer, and 4 had end-stage

lung disease and were not transplant candidates).

No patient had a contraindication due to receiving

ventilatory support for longer than 7 days. Of this

group with contraindications to ECMO, the survival

rate at time of hospital discharge was 25%.

The remaining 25 patients met criteria for ECMO

but did not receive it (see Figure). The mean age of

these 25 patients was 53.5 (SD, 14.3) years, 56%

were female, and 84% had at least 1 major comor-

bidity. The most common comorbidities were chronic

liver disease and chronic lung disease (Table 3). In

most of these patients, ARDS developed in associa-

tion with pneumonia (64%). The mean APACHE II

score was 28.9 (SD, 7.4) and the mean Sequential

Organ Failure Assessment score was 13.2 (SD, 3.6)

on the first day of admission to the intensive care

unit. The median lowest PaO2/FIO

2 ratio for this group

was 66 (IQR, 58-93) mm Hg, with a concomitant

median FIO2 of 0.90 (IQR, 70%-100%) and positive

end-expiratory pressure of 14 (IQR, 12-18) cm H2O.

Most patients (64%) met criteria for ECMO consid-

eration because of isolated severe hypoxemia rather

than uncompensated hypercapnia (16%) or exces-

sive inspiratory pressures (0%). Five patients (20%)

met more than 1 criterion, having severe hypoxemia

combined with either hypercapnia or excessive

Characteristic

Table 2Demographics and clinical features of patients in the intensive care unit who met the criteria for severe ARDS and ECMO consideration

Age, mean (SD), y

Female sex, No. (%) of patients

APACHE II score, mean (SD)

SOFA score, mean (SD)

RESP score, median (IQR)

Pneumonia, No. (%) of patients

Lowest Pao2/Fio2, median (IQR)

Highest Fio2, mean (SD)

Highest PEEP, median (IQR), cm H2O

Highest Pplat, median (IQR), cm H2O

Highest Pco2, mean (SD), mm Hg

Lowest pH, mean (SD)

Lowest tidal volume, mL/kg PBW

51.2 (18.9)

5 (31.2)

27.4 (7.3)

11.6 (46)

-1 (-4 to 2)

8 (50)

67.5 (49.5-75)

95 (70-100)

12 (10.5-16)

34.5 (30.5-36.5)

70.5 (24.5)

7.16 (0.10)

5.56 (0.62)

53.5 (14.3)

14 (56)

28.9 (7.4)

13.2 (3.6)

1 (-2 to 2)

16 (64)

66 (58-93)

90 (70-100)

14 (12-18)

32 (30-36)

61 (20)

7.16 (0.12)

5.32 (0.77)

52.6 (16.0)

19 (46)

28.3 (7.5)

12.6 (4.0)

0 (-2 to 2)

24 (59)

61 (51-91)

90 (70-100)

14 (11-16)

33 (30-36)

67.5 (22.5)

7.16 (0.11)

5.41 (0.72)

Contraindication to ECMO (n = 16)

Eligible for ECMO(n = 25)

All patientsa

(n = 41)

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; ARDS, acute respiratory distress syndrome; ECMO, extracorporeal membrane oxygen-ation; FIO2, fraction of inspired oxygen; ICU, intensive care unit; IQR, interquartile range; PEEP, positive end-expiratory pressure; Pplat, plateau pressure; PBW, predicted body weight; RESP, Respiratory Extracorporeal Membrane Oxygenation Survival Prediction; SOFA, Sequential Organ Failure Assessment.a Includes all patients meeting criteria for severe ARDS and at least 1 criterion for ECMO consideration, subdivided into patients with a contraindication to

ECMO and patients who were eligible for ECMO but did not receive it. Specific demographics of the 5 patients who received ECMO are reported in Table 1.

Figure Flowchart of patients who met criteria for severe acute respiratory distress syndrome (ARDS) and extracorporeal mem-brane oxygenation (ECMO).

5 Patients received ECMO

46 Patients with severe ARDS met the ECMO criteria

25 Patients with no contraindication but did not receive ECMO

16 Patients with contraindication to ECMO


The predicted sur-vival rate with ECMO was 57%; the actual survival rate without

ECMO was 56%.

pressures. Most received lung protective ventilation

with a mean lowest tidal volume per kilogram of

predicted body weight of 5.3 (SD, 0.8). Two patients

who received tidal volumes greater than 6 mL/kg of

predicted body weight had isolated uncompensated

hypercapnia rather than severe

hypoxemia. Approximately one-

half of patients (48%) received

neuromuscular blockade, and 3

(12%) were placed in the prone

position. The median length of

stay in the intensive care unit was

13 (IQR, 9.5-26) days, and

median hospital stay was 21 (IQR,

13-35) days (Table 4). On the day

that the patient met the criteria for severe ARDS and at

least 1 criterion for ECMO, the median RESP score was

1 (IQR, -2 to 2), which corresponds to a predicted

survival rate while receiving ECMO of 57%.19 Our

group of 25 patients who met ECMO eligibility cri-

teria but did not receive ECMO had an actual sur-

vival rate of 56% at the time of hospital discharge.

Discussion Many institutions have developed ECMO pro-

grams for use in the treatment of severe ARDS

despite the lack of high-level evidence supporting

the assumption that the benefits of ECMO outweigh

the risks. With this prospective observational cohort

study, we sought to characterize the features and

outcomes of patients with severe ARDS who might

have received ECMO at other institutions but did not

receive ECMO at our hospital. To our knowledge, no

other prospective studies have been done to assess

the outcomes of patients with ARDS who met crite-

ria to receive ECMO but did not receive it.

Our data support several key conclusions. First,

the unselected population arriving in our tertiary

medical intensive care unit with severe ARDS is older

and has more comorbidities and higher illness sever-

ity scores than did the patients reported in prior case

series describing low mortality rates with the use of

ECMO for H1N1 influenza–related severe ARDS.9-12

Patients in these case series were relatively young,

lacked substantial comorbid conditions, and

potentially would have survived without ECMO.

In other case series evaluating outcomes in criti-

cally ill patients with H1N1 influenza, researchers

reported comparably low mortality rates when ECMO

was not used.16,17 In contrast, very few of our patients

with severe ARDS were young and had isolated

respiratory failure. A strength of our study is that

all critically ill patients with ARDS severe enough

to meet the criteria for ECMO consideration were

Characteristic

Table 3Comorbid conditions of patients in the intensive care unit meeting the criteria for severe acute respiratory distress syndrome (ARDS) and consideration for extracorporeal membrane oxygenation (ECMO)

None

Chronic liver diseaseb

Chronic lung diseasec

Chronic dialysis

Malignant neoplasiad

AIDS

Immunosuppressione

Sickle cell disease

Vasculitis

Otherf

0 (0)

5 (31)

6 (38)

1 (6)

2 (12)

0 (0)

1 (6)

1 (6)

0 (0)

1 (6)

4 (16)

12 (48)

6 (24)

2 (8)

1 (4)

4 (16)

4 (16)

1 (4)

1 (4)

0 (0)

4 (10)

17 (41)

12 (29)

3 (7)

3 (7)

4 (10)

5 (12)

2 (5)

1 (2)

1 (2)



No. (%) of patients

All patientsa

(n = 41)

a Includes all patients meeting criteria for severe ARDS and at least 1 criterion for ECMO consideration, subdivided into patients with a contraindication to ECMO and patients who were eligible for ECMO but did not receive it. Specific demographics of the 5 patients who received ECMO are reported in Table 1.

b Biopsy-proven cirrhosis, portal hypertension, hepatic failure based on Child-Pugh score 10.c Use of home oxygen, chronic hypercapnia PaCO2 50 mm Hg, severe pulmonary hypertension (mean pulmonary artery pressure 40 mm Hg or New York Heart

Association class IV symptoms), chronic ventilator dependence.d Leukemia, lymphoma, solid tumor currently receiving cancer treatment, or known metastatic cancer.e Treatment in the past year with immunosuppressive agents such as azathioprine, mycophenolate, tacrolimus, sirolimus, cyclophosphamide, or corticosteroids

( 15 mg of prednisone daily or equivalent for at least 20 days).f Parkinson disease.


included prospectively. As a result, our cohort was

substantially older, with more organ failures, and

was most likely more representative of a general

population with severe ARDS (Table 5).

Second, although the overall survival rate in the

cohort without ECMO in our study was 44%, which

compares unfavorably with published series of ECMO

in severe ARDS, more than one-third of our patients

with severe ARDS would have been considered ineli-

gible for ECMO owing to a contraindication such as

an inability to receive anticoagulation or a condition

that severely limited the likelihood of benefit from

ECMO (eg, metastatic cancer, anoxic brain injury).

This is a key finding, suggesting that many patients

with severe ARDS may not even be eligible for ECMO

support. Notably, these possibly ineligible patients

had a hospital survival rate of only 25%, which was

substantially lower than the survival rate of patients

without contraindications to ECMO. This low sur-

vival rate is not surprising, because the conditions

that make patients ineligible for ECMO, such as

cerebral hemorrhage, life-threatening bleeding, and

end-stage lung disease, frequently portend poor

prognoses regardless of the treatment strategy. Indeed,

in a recent retrospective analysis of patients with

interstitial lung disease treated with ECMO for acute

respiratory failure, authors reported a 7% survival rate

in those who did not qualify for lung transplant.20

Finally, among the patients who would have been

eligible but did not receive ECMO, we report a survival

Characteristic

Table 4Clinical course and survival rates at time of hospital discharge of patients in the intensive care unit meeting criteria for severe ARDS and ECMO consideration

Adjunct therapies, No. (%) of patients Neuromuscular blockade Prone positioning

ICU length of stay, median (IQR), d

Hospital length of stay, median (IQR), d

Survival, No. (%) of patients

5 (31) 5 (31)1 (6)

20 (6-33)

22 (14.5-43)

4 (25)

12 (48)12 (48) 3 (12)

13 (9.5-26)

21 (13-35)

14 (56)

17 (41)17 (41) 4 (10)

17.5 (7.5-27)

21 (13-35)

18 (44)



All patientsa

(n = 41)

Abbreviations: ARDS, acute respiratory distress syndrome; ECMO, extracorporeal membrane oxygenation; IQR, interquartile range.a Includes all patients meeting criteria for severe ARDS and at least 1 criterion for ECMO consideration, subdivided into patients with a contraindication to

ECMO and patients who were eligible for ECMO but did not receive it. Specific demographics of the 5 patients who received ECMO are reported in Table 1.

Reference

Table 5Comparison of demographics and mortality from severe ARDS in previously published case series

Davies et al9: Australia, New Zealand

Patroniti et al10: Italy

Noah et al11: United Kingdom

Pham et al12: France

Kumar et al17: Canada

Miller et al16: Utah

Peek et al7: CESAR intervention group

Peek et al7: CESAR, control group

Schmidt et al19: ELSO database

Johns Hopkinsa

23

29

28

36

17.3

27

37

45

43

44

56

63

55

59

147

61

75.9

75

NR

66

NR

NR

NR

NR

19.7

25

19.7

19.9

NR

28.9

NR

7

9

9.5

6.8

7

NR

NR

NR

13.2

36

39

34

42

32

34

39.9

40.4

41

53.5

Yes

Yes

Yes

Yes

Yes

Yes

No

No

No

No

100

100

86.3

100

4.2

0

75

0

100

0

68

49

80

123

168

47

90

90

2355

25

Mortality, %PaO2/FIO2

APACHE II score

SOFA score

Age, mean, y

H1N1 influenzaN

ECMO received, %

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; ARDS, acute respiratory distress syndrome; CESAR, conventional ventilatory support vs extracorporeal membrane oxygenation for severe adult respiratory failure; ELSO, Extracorporeal Life Support Organization; ECMO, extracorporeal membrane oxygenation; FIO2, fraction of inspired oxygen; NR, not reported; SOFA, Sequential Organ Failure Assessment.a Data reported here include the 25 patients who were eligible for ECMO but did not receive it, that is, patients with severe ARDS who met at least 1 ECMO

criterion and did not have a contraindication to ECMO.


Actual survival rates without ECMO

were similar to pre-dicted survival

rates with ECMO.

rate of 56%. This study was not a randomized trial;

therefore, we cannot directly compare this survival

rate with the survival rate obtained in patients receiv-

ing ECMO. However, to compensate for this limita-

tion, we calculated the RESP score for each patient

at the time they met the criteria for ECMO consider-

ation to estimate the predicted survival for our

patients had they received ECMO. Based on the

RESP score, predicted survival for our cohort receiv-

ing ECMO was 57%, indicat-

ing that the outcomes without

ECMO are similar to the out-

comes predicted for patients

who received ECMO. Moreover,

the survival rate of 56% deter-

mined in this study is similar

to the survival rate reported in

multiple clinical studies and

the Extracorporeal Life Support

Organization database (Table 5). Although caution

should be applied to any conclusions drawn from

this observational series, we suggest, on the basis

of the data, that the survival rate of patients with

severe ARDS treated without ECMO may be similar

to the survival rate of patients with severe ARDS who

receive ECMO.

Our institution is a high-volume center for treat-

ment of ARDS and use of ECMO. In the calendar

year 2016, more than 60 patients were treated with

ECMO. However, we use ECMO primarily for patients

after cardiotomy or who are experiencing cardiogenic

shock. In our medical intensive care unit, patients with

severe ARDS are typically treated with lung protective

ventilation, antibiotics for infections, conservative

fluid management, adjuncts such as neuromuscular

blockade and prone positioning, and supportive

care strategies. Fewer than half of the patients with

severe ARDS in the present study were treated with

neuromuscular blockade, and less than 10% were

treated with prone positioning, perhaps because

the implementation of these interventions took

time after their use was supported by findings in

landmark publications.21,22 Our implementation

rates of these adjunct therapies are similar to imple-

mentation rates across the world in 2016, as identi-

fied in the Large Observational Study to Understand

the Global Impact of Severe Acute Respiratory Fail-

ure (LUNG-SAFE) study.15 Furthermore, had these

proven therapies been used more widely, it is possi-

ble that our outcomes without ECMO would have

been even better.

This study has several limitations. Although data

were collected prospectively to capture all patients

who met the criteria for ECMO consideration, this is

an observational study of only 46 patients. As such,

no attempt was made to randomly assign patients

to receive ECMO or usual care. Most other ECMO

case series and case-control series are from the H1N1

influenza population; thus, it is difficult to draw

comparisons to our study. In addition, the RESP

score was developed on the basis of patients who

received ECMO, even though only pre-ECMO vari-

ables were included in the model. This score has

not been validated for patients who did not receive

ECMO, although it does appear to be a relevant

instrument for predicting survival in patients receiv-

ing ECMO for severe acute respiratory failure.

Finally, 5 patients in our sample did receive

ECMO. These patients were substantially younger,

but they had more severely impaired gas exchange

(mean PaO2/FIO

2 ratio, 52.6 mm Hg; PCO

2, 102.2

mm Hg), worse respiratory mechanics (median pla-

teau pressure, 44 [IQR, 41-46] cm H2O; mean tidal

volume, 3.76 [SD, 0.70] mL/kg predicted body

weight), and adjunctive therapies. The mortality rate

for the patients receiving ECMO was 80%, which is

higher than mortality rates reported for prior ECMO

series (see Table 5). In all 5 patients, ECMO was

offered as a late salvage therapy by the clinical team

caring for them. Three of the patients who received

ECMO had received mechanical ventilatory support

for at least 7 days, which could have been considered

a contraindication to ECMO.6 The high mortality rate

of these patients further emphasizes the need to iden-

tify the appropriate patient population for ECMO.

Conclusions The general patient population with severe ARDS

in this study was older and had more comorbid con-

ditions than the patient populations in many of the

case reports of favorable outcomes of severe ARDS

treated with ECMO. The reports of low mortality

rates in those case series could be attributable to the

patients’ young age and lack of comorbid conditions

rather than to the use of ECMO. Almost all of our

institution’s patients with severe ARDS were treated

medically, even if they were eligible for ECMO. Their

actual survival rate without ECMO was similar to their

predicted survival rates with ECMO. A new random-

ized controlled trial evaluating ECMO versus con-

ventional management for severe ARDS is currently

enrolling patients (ClinicalTrials.gov identifier:

NCT01470703). However, until reduced mortality

with ECMO is demonstrated in patients with acute

respiratory failure in future randomized controlled

trials, we suggest, on the basis of our results, that it


is reasonable to treat patients with severe ARDS with

usual best-care clinical practices without ECMO.

ACKNOWLEDGMENTSWe thank Bernice Frimpong and Karen Oakjones- Burgess for invaluable assistance with screening and data management.

FINANCIAL DISCLOSURESThis research was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health (Award T32HL007534).

SEE ALSO For more about ECMO, visit the Critical Care Nurse web-site, www.ccnonline.org, and read the Cochrane Review Summary by Sandoval, “Extracorporeal Membrane Oxygenation for Critically Ill Adults” (December 2016).

REFERENCES 1. Ventetuolo CE, Muratore CS. Extracorporeal life support in

critically ill adults. Am J Respir Crit Care Med. 2014; 190(5): 497-508.

2. Combes A, Leprince P, Luyt C-E, et al. Outcomes and long-term quality-of-life of patients supported by extracorporeal membrane oxygenation for refractory cardiogenic shock. Crit Care Med. 2008;36(5):1404-1411.

3. Brogan TV, Thiagarajan RR, Rycus PT, Bartlett RH, Bratton SL. Extracorporeal membrane oxygenation in adults with severe respiratory failure: a multi-center database. Inten-sive Care Med. 2009;35(12):2105-2114.

4. Zapol WM, Snider MT, Hill J, et al. Extracorporeal membrane oxygenation in severe acute respiratory failure: a random-ized prospective study. JAMA. 1979;242(20):2193-2196.

5. Morris AH, Wallace CJ, Menlove RL, et al. Randomized clini-cal trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2 removal for adult respiratory distress syndrome. Am J Respir Crit Care Med. 1994;149(2):295-305.

6. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J Med. 2011;365(20):1905-1914.

7. Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and eco-nomic assessment of conventional ventilatory support ver-sus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised con-trolled trial. Lancet. 2009;374(9698):1351-1363.

8. Kahn JM, Goss CH, Heagerty PJ, Kramer AA, O’Brien CR, Rubenfeld GD. Hospital volume and the outcomes of mechanical ventilation. N Engl J Med. 2006;355(1):41-50.

9. Davies A, Jones D, Bailey M, et al. Extracorporeal mem-brane oxygenation for 2009 influenza A(H1N1) acute respi-ratory distress syndrome. JAMA. 2009;302(17):1888-1895.

10. Patroniti N, Zangrillo A, Pappalardo F, et al. The Italian ECMO network experience during the 2009 influenza A(H1N1) pan-demic: preparation for severe respiratory emergency out-breaks. Intensive Care Med. 2011;37(9):1447-1457.

11. Noah MA, Peek GJ, Finney SJ, et al. Referral to an extracor-poreal membrane oxygenation center and mortality among patients with severe 2009 influenza A(H1N1). JAMA. 2011; 306(15): 1659-1668.

12. Pham T, Combes A, Rozé H, et al. Extracorporeal membrane oxygenation for pandemic influenza A(H1N1)–induced acute respiratory distress syndrome. Am J Respir Crit Care Med. 2013;187(3):276-285.

13. Zangrillo A, Biondi-Zoccai G, Landoni G, et al. Extracorporeal membrane oxygenation (ECMO) in patients with H1N1 influenza infection: a systematic review and meta-analysis including 8 studies and 266 patients receiving ECMO. Crit Care. 2013;17(1):R30.

14. Rubenfeld GD, Caldwell E, Peabody E, et al. Incidence and outcomes of acute lung injury. N Engl J Med. 2005; 353(16): 1685-1693.

15. Bellani G, Laffey JG, Pham T, et al. Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA. 2016;315(8):788-800.

16. Miller RR III, Markewitz BA, Rolfs RT, et al. Clinical findings and demographic factors associated with ICU admission in utah due to novel 2009 influenza A(H1N1) infection. Chest. 2010;137(4):752-758.

17. Kumar A, Zarychanski R, Pinto R, et al. Critically ill patients with 2009 influenza A(H1N1) infection in Canada. JAMA. 2009; 302(17):1872-1879.

18. Force ADT, Ranieri VM, Rubenfeld GD, et al. Acute respira-tory distress syndrome: the Berlin definition. JAMA. 2012; 307(23):2526-2533.

19. Schmidt M, Bailey M, Sheldrake J, et al. Predicting survival after extracorporeal membrane oxygenation for severe acute respiratory failure. The Respiratory Extracorporeal Membrane Oxygenation Survival Prediction (RESP) score. Am J Respir Crit Care Med. 2014;189(11):1374-1382.

20. Trudzinski FC, Kaestner F, Schäfers H-J, et al. Outcome of patients with interstitial lung disease treated with extracor-poreal membrane oxygenation for acute respiratory failure. Am J Respir Crit Care Med. 2015;193(5):527-533.

21. Guerin C, Reignier J, Richard JC, et al. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):2159-2168.

22. Papazian L, Forel JM, Gacouin A, et al. Neuromuscular block-ers in early acute respiratory distress syndrome. N Engl J Med. 2010;363(12):1107-1116.


Cardiovascular Critical Care


Background Intravenous fluid boluses are administered to patients in shock to improve tissue hypoperfusion. However, fluid boluses result in clinically significant stroke volume increases in only about 50% of patients. Hemodynamic responses to passive leg raising mea-sured with invasive and minimally invasive methods are accurate predictors of fluid responsiveness. However, few studies have used noninvasive blood pressure mea-surement to evaluate responses to passive leg raising.Objective To determine if passive leg raising–induced increases in pulse pressure or systolic blood pressure can be used to predict clinically significant increases in stroke volume index in healthy volunteers. Methods In a repeated-measures study, hemodynamic measurements were obtained in 30 healthy volunteers before, during, and after passive leg raising. Each partici-pant underwent the procedure twice.Results In the first test, 20 participants (69%) were respon-ders (stroke volume index increased by ≥ 15%); 9 (31%) were nonresponders. In the second test, 15 participants (50%) were responders and 15 (50%) were nonresponders. A passive leg raising–induced increase in pulse pressure of 9% or more predicted a 15% increase in stroke volume index (sensitivity, 50%; specificity, 44%). There was no association between passive leg raising–induced changes in systolic blood pressure and fluid responsiveness.Conclusion A passive leg raising–induced change in stroke volume index measured by bioreactance differenti-ated fluid responders and nonresponders. Pulse pressure and systolic blood pressure measured by oscillometric noninvasive blood pressure monitoring were not sensi-tive or specific predictors of fluid responsiveness in healthy volunteers. (American Journal of Critical Care. 2018; 27:228-237)

NONINVASIVE BLOOD PRESSURE MONITORING

AND PREDICTION OF FLUID

RESPONSIVENESS TO

PASSIVE LEG RAISINGBy Joya D. Pickett, RN, PhD, ARNP-CNS, CCNS, ACNS-BC, CCRN, Elizabeth Bridges, RN,

PhD, CCNS, Patricia A. Kritek, MD, EdM, and JoAnne D. Whitney, RN, PhD, CWCN



Static parameters such as central venous pressure are not predictive of fluid responsiveness.

Administration of intravenous fluid boluses is often one of the first interventions considered for patients in shock with signs of hypoperfusion.1-4 However, only about 50% of patients respond to fluid administration with a clinically significant (10%-15%) increase in stroke volume (SV) or stroke volume index (SVI).5-12 Administration of fluids to patients whose SV does not improve may exacerbate

pulmonary edema, precipitate respiratory failure, and prolong mechanical ventilation.12-15 Alternatively, undertreated hypovolemia may lead to inappropriate use of vasopressors and exacerbate organ hypoperfusion and ischemia.12 To avoid the deleterious effects associated with fluid overload and undertreated hypovolemia, it is important to predict which patients’ SV will increase in response to fluid administration.3,12

Traditionally, static parameters such as central

venous pressure have been used to guide fluid admin-

istration, but these parameters do not predict fluid

responsiveness.3,12,16,17 Dynamic parameters, such as

pulse pressure (PP) variation measured from an

arterial catheter, are highly predictive of response

to a fluid bolus.12,18,19 However, these indices can

be used only in patients who are fully supported

with mechanical ventilation and receiving ade-

quate tidal volumes (8 mL/kg) or patients without

cardiac arrhythmias.18

Passive leg raising (PLR) is a reliable alternative

method to predict fluid responsiveness in patients

who are spontaneously breathing, are receiving

mechanical ventilation, or have cardiac arrhyth-

mias.20 Semirecumbent PLR is performed by lifting

the legs to a 45° angle while lowering the head and

upper trunk from a 45° semirecumbent position to

the supine (flat) position.21 PLR causes a transient,

reversible autotransfusion, temporarily increasing

preload and thus mimicking a fluid bolus.13,22-24 The

PLR-induced increase in preload will induce a

clinically significant increase in SV if both the right

and left ventricles are functioning on the ascending

portion of the Frank-Starling curve.12 If a patient

responds to PLR with an increase in SV or its surro-

gates, the patient would most likely respond to a

fluid bolus.

Rapid fluid administration is recommended

during the first few hours after onset of symptoms in

patients with sepsis and septic shock.1,2,4 PLR-induced

changes in SV, PP, and systolic blood pressure (SBP)

measured with direct arterial monitoring are accurate

indicators of fluid responsiveness.6,8,20,23,25 However,

these parameters require invasive

monitoring. Oscillometric nonin-

vasive blood pressure (NIBP)

monitoring is readily available at

the bedside and is often the initial

method used to measure the

response to fluid administration.

In only 1 previous study have

PLR-induced changes in PP and

SBP as determined by NIBP moni-

toring been evaluated.9 However, the study included

central intravenous catheter placement, which may

delay initial treatment. The purpose of our study was

to determine if PLR-induced changes in PP and SBP

measured by oscillometric NIBP monitoring are sen-

sitive and specific indicators of a clinically significant

increase in SVI in healthy volunteers.

Methods Design

The study was a single-group repeated-measures

design. We measured the following parameters

noninvasively before, during, and after PLR: SBP,

diastolic blood pressure (DBP), mean arterial pres-

sure (MAP), heart rate, SVI, and cardiac index. We

repeated PLR with the same participants after a

5-minute washout period.

SampleWe enrolled a convenience sample of 30 vol-

unteers who reported no major health problems

About the AuthorsJoya D. Pickett is a critical care clinical nurse specialist at Swedish Medical Center, Seattle, Washington. Elizabeth Bridges is a professor at University of Washington School of Nursing and a clinical nurse researcher at University of Washington Medical Center, University of Washington Medicine, Seattle, Washington. Patricia A. Kritek is medi-cal director of critical care and an associate professor in the Division of Pulmonary and Critical Care Medicine at University of Washington Medical Center. JoAnne D. Whitney is a professor at the University of Washington School of Nursing and endowed professor in critical care nursing at Harborview Medical Center, University of Washington Medicine.

Corresponding author: Joya D. Pickett, Swedish Medical Center, 747 Broadway, Seattle, WA 98122 (email: [email protected]).


SVI was measured by using a noninva-

sive continuous car-diac monitoring

system that uses bio-reactance technology.

(Figure 1). We included or excluded candidates on

the basis of their responses to a health history ques-

tionnaire. Candidates included were aged 18 years

or older and had a regular pulse rhythm, pulse rate

of 60/min to 120/min, and SBP of 90 to 160 mm

Hg. Exclusion criteria were self-reported history of

pulmonary edema, mitral or

aortic stenosis, cardiac dysrhyth-

mias, peripheral vascular disease,

musculoskeletal deformities (eg,

limb amputations), implanted

devices (eg, pacemakers), preg-

nancy, and inability to lie fl at

with the legs elevated.

For the power analysis, we

used the results of a meta-

analysis of PLR-induced changes

in radial artery PP to predict

fl uid responsiveness.20 With an level of .05 and

level of .80, a sample size of 30 participants would

provide ± 15% precision in measurement of the

response to PLR.

Variables and Measurement InstrumentsBlood Pressure Measurements. We measured SBP,

DBP, MAP, and heart rate from the brachial artery

by using oscillometric NIBP monitoring (Critikon

Dinamap, GE Healthcare). Oscillometric NIBP

monitoring is accurate in a wide variety of clinical

situations.26 We used appropriately sized adult (arm

circumference 27-34 cm) and extra large (arm circum-

ference > 45 cm) blood pressure (BP) cuffs.

Stroke Volume Index. We measured SVI with a

noninvasive continuous cardiac output monitoring

system that uses bioreactance technology (NICOM,

Cheetah Medical).5 This system involves 4 electrodes

placed on the thorax. An alternating electric current

is passed between the outer pair of electrodes, and

the resulting voltage signal is sensed by the inner pair

of electrodes. Comparison of phase shifts between

the current and the voltage signal provides an instan-

taneous recording that is proportional to aortic blood

fl ow. The frequency shift is used to determine the SVI.

The signal is averaged every 10 seconds and recorded

as a digital display.5

The accuracy of this monitoring system has been

demonstrated in various patient populations. Accord-

ing to Critchley and Critchley,27 acceptance of a new

technique should require a limit of agreement of up

to 30% between 2 devices. A study of cardiac output

monitoring in 110 critically ill patients revealed an

error rate of 9% to 20% (depending on whether

cardiac output was stable or increasing) between

continuous measurement with the noninvasive

bioreactance device and measurement with ther-

modilution via a pulmonary artery catheter.28

Procedure We obtained approval from our institution’s

Human Subjects Division. Participants gave consent,

and we conducted the study in accordance with the

ethical standards of the Declaration of Helsinki.

Before each test, the noninvasive cardiac output

monitoring system underwent internal calibration.

We prepared the skin on each participant’s anterior

chest and abdomen according to the monitoring

system manufacturer’s instructions. We placed elec-

trodes on the thorax (2 on the shoulders, 2 on the

abdomen). We obtained NIBP measurements in

each arm and used the reading from the arm with

the highest SBP.29 We positioned the arm at the

level of the phlebostatic axis.30

The procedure began after the participant had

undergone a 10-minute stabilization period in the

baseline position (supine with the backrest elevated

45° by a wedge pillow and legs horizontal on the

bed). We then moved the participant into the PLR

position by removing the wedge pillow from the

upper torso and lowering the head and torso to hori-

zontal while simultaneously elevating the legs to a

45° angle with the support of the wedge pillow. We

Figure 1 CONSORT diagram representing the study enrollment and description of participants.

Assessed for study eligibility

N = 31

Total participants N = 30

Test A: n = 29(1 participant’s sensor

detached)

Responders n = 20

Nonresponders n = 9

Excluded n = 1

Test B: n = 30 (All participants repeated

and completed test)

Responders n = 15

Nonrespondersn = 15


Figure 2 Diagram of procedure for passive leg raising.

Abbreviations: HOB, head of bed; PLR, passive leg raise.

3 min 3 min 3 min 5-min washout

Baseline During PLR After PLR

HOB45°

Legs45°

HOB45°

Characteristic

Table 1Characteristics of study participants

40 (14)35 (13)

12 (63) 7 (37)

3 (27) 8 (73)

67 (4)69 (3)

178 (51)184 (42)

1.90 (0.24)1.98 (0.22)

38 (14)38 (15)

11 (61) 7 (39)

9 (82) 2 (18)

68 (3)68 (4)

185 (42)163 (45)

1.96 (0.19)1.86 (0.29)

Age, mean (SD), y Responder Nonresponder

Sex, No. (%) Female, 19 (63) Responder Nonresponder Male, 11 (37) Responder Nonresponder

Height, mean (SD), in Responder Nonresponder

Weight, mean (SD), lb Responder Nonresponder

Body surface area, mean (SD), m2

Responder Nonresponder

Test B (n = 30) Test A (n = 29)

SI conversion factors: To convert height to centimeters, multiply by 2.54; to convert weight to kilograms, multiply by 0.45.

activated the NIBP readings immediately after plac-

ing the participant in this position. After the

3-minute PLR procedure was complete, we returned

the participant to the baseline position. The partici-

pant walked in place for a washout period of 5 min-

utes. We then repeated the procedure, designating the

first procedure as test A and the second as test B (Fig-

ure 2). Hemodynamic measurements were obtained

at baseline, immediately after the participant was

placed in the PLR position, and after the PLR proce-

dure. During each phase we took 3 measurements

at approximately 1 minute, 1 minute 40 seconds,

and 2 minutes 20 seconds.

Statistical Analysis We used descriptive statistics to summarize

the sample demographics. Continuous data were

expressed as means (SD). On the basis of a litera-

ture review, we classified each participant a priori

as a fluid responder if the SVI increased by 15% or

more in response to PLR.3,20,22 If the PLR-induced

change in SVI was less than 15%, we classified the

participant as a fluid nonresponder. We calculated

the PLR-induced increase in SVI according to the

proprietary protocol of the noninvasive continuous

cardiac output monitoring device.

We used the mean and standard deviation of

3 repeated measurements to calculate the variables

(eg, PP, SBP, heart rate). The PP was calculated by

subtracting DBP from SBP. We used paired t tests to

compare hemodynamic measurements before and

after the PLR procedure. We used the independent

t test to compare the measurements of responders

and nonresponders. We used the 2 statistic to com-

pare responders and nonresponders according to

SVI, PP, and SBP.

We analyzed the sensitivity and specificity of

PLR-induced changes by using cutoff values based

on a priori studies of fluid responsiveness: PP change

of 9% or greater20 and SBP change of 9% or greater.9

We also evaluated post hoc cutoff values by using

the Youden index, a statistic used to weigh the pro-

portion of false-positive and false-negative results

to identify the optimum cutoff point.31 We analyzed

the data with statistics software (SPSS version 19,

IBM). All tests of significance were 2-tailed, and we

set the level of significance at .05.

Results Sample

We enrolled 30 healthy volunteers who met the

inclusion criteria. All participants completed the study

without complication. We excluded 1 potential

participant from the study because of an implanted

device. In test A, 1 participant’s noninvasive cardiac

output monitoring sensor detached and the response

to PLR could not be measured. The baseline charac-

teristics of responders and nonresponders were not

significantly different (Table 1).

PLR-Induced Changes in SVI Of 29 participants in test A, 20 (69%) were

responders and 9 (31%) were nonresponders. Of 30

participants in test B, 15 (50%) were responders and

15 (50%) were nonresponders. In test A, the mean

(SD) PLR-induced change in SVI was 26% (10%) in

responders and 7% (6%) in nonresponders (P < .001).

The results of test B were similar. Tables 2 and 3 show

the hemodynamic variables of responders and non-

responders in tests A and B.


Prediction of Fluid Responsiveness: PP In test A, the PLR-induced change in PP was not

significantly different between responders and non-

responders (mean [SD]: responders, 9% [11%]; non-

responders, 8% [11%]; P = .90). However, in responders

the PLR-induced absolute increase in PP was signifi-

cantly higher than the baseline PP (mean [SD]: base-

line, 47 [11] mm Hg; during PLR, 51 [13] mm Hg;

P < .001). SVI increased in all but 1 participant. How-

ever, the association between the change in PP and

change in SVI was not significant (r = 0.03, P = .88).

Using the a priori threshold of a 9% or greater

change in PP, we calculated that the sensitivity of

the PLR-induced increase in PP in test A was 50%

(95% CI, 27%-73%) and specificity was 44% (95%

CI, 14%-79%). The area under the curve was 0.49

(95% CI, 0.25-0.74; P = .96), indicating that the

PLR-induced change in PP did not differentiate

between responders and nonresponders. Analysis

using the Youden index cutoff of 6% yielded a

sensitivity of 70% and a specificity of 44%. We

found similar results in test B (Table 4). Analysis

using the a priori threshold of a 9% or greater

increase in PP yielded a sensitivity of 47% and a

specificity of 67%. Using the Youden index cutoff of

2% revealed a sensitivity of 60% and a specificity of

46%. The PLR-induced changes in SVI and PP were

not associated (r = 0.10, P = .61).

Prediction of Fluid Responsiveness: SBP In both tests, SBP decreased from baseline

values in both responders and nonresponders. The

PLR-induced changes in SBP were similar in test A

(mean [SD]: responders, −4% [4%]; nonrespond-

ers, −4% [6%]; P = .99) and test B (mean [SD]:

responders, −4% [4%]; nonresponders, −6 %

[4%]; P = .12). The PLR-induced increase in SVI

and the change in SBP were not associated in test

A (r = 0.12, P = .53) or in test B (r = 0.23, P = .17).

Figure 3 depicts test A results.

Variable

Table 2Hemodynamic variables of participants classified as responders (n = 20) and nonresponders (n = 9) in test Aa

Heart rate, beats per minute Responder Nonresponder

Systolic blood pressure, mm Hg Responder Nonresponder

Diastolic blood pressure, mm Hg Responder Nonresponder

Mean arterial pressure, mm Hg Responder Nonresponder

Pulse pressure, mm Hg Responder Nonresponder

Cardiac indexe


Stroke volume indexh


69 (9)64 (11)

114 (16)115 (16)

68 (8)71 (9)

84 (10)87 (10)

47 (11)44 (12)

3.4 (0.60)3.6 (0.38)

53 (8) 57 (11)

-1 (7)-1 (7)

-4 (4)-4 (6)

-12 (6)-13 (6)

-12 (6)-11 (6)

9 (11)8 (11)

26 (12)g

7 (6)

26 (10)g

7 (6)

69 (10)64 (11)

113 (18)111 (15)

66 (7)69 (8)c

84 (10)84 (12)d

47 (13)43 (11)

3.8 (0.64)f

5.0 (4.0)

55 (9) 59 (12)

68 (10)63 (12)

109 (17)b

109 (15)

58 (6)b

62 (11)b

74 (10)b

77 (13)b

51 (13)b

47 (11)

4.3 (0.64)b

3.9 (0.47)

63 (10)i 62 (14)

Baseline During PLR After PLR % Change in response

to PLR

Abbreviation: PLR, passive leg raising. a Data are expressed as mean (SD).b During PLR versus baseline: P = .001. c After PLR versus baseline: P = .04.d After PLR versus baseline: P = .05.e Calculated as cardiac output in liters per minute divided by body surface area in square meters.f After PLR versus baseline: P = .003.g Responder versus nonresponder: P < .001.h Calculated as stroke volume in milliliters per beat divided by body surface area in square meters.i During PLR versus baseline: P = .002.

www.ajcconline.org

Table 4Accuracy of pulse pressure for evaluating noninvasive blood pressure measurements of the response to passive leg raising

Statistic

0.61

9

47

67

1.40

0.80

58

56

0.49

9

50

44

0.90

1.12

67

29

Area under curve

Cutoff, %

Sensitivity, %

Specificity, %

Positive likelihood ratio

Negative likelihood ratio

Positive predictive value, %

Negative predictive value, %

Test B Test A

% Change in pulse pressure

Using the a priori threshold of a 9% or greater

PLR-induced increase in SBP, we calculated that the

change in SBP was not predictive of fluid responsive-

ness (area under the curve, 0.42; 95% CI, 0.18-0.66;

P = .48). Test B yielded similar results.

PLR-Induced Changes in Other Hemodynamic Variables

The responses of other hemodynamic variables

are summarized in Tables 2 and 3. The only signifi-

cant change was in the cardiac index. In both test A

and test B, the PLR-induced increase in cardiac index

in responders was significantly higher than in non-

responders (P < .001).

Variable

Table 3Hemodynamic variables of participants classified as responders (n = 15) and nonresponders (n = 15) in test Ba

Heart rate, beats per minute Responder Nonresponder

Systolic blood pressure, mm Hg Responder Nonresponder

Diastolic blood pressure, mm Hg Responder Nonresponder

Mean arterial pressure, mm Hg Responder Nonresponder

Pulse pressure, mm Hg Responder Nonresponder

Cardiac indexi


Stroke volume indexm


69 (13) 64 (9)

109 (12)c

122 (17)

65 (7) 69 (7)

81 (8)f

88 (10)

44 (10)g

53 (13)

3.3 (0.80)3.9 (0.46)

47 (9)n

59 (11)

71 (13)66 (10)b

109 (12)f

121 (17)

65 (6)70 (8)

81 (9)88 (11)

44 (10)52 (14)

3.6 (0.81)k

3.8 (0.47)

52 (11)o 58 (11)

68 (11)64 (9)

105 (12)d

114 (15)e

58 (5)d

60 (9)d

73 (6)d

78 (11)d

47 (10)h 53 (11)

4.1 (0.82)j

4.0 (0.43)

61 (12)d

63 (11)

Baseline During PLR After PLR % Change in response

to PLR

Abbreviation: PLR, passive leg raising.a Data are expressed as mean (SD). b After PLR versus baseline: P = .045.c Responder versus nonresponder: P = .03. d During PLR versus baseline: P = .001. e During PLR versus baseline: P < .001.f Responder versus nonresponder: P = .04. g Responder versus nonresponder: P = .05. h During PLR versus baseline: P = .04.i Calculated as cardiac output in liters per minute divided by body surface area in square meters.j During PLR versus baseline: P = .05.k After PLR versus baseline: P = .001.l Responder versus nonresponder: P < .001. m Calculated as stroke volume in milliliters per beat divided by body surface area in square meters.n Responder versus nonresponder: P = .004. o After PLR versus baseline: P < .001.

-0.59 (6) -0.15 (4)

-4 (4) -6 (4)

-11 (6) -13 (9)

-10 (6) -11 (7)

7 (10) 3 (9)

25 (12)l

3 (7)

31 (12)l 5 (6)


Oscillometric NIBP monitoring has an

inherent time lag . . . thus missing the

peak effect of the PLR maneuver.

Discussion The main fi nding of our study is that PP measured

by NIBP monitoring is not a sensitive or specifi c pre-

dictor of fl uid responsiveness in

healthy volunteers. Approxi-

mately 50% of the participants

in our study had a 15% or

greater PLR-induced increase in

SVI, consistent with the result

of a previous study of healthy

volunteers.32 Participants’ SBP

decreased in response to PLR in

both of our tests and could not

be used to detect fl uid respon-

siveness. The general decrease in

SBP from baseline to maneuver is contrary to the

expected response to PLR.

The PLR-induced decreases in SBP, DBP, and MAP

in our study are congruent with the results of other

studies in healthy volunteers.32-36 In a study of the

effects of PLR on central hemodynamics in 50 healthy

volunteers, central aortic pressures (measured with

radial artery applanation tonometry) and brachial

artery SBP, DBP, MAP, and PP (all measured with

NIBP monitoring) decreased 1 minute after PLR

performed with 60° leg and head elevations.33 These

fi ndings suggest brachial artery dilation. Researchers

in several other studies have noted brachial artery

dilation in healthy volunteers in response to PLR.37-40

Various researchers have concluded that the decrease

in BP is most likely due to activation of low-pressure

baroreceptors as a result of intrathoracic blood

pooling,37,38,41 which inhibits sympathetic vasomo-

tor discharge (causing a decrease in vascular tone),

or activation of arterial baroreceptors in addition to

low-pressure baroreceptors.42,43 Moreover, changes

in arterial tone maintain a constant MAP despite vari-

ation in cardiac output.42,44 Therefore, PLR may cause

simultaneous but opposing refl exive responses result-

ing in an increase in cardiac output and simultaneous

dilation of peripheral arteries, which combined may

contribute to the BP response.41

Responses to PLR may differ in critically ill

patients receiving mechanical ventilation and in

healthy volunteers. The results of our study con-

trast with those of a meta-analysis of PLR that

included direct arterial BP measurements in criti-

cally ill patients receiving vasopressors.20 In the

meta-analysis, pooled data from 4 studies showed

that a PLR-induced increase in direct radial artery

PP (threshold, 9%-12%) was predictive of a response

to a fl uid bolus with a sensitivity of 60% (95% CI,

47%-71%) and a specifi city of 86% (95% CI, 75%-

94%). In a study of 39 critically ill patients, Boulain

et al23 concluded that the arteries of the upper

limbs dilate in response to PLR. However, because

of increases in positive intrathoracic pressure, mechan-

ical ventilation reduces the stretch of the baroreceptors

in the pulmonary vessels, thus attenuating barorecep-

tor stimulation and the subsequent arterial dilation.

Similarly, Delerme et al36 and Geerts et al45 suggested

that a supine PLR technique (lifting the legs pas-

sively from horizontal to a 30° to 45° elevation

while the head and upper torso remain fl at) main-

tains the level of the heart and baroreceptors and

thus may limit baroreceptor activation. Because the

participants in our study were spontaneously breath-

ing and not receiving mechanical ventilation, the

increase in vasodilation may have been relatively

greater, resulting in a decrease in BP in comparison

with patients receiving mechanical ventilation. Bou-

lain et al23 also suggested that in patients in shock,

administration of -adrenergic catecholamines, which

Figure 3 Relationship between percentage changes in stroke volume index (SVI) and systolic blood pressure induced by passive leg raising (PLR) in test A. The horizontal line rep-resents the a priori threshold of a 15% increase in SVI. Partici-pants with PLR-induced increases in SVI over this threshold were classifi ed as responders (triangles). Participants with PLR-induced increases in SVI below this threshold were classi-fi ed as nonresponders (asterisks). The vertical line represents the a priori threshold of a 9% change in SBP induced by PLR.

-20 -10

-10

0

10

20

30

40

50

0 10 20

Test A, % change in SBP

Test

A, %

ch

ang

e in

SV

I


PLR-induced changes in oscillometric NIBP measurements of PP and SBP are not recom-mended for predicting fluid responsiveness in healthy volunteers.

result in venous vasoconstriction, may shift venous

blood from an unstressed to a stressed volume and

thus amplify the preload augmentation of the PLR

response.

Another explanation for the differences in results

may be the limitations of NIBP monitoring. NIBP

measurements can be obtained only after the BP cuff

is inflated to approximately 160 to 200 mm Hg, with

variable peak inflation pressure depending on the

participant’s initial SBP reading. This constraint may

create a lag time and variability between the peak

effect of the PLR maneuver and the time of mea-

surement. If this occurs, the PLR peak effect may be

missed, potentially causing a false-negative response.

Additionally, oscillometric NIBP monitoring devices

directly measure the MAP and extrapolate the SBP

and DBP. PP is determined by both SBP and DBP,

both of which are extrapolated measurements, poten-

tially increasing measurement error in PP.

The noninvasive cardiac output monitoring

device that we employed uses bioreactance to

measure responses to PLR. Validation studies have

found the device to be accurate and reliable,28,46

and another study validated the device’s ability to

measure the response to PLR.47 Researchers in 1

study concluded that bioreactance was not reliable

for estimating cardiac output in response to PLR in

critically ill patients.48 In that study, investigators

averaged 3 thermodilution boluses from the ref-

erence method and compared the result with 1

instantaneous bioreactance value. In other stud-

ies, satisfactory concordance was observed when

10 minutes’ worth of bioreactance values were com-

pared with the results of thermodilution boluses.28,46,47

Averaging cardiac output data over 10 minutes adjusts

for the differences in time responses between devices

and most likely explains much of the difference

in the study findings.49

Nursing ImplicationsFluid responsiveness alone is not an indication

to administer fluid boluses.50 Fluid resuscitation is

indicated in patients with evidence of inadequate

tissue perfusion (eg, MAP ≤ 65 mm Hg, increased

serum lactate level1,50,51). Nurses must identify the

need for fluid resuscitation as well as a patient’s

likelihood of increasing SV in response to fluid

administration (fluid responsiveness).

The results of 2 systematic reviews and meta-

analyses suggested that PLR-induced changes in SV

and its surrogates are reliable predictors of fluid

responsiveness.20,25 The PLR-induced change in the

invasive radial artery PP is a weaker, but also signifi-

cant, predictor of fluid responsiveness. These

results were independent of the type of measure-

ment device used (eg, esophageal Doppler imag-

ing, echocardiography, bioreactance) to measure

the PLR-induced change. However, the results of

our study in healthy volunteers do not support the

use of PLR-induced changes in NIBP measure-

ments. Thus, the use of NIBP monitoring as a part

of PLR is not recommended.

Study Limitations and Future Research A limitation of this study is that it was not per-

formed in critically ill patients and therefore conclu-

sions cannot be generalized to the critically ill. We

did not measure other stressors, such as pain, that

may change the cardiopulmonary response to PLR.

Measuring these variables

may have helped to distin-

guish causes of the change

in hemodynamic variables

in contrast to the PLR-

induced changes.

Future studies need to

focus on the population of

critically ill patients. Other

considerations for future

research include compari-

son studies of participants

with different types of

shock (eg, septic vs hemor-

rhagic shock). Few studies in critically ill patients

have used a 60° leg elevation, which may increase

diagnostic accuracy in patients with certain conditions

(eg, hypovolemia). Whether the volume status of a

patient changes the volume of autotransfusion by

PLR is unknown.45 Studies comparing normovolemic

with hypovolemic patients may increase the under-

standing of differences between conditions.

Conclusion PLR-induced changes in oscillometric NIBP mea-

surements of PP and SBP were not sensitive or specific

predictors of fluid responsiveness in healthy volun-

teers and are not recommended.

FINANCIAL DISCLOSUREThis study received partial support from the University of Washington School of Nursing Hester McLaws Nursing Scholarship Award.

SEE ALSO For more about passive leg raising, visit the Critical Care Nurse website, www.ccnonline.org, and read the article by Pickett et al, “Passive Leg-Raising and Predic-tion of Fluid Responsiveness: Systematic Review” (April 2017).


REFERENCES1. Dellinger RP, Levy MM, Rhodes A, et al; Surviving Sepsis

Campaign Guidelines Committee Including the Pediatric Subgroup. Surviving Sepsis Campaign: international guide-lines for managment of severe sepsis and septic shock, 2012. Intensive Care Med. 2013;39(2):165-228.

2. Marik PE. Early management of severe sepsis. Chest. 2014; 145(6):1407-1418.

3. Marik PE. Techniques for assessment of intravascular vol-ume in critically ill patients. J Intensive Care Med. 2009; 24(5): 329-337.

4. Surviving Sepsis Campaign. Updated bundles in response to new evidence. http://www.survivingsepsis.org/SiteCol-lectionDocuments/SSC_Bundle.pdf. Updated April 2015. Accessed February 3, 2018.

5. Marik PE, Levitov A, Young A, Andrews L. The use of biore-actance and carotid Doppler to determine volume respon-siveness and blood flow redistribution following passive leg raising in hemodynamically unstable patients. Chest. 2013;143(2):364-370.

6. Préau S, Saulnier F, Dewavrin F, Durocher A, Chagnon JL. Passive leg raising is predictive of fluid responsiveness in spontaneously breathing patients with severe sepsis or acute pancreatitis. Crit Care Med. 2010;38(3):819-825.

7. Biais M, Vidil L, Sarrabay P, Cottenceau V, Revel P, Sztark F. Changes in stroke volume induced by passive leg raising in spontaneous breathing patients: comparison between echocardiography and Vigileo/FloTrac device. Crit Care. 2009; 13(6):R195.

8. Lakhal K, Ehrmann S, Runge I, et al. Central venous pressure measurements improve the accuracy of leg raising-induced change in pulse pressure to predict fluid responsiveness. Intensive Care Med. 2010;36(6):940-948.

9. Lakhal K, Ehrmann S, Benzekri-Lefèvre D, et al. Brachial cuff measurements of blood pressure during passive leg raising for fluid responsiveness prediction. Ann Fr Anesth Reanim. 2012;31(5):e67-72.

10. Monnet X, Bleibtreu A, Ferré A, et al. Passive leg-raising and end-expiratory occlusion tests perform better than pulse pressure variation in patients with low respiratory system compliance. Crit Care Med. 2012;40(1):152-157.

11. Maizel J, Airapetian N, Lorne E, Tribouilloy C, Massy Z, Slama M. Diagnosis of central hypovolemia by using pas-sive leg raising. Intensive Care Med. 2007;33(7):1133-1138.

12. Marik PE, Monnet X, Teboul JL. Hemodynamic parameters to guide fluid therapy. Ann Intensive Care. 2011;1(1):1.

13. Thiel SW, Kollef MH, Isakow W. Non-invasive stroke vol-ume measurement and passive leg raising predict volume responsiveness in medical ICU patients: an observational cohort study. Crit Care. 2009;13(4):R111.

14. Upadya A, Tilluckdharry L, Muralidharan V, Amoateng-Adjepong Y, Manthous CA. Fluid balance and weaning outcomes. Intensive Care Med. 2005;31(12):1643-1647.

15. Rosenberg AL, Dechert RE, Park PK, Bartlett RH; NIH NHLBI ARDS Network. Review of a large clinical series: association of cumulative fluid balance on outcome in acute lung injury: a retrospective review of the ARDSnet trial volume study cohort. J Intensive Care Med. 2009;24(1):35-46.

16. Bridges EJ. Arterial pressure-based stroke volume and func-tional hemodynamic monitoring. J Cardiovasc Nurs. 2008;23(2):105-112.

17. Marik PE, Cavallazzi R. Does the central venous pressure predict fluid responsiveness? An updated meta-analysis and a plea for some common sense. Crit Care Med. 2013; 41(7): 1774-1781.

18. Marik PE, Cavallazzi R, Vasu T, Hirani A. Dynamic changes in arterial waveform derived variables and fluid responsiveness in mechanically ventilated patients: a systematic review of the literature. Crit Care Med. 2009;37(9):2642-2647.

19. DeBacker D, Heenen S, Piagnerelli M, Koch M, Vincent JL. Pulse pressure variations to predict fluid responsiveness: influence of tidal volume. Intensive Care Med. 2005;31(4): 517-523.

20. Cavallaro F, Sandroni C, Marano C, et al. Diagnostic accuracy of passive leg raising for prediction of fluid responsiveness in adults: systematic review and meta-analysis of clinical studies. Intensive Care Med. 2010;36(9):1475-1483.

21. Jabot J, Teboul JL, Richard C, Monnet X. Passive leg raising for predicting fluid responsiveness: importance of the pos-tural change. Intensive Care Med. 2009;35(1):85-90.

22. Monnet X, Rienzo M, Osman D, et al. Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med. 2006;34(5):1402-1407.

23. Boulain T, Achard JM, Teboul JL, Richard C, Perrotin D, Ginies G. Changes in BP induced by passive leg raising predict response to fluid loading in critically ill patients. Chest. 2002;121(4):1245-1252.

24. Teboul JL, Monnet X. Prediction of volume responsiveness in critically ill patients with spontaneous breathing activity. Curr Opin Crit Care. 2008;14(3):334-339.

25. Cherpanath TG, Hirsch A, Geerts BF, et al. Predicting fluid respon-siveness by passive leg raising: a systematic review and meta-analysis of 23 clinical trials. Crit Care Med. 2016; 44(5): 981-991.

26. Lakhal K, Ehrmann S, Runge I, et al. Tracking hypotecxnsion and dynamic changes in arterial blood pressure with bra-chial cuff measurements. Anesth Analg. 2009;109(2):494-501.

27. Critchley LA, Critchley JA. A meta-analysis of studies using bias and precision statistics to compare cardiac output mea-surement techniques. J Clin Monit Comput. 1999; 15(2): 85-91.

28. Squara P, Denjean D, Estagnasie P, Brusset A, Dib JC, Dubois C. Noninvasive cardiac output monitoring (NICOM): a clinical validation. Intensive Care Med. 2007;33(7):1191-1194.

29. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Pro-fessional and Public Education of the American Heart Asso-ciation Council on High Blood Pressure Research. Circulation. 2005;111(5):697-716.

30. Obtaining accurate noninvasive blood pressure measure-ments in adults. Crit Care Nurse. 2016;36(3):e12-16.

31. Fluss R, Faraggi D, Reiser B. Estimation of the Youden index and its associated cutoff point. Biom J. 2005;47(4):458-472.

32. Keller G, Cassar E, Desebbe O, Lehot JJ, Cannesson M. Ability of pleth variability index to detect hemodynamic changes induced by passive leg raising in spontaneously breathing volunteers. Crit Care. 2008;12(2):R37.

33. Kamran H, Salciccioli L, Gusenburg J, et al. The effects of passive leg raising on arterial wave reflection in healthy adults. Blood Press Monit. 2009;14(5):202-207.

34. Kamran H, Salciccioli L, Kumar P, et al. The relation between blood pressure changes induced by passive leg raising and arterial stiffness. J Am Soc Hypertens. 2010;4(6):284-289.

35. Delerme S, Renault R, Le Manach Y, et al. Variations in pulse oximetry plethysmographic waveform amplitude induced by passive leg raising in spontaneously breathing volunteers. Am J Emerg Med. 2007;25(6):637-642.

36. Delerme S, Castro S, Freund Y, et al. Relation between pulse oximetry plethysmographic waveform amplitude induced by passive leg raising and cardiac index in spon-taneously breathing subjects. Am J Emerg Med. 2010; 28(4):505-510.

37. London GM, Pannier BM, Laurent S, Lacolley P, Safar ME. Brachial artery diameter changes associated with cardio-pulmonary baroreflex activation in humans. Am J Physiol. 1990; 258(3 Pt 2):H773-777.

38. Roddie IC, Shepherd JT, Whelan RF. Reflex changes in vaso-constrictor tone in human skeletal muscle in response to stimulation of receptors in a low-pressure area of the intra-thoracic vascular bed. J Physiol. 1957;139(3):369-376.

39. Kamran H, Salciccioli L, Namana V, et al. Passive leg raising induced brachial artery dilation: is an old technique a simpler method to measure endothelial function? Atherosclerosis. 2010;212(1):188-192.

40. Bapat M, Musikantow D, Khmara K, et al. Comparison of passive leg raising and hyperemia on macrovascular and microvascular responses. Microvasc Res. 2013;86:30-33.

41. Cherpanath TG, Aarts LP, Groeneveld JA, Geerts BF. Defin-ing fluid responsiveness: a guide to patient-tailored volume titration. J Cardiothorac Vasc Anesth. 2014;28(3):745-754.

42. Mamontov OV, Kalinichenko AN, Conrady AO, Shlyakhto EV. Cardiopulmonary reflex influence on the system hemody-namic rapid regulation mechanisms. Comput Cardiol. 2008; 35:801-804.

43. Mark AL, Abboud FM, Fitz AE. Influence of low- and high-pressure baroreceptors on plasma renin activity in humans. Am J Physiol. 1978;235(1):H29-33.

44. Monnet X, Teboul J. Assessment of volume responsiveness during mechanical ventilation: recent advances. Crit Care. 2013;17(2):217.


45. Geerts B, de Wilde R, Aarts L, Jansen J. Pulse contour analy-sis to assess hemodynamic response to passive leg raising. J Cardiothorac Vasc Anesth. 2011;25(1):48-52.

46. Raval NY, Squara P, Cleman M, Yalamanchili K, Winklmaier M, Burkhoff D. Multicenter evaluation of noninvasive car-diac output measurement by bioreactance technique. J Clin Monit Comput. 2008;22(2):113-119.

47. Benomar B, Ouattara A, Estagnasie P, Brusset A, Squara P. Fluid responsiveness predicted by noninvasive bioreactance-based passive leg raise test. Intensive Care Med. 2010;36(11):1875-1881.

48. Kupersztych-Hagege E, Teboul JL, Artigas A, et al. Bioreac-tance is not reliable for estimating cardiac output and the effects of passive leg raising in critically ill patients. Br J Anaesth. 2013;111(6):961-966.

49. Squara P. Bioreactance for estimating cardiac output and the effects of passive leg raising in critically ill patients. Br J Anaesth. 2014;112(5):942.

50. Marik PE. Fluid responsiveness and the six guiding principles of fl uid resuscitation. Crit Care Med. 2016;44(10):1920-1922.

51. Garcia X, Pinsky MR. Clinical applicability of functional hemodynamic monitoring. Ann Intensive Care. 2011;1:35.


Your ticket to a great job offer.

15-0876

Search daily job postings

on the homepage.

You’re admitted to AACN’s Official Career Center.

Designed as a comprehensive career resource for nurses of all levels, the Career Center enables you to explore job postings by specialty area, location, and hospital/facility.

AACNCareerCenter.org

Post your CV or résumé today.

Choose from the best career

opportunities in nursing.

Start now. The perfect job

may be waiting for you.

Brief Report


Background Unplanned admissions of patients to inten-sive care units from medical-surgical units often result from failure to recognize clinical deterioration. The early warning score is a clinical decision support tool for nurse surveillance but must be communicated to nurses and implemented appropriately. A communication process including collaboration with experienced intensive care unit nurses may reduce unplanned transfers.Objective To determine the impact of an early warning score communication bundle on medical-surgical trans-fers to the intensive care unit, rapid response team calls, and morbidity of patients upon intensive care unit transfer.Methods After an early warning score was electronically embedded into medical records, a communication bun-dle including notification of and telephone collaboration between medical-surgical and intensive care unit nurses was implemented. Data were collected 3 months before and 21 months after implementation.Results Rapid response team calls increased nonsignifi-cantly during the study period (from 6.47 to 8.29 per 1000 patient-days). Rapid response team calls for patients with early warning scores greater than 4 declined (from 2.04 to 1.77 per 1000 patient-days). Intensive care unit admissions of patients after rapid response team calls significantly declined (P = .03), as did admissions of patients with early warning scores greater than 4 (P = .01), suggesting that earlier intervention for patient deterio-ration occurred. Documented reassessment response time declined significantly to 28 minutes (P = .002). Conclusion Electronic surveillance and collaboration with experienced intensive care unit nurses may improve care, control costs, and save lives. Critical care nurses have a role in coaching and guiding less experienced nurses. (American Journal of Critical Care. 2018; 27:238-242)

EARLY WARNING SCORE COMMUNICATION BUNDLE:A PILOT STUDYBy Cheryl Gagne, RN, DNP, CNEA, and Susan Fetzer, RN, PhD

E RBEvidence-Based Review on pp 243-244



Collaborative support from seasoned colleagues improves outcomes.

Up to 40% of intensive care unit (ICU) admissions may be avoidable,1 with nearly half of the admissions from medical-surgical units related to deterioration in the patient’s condition since the admitting diagnosis.2 Although patients exhibit clinical signs of deterioration up to 8 hours before a cardiac arrest,3 unplanned transfers to the ICU often result from failure to recognize clinical deterioration.4

Failure to rescue, described by Silber and colleagues5 in 1992, has been attributed to an inability to recognize early signs of deterioration.

The Institute for Healthcare Improvement

describes a “bundle of care” as 3 to 5 evidence-based

interventions uniformly applied together to improve

patient outcomes.6 To improve timely assessment of

and interventions for medical-surgical patients whose

condition was deteriorating, we created a communi-

cation bundle consistent with the Institute for Health-

care Improvement’s guidelines. The communication

bundle includes an electronically embedded early

warning score (EWS), an electronic medical record

(EMR) nurse alert, a pager alert to an experienced

ICU nurse, and a telephone consultation between a

medical-surgical nurse and an ICU nurse. We con-

ducted a pilot study to determine the effectiveness of

the communication bundle on ICU patient admis-

sions, patient morbidity, and rapid response team

(RRT) calls from medical-surgical units.

BackgroundIn the late 1990s, scoring systems arose as a

way to quickly warn of deterioration. Patient EWSs

are derived from physiological parameters, such as

blood pressure, heart rate, and temperature, gleaned

from nursing assessments. A predetermined score

prompts specific nursing action, such as as-needed

respiratory treatment, to ameliorate further deterio-

ration. Smith and others7 conducted a systematic

review of EWS outcomes of adult medical-surgical

patients. In 4 studies, at least a 50% increase in RRT

calls following EWS implementation was reported.

In 5 studies, the impact of EWS systems on ICU use

was measured; a significant increase in ICU admis-

sions was reported in 3 studies, and no difference

was found in 2 studies.

An EWS embedded in the EMR allows tracking,

trending, and automatic triggering by patient status.

Although the EWS provides timely recognition of

deterioration, information must be communicated

to a nurse with an appreciation for the urgency of

the situation and the knowledge to take action. The

lack of communication and action may explain the

conflicting findings reported by Smith and others.7

To improve communication, Bailey and colleagues8

added an EWS alert to the medical-surgical charge

nurse’s pager but could not explain a lack of improve-

ment in patients’ out-

comes, suggesting that

a more integrated

approach to identify

interventions is needed.

Jackson and others9

reported a strong nega-

tive correlation between medical-surgical nurses’

years of experience and reluctance to activate an RRT

call. They concluded that “collaborative support

from seasoned colleagues” targeting inexperienced

medical-surgical nurses could improve outcomes

of patients whose condition is deteriorating.9

Experienced ICU nurses serving on RRTs are

trained to recognize and intervene in situations of

clinical deterioration. We hypothesized that an EWS

communication bundle including an experienced ICU

nurse would decrease transfers, RRT calls, and morbid-

ity of medical-surgical patients transferred to the ICU.

Methods In this interrupted time-series study, measure-

ments recorded before and after intervention provided

EWS, RRT, and admission data on medical-surgical

patients transferred to the ICU.

Setting and SampleA 189-bed, Magnet-designated community hos-

pital in the Northeast with 97 medical-surgical beds

and 11 ICU beds was the setting for the study. We

studied inpatient medical-surgical patients with an

EWS greater than 4, indicating clinical deterioration.

InterventionWe embedded a 7-item modified EWS system10

with possible scores ranging from 0 to 21 into the

About the AuthorsCheryl Gagne is vice president and Susan Fetzer is a nurse researcher in patient care services at Southern New Hamp-shire Medical Center, Nashua, New Hampshire.

Corresponding author: Susan Fetzer, RN, PhD, Southern New Hampshire Medical Center, 8 Prospect Street, Nashua, New Hampshire, 03061 (email: [email protected]).


Both transfer of medical-surgical

patients to the ICU and ICU admis-

sions of RRT patients declined.

EMR (Table 1). In this system, EWS parameters are

manually entered into the EMR, and scores are auto-

matically generated. A score greater than 4 activates

a red exclamation point next to the patient’s name on

the EMR, visible to the nurse and

unit secretary. An electronic page

including room, medical record

number, and EWS score simulta-

neously notifies an experienced

ICU nurse preassigned to respond

to RRT calls. The ICU nurse

reviews the EMR and contacts the

patient’s nurse by telephone. The

nurses discuss the EWS alert and

develop a plan. The collaboration

results in reassessment of at least 1

EWS parameter and may include notifying a pro-

vider, administering as-needed medications, or fol-

lowing established protocols for interventions.

MeasuresOutcome measures included the number of

patients transferred to the ICU from the medical-

surgical units, the number of RRT calls per 1000

patient-days, ICU admissions, and RRT calls to

patients with an EWS greater than 4. Communica-

tion bundle response time was defined as the differ-

ence between EWS alert and EMR documentation

of the patient’s reassessment.

ProcedureThe institutional review board granted approval,

and the information technology department provided

quarterly outcome data. We obtained preimplemen-

tation data before embedding the EWS into the EMR.

The communication bundle was hardwired into 4

medical-surgical units and the ICU from January

through December 2015, with postimplementation

data collected from January through June 2016. We

used statistics software (SPSS v24, IBM) to generate

descriptive statistics and conduct analyses of variance.

Results The number of RRT calls before implementation

was 6.47 per 1000 patient-days. Among patients with

an EWS greater than 4, RRT calls numbered 2.04 per

1000 patient-days (Table 2). Medical-surgical trans-

fers accounted for 21.3% of ICU admissions, with

nearly half (18 of 41, or 44%) of medical-surgical

patients who had received an RRT call transferred

to the ICU.

We analyzed 6 quarters of postimplementation

data. After implementation, RRT calls increased

from 6.47 to 8.29 per 1000 patient-days, although

this finding was not significant. However, calls for

patients with an EWS greater than 4 declined. With

the exception of quarter 6, medical-surgical trans-

fers to ICU declined. ICU admissions of patients

who had received an RRT call declined significantly

(P = .03), as did ICU admissions of patients with an

EWS greater than 4 (P = .01). EWS response time

decreased significantly (P = .002), indicating a success-

ful implementation of the communication bundle.

Discussion After implementation, RRT calls increased, a

finding similar to that of Kollef et al,2 who also

found an increase in the number of RRT calls with

EWS alerts. However, in our study, the increase in

RRT calls occurred in patients with lower EWS

scores. Fewer patients who received an RRT call

exhibited EWS scores greater than 4. These findings

suggest that deterioration was identified sooner,

prompting earlier intervention. As a result, the

Item

Table 1Early warning scoring systema

Central nervous system

Respiratory rate, breaths per minute

Heart rate, beats per minute

Systolic blood pressure, mm Hg

Body temperature, ºC

Oxygen saturation with therapy, %

Urine output in 2 h, mL/h

a The composite score for all 7 items ranges from 0 to 21. The higher the score, the greater the physiological deterioration, with a composite score greater than 4 generating a communication alert.

No response

>30

>190

>220

>40

Respond to pain

111-190

201-220

38.6-40.0

Respond to voice

21-30

101-110

181-200

37.6-38.5

Alert

8-20

91-100

101-180

35.1-37.5

>94

40-90

81-100

Confused, agitated

71-80

34-35

91-93

<8

<40

<70

<34

<90

<30

3210123

Early warning score


number of ICU admissions declined, although the

unexplained increase during quarter 6 contributed

to the nonsignificance of this outcome measure. The

decline in ICU patients with EWS scores greater than

4 admitted to the ICU reflects a decrease in transfer

morbidity. The communication bundle may have

resulted in more effective interventions by medical-

surgical nurses. The addition of ICU nurse collabo-

ration may be a facilitator.

Dummett and others11 described the EWS as

enhancing situational awareness, a first step in pro-

ducing cognitive change. The decreased EWS response

time may reflect this type of cognitive change.

LimitationsThe fact that we used only 1 quarter of preim-

plementation data for comparison may be a limita-

tion because patients’ acuity can be influenced by

seasonal changes. Although the communication

bundle resulted in earlier interventions, the effect

on patients’ survival is not known.

Conclusion The EWS is a clinical measurement tool that

complements nursing judgment. Electronic surveil-

lance together with experienced ICU nurse collabora-

tion has the potential to improve care, preserve

health care dollars, and save lives. An electronically

embedded EWS together with a communication

bundle reduced ICU admissions from

medical-surgical units and decreased patient mor-

bidity. Obtaining the perspective of nurses imple-

menting the EWS communication bundle is

warranted.

ACKNOWLEDGMENTThe authors appreciate the support of Andrew Watt, MD.


REFERENCES1. Robertson LC, Al-Haddad M. Recognizing the critically ill

patient. Anaesth Intensive Care Med. 2013;14(1):11-14. 2. Kollef MH, Chen Y, Heard K, et al. A randomized trial of

real-time automated clinical deterioration alerts sent to a rapid response team. J Hosp Med. 2014; 9(7):424-429.

3. Mapp ID, Davis LL, Krowchuk H. Prevention of unplanned intensive care unit admissions and hospital mortality by early warning systems. Dimens Crit Care Nurs. 2013; 32(6): 300-309.

4. Alam N, Hobbelink EL, Tienhoven AJ, van de Ven PM, Jansma EP, Nanayakkara PW. The impact of the use of the Early Warning Score (EWS) on patient outcomes: a systematic review. Resuscitation. 2014;85(5):587-594.

5. Silber JH, Williams SV, Krakauer H, Schwartz JS. Hospital and patient characteristics associated with death after surgery: a study of adverse occurrence and failure to rescue. Med Care. 1992;30(7):615-629.

6. Resar R, Griffin FA, Haraden C, Nolan TW; Institute for Healthcare Improvement. Using care bundles to improve health care quality. IHI Innovation Series. http://www.ihi .org/resources /Pages/IHIWhitePapers/UsingCareBundles .aspx. Published 2012. Accessed January 29, 2018.

7. Smith ME, Chiovaro JC, O’Neil M, et al. Early warning system scores for clinical deterioration in hospitalized patients: a systematic review. Ann Am Thorac Soc. 2014; 11(9):1454-1465.

8. Bailey TC, Chen Y, Mao Y, et al. A trial of a real-time alert for clinical deterioration in patients hospitalized on general medical wards. J Hosp Med. 2013;8(5):236-242.

Outcome measure

Table 2Outcomes before and after implementation of an early warning score communication bundle

Rapid response calls Total number Number per 1000 patient-days

Rapid response calls with EWS >4 per 1000 patient-days

ICU admissions from medical- surgical units, No. (%)

ICU admissions from medical- surgical units after RRT,a No. (%)

ICU admissions from medical- surgical units with EWS >4 per

1000 patient-daysb

EWS communication response time,c minutes

Abbreviations: EWS, early warning score; ICU, intensive care unit; NA, not applicable; Q, quarter; RRT, rapid response team.a df = 6, F = 8.794, P = .03.b df = 6, F = 15.653, P = .01.c df = 5, F = 34.714, P = .002.

568.29

1.77

46 (17.8)

11 (20)

0.75

28

79 11.78

2.09

82 (31.4)

15 (19)

0.44

29

54 8.71

2.08

54 (20.3)

14 (26)

0.62

33

619.02

1.77

59 (21.5)

16 (26)

0.74

27

619.04

2.57

55 (20.5)

19 (31)

0.86

39

527.40

3.70

64 (24.6)

17 (33)

1.00

85

416.47

2.04

57 (21.3)

18 (44)

1.56

NA

Q7Q6Q5Q4Q3Q2Q1

After implementation, by quarterBefore

implementation


9. Jackson S, Penprase B, Grobbel C. Factors influencing reg-istered nurses’ decision to activate an adult rapid response team in a community hospital. Dimens Crit Care Nurs. 2016; 35(2):99-107.

10. Moon A, Cosgrove JF, Lea D, Fairs A, Cressey DM. An eight year audit before and after the introduction of modified early warning score (MEWS) charts, of patients admitted to a tertiary referral intensive care unit after CPR. Resuscitation. 2011;82(2):150-154.

11. Dummett BA, Adams C, Scruth E, Liu V, Guo M, Escobar GJ. Incorporating an early detection system into routine clinical

practice in two community hospitals. J Hosp Med. 2016;11 (Suppl 1):S25-S31.


THE DEFINITIVE RESOURCEfor critical care nursing

and practice

• NEW! corresponds to the most recent CCRN®

current literature, national and international guidelines, clinical studies, AACN/ANA scope of practice, and ECCO 3.0.

• NEW! addresses the healthcare needs of today’s patients, including pain management, palliative care, and end-of-life care; nutritional support; transplantation; and key AACN initiatives.

• NEW! Clinical Pearls Key Concepts highlight important content and serve as a convenient quick reference.

• NEW!• with an emphasis on clinical nursing practice, prepare you for the full range of knowledge you will encounter on the CCRN exam and in practice.

American Association of Critical-Care Nurses; and Tonja M. Hartjes, DNP, ACNP-BC, FNP-BC, CCRN-CSC, FAANP

October 2017 ISBN: 978-1-4557-1065-2

NEW EDITION!

Evidence-Based Review and Discussion PointsBy Ronald L. Hickman, RN, PhD, ACNP-BC

Evidence-Based Review (EBR) is the journal club feature in the American Journal of Critical Care. In a journal club, attend ees review and critique published research articles: an important first step toward integrating evidence-based practice into patient care. General and specific questions such as those outlined in the “Discussion Points” box aid journal club participants in probing the quality of the research study, the appropriateness of the study design and methods, the validity of the conclusions, and the implications of the article for clinical practice. When critically appraising this issue’s EBR article, “Early Warning Score Communication Bundle: A Pilot Study” (pp 238-242), consider the questions and discussion points outlined in the “Discussion Points” box.

F ailure to recognize early signs of deteriora-

tion can result in unplanned and unneces-

sary admissions to an intensive care unit

(ICU) for acutely ill patients convalescing in the

hospital. Clinical deterioration is likely to occur up

to 8 hours before decisions to initiate a transfer

to an ICU. However, failure to recognize a patient’s

clinical deterioration accounts for 40% of the

unplanned and avoidable transfers to an ICU.

Early warning scores (EWSs), systematic

methods for early detection of a patient’s risk

for clinical deterioration, are often derived from

physiological parameters and nursing assessments

to prompt nursing action. The adoption of elec-

tronic medical records by health care systems has

provided an opportunity for the integration of

EWSs as a clinical decision support tool to mini-

mize unplanned transfers and avoid ICU trans-

fers. However, one of the significant

limitations of EWSs integrated into

electronic medical records has been the

inability to alert nurses of the urgency

of the situation and facilitate commu-

nication to a rapid response team.

To address the current limitations

of EWSs integrated into electronic medi-

cal records, the authors have developed

an EWS communication bundle, which

they hypothesized would decrease trans-

fers to an ICU, rapid response team calls,

and morbidity of medical-surgical patients.

This study had an interrupted time-series

design with pretest and posttest measure-

ments and was conducted in a 189-bed

community hospital. A 7-item modified

EWS system was embedded into the elec-

tronic medical record, and scores greater

than 4 indicated evidence of a patient’s

clinical deterioration. For patients with

EWSs greater than 4, the communication

bundle was activated and included 2

alerts: (1) a red exclamation mark next

to the patient’s name in the electronic

medical record and (2) a simultaneous

page sent to the rapid response team’s

nurse. The activation of the communica-

tion bundle was designed to facilitate the

reassessment of the patient, notifying a

provider, and prompt the administration

of medications or initiation of protoco-

lized interventions.

Investigator Spotlight

This feature briefly describes the personal journey and background story of the EBR article’s investigators, discussing the circumstances that led them to undertake the line of inquiry represented in the research article featured in this issue.

Cheryl Gagne, RN, DNP, CNEA, is the vice president for patient

care services at Southern New Hampshire Medical Cen-

ter in Nashua, New Hampshire. She has been a registered

nurse for more than 35 years and practiced for more than

20 years as a critical care nurse.

Gagne says that she has been motivated to influence

the care of the critically ill through personal and profes-

sional experiences. “From my first assignment as a nurse

through today, I relentlessly pursue solu-

tions to challenges patients, and their

nurses, face on a daily basis,” she shares.

She adds that her interest in early warn-

ing scores was influenced by an endur-

ing childhood experience where her

brother acquired an acute but devastat-

ing illness, since which she has been

compelled to find a solution to help

patients like her brother.

According to Dr Gagne, clinical decision support is a

promising method that will affect the delivery and quality

of health care. Clinical decision support will become the

norm in electronic medical records and, as algorithms

become more sophisticated, we will be able to examine

everything from nursing care intensity to early detection

of infection, predicts Gagne. “I look forward to seeing how

early warning scores and similar tools will evolve and impact

the future of health care,” she adds.

©2018 American Association of Critical-Care Nurses, doi:https://doi.org/10.4037/ajcc2018196

Cheryl Gagne


About the AuthorRonald L. Hickman is an associate professor and assistant dean for nursing research, Case Western Reserve University, Cleveland, Ohio.

Discussion Points

A. Description of the Study

What is the significance of the problem

posed by the authors?

What is the purpose of the study?

B. Literature Evaluation

What evidence is provided in support of early

warning scores in hospitalized adults?

What was identified as a major barrier

related to the implementation of early warning

scores?

C. Methods and Design

Describe how the authors conceived the

early warning score with a communication

bundle.

Describe what data were collected for this

publication.

D. Results

What were the major findings of this study?

What was one of the limitations of the study

cited by the authors?

How can you use the findings of this study to

improve the quality of nursing care at your

facility?

The authors report that electronic surveillance

of a patient’s clinical deterioration can be feasibly

integrated into an electronic medical record and

minimize unplanned and avoidable ICU transfers.

Based on the results of their study, the authors were

able to demonstrate that the use of an EWS commu-

nication bundle significantly reduced ICU transfers

for patients who met the alert criterion. They also

noted a slight increase in the number of rapid response

team calls, which was not statistically significant.

The authors conclude that an EWS communication

bundle is an effective strategy for surveillance and

promotes collaboration among health care providers

to optimize the quality of patient care.

Information From the AuthorsCheryl Gagne, RN, DNP, CNEA, lead author on this

EBR article, provides personal and professional per-

spectives that influenced the implementation of the

EWS communication bundle. Dr Gagne comments

that she had a longstanding desire to identify innova-

tive solutions to improve patient care, which motivated

her to investigate the effects of an EWS communica-

tion bundle.

Every day, nurses are assessing patients while bal-

ancing the competing demands of delivering high-qual-

ity nursing care. “Of the hundreds of clinical decisions

nurses make every shift, the decision that results in

earlier rather than later recognition of deterioration

is perhaps one of the most important,” comments

Dr Gagne. “As nurses, we intervene on behalf of our

patients throughout the day, and an EWS can be a

seamless, electronically driven strategy to identify

and respond to subtle and progressive changes in a

patient’s condition,” she says.

From Dr Gagne’s perspective, the EWS communica-

tion bundle is a solution that enhances nurse-to-nurse

collaboration. She remarks that “through the interac-

tions between the bedside nurse and [the] critical care

nurse on the response rapid response team prompted

by the EWS alert, our patients are received effective care

earlier and [are] being transferred less to the ICU.”

An unintended consequence of the EWS commu-

nication bundle is the increased confidence of the

nurses in recognizing and initiating appropriate nurs-

ing care for patients with declining health. “Our

bedside nurses are experiencing more confidence

because of their ability to more quickly respond to

and even reverse signs of impending deterioration of

their patients’ condition,” she mentions.

Implications for PracticeGagne encourages readers of the American Journal

of Critical Care to consider implementing the use of

EWS with a communication bundle to potentially

improve care. Dr Gagne and her coauthor comment

that the EWS is a clinical measurement that comple-

ments nursing judgment; however, they recognize the

limitations of the current state of the science regarding

early warning assessment tools. Gagne comments that

“the most obvious unresolved early warning score issue

is the lack of reliability and validity of available tools.”

Despite the lack of evidence of the reliability and valid-

ity of early warning scores across clinical populations,

Dr Gagne envisions that such early warning scores

will be routinely embedded in electronic medical

records and provide health care providers with robust

clinical decision support to detect and minimize a

patient’s clinical deterioration.


Brief Report


Background Exposure to bright light has alerting effects. In nurses, alertness may be decreased because of shift work and high work pressure, potentially reducing work performance and increasing the risk for medical errors. Objectives To determine whether high-intensity dynamic light improves cognitive performance, self-reported depressive signs and symptoms, fatigue, alertness, and well-being in intensive care unit nurses.Methods In a single-center crossover study in an inten-sive care unit of a teaching hospital in the Netherlands, 10 registered nurses were randomly divided into 2 groups. Each group worked alternately for 3 to 4 days in patients’ rooms with dynamic light and 3 to 4 days in control lighting settings. High-intensity dynamic light was administered through ceiling-mounted fluorescent tubes that delivered bluish white light up to 1700 lux during the daytime, versus 300 lux in control settings. Cognitive performance, self-reported depressive signs and symptoms, fatigue, and well-being before and after each period were assessed by using validated cognitive tests and questionnaires. Results Cognitive performance, self-reported depressive signs and symptoms, and fatigue did not differ signifi-cantly between the 2 light settings. Scores of subjective well-being were significantly lower after a period of working in dynamic light.Conclusions Daytime lighting conditions did not affect intensive care unit nurses’ cognitive performance, per-ceived depressive signs and symptoms, or fatigue. Per-ceived quality of life, predominantly in the psychological and environmental domains, was lower for nurses working in dynamic light. (American Journal of Critical Care. 2018; 27:245-248)

EFFECT OF DYNAMIC LIGHT APPLICATION ON COGNITIVE PERFORMANCE AND WELL-BEING OF INTENSIVE CARE NURSESBy Koen S. Simons, MD, Enzio R. K. Boeijen, RN, BSc, Marlies C. Mertens, PhD, Paul Rood, RN, MSc, Cornelis P.C. de Jager, MD, PhD, and Mark van den Boogaard, RN, PhD


Exposure to environmental light has profound effects on humans in health and dis-ease.1 Apart from entraining the biological clock, light exposure also influences cog-nitive performance and alertness.2 Alertness may be decreased in nursing care because of variable shifts and high work pressure.3 Stimulating alertness by means of bright light therapy can improve performance4,5 and in health care settings may lead to

better patient care, possibly by reducing the number of errors. Studies on the alerting effect of lighting therapy on the performance of nurses are scarce. Some results suggest a beneficial effect of a brief exposure to bright light6 or improved alertness in nurses exposed to bright light, compared with dim light conditions, during the night shift.7

Bright light may improve alert-

ness and reduce medical errors.


Methods The intensive care unit (ICU) of the Jeroen Bosch

Hospital, ‘s-Hertogenbosch, the Netherlands, a teach-

ing hospital with a 16-bed mixed medical-surgical

ICU, has a dynamic lighting system. This system,

the dynamic lighting application (DLA), offers bright

light intensities (up to 1700 lux, compared with 300

lux in standard settings) and colors according to a

fixed rhythm via conventional, ceiling-mounted flu-

orescent tubes inside patients’ rooms.

Details of a DLA study with ICU patients

have been published.8

We performed a crossover study

to determine the effect of daytime

DLA on cognitive function and self-

reported measures of well-being in 10

ICU nurses. Participants were divided

into 2 groups. In the first group, participants were

scheduled to work at least 3 to a maximum of 4

consecutive daytime shifts in patients’ rooms where

the light was set to DLA. After a washout period,

the participants were subsequently scheduled to 3

to 4 daytime shifts in patients’ rooms with standard

lighting settings. In the second group, the partici-

pants followed the same schedule, starting with

standard lighting settings and ending with DLA.

Before and directly after each test period, partic-

ipants were asked to undergo cognitive performance

tasks by using 2 validated computer-based tests, the

Test of Divided Attention (TODA)9 and the Test of

Selective Sustained Attention (TOSSA).10 In the TODA,

participants listen to groups of beeps and are requested

to distinguish the groups of 3 beeps while simultane-

ously determining the correctness of a sum. TODA

scores range from 0% to 100% and are a measure of

the functioning of divided attention. In the TOSSA,

participants listen to groups of beeps and are asked

to distinguish the groups of 3 beeps by pushing the

space bar on the computer keyboard as quickly as

possible. TOSSA produces 3 scores: concentration

strength, detection strength, and response inhibition

strength.10 Scores for each of the 3 range from 0% to

100%. To minimize the influence of a learning effect

on the performance on these tests, participants com-

pleted a practice session of both tests 1 time before

the actual study started.

Information on subjective well-being, mental

health, and sleep quality of the participants was

assessed by using validated questionnaires11-15: the

Center for Epidemiologic Studies Depression scale

(CES-D), the Fatigue Assessment Scale (FAS), and

the World Health Organization Quality of Life abbre-

viated version (WHOQOL-BREF). An additional ques-

tionnaire (a diary) was developed to obtain baseline

characteristics and subjective measures of alertness

and fatigue (Table 1); the diary was filled in daily by

using a numeric rating scale. Syntax formats were

used to analyze the outcomes of the CES-D, the FAS,

and the WHOQOL-BREF. The first and last measure-

ments during the working period were used to com-

pute difference scores. The delta (di) of these scores

was used for comparisons between DLA and the

control period.

About the AuthorsKoen S. Simons is an intensivist, Department of Intensive Care Medicine, Jeroen Bosch Hospital, ‘s-Hertogenbosch, the Netherlands, and Department of Intensive Care, Rad-boud University Medical Center, Nijmegen, the Nether-lands. Enzio R. K. Boeijen is a student in nursing sciences, Paul Rood is a PhD student, and Mark van den Boogaard is an assistant professor, Department of Inten-sive Care, Radboud University Medical Center. Cornelis P. C. de Jager is an intensivist, Department of Intensive Care Medicine, Jeroen Bosch Hospital. Marlies C. Mertens is a psychologist, Department of Medical Psychology, Jeroen Bosch Hospital and at Eindhoven Corporation of Primary Health Care Centers (SGE), Eindhoven, the Netherlands.

Corresponding author: Koen S. Simons, MD, Department of Intensive Care Medicine, Jeroen Bosch Ziekenhuis, Henri Dunantstraat 1, 5223 GZ ’s-Hertogenbosch, the Netherlands (email: [email protected]).

Results The mean age of participants was 34 years, and

70% were female. Participants worked a similar num-

ber of hours in the DLA period and the control period.

Scores on the TODA and all outcome measures of

the TOSSA did not change significantly in both peri-

ods after each testing period (Table 2). Mean CES-D

and FAS scores did not change significantly during

either the DLA period or the control period. Total

scores of the WHOQOL-BREF decreased from 109.7

to 108.0 during the testing period in the DLA period

and improved from 103.9 to 112.5 in the control

period; the difference score during the testing period

was -1.7 in the DLA period and +8.6 in the control

period (P = .01). Significant changes occurred in the

domain of psychological health and environment

(Table 2).

DLA was generally well tolerated by nurses. Two

participants reported headaches due to constantly

squinting their eyes because of the DLA brightness.

Discussion In this crossover study, we found no differences

in cognitive performance, fatigue, or depressive

feelings of ICU nurses when they worked in an envi-

ronment of enhanced light levels compared with

when they worked in an environment with normal

Type of question

Table 1“Diary” questionnaire consisting of baseline characteristics and subjective questions about well-being

General

Specific

Dailya

AgeSexMarital statusWorking hours during research period

Physical health problemsPsychiatric health problemsUse of (psychotropic) medicationSleep experienceExperience working in lighting and type of lighting

Topic Numeric Rating Scale, 0-10Sleeping quality Bad - good Feeling dull Not dull - very dull Feeling good Bad - goodSubjective sleep Minimum hours slept - duration maximum hours sleptActivity Not active - very activePositive thoughts Negative thoughts - positive thoughtsPositive events Negative events - positive events

Topic

a Daily scores were filled in using a numeric rating scale from 0 to 10.

Variable

Table 2Clinical outcomes of dynamic light application (DLA)a

Cognitive performance TODA, mean (SD) TOSSA, mean (SD) Concentration strength Detection strength Response inhibition strength

Mental health CES-D, mean (SD) FAS, mean (SD) WHOQOL-BREF, mean (SD) Overall QOL and health Domain 1: Physical health Domain 2: Psychological health Domain 3: Social relationships Domain 4: Environment Diary, median (IQR) Sleep quality Feeling dull Feeling good Subjective sleep duration Activity Positive thoughts Positive events

Abbreviations: CES-D, Center for Epidemiologic Studies Depression scale; FAS, Fatigue Assessment Scale; IQR, interquartile range; TODA, Test of Divided Attention; TOSSA, Test of Selective Sustained Attention; WHOQOL-BREF, World Health Organization Quality of Life abbreviated version.

a Data were collected directly before and after the specified testing period. b Difference between the scores.c P = .008 for difference between DLA before and Control before.d P = .006 for difference between DLA after and Control after.

.06

.92

.89

.87

.61 .40< .01 .10 .31 .05 .08 .01

.78

.15

.61

.95

.40

.34

.86

3.3

3.83.57.3

0.0-0.48.60.20.50.60.50.9

-0.30.6-0.4-1.00.40.00.2

94.5 (8.1)

95.4 (7.3)96.5 (5.7)98.8 (2.1)

2.4 (2.5) 17.4 (3.5)112.5 (6.0) 9.0 (0.9) 16.8 (1.0) 16.8 (1.0) 18.0 (1.6) 17.7 (1.2)

8.0 (6.5-8.0)8.0 (6.0-8.0)8.0 (7.0-9.0)6.0 (5.0-7.0)7.0 (7.0-8.5)8.0 (7.0-9.0)9.0 (7.0-9.0)

91.3 (10.9)

91.6 (19.1)92.9 (17.6)98.5 (3.5)

2.4 (2.8) 17.8 (3.3)103.9 (7.4) 8.8 (1.1) 16.3 (1.3) 16.2 (1.4) 17.5 (1.6) 16.8 (1.1)

8.5 (5.8-10.0)7.5 (5.8-8.3)8.5 (7.0-9.0)7.0 (6.0-8.3)8.0 (7.0-8.0)9.0 (7.8-10.0)8.0 (7.0-10.0)

-0.7

0.80.65.0

0.4 0.2-1.7-0.6 0.2-0.3-0.3-0.2

-0.2-1.9-0.8-1.2-0.4 0.8 0.3

93.1 (11.0)

95.0 (9.0)95.8 (8.3)99.2 (1.1)

2.3 (2.3)17.1 (2.9)

108.0 (4.9)d

8.6 (0.8)16.9 (0.8)15.9 (0.8)16.9 (0.9)16.8 (1.2)

7.0 (7.0-8.3)6.0 (2.8-9.0)8.0 (7.3-9.0)5.5 (5.0-7.0)7.5 (6.8-8.0)8.0 (7.8-8.5)8.0 (6.3-8.3)

93.9 (9.7)

94.2 (9.2)95.2 (8.3)99.0 (1.3)

1.9 (2.4) 16.9 (2.5)

109.7 (4.9)c

9.2 (0.8)16.7 (0.8)16.2 (0.8)17.2 (1.3)17.1 (1.4)

8.0 (7.0-9.0)8.0 (6.8-8.3)9.0 (8.0-9.0)8.0 (6.8-8.3)8.0 (6.8-9.0)8.0 (7.3-9.3)8.0 (6.5-9.0)

Pdibdi

b AfterAfter BeforeBefore

Scores in control groupScores in DLA group


light levels. Interestingly, we found a significant

increase in subjective well-being, specifically in the

psychological and environmental domains of the

WHOQOL-BREF, in participants working in control

lighting settings compared with participants work-

ing in rooms with DLA.

Several reasons may explain our findings. First,

although exposure to bright light has beneficial effects

in specific conditions, it is only one of many factors

that influence cognitive performance and well-being.

Shift work, personal stress, physical complaints, and

sleep quality are all major contributors to subjective

well-being and occur frequently among nurses.3

These contributors may play such a large role that

the light therapy itself cannot alter feelings of cogni-

tive performance, well-being, depression, or fatigue.

Second, the difference in light exposure between the

DLA and the control periods may be too small for

an additional alerting effect of DLA to be detected,

because of a sufficient magnitude of lighting inten-

sity in the control group and because nursing care

also involves activity outside the brightly lit patients’

rooms, thereby diluting the total light exposure in

the DLA group. Third, differences in perceived well-

being at baseline between the 2 groups, as indicated

by the WHOQOL-BREF domain scores, might be

related to expected benefits of working in a DLA

environment or might just be serendipitous find-

ings. An adverse effect of DLA, thereby preventing

an improvement in quality of life, as found in the

control period, appears to be unlikely, yet cannot

be entirely excluded. Finally, we did not measure

the number of errors. Although we did not find ben-

eficial effects of exposure to bright light on alertness

or well-being, beneficial effects on reduction in the

number of errors still may be possible; however,

such effects seem unlikely.

Our study had several limitations. For practical

reasons, we were unable to increase the duration of

the daytime shifts for longer than 4 consecutive days.

Shift work is associated with disruption of circadian

rhythm, and realignment of circadian rhythmicity

may require up to 1 month.16 Second, we did not take

into account the type of patient and the workload of

the participants, variables that may have influenced

subjective well-being and cognitive performance.

Up to now, most research on effects of environ-

mental light has been performed in experimental

settings. In our study, we found no improvement in

cognitive performance or psychological function-

ing in a real working environment. Future research

should focus on longer duration of light exposure

and possibly higher light intensities in the whole

work area, especially during conditions of dim light,

and on effects of light therapy on the number of

errors in daily ICU nursing.

ACKNOWLEDGMENTThis study was performed at Jeroen Bosch Ziekenhuis, ’s-Hertogenbosch, the Netherlands.


REFERENCES1. Münch M, Bromundt V. Light and chronobiology: implica-

tions for health and disease. Dialogues Clin Neurosci. 2012; 14(4):448-453.

2. Chellappa SL, Steiner R, Blattner P, Oelhafen P, Götz T, Cajo-chen C. Non-visual effects of light on melatonin, alertness and cognitive performance: can blue-enriched light keep us alert? PLoS One. 2011;6(1):e16429.

3. Dall’Ora C, Ball J, Recio-Saucedo A, Griffiths P. Characteristics of shift work and their impact on employee performance and wellbeing: a literature review. Int J Nurs Stud. 2016; 57:12-27.

4. Viola AU, James LM, Schlangen LJ, Dijk DJ. Blue-enriched white light in the workplace improves self-reported alertness, performance and sleep quality. Scand J Work Environ Health. 2008;34(4):297-306.

5. Daurat A, Aguirre A, Foret J, Gonnet P, Keromes A, Benoit O. Bright light affects alertness and performance rhythms during a 24-h constant routine. Physiol Behav. 1993; 53(5): 929-936.

6. Tanaka K, Takahashi M, Tanaka M, et al. Brief morning expo-sure to bright light improves subjective symptoms and perfor-mance in nurses with rapidly rotating shifts. J Occup Health. 2011;53(4):258-266.

7. Yoon IY, Jeong DU, Kwon KB, Kang SB, Song BG. Bright light exposure at night and light attenuation in the morning improve adaptation of night shift workers. Sleep. 2002; 25(3):351-356.

8. Simons KS, Laheij RJ, van den Boogaard M, et al. Dynamic light application therapy to reduce the incidence and duration of delirium in intensive-care patients: a randomised controlled trial. Lancet Respir Med. 2016;4(3):194-202.

9. Kovács F. TODA Test of Divided Attention: Manual [in Dutch]. Voorhout, the Netherlands: Pyramid Productions; 2009:1-23. http://www.how-psychology-tests-brain-injury.com/support-files/toda_handleiding.pdf. Accessed February 28, 2018.

10. Kovács F. TOSSA Test of Sustained Selective Attention: Man-ual. Version 4. Voorhout, the Netherlands: Pyramid Productions; 2016:1-59. http://pyramidproductions.nl.server41.firstfind.nl/Bijlage/TOSSA_manual.pdf. Accessed February 28, 2018.

11. Bouma JR, Ranchor AV, Sanderman R, van Sonderen E. Het meten van symptomen van depressie met de CES-D: een han-dleiding [in Dutch]. 2nd ed. Groningen, the Netherlands: Research Institute SHARE, University of Groningen; 2012.

12. Schroevers MJ, Sanderman R, van Sonderen E, Ranchor AV. The evaluation of the Center for Epidemiologic Studies Depression (CES-D) scale: depressed and positive affect in cancer patients and healthy reference subjects. Qual Life Res. 2000;9(9):1015-1029.

13. De Vries J, Michielsen H, Van Heck GL, Drent M. Measuring fatigue in sarcoidosis: the Fatigue Assessment Scale (FAS). Br J Health Psychol. 2004;9(pt 3):279-291.

14. Michielsen HJ, De Vries J, Van Heck GL. Psychometric qualities of a brief self-rated fatigue measure: the Fatigue Assessment Scale. J Psychosom Res. 2003;54(4):345-352.

15. The WHOQOL Group. Development of the World Health Orga-nization WHOQOL-BREF quality of life assessment. Psychol Med. 1998;3:551-558.

16. Arendt J. Shift work: coping with the biological clock. Occup Med (Lond). 2010;60(1):10-20.


THE VALUE OF LEAD aVR: A FREQUENTLY NEGLECTED LEADBy Salah S. Al-Zaiti, RN, PhD, CRNP, Teri M. Kozik, RN, PhD, CNS, CCRN, Michele M. Pelter, RN, PhD, and Mary G. Carey, RN, PhD, CNS

©2018 American Association of Critical-Care Nurses doi:https://doi.org/10.4037/ajcc2018523

Salah S. Al-Zaiti is an assistant professor at the Department of Acute and Tertiary Care Nursing, University of Pittsburgh, Pennsylva-nia. Teri M. Kozik is a nurse researcher at St Joseph’s Medical Center, Stockton, California. Michele M. Pelter is an assistant profes-sor at the Department of Physiological Nursing, University of California, San Francisco, California. Mary G. Carey is associate director for clinical nursing research, Strong Memorial Hospital, Rochester, New York.

ECG Puzzler A regular feature of the American Journal of Critical Care, the ECG Puzzler addresses electrocardiogram (ECG) interpretation for clinical practice. We welcome letters regarding this feature.

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Yes No NA

Scenario: The following 12-lead electrocardiogram

(ECG) was obtained from a 71-year-old woman who

called 9-1-1 for new onset acute chest pain associated

with dyspnea. She had no significant cardiac history.

The paramedics transmitted this prehospital

ECG to a medical command physician, request-

ing specific medical directions. What is your

assessment of this ECG?


Interpretation Questions:

1. Is the ECG properly calibrated (10 mm) and are leads properly placed?

If no, interpret cautiously.

2. Is this a sinus rhythm (one P wave preceding every QRS complex)?

If no, check for number of P waves in relation to QRS complexes.

3. Is the heart rate (R-R interval) normal (60-100/min)?

If no, check for supraventricular or ventricular arrhythmias.

4. Is the QRS complex narrow (duration ≤ 110 milliseconds [ms] in V1)?

If no, check for bundle branch blocks (BBBs), pacing, or ventricular arrhythmia.

5. Is the ST segment deviated (≥ 2 mm in V2-V

3, or ≥ 1 mm in other leads)?

If yes, check for similar deviations in contiguous cardiac territories.

6. Is the T wave inverted in relation to the QRS (> 5 mm)?

If yes, check for ST deviation or conduction abnormalities.

7. Is the QT interval lengthened (> 450 ms [men] or > 470 ms [women])?

If yes, check for ventricular arrhythmias or left ventricular hypertrophy.

8. Is R- or S-wave amplitude enlarged (S wave V1 + R wave V

5 > 35 mm)?

If yes, check for axis deviation or other chamber hypertrophy criteria.

II

III aVF

aVL

aVR V1

V4

V2

V5

V3

V6

I

Interpretation Sinus tachycardia with ST-segment elevation in lead

aVR and profound ST-segment depression in multiple leads,

consistent with severe left main coronary artery disease and/or

possibly occlusion of the left anterior descending coronary

artery (LAD). These findings are consistent with high-risk

acute coronary syndrome (ACS), and the patient should be

transported emergently to a tertiary care hospital that can

provide primary percutaneous coronary intervention.

Rationale Lead aVR is the augmented unipolar right arm lead,

which is the only lead that opposes the direction of the

main cardiac vector. This “view” makes lead aVR a valuable

lead for diagnosing not only ACS, but cardiac arrhythmias

with retrograde conduction (eg, junctional rhythm, ven-

tricular tachycardia), yet this lead has historically received

the least attention during ECG evaluation. Some ECG

experts ironically call the standard ECG the “11-lead

ECG” to prompt clinicians to consider examining lead

aVR in their evaluation. Within the context of ACS,

ST-segment elevation in lead aVR is frequently associated

with LAD occlusion proximal to the first septal perforator

(sensitivity and specificity >80%), with high probability of

multivessel disease. Importantly, this ST-segment pattern

is a strong predictor of hospital death, recurrent ischemia,

and heart failure, suggesting that early angiography, with

intervention if indicated, may improve clinical outcomes.

Management In this case study, the patient was sent to the catheter-

ization laboratory, which revealed a 90% occlusion of the

LAD and 70% occlusion of the left circumflex coronary artery,

which is consistent with the ST-segment pattern seen on

this ECG. Both vessels were successfully stented. The patient

was discharged home a few days after the procedure, and

no adverse events were reported within 30 days of the initial

presentation. Clinical training curriculums should empha-

size the value of examining all 12 leads of the standard ECG,

including lead aVR, which is often forgotten.


Answers:1. The calibration marks indicate proper gain (10 mm/mV).

2. Yes, there is 1 P wave per QRS complex.

3. No, the heart rate is tachycardiac at 103/min.

4. Yes, the QRS duration is normal.

5. Yes, the ST segment is elevated in lead aVR, and depressed in leads I, II, aVF, aVL, and V2 throughV

6.

6. Yes, the T wave is inverted in lead aVL.

7. No, the QT interval is not lengthened.

8. No, no signs of chamber hypertrophy are apparent.

II

III aVF

aVL

aVR V1

V4

V2

V5

V3

V6

I

The American Journal of Critical Care offers an exclusively online publication process that disseminates cutting-edge research on high acuity and critical care in the most timely and efficient manner possible. The abstracts below represent full-text articles available exclusively on the American Journal of Critical Care Web site, www.ajcconline.org. These OnlineNOW articles are fully peer reviewed, edited, formatted, and citable. Reprints of the full-text articles are available by calling (800) 899-1712 or (949) 362-2050 (ext 532) or by e-mailing [email protected].

OnlineNOW

2018 NATIONAL TEACHING INSTITUTE RESEARCH ABSTRACTS The 2018 National Teaching Institute (NTI) Research Abstracts are now available online. You can search, download, and print the 2018 NTI Research Abstracts at your convenience. Go to the AJCC website at www.ajcconline.org and click on the May issue. The 2018 research abstracts are listed in the OnlineNOW offerings. Abstracts can be searched using keywords or author names. (American Journal of Critical Care. 2018;27:e1-e17)

©2018 American Association of Critical-Care Nurses doi:https://doi.org/10.4037/ajcc2018805

Do you have a QR scanner app on your iPhone or Android? Scan this QR code with your phone to access this article instantly.


NEW HAMPSHIREManchesterHorizons 2018 - Call for Poster AbstractsDate: October 9-11, 2018. Place: Manchester, NH. Sponsor: Horizon Chapter of AACN. Contact: Pat Rosier. E-mail: [email protected]. Deadline for abstracts: May 1, 2018.

For AJCC Education Directory submission information, phone (800) 809-2273, ext 532, or e-mail [email protected].

Education DirectorySave Time. Learn More.We know you’re busy — busy at work, busy at home, busy all around.

most important articles in each issue of the American Journal of Critical Care.

Now you can get an early look at what’s coming in each issue of the American Journal of Critical Care — before the journal arrives in your mailbox — with eTOCs (electronic tables of contents).

Go toajcc.aacnjournals.org/cgi/alerts and sign up today

Watch Jillian’s story at

renalacute.com/stories

CRRTBUILT FOR my ICU.

When Jillian Kouns decided to make a career change

and become an ICU nurse 6 years ago, she was

amazed to see the difference that CRRT with the

PRISMAFLEX System could make for her patients,

but she really wanted to know more about the

therapy. Working closely with her Baxter rep, Amy,

she realizes she always has someone to count on,

whether that’s help with troubleshooting, getting the

latest updates on the therapy, or just showing her

new best practices. Because at Baxter, supporting

RNs like Jillian to deliver the best CRRT program

possible is our priority.

COMPLETE SUPPORT | PATIENT SAFETY | FLEXIBILITY

Baxter and Prismaflex are registered trademarks of Baxter International Inc. or its subsidiaries.USMP/MG120/18-0006a 02/18

The PRISMAFLEX Control Unit is intended for:

Continuous Renal Replacement Therapy (CRRT) for patients weighing 20 kilograms or more with acute renal failure and/or fluid overload.

Therapeutic Plasma Exchange (TPE) therapy for patients weighing 20 kilograms or more with diseases where fluid removal of plasma components is indicated.

Rx Only. For safe and proper use of this device, refer to the Operator’s Manual.

Baxter’s support has been such a service for our staff. They have provided amazing education and helped our leadership to continueto carry that on.”

Jillian, RN