Postoperative Emergency Response Team Activation at a Large Tertiary Medical Center

ORIGINAL ARTICLE

Postoperative Emergency Response TeamActivation at a Large Tertiary Medical CenterToby N. Weingarten, MD; Sam J. Venus, MD; Francis X. Whalen, MD;Brittany J. Lyne, SRNA; Holly A. Tempel, SRNA; Sarah A. Wilczewski, SRNA;Bradly J. Narr, MD; David P. Martin, MD, PhD; Darrell R. Schroeder, MS; andJuraj Sprung, MD, PhD

Abstract

Objective: To study characteristics and outcomes associated with emergency response team (ERT) activation in post-surgical patients discharged to regular wards after anesthesia.Patients and Methods: We identified all ERT activations that occurred within 48 hours after surgery from June 1,2008, through December 31, 2009, in patients discharged from the postanesthesia care unit to regular wards. For eachERT case, up to 2 controls matched for age (�10 years), sex, and type of procedure were identified. A chart review wasperformed to identify factors that may be associated with ERT activation.Results: We identified 181 postoperative ERT calls, 113 (62%) of which occurred within 12 hours of discharge fromthe postanesthesia care unit, for an incidence of 2 per 1000 anesthetic administrations (0.2%). Multiple logisticregression analysis revealed the following factors to be associated with increased odds for postoperative ERT activation:preoperative central nervous system comorbidity (odds ratio [OR], 2.53; 95% confidence interval [CI], 1.20-5.32;P�.01), preoperative opioid use (OR, 2.00; 95% CI, 1.30-3.10; P�.002), intraoperative use of phenylephrine infusion(OR, 3.05; 95% CI, 1.08-8.66; P�.04), and increased intraoperative fluid administration (per 500-mL increase, OR,1.06; 95% CI, 1.01-1.12; P�.03). ERT patients had longer hospital stays, higher complication rates, and increased30-day mortality compared with controls.Conclusion: Preoperative opioid use, history of central neurologic disease, and intraoperative hemodynamicinstability are associated with postoperative decompensation requiring ERT intervention. Patients with theseclinical characteristics may benefit from discharge to progressive or intensive care units in the early postoperativeperiod.

© 2012 Mayo Foundation for Medical Education and Research � Mayo Clin Proc. 2012;87(1):41-49

mppa

waethctt

PTtBrtv

FAFBsaMaCgional Medical Center, Or-lando, FL (S.J.V.).

E mergency response teams (ERTs) have beenintroduced by hospitals to evaluate and man-age hospitalized patients whose condition is

acutely deteriorating. Patients assessed as clini-cally stable and able to be managed on regularwards (ie, standard nursing wards) may experi-ence acute deterioration,1,2 and the ERT is de-signed to promptly deliver care to these patients.Identification of characteristics that can predictpostoperative adverse events would be desirablebecause early intervention may prevent more se-vere complications.2 Critical analysis of types andcauses of ERT activation may provide clues to ear-lier identification and, potentially, prevention ofimpending complications.

Because ERT systems are costly,3 some authorshave proposed preemptive triage of higher-acuitypatients to intensive care units (ICUs) or progressivecare units (eg, step-down units).4 The problem withsuch an approach is that sudden postoperative ad-verse events can occur even in patients whose con-dition was stable in the postanesthesia recovery unit
(PACU) and who fulfilled discharge criteria for dis- t
Mayo Clin Proc. � January 2012;87(1):41-49 � doi:10.1016/j.mayocpwww.mayoclinicproceedings.org

issal to regular wards.5 It is unknown whetherostoperative adverse events can be predicted fromatients’ comorbidities or other aspects of perioper-tive management.

Our objective was to examine factors associatedith the need for ERT activation after dismissal from

nesthetic care. Because we were primarily inter-sted in perioperative factors, our study was limitedo ERTs activated within the first 48 postoperativeours. Identification of factors associated with in-reased risk for these events may help to betterriage patients and minimize adverse postopera-ive outcomes.

ATIENTS AND METHODShe approval for review of medical records was ob-

ained from the Mayo Clinic Institutional Reviewoard in Rochester, MN. This study employed a ret-ospective case-control design that assessed poten-ial factors associated with the need for an ERT acti-ation after either surgery or diagnostic procedures
hat required anesthesia.
.2011.08.003 � © 2012 Mayo Foundation for Medical Education and

rom the Department ofnesthesiology (T.N.W.,.X.W., B.J.L., H.A.T., S.A.W.,.J.N., D.P.M., J.S.) and Divi-ion of Biomedical Statisticsnd Informatics (D.R.S.),ayo Clinic, Rochester, MN;

nd Department of Criticalare Medicine, Orlando Re-

Research 41

MAYO CLINIC PROCEEDINGS

42

The Department of Anesthesiology at MayoClinic prospectively maintains a log of all ERT acti-vations that occur at 2 Mayo Clinic-affiliated hospi-tals in Rochester: Saint Marys and Methodist. Usingthis log, we identified adult patients who requiredERT activation within 48 hours of discharge fromthe PACU to regular wards from June 1, 2008, toDecember 31, 2009. (Regular ward refers to a stan-dard nursing ward where patient vital signs are as-sessed at protocol-defined intervals as well as whenclinically indicated. Care may include continuouspulse oximetry but not invasive monitoring as in anICU or progressive care unit setting.) By June 1,2008, the ERT activation log system had been fullyimplemented and was capturing 100% of events.The 48-hour time window was selected to allowidentification of factors directly related to the intra-operative course. Surgical patients discharged tomonitored wards were excluded. (Monitored ward isdefined as an advanced patient care ward where pa-tient vital signs are continually monitored as in anICU or other specialized patient care areas wherecontinual monitoring is indicated, eg, progressivecare unit.) Patients who underwent cardiac catheter-ization, bronchoscopy, or childbirth were excluded.For each patient who required ERT activation, weused the Mayo Clinic medical record database toidentify potential controls of the same sex and sim-ilar age (�10 years) who underwent the same pro-cedure (as determined from International Classifica-tion of Diseases, Ninth Revision procedure codes)during the study period and did not have ERT acti-vation in the first 48 postoperative hours. Fromthese pools of potential controls, we randomly se-lected up to 2 controls for each ERT patient. Forcases in which fewer than 2 potential controls couldbe identified, we did not select alternative controls.

Indications for ERT ActivationsAt our institution an ERT consists of either a rapidresponse team (RRT) or code team. An RRT call canbe initiated by any health care team member con-cerned about the acutely deteriorating medical con-dition of a patient. Typically at our institution, RRTcalls are prompted as described by others,6 and in-dications include the following: decline in oxyhe-moglobin saturation (assessed by either pulse oxim-etry or clinical assessment), bradypnea, tachypnea,profound bradycardia or tachycardia, hypotension,concern for possible heart attack (“chest pain”),stroke (acute neurologic deficits), or acute mentalstatus changes (agitation, delirium). However, anRRT call can be initiated for any other indication atthe discretion of the health care team member. TheRRT consists of an attending physician board-certi-fied in critical care medicine, a critical care fellow or
senior anesthesia resident, a respiratory therapist,
Mayo Clin Proc. �

and a critical care registered nurse. Code team acti-vations are reserved for immediate life-threateningevents (cardiopulmonary arrest, severe respiratorycompromise that is assessed to require tracheal in-tubation and mechanical ventilation) or profoundlyunstable cardiac conditions (possibly requiring car-dioversion, defibrillation). The code team is similarin structure to the RRT but consists of an additionalcritical care registered nurse, an internal medicineresident, and a pharmacist. The level of ERT (RRT orcode team) activation is left to the discretion of theindividual health care team member (ie, nurse car-ing for an unstable patient); thus it is prone to sub-jectivity. Therefore, the level of activation occasion-ally is erroneous (making a “mistake on the safeside” by activating the code team when the RRTwould have been more appropriate). Regardless, theroles and capabilities of these 2 teams are closelyinterrelated. Because our interest was to examinepatients who had an acute postoperative deteriora-tion, we analyzed all ERT interventions, regardlessof the level of activation.

Data AbstractionElectronic medical records were abstracted for de-mographics; comorbid conditions; preoperative, in-traoperative, and postoperative variables; postoper-ative course and complications; and details of theERT activation. Comorbid conditions were definedaccording to definitions used for numerous out-come studies at Mayo Clinic,7 including cardiovas-cular disease (coronary artery disease [myocardialinfarction, coronary stent placement, or cardiac by-pass surgery], congestive heart failure or cardiomy-opathy [ejection fraction �40%], the potential forcardiac dysrhythmia [atrial fibrillation or flutter, im-planted pacemaker and/or automated defibrillator],arterial hypertension [medically treated], peripheralvascular disease), central neurologic disease (historyof seizures, dementia, stroke, or transient ischemicattacks), pulmonary disease (asthma, chronic ob-structive or restrictive pulmonary disease, pulmo-nary hypertension, obstructive sleep apnea), diabe-tes mellitus (medically treated), kidney disease, andpreoperative scheduled use of opioids and/or ben-zodiazepines. Overall physical status was assessedfrom the American Society of Anesthesiologists(ASA) Physical Status score.

The anesthetic record was reviewed for anesthe-sia duration, anesthetic method (general, regional),urgency (elective, emergent), muscle relaxant use,blood transfusion, fluid administration, and periop-erative complications. Complications included he-modynamic instability or systemic arterial hypoten-sion (inferred by use of vasopressor infusion as a
surrogate for recalcitrant hypotension), need for car-
January 2012;87(1):41-49 � doi:10.1016/j.mayocp.2011.08.003www.mayoclinicproceedings.org

tgctrd(wld.Aw

RDtrlvoa1aicttorthl(ldi

tgctmvrpi

ANESTHESIA AND EMERGENCY TEAM ACTIVATION

dioversion, adverse respiratory events, or other se-vere perioperative complication.

All data were collected and managed usingREDCap electronic data capture tools hosted atMayo Clinic.8 ERT notes were abstracted and sup-plemented by review of the medication administra-tion record, ICU admission note, subsequent progressnotes, and discharge summaries. Data abstracted in-cluded the probable primary cause of the ERT activa-tion: hypotension from hypovolemia or distributiveshock, respiratory cause, cardiac cause, hyperten-sive crisis, neurologic causes such as mental statuschanges, uncontrolled pain, psychiatric reasons, ordrug reactions. The following interventions were re-corded: respiratory (tracheal intubation, applica-tion of a noninvasive ventilatory device such as acontinuous positive airway pressure device, bron-chodilator administration), cardiac (cardiopulmo-nary resuscitation, defibrillation or cardioversion,administration of vasoactive drugs, nitroglycerin,antiarrhythmic drugs, or diuretics), intravenousfluid bolus or blood product administration, glu-cose administration, or administration of analgesics,sedatives, naloxone, flumazenil, or antipsychotics.Immediate outcome of the ERT was categorized asfollows: remained in the previous hospital settingwith or without intervention, transfer to a moni-tored ward, transfer to the operating room for ex-ploration or treatment, or death.

OutcomesPostoperative complications that occurred withinthe first 30 postoperative days were reported. Infor-mation was obtained from the medical records fromthe index hospitalization, rehospitalization, or out-patient visits. A 30-day mortality rate was calcu-lated. Postoperative complications included myo-cardial infarction, cerebrovascular event, respiratoryfailure requiring tracheal reintubation, acute kidneyinjury (serum creatinine increase �1 mg/dL andabove 1.5 mg/dL [to convert to �mol/L, multiply by88.4]), thromboembolic event, sepsis or multiorganfailure, blood transfusion requirement, or death.Causes of death were recorded. Total days in theICU and hospital were recorded.

Statistical AnalysesData are summarized using mean � SD or medianwith interquartile range (IQR) for continuous vari-ables and frequency percentage for nominal vari-ables. To estimate the incidence of ERT activation, adenominator was obtained using information pro-vided by the revenue accounting office; it was basedon patient counts after excluding all surgical or pro-cedural categories of patients expected to be admit-
ted to monitored wards (eg, cardiac, major vascular, a

thoracic). Of note, this denominator was not re-duced to account for other planned or unplannedadmissions to monitored wards in patients under-going operations that do not routinely require thislevel of postoperative care. Two sets of analysescomparing characteristics between ERT cases andcontrols were performed. One analysis included allERT cases and compared characteristics betweencases and controls using the 2-sample t test for con-inuous variables and the Fisher exact test for cate-orical variables. The other analysis excluded ERTases for which no matched controls could be iden-ified and was performed using conditional logisticegression, taking into account the matched studyesign. Characteristics found to have evidenceP�.05) of an association in univariate analysesere included as explanatory variables in a multiple

ogistic regression model with ERT activation as theependent variable. In all cases, 2-tailed P values of

05 or less were considered statistically significant.nalyses were performed using SAS statistical soft-are (Version 9.2, SAS Institute, Inc., Cary, NC).

ESULTSuring the study period approximately 95,000 pa-

ients underwent surgery or diagnostic proceduresequiring anesthesia and were discharged to a regu-ar ward. Of those, 181 patients required ERT acti-ation within 48 hours; therefore, the estimated ratef ERT activation in this population was 2 per 1000nesthetic administrations (0.2%). Of these events,68 (93%) were RRT, and 13 (7%) were code teamctivations. Of the code team activations, 6 met thenstitutional definition of a code (5 patients receivedardiopulmonary resuscitation, and in one patienthe trachea was intubated to protect the airway inhe context of acute mental status changes). In thether 7 patients, the code team was activated foreasons other than cardiopulmonary collapse (eg,ransient syncope, hypotension). Of the entire co-ort, 81 patients (45%) underwent general or uro-

ogic surgery; 62 (34%), orthopedic surgery; 1910%), gynecologic surgery; 11 (6%), otolaryngo-ogical surgery; 5 (3%), neurosurgery; and l3 (7%),iagnostic procedures or minor operations requir-

ng anesthetic care outside the operating room.The majority (N�113; 62.4%) of ERT interven-

ions occurred during the first 12 hours after sur-ery, and more than three-fourths (142; 78.5%) oc-urred within the first 24 hours (Figure). The meanime from ERT activation to team arrival was 4�2inutes. Table 1 describes the reasons for ERT acti-

ation and types of interventions. The most frequenteasons for ERT activation were hypotension in 58atients (32%), cardiac in 36 (20%), and pulmonary

n 31 (17%). In all cases a physician on the ERT
ssessed the patient and made decisions regarding
.2011.08.003 43


44

treatment and disposition. Immediate outcomes ofERT activation are summarized in Table 1.

The median ICU length of stay of patientstransferred to a monitored ward was 2 days (IQR,2-4 days). Two patients died during code teamcalls, both from massive saddle pulmonary em-boli associated with a cardiac arrest. The admin-istration of a fluid bolus (N�63; 35%) was themost common overall intervention, followed bynaloxone administration (N�16; 9%) and opioidadministration (N�12; 7%) (Table 1).

Using previously described matching criteria,we identified 318 controls for these 181 ERT pa-tients. For 154 ERT patients (85%) we identified 2matched controls; for 10 ERT patients (6%) only 1control could be identified, and for 17 ERT patients(9%) no controls could be identified because thepatient underwent an operation that was uncom-mon or that rarely requires anesthesia. Clinical anddemographic features of ERT cases and controls aresummarized in Table 2. ERT and control groupswere similar with the exceptions of higher rates ofcentral neurologic diseases and preoperative use ofopioid analgesics in ERT patients. Among 18 ERTpatients (10%) with preexisting central neurologicdiseases, 5 (28%) had an ERT activation for worsen-ing neurologic status. Among 55 ERT patients(30%) who used opioids, 7 (13%) received nalox-one during the ERT intervention.

Surgical characteristics in the control and ERTgroups are shown in Table 3. The perioperativecourse was similar between groups except that more

18

Time from end of anesthesia (h)24 30 36 42 48

quency of time to emergency response team

ERT patients had intraoperative hemodynamic in-

Mayo Clin Proc. �

stability as reflected by an increased use of phenyl-ephrine infusion and greater intravenous fluid to-tals. Of the 10 ERT patients (6%) who receivedintraoperative phenylephrine infusion, 8 requiredERT intervention for later hemodynamic instability(5, hypotension; 3, cardiac).

The intraoperative course in both groups wasdevoid of any serious complications. The PACUcourse was generally unremarkable, except for 1ERT patient who required a transient phenylephrineinfusion and fluid bolus but was subsequently dis-charged to the regular nursing ward. No complica-tions (tracheal intubation, aspiration, laryngo-spasm, pulmonary edema, or seizures) occurred inthe PACU, and all patients were transferred to reg-ular wards. Most patients were discharged to regularwards with supplemental oxygen (188 ERT patients[61%] and 110 controls [50%]; P�.78).

From multivariate analysis, the following fac-tors were found to be significantly associated withERT activation (Table 4): preoperative central ner-vous system comorbidity (odds ratio [OR], 2.53;95% confidence interval [CI], 1.20-5.32; P�.01),preoperative scheduled opioid use (OR, 2.00; 95%CI, 1.30-3.10; P�.002), intraoperative use of phen-ylephrine infusion (OR, 3.05; 95% CI, 1.08-8.66;P�.04), and greater intraoperative fluid adminis-tration (per 500-mL fluid bolus, OR, 1.06; 95%CI, 1.01-1.12; P�.03). Similar results were ob-tained from the matched-set analysis that ex-cluded the 17 ERT cases (9%) with no matchedcontrols (Table 4).

The length of stay was longer for patients whorequired ERT activation (median, 4 days [IQR, 3-8days] vs 3 days (IQR, 1-4 days); P�.001). Compli-cation rates during hospitalization, including 30-day mortality rates, were higher among patientswho required ERT activation compared with con-trols (Table 5).

DISCUSSIONRecovery from surgery or diagnostic procedures re-quiring anesthesia may be associated with compli-cations. To avoid serious morbidities, it is importantto anticipate potential problems so that preemptivemeasures can be implemented. Our study demon-strated that preoperative scheduled use of opioids,history of central neurologic disease, and hemody-namic instability during surgery were characteristicsindependently associated with unexpected deterio-ration within the first 48 postoperative hours. Themajority of ERT calls occurred within the first 24postoperative hours, most often for hypotension,mental status changes, or respiratory problems. De-spite the fact that some patients in the ERT cohortrequired minimal to no intervention, ERT patients

100

80

60

40

20

00 6 12

Perc

enta

ge o

f ER

T ca

lls

FIGURE. Cumulative fre(ERT) activation.

had more severe in-hospital complications.


ipdcidirs


ERTs have been established to intervene quicklyto mitigate sudden deterioration of patients’ health.Although this initiative intuitively appears effective,ERT outcomes are difficult to assess. ERT interven-tion may reduce ICU resource utilization9 and im-mediate mortality after cardiac arrest.3,6,10-13 How-ever, a large meta-analysis did not find evidence thatthese initiatives reduce the overall in-hospital mor-tality.14 In contrast, introduction of ERTs wasfound to increase long-term survival in surgicalpatients.15 This discrepancy14,15 may be relatedto differences in disease complexity between sur-gical and medical patients. Surgical critical events maybe associated with more reversible causes (bleeding,oversedation, hypotension), whereas medical criticalevents may be related to advanced or terminal condi-tions. If this is true, then implementation of ERTs, bet-ter triage of higher-risk patients, or both may improvepostoperative morbidity and mortality.

Characteristics used to predict ERT activationhave been reported only once.16 A small case-con-trol study identified ASA Physical Status class 3 orgreater and after-hours surgery as predictors. As inour study, hypotension and decreased level of con-sciousness were the main reasons for ERT activa-tion.16 That study used ASA Physical Status as asurrogate for comorbidity,16 but the small cohortprecluded examining the relationship between spe-cific comorbidities and intraoperative events withearly postoperative complications. In our larger se-ries, we identified 3 markers for increased postop-erative ERT activation.

The first predictor was preoperative opioid ther-apy that led to respiratory depression and overseda-tion (as evidenced by a higher rate of naloxone ad-ministration). Other investigators also reported thatpatient-controlled analgesia17 and intravenous mor-phine have been associated with higher rates of ERTactivation for respiratory depression.18 Further-more, it is known that opioid-tolerant patients post-operatively report greater intensity of pain and usemore opioid analgesics than opioid-naïve pa-tients.19 A retrospective case-control series foundthat 47.8% of opioid-tolerant patients experiencedpostoperative moderate to severe sedation com-pared with 18.5% of opioid-naïve patients.20 Ahigher proportion of these patients in our study re-ceived naloxone (N�7 [13%] vs N�9 [7%]), fur-ther suggesting the need for increased vigilance forsigns of oversedation.

The second predictor was the use of phenyleph-rine infusions and greater intravenous fluid admin-istration. At our institution we correct brief episodesof hypotension with boluses of ephedrine or phen-ylephrine but typically initiate phenylephrine infu-sions, in addition to fluid boluses, when hypoten-
sion persists. Of note, these patients received more H

ntraoperative fluids (crystalloids and colloids) com-ared with the average amounts given to the remain-er of ERT patients (see footnote to Table 3). Weonsequently used the initiation of “phenylephrinenfusion” as a surrogate for more pronounced hemo-ynamic instability. At our institution, phenylephrine

nfusion is considered a temporizing measure to cor-ect for factors associated with perioperative hypoten-ion, and it is widely viewed as a benign intervention.

TABLE 1. Causes for Emergency Response Team (ERInterventions During ERT Call

Characteristic

Causes for ERT activation

Hypotension

Cardiac

Pulmonary

Neurologic

Pain/psychiatric issues

Drug interactions

Hypertension

Epistaxis

Specific interventions

Chest compressions

Defibrillation/cardioversion

Tracheal intubation

Noninvasive ventilation

Other interventions and medications administered

Intravenous fluid bolus

Naloxone

Opioids

Blood transfusion

Vasopressors (epinephrine, phenylephrine)

Bronchodilator

Benzodiazepine

Anticholinergic (atropine)

Flumazenil

Antiarrhythmic (amiodarone)

Diuretic (furosemide)

50% glucose intravenous bolus

Antihypertensive (labetalol, hydralazine)

Nitroglycerin

Immediate outcome

Death

Transfer to surgery

Transfer to monitored ward

Treatment on regular ward

Continued observation on regular ward

Data are presented as No. (percentage).

T) Activation and Specific

Patients (N�181)

58 (32)

36 (20)

31 (17)

23 (13)

17 (9)

12 (7)

3 (2)

1 (1)

5 (3)

2 (1)

5 (3)

9 (5)

63 (35)

16 (9)

12 (7)

10 (6)

5 (3)

5 (3)

3 (2)

2 (1)

2 (1)

2 (1)

1 (1)

2 (1)

1 (1)

1 (1)

2 (1)

4 (2)

71 (39)

57 (32)

47 (26)

owever, among ERT patients who required phenyl-

.2011.08.003 45

fi

osis,


46

ephrine infusion, postoperative hypotension was themost common indication for ERT activation. There-fore, our perception that a phenylephrine infusion isnot a marker for subsequent instability, even after astable hemodynamic course in the PACU, needs to bereexamined. Another study demonstrated that intra-operative hypotension predicts postoperative adverseevents.21

A third predictor was a history of central neuro-logic disease. Preexisting cognitive dysfunction is astrong predictor of postoperative delirium.22 In our

TABLE 2. Demographics and Preoperative ComorbidiRequired Emergency Response Team (ERT) Activation

Variable C

Age (y), mean � SD

Body mass index (kg/m2), mean � SD

Male sex

ASA Physical Status

1-2

3-4

Preoperative creatinine (mg/dL), mean � SDc

Comorbidities

Cardiovasculard

Hypertension

Coronary artery disease

Atrial fibrillation or flutter

Pacemaker/internal defibrillator

Peripheral vascular disease

Congestive heart failure

Respiratoryd

Obstructive sleep apnea

Preoperative use of CPAP

Severe chronic lung diseasee

Asthma

Central nervous systemd

Stroke or transient ischemic attacks

Seizures

Dementia

Diabetes mellitus

Preoperative scheduled medication use

Opioids

Benzodiazepines

a Data are presented as No. (percentage) unless indicated otherCPAP � continuous positive airway pressure.b P values are from t tests and Fisher exact test for categorical vc Data are missing for 59 patients. To convert mg/dL to �mol/Ld Patients may have more than one comorbidity within the givenexceed the total for the overall category.e Includes chronic obstructive pulmonary disease, pulmonary fibr

patients, acute postoperative decline in neurologic

Mayo Clin Proc. �

unction was a frequent cause for ERT activation. Sim-larly, Lee et al16 reported a decline in mental status asa common reason for ERT activation; however, theydid not comment on preoperative neurologic diseaseas a risk factor. The relatively small number of pa-tients in our study with this condition precludesus from making a general recommendation re-garding preemptive triage of these patients to ahigher level of postoperative care. Furthermore,the hospital outcomes of these patients weregood; more than 50% did not require transfer to a

in Controls and in Patients Who Subsequently

rol (N�318)a ERT (N�181) P valueb

9.4�17.0 59.9�17.4 .77

9.0�6.3 28.3�6.4 .23

152 (48) 88 (49) .93

.12

198 (62) 99 (55)

120 (38) 82 (45)

1.2�2.3 1.1�0.7 .69

146 (46) 94 (52) .23

141 (44) 82 (45)

38 (12) 28 (15)

20 (6) 12 (7)

5 (2) 10 (6)

10 (3) 10 (6)

12 (4) 9 (5)

60 (19) 45 (25) .14

31 (10) 21 (12)

18 (6) 4 (2)

10 (3) 14 (8)

24 (8) 16 (9)

14 (4) 18 (10) .02

10 (3) 8 (4)

2 (1) 6 (3)

2 (1) 6 (3)

55 (17) 29 (16) .80

57 (18) 55 (30) .002

36 (11) 19 (10) .88

ASA � American Society of Anesthesiologists;

les.iply by 88.4.ory; therefore, the sum of the numbers within the category may

and pulmonary hypertension.

ties

ont

5

2

wise.

ariab, multcateg

higher level of care after ERT activation.


fusion


The PACU course was generally uneventful inour patients, which is not surprising because to beincluded in our study, they needed to be in suffi-ciently stable condition to warrant discharge toregular wards.5 A study of respiratory and cardio-vascular events in the PACU demonstrated thattachycardia and hypertension were predictive of un-

TABLE 3. Anesthetic and Intraoperative CharacteristicResponse Team (ERT) Activationa

Characteristic Con

Emergency procedure

Type of anesthesia

General

Neuroaxial

Peripheral nerve block � sedation

Airway management

Endotracheal intubation

Laryngeal mask airway

Mask or nasal cannula

Intraoperative use

Nondepolarizing muscle blockers

Blood transfusion

Crystalloids/colloids (L), mean � SD

Phenylephrine infusion

Antihypertensives

Adverse intraoperative events

Bronchospasm

Hypoxemia

Anesthetic duration (h), mean � SD

a Data are presented as No. (percentage) unless indicated otherb ERT patients (n�10) who had intraoperative phenylephrine in

TABLE 4. Multivariate Analysis of Factors Associated

All ERT

Odds ratio

Preoperative

Central nervous system comorbidityb 2.53

Preoperative scheduled opioid use 2.00

Intraoperative

Phenylephrine infusion 3.05

Crystalloids/colloidsc 1.06

a Characteristics found to have a significant univariate associationmodel. In addition to an analysis that included all ERT cases and ccases that did not have any matched controls. The matched-set aaccount the matched-set study design. CI � confidence intervab Preoperative central nervous system comorbidities include cerc
Odds ratios are for a 500-mL increase.

planned ICU admission and increased mortality,demonstrating a relationship between early postop-erative clinical deterioration and long-term out-comes.23 Furthermore, this study revealed a strongtendency for patients to exhibit similar cardiovascu-lar events intraoperatively and in the PACU.23 Inour study, patients with intraoperative hypotension

Controls and in Patients Who Required Emergency

(N�318) ERT (N�181) P value

(5) 7 (4) .66

.66

(81) 145 (80)

(11) 18 (10)

(8) 18 (10)

.93

(76) 136 (75)

(4) 9 (5)

(19) 36 (20)

(42) 87 (48) .26

(7) 18 (10) .24

1.5 2.6�2.0 .04

(2) 10 (6)b .03

(10) 12 (7) .23

(2) 6 (3) .32

(1) 1 (1) .92

1.7 3.4�2.1 .57

received 3.3�1.6 L of fluids intraoperatively.

Emergency Response Team (ERT) Activationa

and controls Matched-set analysis

5% CI P value Odds ratio 95% CI P value

0-5.32 .01 2.44 1.07-5.58 .04

0-3.10 .002 1.95 1.18-3.24 .01

8-8.66 .04 3.85 1.25-11.84 .02

1-1.12 .03 1.12 1.04-1.20 .003

Tables 2 and 3) were included in the multiple logistic regressionls, a matched-set analysis was performed that excluded 17 ERT

is was performed using conditional logistic regression, taking into

ascular disease, seizures, and dementia.

s in

trol

16

257

36

25

243

14

61

135

22

2.2�

6

31

6

2

3.3�

wise.

With

cases

9

1.2

1.3

1.0

1.0

(seeontronalys

l.ebrov

.2011.08.003 47


48

had unremarkable PACU stays, and yet some expe-rienced subsequent hemodynamic instability. Thissuggests that the aggressive volume replacementcombined with intensive monitoring maintainedhemodynamic stability throughout the PACU stay.However, the underlying comorbidities or condi-tions that led to subsequent instability after dis-charge to a regular ward were either not manifest ornot recognized in the PACU.

This is a retrospective study with all inherentlimitations. Because we limited our study to ERTcalls during the first 48 postoperative hours, wecould have missed significant complications that oc-curred outside that time frame but still were relatedto anesthesia and surgery. Despite the existence ofwell-defined clinical criteria for ERT activation, con-siderable subjectivity is involved. More specifically,activation is triggered typically by allied health careprofessionals with varied levels of training; there-fore, some ERT activations were initiated for non-critical reasons (eg, epistaxis), which affects the abil-ity to identify factors associated with increased odds

TABLE 5. Outcomes for Controls and for Patients WhoDuring Hospitalizationa

Complicationsb

Control(N�318

No. of patie

Myocardial infarction 1 (�1

Stroke 1 (�1

Mechanical ventilationc 2 (1)

Renal failure 7 (2)

Deep vein thrombosis 1 (�1

Pulmonary embolus 1 (�1

Sepsis/multiorgan failure 3 (1)

Need for blood transfusion 3 (1)

Mortality at 30 dd

Alive 315 (99)

Dead 3 (1)

a Data are presented as No. (percentage).b These complications represent all complications that occurreddischarge and are not limited to those occurring at the time ofc Mechanical ventilation was implemented for patients who requmonary arrest, acute lung injury, pneumonia, sepsis, or multisystd All deaths in the control group, and 1 death in the ERT group, oin the ERT group included cardiac arrest during the time of the raemboli as determined on autopsy; pneumonia in 1; multiorganout-of-hospital deaths were not determined. The 3 control plymphoma who had cardiac arrest after readmission followinscleroderma (with cardiac and pulmonary involvement) who h62-year-old woman with renal cell carcinoma and multiple pulmalignant effusion under anesthesia. The ERT patient who dieda left hemimandibulectomy.

for more severe events. This is reflected in our data

Mayo Clin Proc. �

in that 47 ERT activations (26%) resulted in no in-tervention by the team. This is also the reason whywe decided to include cases from both RRT and codeteam activations; ie, the reasons why one team wasactivated over the other are at times arbitrary. Re-gardless, we think that activation of either team rep-resents an acute unexpected health decompensationand thus is relevant for study. We believe our re-ported incidence represents an underestimate be-cause invariably some patients were discharged tothe ICU for other reasons but were still included inthe denominator. However, our incidence of 0.2% issimilar to a previous report.16

CONCLUSIONPatients with preoperative central neurologic dis-ease, scheduled preoperative opioid use, or intraop-erative hemodynamic instability were found to be atincreased risk for ERT activation within 48 postop-erative hours. More vigilant monitoring of such pa-tients may be warranted in the immediate postoper-

Emergency Response Team (ERT) Intervention

ERT (N�181)

P valueNo. of patients

5 (3) .03

6 (3) .01

16 (9) �.001

12 (7) .03

3 (2) .14

2 (1) .30

15 (8) �.001

20 (11) �.001

.02

173 (96)

8 (4)

the time of postanesthesia recovery unit discharge to hospitalRT intervention.entilatory support for respiratory failure arising from cardiopul-

rgan failure.d following hospital discharge. Causes of the 7 in-hospital deaths

esponse team activation in 2 patients, due to massive pulmonaryin 3; and acute hemorrhage in 1. The specific causes of the 4

s were as follows: an 85-year-old woman with non-Hodgkinction of a tongue lesion; a 57-year-old man with advanceddergone a muscle biopsy under monitored anesthetic care; ary emboli who died after placement of a pleural catheter forf hospital was a 94-year-old woman who previously underwent

Had

)

nts

)

)

)

)

fromthe Eired vem occurrepid r

failureatientg resead unmonaout o

ative period.


1

1

1

1

1

1

1

2

2

2

2


ACKNOWLEDGEMENTWe are thankful to Andrew Hanson for assistancewith data management and statistical analyses.

Grant Support: This project was supported by the Depart-ment of Anesthesiology, Mayo Clinic, Rochester, MN, andNational Institutes of Health/National Center for ResearchResources Clinical and Translational Science Awards GrantNumber UL1 RR024150.Its contents are solely the responsibility of the authors anddo not necessarily represent the official views of the Na-tional Institutes of Health.

Correspondence: Address to Juraj Sprung, MD, PhD, De-partment of Anesthesiology, Mayo Clinic, 200 First St SW,Rochester, MN 55905 ([email protected]).

REFERENCES1. George AL Jr, Folk BP III, Crecelius PL, Campbell WB. Pre-

arrest morbidity and other correlates of survival after in-hospital cardiopulmonary arrest. Am J Med. 1989;87(1):28-34.

2. Schein RM, Hazday N, Pena M, Ruben BH, Sprung CL. Clinicalantecedents to in-hospital cardiopulmonary arrest. Chest.1990;98(6):1388-1392.

3. Dacey MJ, Mirza ER, Wilcox V, et al. The effect of a rapidresponse team on major clinical outcome measures in a com-munity hospital. Crit Care Med. 2007;35(9):2076-2082.

4. Litvak E, Pronovost PJ. Rethinking rapid response teams. JAMA.2010;304(12):1375-1376.

5. Aldrete JA. The post-anesthesia recovery score revisited. J ClinAnesth. 1995;7(1):89-91.

6. Bellomo R, Goldsmith D, Uchino S, et al. Prospective con-trolled trial of effect of medical emergency team on postop-erative morbidity and mortality rates. Crit Care Med. 2004;32(4):916-921.

7. Hosking MP, Warner MA, Lobdell CM, Offord KP, Melton LJ.Outcomes of surgery in patients 90 years of age and older.JAMA. 1989;261(13):1909-1915.

8. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG.Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providingtranslational research informatics support. J Biomed Inform.2009;42(2):377-381.

9. Baxter AD, Cardinal P, Hooper J, Patel R. Medical emergencyteams at The Ottawa Hospital: the first two years. Can J
Anaesth. 2008;55(4):223-231.

10. Bellomo R, Goldsmith D, Russell S, Uchino S. Postoperativeserious adverse events in a teaching hospital: a prospectivestudy. Med J Aust. 2002;176(5):216-218.

11. Bellomo R, Goldsmith D, Uchino S, et al. A prospectivebefore-and-after trial of a medical emergency team. Med JAust. 2003;179(6):283-287.

12. DeVita MA, Braithwaite RS, Mahidhara R, Stuart S, Foraida M,Simmons RL. Use of medical emergency team responses toreduce hospital cardiopulmonary arrests. Qual Saf Health Care.2004;13(4):251-254.

3. Jones D, Bellomo R, Bates S, et al. Long term effect of amedical emergency team on cardiac arrests in a teachinghospital. Crit Care. 2005;9(6):R808-R815.

4. Chan PS, Jain R, Nallmothu BK, Berg RA, Sasson C. Rapidresponse teams: a systematic review and meta-analysis. ArchIntern Med. 2010;170(1):18-26.

5. Jones D, Opdam H, Egi M, et al. Long-term effect of a medicalemergency team on mortality in a teaching hospital. Resusci-tation. 2007;74(2):235-241.

6. Lee A, Lum ME, O’Regan WJ, Hillman KM. Early postoperativeemergencies requiring an intensive care team intervention: therole of ASA physical status and after-hours surgery. Anaesthe-sia. 1998;53(6):529-535.

7. Overdyk FJ, Carter R, Maddox RR, Callura J, Herrin AE, Hen-riquez C. Continuous oximetry/capnometry monitoring re-veals frequent desaturation and bradypnea during patient-controlled analgesia. Anesth Analg. 2007;105(2):412-418.

8. Fecho K, Joyner L, Pfeiffer DL. Opioids and code blue emer-gencies [abstract]. Anesthesiology. 2008;109:A34. http://www.asaabstracts.com/strands/asaabstracts/abstract.htm;jsessionid�

DA80E5E3297E82419E075E6EC3F84D34?year�2008&index�

15&absnum�1555. Accessed September 1, 2011.9. Ready LB. Acute pain: lessons learned from 25,000 patients.

Reg Anesth Pain Med. 1999;24(6):499-505.0. Rapp SE, Ready LB, Nessly ML. Acute pain management in

patients with prior opioid consumption: a case-controlled ret-rospective review. Pain. 1995;61(2):195-201.

1. Kheterpal S, O’Reilly M, Englesbe MJ, et al. Preoperative andintraoperative predictors of cardiac adverse events after gen-eral, vascular, and urological surgery. Anesthesiology. 2009;110(1):58-66.

2. Robinson TN, Raeburn CD, Tran ZV, Angles EM, Brenner LA,Moss M. Postoperative delirium in the elderly: risk factors andoutcomes. Ann Surg. 2009;249(1):173-178.

3. Rose DK, Cohen MM, DeBoer DP. Cardiovascular events inthe postanesthesia care unit: contribution of risk factors. An-
esthesiology. 1996;84(4):772-781.
.2011.08.003 49

mailto:[email protected]

http://www.asaabstracts.com/strands/asaabstracts/abstract.htm;jsessionid=DA80E5E3297E82419E075E6EC3F84D34?year=2008&index=15&absnum=1555




Documents

Postoperative Emergency Response Team Activation at a Large Tertiary Medical Center