�������� ����� ��

Esophageal button battery ingestions: Decreasing time to operative interven-tion by level I trauma activation

Robert T. Russell, Russell L. Griffin, Elizabeth Weinstein, Deborah F.Billmire

PII: S0022-3468(14)00062-1DOI: doi: 10.1016/j.jpedsurg.2014.01.050Reference: YJPSU 56682

To appear in: Journal of Pediatric Surgery

Received date: 15 October 2013Revised date: 27 January 2014Accepted date: 29 January 2014

Please cite this article as: Russell Robert T., Griffin Russell L., Weinstein Elizabeth,Billmire Deborah F., Esophageal button battery ingestions: Decreasing time to opera-tive intervention by level I trauma activation, Journal of Pediatric Surgery (2014), doi:10.1016/j.jpedsurg.2014.01.050

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

Esophageal Button Battery Ingestions: decreasing time to operative intervention by level I

trauma activation

Robert T. Russell, MD, MPH1, Russell L. Griffin, PhD2, Elizabeth Weinstein, MD,3 and Deborah

F. Billmire, MD4

Affiliations: 1University of Alabama at Birmingham, Division of Pediatric Surgery, Children’s

of Alabama, Birmingham, AL; 2University of Alabama at Birmingham, School of Public Health,

Department of Epidemiology, Birmingham, AL; 3Division of Emergency Medicine, JW Riley

Hospital for Children, Indianapolis, IN; 4Division of Pediatric Surgery, JW Riley Hospital for

Children, Indianapolis, IN

Address Correspondence to: Robert T. Russell, Division of Pediatric Surgery, Children’s of

Alabama, 1600 7th Avenue South, Lowder Building Suite 300, Birmingham, AL 35233,

[robert.russell@childrensal.org], 205-638-5339, Fax 205-975-4972

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

Abstract

Purpose: The incidence of button battery ingestions is increasing and injury due to esophageal

impaction begins within minutes of exposure. We changed our management algorithm for

suspected button battery ingestions with intent to reduce time to evaluation and operative

removal.

Methods: A retrospective study was performed to identify and evaluate time to treatment and

outcome for all esophageal button battery ingestions presenting to a major children’s hospital

emergency room from February 1, 2010 through February 1, 2012. During the first year,

standard emergency room triage (ST) was used. During the second year, the triage protocol was

changed and Trauma I triage (TT) was used.

Results: 24 children had suspected button battery ingestions with 11 having esophageal

impaction. One esophageal impaction was due to 2 stacked coins. Time from arrival in

emergency room to battery removal was 183 minutes in ST group (n=4) and 33 minutes in TT

group (n=7) (p=0.04). One patient in ST developed a tracheoesophageal fistula. There were no

complications in the TT group.

Conclusions: The use of Trauma 1 activations for suspected button battery ingestions has led to

more expedient evaluation and shortened time to removal of impacted esophageal batteries.

Key Words: button battery, level I trauma, esophagus

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

Introduction

There are several recent reports of button battery related esophageal injuries, including

erosive esophagitis, tracheoesophageal fistula, hemorrhagic complications, and even death.1-4

From 1990 to 2009, there were an estimated 66,000 battery related emergency department (ED)

visits in the United States in children < 18 years of age. The number and rate of visits increased

significantly during the twenty-year time period, which was due to a significant increase in

battery-related ED visits among children ≤ 5 years.5 A report by Litovitz et al in 2010 also

suggests that fatal and severe button battery ingestions are increasing, and this trend has been

associated with the growing utilization of 3 volt, 20-mm lithium button batteries.6

Recommendations from this report suggest removal of the button battery from the esophagus

within 2 hours from ingestion to limit tissue exposure. Due to the greater voltage and current of

lithium cells, longer impactions can result in more severe tissue damage leading to further

complications. Although time to presentation and duration of procedure are not subject to

control, time from presentation to operative removal may be improved with a standardized

approach. We changed our management algorithm for suspected button battery ingestions with

intent to reduce time to evaluation and operative removal. The goal of this study was to examine

the time to evaluation/treatment and outcomes of patients that were evaluated at our facility one

year prior to and one year after the change of button battery ingestion management algorithm.

Methods

Children presenting to a free standing children’s hospital with a level 1 trauma center

who had known or suspected history of button battery ingestion were included. Known patients

included those transferred from an outside facility with radiographic documentation of battery

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

ingestion. Suspected patients included those who presented to the triage desk from home with a

history of possible ingestion. Prior to February 1, 2010, standard triage (ST) was utilized for

initial evaluation of a suspected button battery ingestion in the emergency room. This involved

standard emergency room registration, initial nursing triage for a brief history of the presenting

problem and medical history followed by triage for physician evaluation. Confirmation of button

battery ingestion prompted immediate consultation with a surgical endoscopist (pediatric

surgeon, pediatric gastroenterologist, or pediatric ENT). After February 1, 2010, Trauma I triage

activation (TT) was utilized to facilitate more expedient evaluation of patients with suspected

button battery ingestion. TT entails immediate notification of the pediatric trauma team, the

anesthesia team, the radiology technician, and the operating room charge nurse of the arrival of

trauma patients to the emergency department. This allows immediate response of the pediatric

trauma surgeon and immediate confirmation of presence and site of battery by imaging.

Availability of an operating room for potential operative intervention is also secured. After IRB

approval, a retrospective chart review was performed to identify all button battery ingestions

from February 1, 2010 through February 1, 2012. To ensure complete capture of button battery

ingestions, we searched the medical records utilizing Current Procedural Terminology (CPT)

codes for endoscopic retrieval of foreign body (43215, 43247). We reviewed each chart

associated with these CPT codes and eliminated all foreign bodies that were not suspected button

batteries. We then collected and analyzed data including type of battery or foreign body,

location of impaction, time of ingestion, time to first medical attention, time to arrival to our

facility, time to operative intervention, method of removal, operative findings, post-operative

complications, and follow up.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

Cases were divided into two groups for evaluation—standard triage (ST) and Trauma I

triage (TT). To compare the two management algorithms, we calculated the time from arrival to

our facility to the time of arrival to the operating room as a measure to measure improvement in

treatment times. Demographic and clinical characteristics of children treated were compared

between time periods using a Fisher’s exact chi-square test and Wilcoxon rank-sums test for

categorical and continuous variables, respectively.

Results

From February 1, 2010 to January 31, 2012, there were a total of twenty-one patients

requiring surgical evaluation for button battery ingestion. Table 1 demonstrates the distribution

of all patients during the time periods of standard and trauma triage. There was no difference in

the median age, impaction site, battery type, or median time from suspected ingestion to hospital

arrival between groups. The most common presenting symptoms were vomiting, drooling, or

dysphagia, however this was in a minority of the population. There was a significant difference

in median time from arrival to the operating room between the ST and TT group (169 vs. 50

minutes; p=0.04) One of the ST patients returned seven days later with acute symptoms from a

developing acquired tracheoesophageal fistula (TEF) at the level of the right mainstem bronchus

secondary to battery impaction. He was readmitted to the hospital one week after battery

retrieval and was hospitalized for a total of 53 days with successful non-operative management

of the acquired TEF. None of the patients in the TT group developed complications secondary to

battery ingestion.

Since the most devastating outcomes occur in those patients with impacted esophageal

button batteries, we chose to focus on those patients in with a foreign body impacted in the

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

esophagus. This further limited our groups to 4 ST patients and 6 TT patients. However

between these two groups, there was no difference between time periods in regards to gender,

age, impaction site, and time from battery ingestion to hospital arrival (Table 2). 80% of the

impacted objects were 20 mm button batteries. One patient had a 12 mm button battery lodge at

the gastroesophageal junction requiring removal and one patient had two stacked coins removed

from the upper esophagus. There was no difference in retrieval methods of impacted batteries

between time periods. One patient required a combination of flexible and rigid esophagoscopy,

but all other patients were managed with either flexible or rigid esophagoscopy depending on

surgeon preference. The median time from hospital arrival to the OR was significantly decreased

by over two hours in the TT group (median, 184 vs. 33, p=0.04).

Discussion

Esophageal button battery impaction can lead to surrounding tissue injury and further

complications in as little as two hours. These injuries are almost exclusively due to 20mm

lithium batteries as batteries 12 mm and less are easily passed, and even 20 mm batteries will not

cause injury if they are able to reach the stomach without impaction. Several mechanisms

leading to tissue damage have been proposed: (1) generation of an external current causing

electrolysis of tissue fluids when placed in a conductive medium; (2) liquefactive necrosis

secondary to leakage of alkaline electrolyte; and (3) tissue necrosis from pressure phenomenon.7

The first mechanism is the most important for the most dangerous and more prevalent 20-mm

lithium batteries. These batteries do not contain an alkaline electrolyte, but they do generate

more current due to their increased voltage and capacitance in comparison with other button

batteries. This diagnosis is not always made by history as the symptoms of ingestion and the

age of the population leads to vague symptoms that may lead to a delay in diagnosis. A chest

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

and/or abdominal radiograph are simple diagnostic modalities to confirm diagnosis. Endoscopic

removal from the esophagus by esophagoscopy with visualization of the impaction site should be

performed. Airway evaluation should be considered when esophageal injury is identified or

when the negative pole of the battery faces anteriorly in the esophagus. Complications from

ingestion and esophageal impaction include esophageal stricture, esophageal perforation, vocal

cord paralysis, tracheoesophageal fistula, spondylodicitis, and aortoesophageal fistula.1, 2, 6, 7

Recent literature has highlighted the increasing epidemiologic trends and hazards

associated with button battery ingestions. Sharpe et al. detailed the significant increase in

national battery-related emergency department (ED) visits from 1990 to 2009. There were

almost 66,000 battery-related ED visits in children < 18 years of age in the United States during

this time period. The number and rate of these visits increased significantly during this time

period and the increase was primarily seen in the group of children ≤ 5 years of age.5 Litovitz et

al. reported a 6.7-fold increase in the percentage of button battery ingestions resulting in severe

or fatal outcomes from 1985 to 2009. Outcomes were significantly worse for children who

ingested large-diameter lithium cells (≥ 20mm) and in those children younger than 4 years.6 This

report led to a revised National Battery Ingestion Hotline guideline for management of button

battery ingestions. (www.poison.org/battery/guideline.asp) These guidelines suggest immediate

endoscopic removal of batteries lodged within the esophagus and careful assessment of the

mucosa for the depth and extent of tissue damage. These studies detail important epidemiologic

trends showing increasing number of visits related to button batteries and increasing serious

outcomes related to these ingestions. This highlights the importance of this study to assess an

effort at a “quality improvement” effort to improve evaluation and treatment times in an attempt

to reduce the risk of serious outcomes.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

A Level I trauma center provides the highest level of surgical care to trauma patients. It

has a full range of specialists and resources available 24 hours a day and admits a minimum

required annual volume of severely injured patients. Level I trauma centers utilize specific triage

criteria to designate different levels of expected severity of injury based on mechanism of injury

and certain physiologic criteria. At our center, the trauma activations are categorized in a two-

tiered system: Level I (severe mechanism, penetrating injury, and/or significantly altered

physiologic criteria) activations, which require an immediate full trauma team response, and

Level II activations, which are initially managed and appropriately triaged based on injuries by

emergency department physicians. We utilized this already existing mechanism of Level I

trauma activation by adding suspected button battery ingestion as one of the Level I triage

criteria to require immediate trauma team evaluation in the emergency department. Those

patients coming from another facility with a known impaction have the team on standby before

arrival. Those patients presenting to the triage desk are immediately placed into a room. The

radiology technician is already present and prompt imaging confirms presence and site of

impaction without need for transport to the radiology department. The operating room nurse

holds a room available until cleared by the trauma surgeon. Although this did result in

overtriage for those found to have no ingestion or a gastric battery, no patient with an esophageal

battery had delay in diagnosis or treatment in our facility. This algorithm change resulted in a

significant decrease in time from presentation to the ED to time to the OR.

Interestingly, however, during the same time frame we noted two trends: an overall

increase in incidence of ingestion, and a decrease in time from suspected ingestion to time to

presentation in the ED. We hypothesize that this decrease in patient response time may be

secondary to increased public awareness and outreach, both nationally and locally. Nationally,

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

efforts by organizations such as the Button Battery Task Force, the National Capitol Poison

Center, and the Consumer Product Safety Commission were ongoing during this time.

Improving product safety standards, voluntary industry efforts, including child resistant

packaging and warning label stickers, and legislative efforts are ongoing to aid in reducing

potential exposures and the hazard related to these batteries.7 During the final portion of our

post-algorithm change time frame, local education and outreach regarding button battery dangers

was conducted by our EMS agency, our state Emergency Medical Services for Children

program, and a regional healthcare coalition. Specifically, these efforts consisted of news

releases, newsletters, and social media blitzes. These modalities were further enhanced by local

media coverage including local news and radio and web distribution. However, though we

believe that this outreach has had impact, the time frame of this outreach is insufficient to

explain the change in response time. Finally, regional EMS Medical direction has subsequently

developed a destination protocol directing transport for known or suspected button battery

ingestion to a facility prepared for emergent removal and can decrease the time spent due to

interfacility transfer.

This study evaluates the importance of improving the efficiency of evaluation and

management of suspected esophageal button battery ingestions at the hospital level. Since this is

a time-sensitive issue, we believed that immediate activation of the trauma team and operating

room staff would lead to improved times to removal of the esophageal foreign body. We have

demonstrated that utilization of this system has led to a more expedient evaluation and operative

removal of suspected esophageal button batteries. Due to our small numbers and low

complication rate in both groups, we were not able to demonstrate a difference in complications

between our two groups.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

This study has several limitations. This is a retrospective study from a single institution

including a small population of patients. We must recognize the unmeasurable “study effect”

that parallels any new change in algorithm and focuses attention on this problem. However, this

also reinforces education of all personnel involved in the process as to the importance and time-

sensitive nature of this problem. The change in management instituted in this study may not be

feasible at all pediatric hospitals depending on the immediate availability of pediatric surgeons,

anesthesia, and operating room staff. Classification of these ingestions as Level I trauma

activations led to a more expedient evaluation prompting quicker transitions to the operating

room for those requiring urgent endoscopy for battery removal. In 2012 the 11th

International

Conference for Injury Prevention and Safety Promotion highlighted the increasing public health

issue surrounding button batteries and the risk for significant and irreversible injury occurring in

as little as 2 hours post ingestion. Participants including the US Consumer Product

Administration and SafeKids international called for a “multi-pronged” solution, including

industry changes and design safety, physician recognition and response and public awareness

campaigns. Our trauma activation algorithm change, in conjunction with community based

outreach and education initiatives and changes in EMS destination protocols represent a novel

and effective response to this growing public health hazard.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

References

1. Gordon AC, Gough MH. Oesophageal perforation after button battery ingestion. Ann R Coll

Surg Engl. Sep 1993;75(5):362-364.

2. Samad L, Ali M, Ramzi H. Button battery ingestion: hazards of esophageal impaction. J

Pediatr Surg. Oct 1999;34(10):1527-1531.

3. Slamon NB, Hertzog JH, Penfil SH, Raphaely RC, Pizarro C, Derby CD. An unusual case of

button battery-induced traumatic tracheoesophageal fistula. Pediatr Emerg Care. May

2008;24(5):313-316.

4. Banerjee R, Rao GV, Sriram PV, Pavan Reddy KS, Reddy DN. Button battery ingestion.

Indian J Pediatr. Feb 2005;72(2):173-174.

5. Sharpe SJ, Rochette LM, Smith GA. Pediatric battery-related emergency department visits in

the United States, 1990-2009. Pediatrics. Jun 2012;129(6):1111-1117.

6. Litovitz T, Whitaker N, Clark L, White NC, Marsolek M. Emerging battery-ingestion hazard:

clinical implications. Pediatrics. Jun 2010;125(6):1168-1177.

7. Jatana KR, Litovitz T, Reilly JS, Koltai PJ, Rider G, Jacobs IN. Pediatric button battery

injuries: 2013 task force update. Int J Pediatr Otorhinolaryngol. Sep 2013;77(9):1392-1399.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

Table 1. Comparison of demographic and clinical characteristics between Standard Triage (ST) and Level

I Trauma (TT) triage among children ingesting a battery

ST

(n=7)

TT

(n=14) p-value*

Gender (%)

Male 57 43 0.66

Female 43 57

Median age (months) 48.0 (16.0-84.0) 25.5 (14.0-72.0) 0.79

X-ray confirmed impaction site (%)

Esophagus

Stomach

57

43

43

57

0.66

Battery Size

< 20 mm

≥ 20 mm

Other

2

5

0

7

5

2

0.45

Median time from ingestion to hospital arrival

(minutes)

120 (49-840) 75.0 (55-210) 0.34

* Presented as % or median (interquartile range)

† based on Fisher’s exact chi-square and Wilcoxon rank sums test for categorical and continuous variables, respectively

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

Table 2. Comparison of demographic and clinical characteristics between Standard Triage (ST) and Level

I Trauma (TT) among children ingesting a battery with impaction in the esophagus

ST

(n=4)

TT

(n=6) p-value*

Gender

Male 2 4 0.60

Female 2 2

Median age (months) 33 (17-66) 54 (36-84) 0.47

Median time from ingestion to hospital arrival

(minutes)

480 (85-928) 55 (39-75) 0.19

Median time from arrival to OR (minutes)

184 (128-275) 33 (26-50) 0.03

Method of removal

Flexible 1 2 0.71

Rigid 2 4

Combination 1 0

* Presented as number or median (interquartile range)

† based on Fisher’s exact chi-square and Wilcoxon rank sums test for categorical and continuous variables, respectively