CONTEXT

Children presenting to health care facilities with button battery (BB) impaction.

OBJECTIVES

To describe characteristics of children with vascular complications after BB impaction, as well as associated outcomes.

DATA SOURCES

National Capital Poison Center registry and PubMed database from inception to December 2021.

STUDY SELECTION

All reports describing children aged <18 years with vascular, esophageal, or airway complications after BB ingestion.

DATA EXTRACTION

We extracted characteristics including date of publication, age and sex of child, battery type and size, duration and location of impaction, complications, subsequent interventions, and interval between battery removal and death.

RESULTS

A total of 361 cases involved severe complications or death after BB ingestion (321 cases from the National Capital Poison Center registry database, 40 additional cases from PubMed). Nineteen percent (69 of 361) were fatal and 14% (51 of 361) involved vascular injuries. Three-quarters (75%) of vascular complications were aorto-esophageal fistulae and 82% of vascular injuries were not survivable. Fatal vascular cases had significantly longer median impaction time (96 hours versus 144 hours, P <.05) and a wider range of presenting features than survivors.

LIMITATIONS

The total number of cases with vascular complications was small, data reported varied between cases, and no data were available on overall exposure. Long-term morbidity data were not available for the survivors.

CONCLUSIONS

Prolonged BB impaction is a risk factor for vascular complications and death. A high index of suspicion is required for children representing with hematemesis after BB impaction, with prompt transfer to a tertiary center because vascular surgical intervention may offer a chance of survival.

Ingestion of button batteries (BBs) is associated with a risk of serious injury and, in some cases, death, particularly in younger children.18  Owing to widespread use of BBs in household devices, there has been a rising trend in the number of BB ingestions and a concurrent increase in serious injuries.912  Between 1995 and 2015, the rate of foreign body ingestions (FBI) in children aged <6 years almost doubled from 9.5 per 10 000 cases to 18 per 10 000. Concurrently, the proportion of FBI because of BBs increased from 0.14% of FBI cases in 1995 to 8.4% in 2015.13 

The risk of impaction and subsequent complications increases with BBs >20 mm in diameter and in children aged <5 years.7,14,15  Severe complications include vocal cord paresis and paralysis, tracheal stenosis, esophageal perforation, esophageal stenosis, tracheoesophageal fistula, vascular fistula, and hemorrhage. Exsanguination from vascular fistula formation is the most common mode of death reported in fatalities after BB impaction.6,8,1517 

BB impaction results in electrical, mechanical, and caustic damage to underlying esophageal mucosa. Caustic injury, which results from the formation of hydroxide ions, leads to a rise in pH, causing liquefaction and necrosis of esophageal tissue. Epithelial damage begins within 15 minutes after impaction and major corrosive injury can occur in as little as 2 hours.7,8,1720  Tissue necrosis leads to erosion which, over days to weeks, results in esophageal perforation and ultimately formation of fistulae between the esophagus and neighboring structures. Serious complications, including fistula formation between the esophagus and major vessels, may become apparent only after several weeks.21 

Unfortunately, many BB ingestions are unwitnessed or underrecognized. Given that esophageal BB impaction is a surgical emergency, the main goals of initial management are to promptly identify the ingestion and then expedite surgical removal to minimize injury and the risk of complications. Presenting features may be nonspecific, including vomiting, difficulty swallowing, change in feeding pattern, food avoidance, drooling, and abnormal breathing. Therefore, a high index of suspicion should be maintained, particularly in children aged <5 years or those with developmental delay.17 

Although BB ingestions are on the rise and numerous case reports have been published, complications remain rare. A comprehensive review of vascular complications after esophageal BB impaction is therefore needed to describe the associations between risk factors and outcomes.

This systematic review was initiated after the death of a 2-year-old child from uncontrollable bleeding because of injury of an esophageal wall artery caused by BB impaction.22  The objective of this review is to summarize the literature on vascular complications after BB impaction with a focus on prevalence, risk factors, presenting features, management, and outcomes. Through synthesis of this data, we aim to highlight patient populations at risk for vascular injury after BB impaction, with the aim of raising awareness and aiding in the development of protocols for triage, early management, and transfer to pediatric tertiary care centers.

Reports of severe and fatal complications after BB ingestion were identified from cases recorded in the National Capital Poison Center (NCPC) BB registry from 1977 to December 2021.23,24  A comprehensive search of the PubMed database was performed using the terms “BB ingestion” and “disc battery ingestion” in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to identify articles published from inception to December 2021. Duplicate reports were removed after cross-referencing author, year of publication, patient age, and sex with the cases listed in the NCPC database.

All reports, including case reports, describing children aged <18 years who ingested BBs with complications of vascular, esophageal, airway injuries, or death were included. Non-English language studies with no translation, retrospective cohort studies with only aggregate data, reports of coingestions or other pathologies, literature reviews, and management guidelines were excluded. Screening of all records and reports was performed by 1 reviewer. Methodology of study identification is illustrated in Fig 1. Cases were reviewed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidance and a formal review protocol was not prepared.25 

FIGURE 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of study identification.

FIGURE 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of study identification.

Close modal

We extracted characteristics including the date of publication, age and sex of child, battery type and size, duration of exposure, location of impaction, complications, and subsequent interventions. In fatal cases, we recorded the interval between battery removal and death. Categorical variables, including gender distribution and outcomes, are presented as numbers and percentages. Age and exposure time are presented with median values and interquartile ranges (IQRs). Bivariate analyses were performed using χ2 test and Fisher’s exact test for comparison of categorical variables (sex and impaction location, respectively) among patients with different complications. Wilcoxon rank sum test was used for comparison of continuous variables (age and impaction time) among patients with different complications. Statistical analysis was performed using RStudio (2009–2021 RStudio, Public Benefit Corporation). This review was not registered in our institution, and it was exempt from the requirement for research ethics approval.

A total of 321 cases reported in the NCPC database met inclusion criteria and 305 published articles were identified from the literature search. After cross-referencing, 58 articles were removed because they were duplicates of cases in the NCPC database. The abstract screening process resulted in the exclusion of 118 articles. After full-text review, a further 104 articles were excluded (retrospective cohort studies with only aggregate data, reports of injuries that were not severe or fatal, reports with multiple or other pathologies and coingestions, patients aged 18 years and above, duplicated cases in separate reports). A total of 25 articles were included, accounting for 40 additional cases of severe injury or death after BB impaction.

We identified a total of 361 cases with severe complications or death after BB ingestion. Among these, 69 of 361 cases (19%) were not survivable and 292 of 361 (81%) were survivable. Of the total, 51 of 361 (14%) involved vascular injuries and 42 of 51 (81%) of the vascular injuries were fatal. Table 1 summarizes patient characteristics for those with severe and fatal complications of esophageal BB ingestion. Table 2 summarizes characteristics for children with vascular complications.

TABLE 1

Patient Characteristics: All Serious or Fatal Complications After BB Ingestion

All CasesAll, n (%)Fatal, n (%)Nonfatal, n (%)Test Statistic
Cases, n (%) 361 69 (19) 292 (81) N/A 
 NCPC database 321 67 254  
 Literature 40 38  
Female, n (%) 162 (45) 38 (55) 124 (42) P = .06 
Male, n (%) 186 (51) 29 (42) 157 (54)  
Unknown, n (%) 13 (4) 2 (3) 11 (4)  
Age, mo, median (IQR) 18 (12.3–25) 22 (13.8–25) 18 (12–25) P = .3 
Unknown, n  
Impaction time, h, median (IQR) 48 (9.4–168) 96 (28–240) 36 (8–144) P < .05 
Unknown, n 65 26 39  
Impaction location, n (%)     
 Upper esophagus 168 (47) 22 (32) 146 (50) P = .07 
 Mid-esophagus 66 (18) 13 (19) 53 (18)  
 Distal esophagus 26 (7) 6 (9) 20 (7)  
 Esophagus 83 (23) 23 (33) 60 (20)  
 Stomach 3 (0.8) 1 (1.4) 2 (0.7)  
 Intestine 2 (0.3) 1 (1.4) 1 (0.3)  
 Pharynx/larynx 10 (3) 2 (3) 8 (3)  
 Trachea/bronchi 1 (0.3) 0 (0) 1 (0.3)  
 Unknown 2 (0.6) 1 (1.4) 1 (0.3)  
Complications     
 Vascular complications 51 (14% of total) 42 (61% of fatal cases) 9 (3% of nonfatal)  
 Nonvascular complications 310 (86% of total) 27 (39%) 283 (97%)  
All CasesAll, n (%)Fatal, n (%)Nonfatal, n (%)Test Statistic
Cases, n (%) 361 69 (19) 292 (81) N/A 
 NCPC database 321 67 254  
 Literature 40 38  
Female, n (%) 162 (45) 38 (55) 124 (42) P = .06 
Male, n (%) 186 (51) 29 (42) 157 (54)  
Unknown, n (%) 13 (4) 2 (3) 11 (4)  
Age, mo, median (IQR) 18 (12.3–25) 22 (13.8–25) 18 (12–25) P = .3 
Unknown, n  
Impaction time, h, median (IQR) 48 (9.4–168) 96 (28–240) 36 (8–144) P < .05 
Unknown, n 65 26 39  
Impaction location, n (%)     
 Upper esophagus 168 (47) 22 (32) 146 (50) P = .07 
 Mid-esophagus 66 (18) 13 (19) 53 (18)  
 Distal esophagus 26 (7) 6 (9) 20 (7)  
 Esophagus 83 (23) 23 (33) 60 (20)  
 Stomach 3 (0.8) 1 (1.4) 2 (0.7)  
 Intestine 2 (0.3) 1 (1.4) 1 (0.3)  
 Pharynx/larynx 10 (3) 2 (3) 8 (3)  
 Trachea/bronchi 1 (0.3) 0 (0) 1 (0.3)  
 Unknown 2 (0.6) 1 (1.4) 1 (0.3)  
Complications     
 Vascular complications 51 (14% of total) 42 (61% of fatal cases) 9 (3% of nonfatal)  
 Nonvascular complications 310 (86% of total) 27 (39%) 283 (97%)  

N/A, not applicable.

TABLE 2

Patient Characteristics: Vascular Complications After BB Ingestion

Vascular complicationsAll, n = 51 (%)Fatal, n = 42 (%)Nonfatal, n = 9 (%)Test Statistic
Female, n (%) 32 (63) 26 (62) 6 (78) P > .99 (vascular P < .05) 
Age, mo, median (IQR) 24 (16.5–36) 24 (16.3–34.5) 24 (17–36) P = .99 
Impaction time, h, median (IQR)    P < .05 (vascular P < .05) 
 Vascular 96 (24–240) 144 (33.5–240) 11 (7.5–14.5)  
 Nonvascular 36 (8–144) 48 (27–144) 36 (8–144)  
Location, n (%)    N/A 
 Upper esophagus 12 (23.1) 11 (26.2) 1 (11)  
 Mid-esophagus 12 (23.1) 12 (28.6) 1 (11)  
 Distal esophagus 8 (15.4) 6 (14.3) 2 (22)  
 Esophagus 17 (32.7) 11 (26.2) 5 (56)  
 Stomach  
 Intestine 2 (3.8) 1 (2.4)  
 Pharynx/larynx 1 (1.9) 1 (2.4)  
 Trachea/bronchi  
 Unknown  
Vascular fistulae, n (%)    N/A 
 Aorto-esophageal 39 (76) 30 (71) 9 (100)  
 Subclavian artery fistula 3 (6) 3 (7.1)  
 Carotid artery 2 (4) 2 (4.8)  
 Thyroid artery (any), 1 (2) 1 (2.4)  
 Unspecified vessel 6 (12) 6 (14)  
Sentinel presentation, n (%)    N/A 
 Hematemesis 39 (75) 29 (69) 9 (100)  
 Cardiac arrest 13 (31)  
 Hemorrhage, not specified 7 (17)  
 Melena 6 (14)  
 Abdominal pain 3 (7.1)  
 Lethargy 1 (2.4)  
 Shock 3 (7.1)  
 Not specified 1 (2.4)  
Interventions, n (%)    N/A 
 Aortic stent  
 Patch angioplasty  
 Resection + anastomosis  
 Primary closure  
 Vascular plug + anastomosis  
 Surgical repair, not specified  
 Thoracotomy  
 Attempted surgical repair  
 Laparotomy  
 No surgical intervention 29 29 (69)  
Mode of death, n (%)     
 Hemorrhage 42 42 (100) N/A N/A 
Vascular complicationsAll, n = 51 (%)Fatal, n = 42 (%)Nonfatal, n = 9 (%)Test Statistic
Female, n (%) 32 (63) 26 (62) 6 (78) P > .99 (vascular P < .05) 
Age, mo, median (IQR) 24 (16.5–36) 24 (16.3–34.5) 24 (17–36) P = .99 
Impaction time, h, median (IQR)    P < .05 (vascular P < .05) 
 Vascular 96 (24–240) 144 (33.5–240) 11 (7.5–14.5)  
 Nonvascular 36 (8–144) 48 (27–144) 36 (8–144)  
Location, n (%)    N/A 
 Upper esophagus 12 (23.1) 11 (26.2) 1 (11)  
 Mid-esophagus 12 (23.1) 12 (28.6) 1 (11)  
 Distal esophagus 8 (15.4) 6 (14.3) 2 (22)  
 Esophagus 17 (32.7) 11 (26.2) 5 (56)  
 Stomach  
 Intestine 2 (3.8) 1 (2.4)  
 Pharynx/larynx 1 (1.9) 1 (2.4)  
 Trachea/bronchi  
 Unknown  
Vascular fistulae, n (%)    N/A 
 Aorto-esophageal 39 (76) 30 (71) 9 (100)  
 Subclavian artery fistula 3 (6) 3 (7.1)  
 Carotid artery 2 (4) 2 (4.8)  
 Thyroid artery (any), 1 (2) 1 (2.4)  
 Unspecified vessel 6 (12) 6 (14)  
Sentinel presentation, n (%)    N/A 
 Hematemesis 39 (75) 29 (69) 9 (100)  
 Cardiac arrest 13 (31)  
 Hemorrhage, not specified 7 (17)  
 Melena 6 (14)  
 Abdominal pain 3 (7.1)  
 Lethargy 1 (2.4)  
 Shock 3 (7.1)  
 Not specified 1 (2.4)  
Interventions, n (%)    N/A 
 Aortic stent  
 Patch angioplasty  
 Resection + anastomosis  
 Primary closure  
 Vascular plug + anastomosis  
 Surgical repair, not specified  
 Thoracotomy  
 Attempted surgical repair  
 Laparotomy  
 No surgical intervention 29 29 (69)  
Mode of death, n (%)     
 Hemorrhage 42 42 (100) N/A N/A 

N/A, not applicable.

Most of the vascular injuries were aorto-esophageal fistulae, accounting for 71% of vascular injuries in fatal cases, 100% of vascular injuries in nonfatal cases, and 75% of vascular injuries overall. Other vessel injuries were subclavian artery fistulae (6%), carotid artery fistulae (4%), and thyroid artery fistulae (2%). The upper esophagus was the commonest location of impaction overall (47%), and specifically within the fatal and nonfatal cases (32% and 50%, respectively). The specific level of impaction within the esophagus was not reported in 33% of the fatal cases, and 33% of cases with vascular complications.

There was a slight female predominance among the fatal cases (n = 38 of 69, 55%) and a slight male predominance among the nonfatal cases (n = 157 of 292, 54%), but this difference was not statistically significant (P = .06). Within the subset of patients with vascular complications, the majority were female (n = 32 of 51, 63%). Compared with cases without vascular complications, the difference in gender distribution was statistically significant (P < .05). There was no significant difference in gender distribution between the fatal and nonfatal vascular cases (62% vs 78% female, P ≥ .99).

Median age in fatal cases was 22 months (IQR 13.8–25) compared with 18 months (IQR 12–25) in nonfatal cases. There was no statistically significant difference in age distribution between the 2 groups (P = .3). Among those with vascular complications, median age was 24 months (IQR 16.5–36) for fatal cases compared with 21 months (IQR 16.3–34.5) in survivors with no statistically significant difference (P = .99).

Duration of exposure of impacted BB (impaction time) differed significantly among the reported cases. Median impaction time for fatal cases was 96 hours (IQR 28–240), compared with 36 hours (IQR 8–144) for the nonfatal cases, (P < .05). Children with vascular complications had a median exposure time of 96 hours (IQR 24–240), significantly longer than in children with other nonvascular complications (36 hours, IQR 8–144, P < .05). Among children with vascular complications specifically, median impaction time for fatal cases was 144 hours (33.5–240), which was significantly longer than the median impaction time of 11 hours (IQR 7.5–14.5) for the survivors (P < .05).

The most common presentations of vascular fistulae were hematemesis (75%), cardiac arrest (25%), hemorrhage (13%), and melaena (12%). Notably, all the survivors with vascular injuries presented with hematemesis. The broader variety of presenting features was only seen in the nonsurvivors. In two-thirds of survivors, computed tomography (CT) angiography (n = 5 of 9) or CT (n = 1 of 9) was used to characterize vascular injury.

Most children with vascular injuries (n = 42 of 51, 82%) did not survive. Of the surviving cases (n = 9 of 51, 18%), all had aorto-esophageal fistulas. Surgical repair for survivors included aortic stent (n = 2), patch angioplasty (n = 1), resection of fistula and end-to-end anastomosis of the aorta (n = 2), suture closure (n = 1), and vascular plug followed by anastomosis (n = 1). In 2 cases, the mode of surgical repair was not reported. Table 3 summarizes the characteristics of the survivors with vascular complications.

TABLE 3

Patient Characteristics: Survivors of Vascular Injury

SourceAge (mo)SexBB Type and DiameterImpaction LocationImpaction Time (h)BB RemovalPresentation of Vascular ComplicationTime From Removal to Representation (D)Vascular Imaging ModalityVascular ComplicationManagement
Duell 201723  24 Lithium <20 mm Esophagus Unknown Removal, unspecified Hematemesis 13 CT Aorto-esophageal fistula Surgical repair, unspecified 
CTV 202023  18 Lithium 20 mm Esophagus Unknown Surgical Hematemesis N/A, primary presentation Unknown Aorto-esophageal fistula Surgical repair, unspecified 
Spiers 201132  Lithium 20 mm Distal esophagus 14 Endoscopic Hematemesis 27 CT angiography Aorto-esophageal fistula Resection, end-to-end anastomosis 
Granata 201833  36 Lithium, size unknown Esophagus Laparotomy Hematemesis, shock N/A, primary presentation Aortogram Aorto-esophageal fistula Vascular stent 
Mahajan 201834  36 Type and size unknown Distal esophagus Unknown Endoscopic Hematemesis 30 CT angiography Aorto-esophageal fistula Primary closure 
Bartkevics 201935  12 Lithium 20 mm Esophagus Unknown Endoscopic Hematemesis 17 MRI followed by CT angiography Aorto-esophageal fistula Resection, end-to end anastomosis 
Sinclair 202036  72 Type unknown 21 mm Mid-esophagus Endoscopic Hematemesis 25 Cardiac catheterization, angiography Aorto-esophageal fistula Vascular stent 
Alreheili 202137  30 Type and size unknown Upper esophagus 16 Endoscopic Hematemesis, melena CT angiography Aorto-esophageal fistula Fistula occluded with vascular plug device 
Wakimoto 2021; Gibbs 202138,39  17 Type unknown 23.5 mm Esophagus Unknown Not removed (BB advanced to small intestine) Hematemesis CT angiogram Aorto-esophageal fistula Patch angioplasty, subsequent repair with intercostal muscle flap 
SourceAge (mo)SexBB Type and DiameterImpaction LocationImpaction Time (h)BB RemovalPresentation of Vascular ComplicationTime From Removal to Representation (D)Vascular Imaging ModalityVascular ComplicationManagement
Duell 201723  24 Lithium <20 mm Esophagus Unknown Removal, unspecified Hematemesis 13 CT Aorto-esophageal fistula Surgical repair, unspecified 
CTV 202023  18 Lithium 20 mm Esophagus Unknown Surgical Hematemesis N/A, primary presentation Unknown Aorto-esophageal fistula Surgical repair, unspecified 
Spiers 201132  Lithium 20 mm Distal esophagus 14 Endoscopic Hematemesis 27 CT angiography Aorto-esophageal fistula Resection, end-to-end anastomosis 
Granata 201833  36 Lithium, size unknown Esophagus Laparotomy Hematemesis, shock N/A, primary presentation Aortogram Aorto-esophageal fistula Vascular stent 
Mahajan 201834  36 Type and size unknown Distal esophagus Unknown Endoscopic Hematemesis 30 CT angiography Aorto-esophageal fistula Primary closure 
Bartkevics 201935  12 Lithium 20 mm Esophagus Unknown Endoscopic Hematemesis 17 MRI followed by CT angiography Aorto-esophageal fistula Resection, end-to end anastomosis 
Sinclair 202036  72 Type unknown 21 mm Mid-esophagus Endoscopic Hematemesis 25 Cardiac catheterization, angiography Aorto-esophageal fistula Vascular stent 
Alreheili 202137  30 Type and size unknown Upper esophagus 16 Endoscopic Hematemesis, melena CT angiography Aorto-esophageal fistula Fistula occluded with vascular plug device 
Wakimoto 2021; Gibbs 202138,39  17 Type unknown 23.5 mm Esophagus Unknown Not removed (BB advanced to small intestine) Hematemesis CT angiogram Aorto-esophageal fistula Patch angioplasty, subsequent repair with intercostal muscle flap 

F, female; M, male; N/A, not available.

The hazards of BB ingestion are well recognized, with numerous cases of severe complications and death documented in the literature. We performed a systematic review containing, to our knowledge, the largest compilation of vascular injuries after esophageal BB impaction. Fortunately, vascular injuries remain rare occurrences in the population generally. Of the 361 cases of severe complications after BB impaction, only 51 (14%) were vascular in nature. However, 61% of children who died after BB impaction had vascular injuries.

These data highlight factors associated with higher risk for development of vascular injuries and death. The median duration of BB impaction in those with vascular complications was 4 days, significantly longer than the median duration of 1.4 days in nonvascular cases. The impaction times were also significantly longer in fatal cases. This difference was most striking when comparing fatal cases of vascular injuries and survivors of vascular injuries. Tissue injury begins within minutes of battery ingestion and serious injury can occur within 2 hours.6,8,17  Time is therefore a crucial factor at presentation because it is associated with development of vascular complications and death. Early recognition of ingestion and rapid coordination of transfer to a pediatric tertiary center for endoscopic removal of the BB are essential to minimize duration of impaction. Survival is associated with shorter impaction times, reinforcing urgency to treat any patient presenting with delayed recognition or prolonged impaction as having an increased risk of serious complications.6,9 

Chest radiographs with anteroposterior and lateral views should be obtained in all patients.17  Once impaction is recognized, caustic mucosal injury may be mitigated by neutralizing tissue pH with sucralfate and honey while endoscopic removal is being organized.26,27  A dose of 10 ml of honey or 1 g of sucralfate may be given every 10 minutes until the BB has been removed. Mitigating measures should not delay emergent removal of the BB.15,17,27 

After removal of the BB, complications can occur after days or weeks because of ongoing alkaline caustic injury.6,19  We recommend that patients with risk factors for serious complications should be identified and a surveillance plan clearly outlined to identify structures at risk for injury and anticipate complications. Families and care providers should be aware that immediate access to a specialist pediatric care facility may be required for management of life-threatening bleeding up to 1 month after BB removal.

Though data are limited, imaging with MRI currently offers the best modality to comprehensively assess the evolution of tissue injury and involvement of associated structures.28,29  Although all previously reported deaths identified by our PubMed and NCPC database search were associated with major vessel injury, some warning signs of which would likely be revealed by MRI, it is of concern that the case which prompted our review was of a fatal injury to a relatively minor artery within the wall of the esophagus.22 

Our data showed the upper and mid-esophagus were the most common locations of impaction among both the larger cohort of severe and fatal complications, and the subgroup with vascular complications. Impaction at the mid-esophagus is a major concern for formation of aorto-esophageal fistula and impaction at the proximal esophagus evokes concerns for injury to the superior and inferior thyroid arteries. In addition to the level of the impaction within the esophagus, the size of the battery and orientation of the negative pole assists in anticipating impaction and vessels and other structures at risk for further injury. Data on BB size and orientation were inconsistently reported in the cases reviewed, but we recommend documenting these parameters in all cases.6,9,14,15,30 

Over 80% of cases with vascular fistulae were fatal. We were able to highlight some differences in presenting features between survivors of vascular complications and those who died (Table 2). The fatal cases presented with clinical features ranging from lethargy, abdominal pain, and hematemesis to hemodynamic decompensation and cardiac arrest. Conversely, hematemesis was the main presenting feature in the 9 surviving cases (Table 3). Fatal cases had a broad range of large and small vessel injuries, whereas all survivors had aorto-esophageal fistulae. The interventions attempted among the fatal cases were mostly limited to thoracotomies and laparotomies, compared with the more specific repairs achieved in the survivors.

Synthesizing this information, we postulate that the longer median impaction time in fatal cases was associated with an increased risk of esophageal perforation and vascular fistula formation. This longer duration and involvement of smaller vessels may also result in more insidious bleeding, which became apparent only once decompensation had occurred with hemorrhagic shock and collapse in the fatal cases. Among the survivors, involvement of large vessels leading to earlier, more obvious bleeding may have led to identification and transfer at an earlier stage, with relative physiologic stability, thus permitting more targeted intervention for hemostasis and vessel repair. It must also be considered that the time to intervention might have differed for children presenting directly to a tertiary center compared with those requiring transfer from peripheral hospitals. This information was not reported in cases we reviewed.

These data further highlight the risks associated with vascular fistulae. Once a sentinel bleed occurs, the risk of exsanguination and death is very high. Sentinel bleeds, as well as “milder” symptoms such as lethargy and abdominal pain, in patients with a history of BB ingestion must be taken very seriously, even if the child appears “well.” The small chance of survival relies on an expedited transfer to a pediatric tertiary center for emergency surgical repair. Pediatricians and emergency providers must be aware that hematemesis, melena, abdominal pain, and vomiting may precede exsanguination and death within a short time frame. Pediatric critical care transport teams and intensivists have a key role in coordinating and expediting transfer to a pediatric tertiary center with appropriate surgical expertise and PICU.

Where there is high suspicion of major vascular involvement at presentation, teams at tertiary centers should weigh the benefits of obtaining additional imaging before intervention against the risk of ongoing vascular injury (in the case of a BB that remains impacted) and catastrophic exsanguination.30  Modalities available include CT angiography, magnetic resonance angiography and catheter angiography. In this time-critical setting, CT angiography is the cross-sectional imaging modality of choice, offering rapid, noninvasive identification of vascular and nonvascular injuries.31  Children presenting with vascular complications may have impacted BB in situ and magnetic resonance angiography would not be safe in such cases.

Despite being the most comprehensive series of vascular injuries reported in the literature, the total number of cases remains small, and the data reported varies between cases. Conclusions drawn from this are therefore limited. Additionally, we do not have long-term follow-up data on the 9 survivors to fully evaluate morbidity after surgical repair. Awareness of the risks of BB ingestion and preventative measures remain the most effective means of harm reduction.

Dr Akinkugbe conceptualized and designed the study, designed the data collection instruments, collected data, conducted the initial analyses, drafted the initial manuscript; and reviewed and revised the manuscript; Drs Wolter and McKinnon conceptualized and designed the study, coordinated and supervised data collection, reviewed and revised the manuscript, and critically reviewed the manuscript for important intellectual content; Drs James, Ostrow, and Everett critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2022-056709.

FUNDING: No external funding.

CONFLICT OF INTEREST DISCLAIMER: The authors have indicated they have no conflicts of interest relevant to this article to disclose.

     
  • BB

    button battery

  •  
  • CT

    computed tomography

  •  
  • FBI

    foreign body ingestions

  •  
  • IQR

    interquartile range

  •  
  • NCPC

    National Capital Poison Center

1
NHS England
.
Risk of death and serious harm from delays in recognizing and treating ingestion of button batteries
.
Available at: https://www.england.nhs.uk/2014/12/psa-button-batteries/. Accessed September 15, 2021
2
Health Canada
.
Battery safety– Electropaedia
.
3
Queensland Health
.
Button batteries– why they’re dangerous and what you can do to keep your kids safe
.
4
National Safety Council
.
Button batteries– National Safety Council
.
5
National Capital Poison Center
.
Batteries cause devastating injuries
.
Available at: https://www.poison.org/articles/button-batteries. Accessed September 15, 2021
6
Leinwand
K
,
Brumbaugh
DE
,
Kramer
RE
.
Button battery ingestion in children: a paradigm for management of severe pediatric foreign body ingestions
.
Gastrointest Endosc Clin N Am
.
2016
;
26
(
1
):
99
118
7
Lerner
DG
,
Brumbaugh
D
,
Lightdale
JR
,
Jatana
KR
,
Jacobs
IN
,
Mamula
P
.
Mitigating risks of swallowed button batteries: new strategies before and after removal
.
J Pediatr Gastroenterol Nutr
.
2020
;
70
(
5
):
542
546
8
Litovitz
T
,
Whitaker
N
,
Clark
L
,
White
NC
,
Marsolek
M
.
Emerging battery-ingestion hazard: clinical implications
.
Pediatrics
.
2010
;
125
(
6
):
1168
1177
9
Akilov
KA
,
Asadullaev
DR
,
Yuldashev
RZ
,
Shokhaydarov
SI
.
Cylindrical and button battery ingestion in children: a single-center experience
.
Pediatr Surg Int
.
2021
;
37
(
10
):
1461
1466
10
Speidel
AJ
,
Wölfle
L
,
Mayer
B
,
Posovszky
C
.
Increase in foreign body and harmful substance ingestion and associated complications in children: a retrospective study of 1199 cases from 2005 to 2017
.
BMC Pediatr
.
2020
;
20
(
1
):
560
11
Mowry
JB
,
Spyker
DA
,
Cantilena
LR
Jr
,
McMillan
N
,
Ford
M
.
2013 Annual Report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 31st Annual Report
.
Clin Toxicol (Phila)
.
2014
;
52
(
10
):
1032
1283
12
Gummin
DD
,
Mowry
JB
,
Spyker
DA
, et al
.
2018 Annual Report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 36th Annual Report
.
Clin Toxicol (Phila)
.
2019
;
57
(
12
):
1220
1413
13
Orsagh-Yentis
D
,
McAdams
RJ
,
Roberts
KJ
,
McKenzie
LB
.
Foreign-body ingestions of young children treated in US emergency departments: 1995–2015
.
Pediatrics
.
2019
;
143
(
5
):
1995
2015
14
Brumbaugh
DE
,
Colson
SB
,
Sandoval
JA
, et al
.
Management of button battery- induced hemorrhage in children
.
J Pediatr Gastroenterol Nutr
.
2011
;
52
(
5
):
585
589
15
Sethia
R
,
Gibbs
H
,
Jacobs
IN
,
Reilly
JS
,
Rhoades
K
,
Jatana
KR
.
Current management of button battery injuries
.
Laryngoscope Investig Otolaryngol
.
2021
;
6
(
3
):
549
563
16
Dunn
JOC
,
Mythen
MG
,
Grocott
MP
.
Physiology of oxygen transport
.
BJA Educ
.
2016
;
16
(
10
):
341
348
17
National Capital Poisons Center
.
National capital poison center button battery ingestion triage and treatment guideline
.
Available at: https://www.poison.org/battery/guideline. Accessed July 22, 2021
18
Singh
A
,
Anand
S
,
Krishnan
N
.
Re: cylindrical and button battery ingestion in children: a single-center experience
.
Pediatr Surg Int
.
2021
;
37
(
10
):
1473
1474
19
Jacobs
IN
,
Jatana
KR
.
Current management of aerodigestive foreign bodies in children
.
Semin Pediatr Surg
.
2021
;
30
(
3
):
151064
20
Eliason
MJ
,
Melzer
JM
,
Winters
JR
, %
Gallagher
TQ
.
Identifying predictive factors for long-term complications following button battery impactions: a case series and literature review
.
Int J Pediatr Otorhinolaryngol
.
2016
;
87
:
198
202
21
Soto
PH
,
Reid
NE
,
Litovitz
TL
.
Time to perforation for button batteries lodged in the esophagus
.
Am J Emerg Med
.
2019
;
37
(
5
):
805
809
22
College of Physicians and Surgeons of Ontario
.
Button battery ingestion
.
23
National Capital Poisons Center
.
Nonfatal button battery ingestions with severe esophageal or airway injury
.
Available at: https://www.poison.org/battery/severecases. Accessed January 22, 2022
24
National Capital Poisons Center
.
Fatal button batter ingestions
.
Available at: https://www.poison.org/battery/fatalcases. Accessed December 22, 2021
25
Page
MJ
,
McKenzie
JE
,
Bossuyt
PM
, et al
.
The PRISMA 2020 statement: an updated guideline for reporting systematic reviews
.
BMJ
.
2021
;
372
:
n71
26
Patel
JK
,
Kobashigawa
JA
.
Cardiology patient page. Heart transplantation
.
Circulation
.
2011
;
124
(
4
):
e132
e134
27
Anfang
RR
,
Jatana
KR
,
Linn
RL
,
Rhoades
K
,
Fry
J
,
Jacobs
IN
.
pH-neutralizing esophageal irrigations as a novel mitigation strategy for button battery injury
.
Laryngoscope
.
2019
;
129
(
1
):
49
57
28
Riedesel
EL
,
Richer
EJ
,
Sinclair
EM
, et al
.
Serial MRI findings after endoscopic removal of button battery from the esophagus
.
AJR Am J Roentgenol
.
2020
;
215
(
5
):
1238
1246
29
Grey
NEO
,
Malone
LJ
,
Miller
AL
, et al
.
Magnetic resonance imaging findings following button battery ingestion
.
Pediatr Radiol
.
2021
;
51
(
10
):
1856
1866
30
Mubarak
A
,
Benninga
MA
,
Broekaert
I
, et al
.
Diagnosis, management, and prevention of button battery ingestion in childhood: a European Society for Paediatric Gastroenterology Hepatology and Nutrition position paper
.
J Pediatr Gastroenterol Nutr
.
2021
;
73
(
1
):
129
136
31
Hahn
LD
,
Prabhakar
AM
,
Zucker
EJ
.
Cross-sectional imaging of thoracic traumatic aortic injury
.
Vasa
.
2019
;
48
(
1
):
6
16
32
Spiers
A
,
Jamil
S
,
Whan
E
,
Forbes
D
,
Gollow
I
,
Andrews
D
.
Survival of patient after aorto-oesophageal fistula following button battery ingestion
.
ANZ J Surg
.
2012
;
82
(
3
):
186
187
33
Granata
A
,
Gandolfo
C
,
Acierno
C
,
Piazza
M
,
Burgio
G
,
Traina
M
.
Button battery removed from the stomach resulting in a missed aorto-esophageal fistula–a multidisciplinary approach to rescuing a very young patient: a case report
.
J Med Case Reports
.
2018
;
12
(
1
):
318
34
Mahajan
S
,
Jaswal
V
,
Kumar
S
, %
Thingnam
S
,
Dogra
N
.
Successful surgical management of an aorto- oesophageal fistula caused by button battery ingestion
.
Eur J Cardiothorac Surg
.
2019
;
55
(
4
):
790
791
35
Bartkevics
M
,
Stankovic
Z
,
Schibli
S
, et al
.
A near miss and salvage management of aorto-esophageal fistula secondary to cell battery ingestion
.
World J Pediatr Congenit Heart Surg
.
2020
;
11
(
1
):
120
122
36
Sinclair
EM
,
Stevens
JP
,
McElhanon
B
, et al
.
Development and repair of aorto-esophageal fistula following esophageal button battery impaction: a case report
.
J Pediatr Surg Case Rep
.
2021
;
66
:
101782
37
Alreheili
KM
,
Almutairi
M
,
Alsaadi
A
,
Ahmed
G
,
Alhejili
A
,
AlKhatrawi
T
.
2-year-old boy who developed an aorto-esophageal fistula after swallowing a button battery, managed using a novel procedure with vascular plug device as a bridge to definitive surgical repair
.
Am J Case Rep
.
2021
;
22
(
1
):
e931013
38
Wakimoto
M
,
Willer
BL
,
Mckee
C
, %
Nafiu
OO
,
Tobias
JD
.
Successful management of an aorto-esophageal fistula following button battery ingestion: a case report and review of the literature
.
Saudi J Anaesth
.
2021
;
15
(
2
):
193
198
39
Gibbs
H
,
Sethia
R
,
McConnell
PI
, et al
.
Survival of toddler with aorto- esophageal fistula after button battery ingestion
.
Case Rep Otolaryngol
.
2021
;
2021
:
5557054