Population-Based Study of Severe Trauma Due to Electrocution in the Calgary Health Region, 1996–2002

Original Article Article original

Population-based study of severe trauma due to electrocution in the Calgary Health Region, 1996–2002

Kevin B. Laupland, MD;*†‡ John B. Kortbeek, MD;*§ Christi Findlay, BA;§ Andrew W. Kirkpatrick, MD;*§ S. Morad Hameed, MD, MPH¶

Background: Severe trauma arising from electrocution is uncommon, and we wished to study the incidence, risk factors and outcome to identify potential areas of prevention. We therefore studied an active population of adult victims of severe electrical trauma in a large area of Canada (the Calgary Health Region [CHR]). Methods: From 3 databases, we collected data on all adult (≥18 yr) residents of the CHR who suffered electrical injury associated with an Injury Severity Score of 12 or greater or died as a result of electrocution between Apr. 1, 1996, and Mar. 1, 2002. Of particular interest were the incidence, the age and sex of the victims, the mechanism, the type of electrical power and the outcome. Results: Ten cases of severe electrical trauma were identified (2.4 per 1 million population annually). Victims’ ages were a mean (and standard deviation) of 38.2 (10.3) years. All 10 victims were male for an annual sex-specific incidence of 4.9 per 1 million people. Men aged 18–49 years were nearly 3 times more likely to be victims of electrocution than older men, although this result was not statistically significant (6.1 v. 2.1 per 1 million annually; relative risk 2.9; 95% confidence interval 0.4–127.5). All episodes of electrocution were unintentional, and 7 were occupationally related. All 10 electrocutions resulted from domestically generated current, and 9 were related to power-line exposure. Overall, 6 patients died for a death rate due to severe electrical trauma of 1.4 per 1 million people annually. Conclu- sion: This population-based study demonstrates that severe electrocution is rare but is associated with death in the majority of cases in this large Canadian region.

Contexte : Les traumatismes graves causés par électrocution sont rares et nous voulions en étudier l’incidence, les facteurs de risque et l’issue afin de définir des domaines possibles de prévention. Nous avons donc étudié une population active d’adultes victimes d’un traumatisme sévère causé par l’électricité dans une grande région du Canada (la Région sanitaire de Calgary [RSC]). Méthodes : Nous avons recueilli dans trois bases des données sur tous les résidents adultes (≥18 ans) de la RSC ayant été victimes d’un traumatisme causé par électrocution et associé à un indice de gravité de la blessure de 12 ou plus, ou qui sont morts électrocutés, entre le 1er avril 1996 et le 1er mars 2002. Nous nous intéressions particulière- ment à l’incidence, à l’âge et au sexe des victimes, au mécanisme de l’incident, au type d’énergie élec- trique et à l’issue. Résultats : Nous avons trouvé 10 cas de traumatismes graves causés par électrocution (2,4 par million d’habitants par année). Les victimes avaient en moyenne (et écart type) 38,2 (10,3) ans. Les 10 victimes étaient de sexe masculin, ce qui représente une incidence annuelle spécifique au sexe de 4,2 par million de personnes. Les hommes de 18 à 49 ans étaient presque trois fois plus susceptibles d’être victimes d’électrocution que les hommes plus âgés, même si ce résultat n’était pas statistiquement significatif (6,1 c. 2,1 par million par année; risque relatif de 2,9; intervalle de confiance à 95 %, 0,4–127,5). Aucun des épisodes d’électrocution n’était intentionnel et 7 étaient reliés au travail. Les 10 électrocutions ont été causées par des circuits domestiques et 9 étaient reliées à une exposition à des lignes de transmission. Au total, 6 patients sont morts, ce qui représente un taux de mortalité at- tribuable à un traumatisme grave causé par électrocution de 1,4 par million de personnes par année. Conclusion : Cette étude représentative démontre que l’électrocution grave est rare, mais qu’elle en- traîne la mort dans la majorité des cas dans cette grande région canadienne.

From the Departments of *Critical Care Medicine, †Medicine, ‡Pathology and Laboratory Medicine and §Surgery, University of Calgary and the Calgary Health Region, Calgary, Alta., and ¶Trauma Services, Vancouver General Hospital, Vancouver, BC

Accepted for publication Nov. 26, 2003. Correspondence to: Dr. S. Morad Hameed, Trauma Services, Vancouver General Hospital, 855 W 12th Ave., Vancouver BC V5Z 1M9; fax 604 875-5348; [email protected]

' 2005 CMA Media Inc. Can J Surg, Vol. 48, No. 4, August 2005 289 Laupland et al

evere trauma resulting from elec- of the CHR who were victims of se- numbers of cases observed in the S trical shock has been character- vere trauma resulting from electrical study as the numerator and local ized by multisystem involvement and injury between Apr. 1, 1996, and population demographic data as the high death rates.1 Electrical trauma Mar. 31, 2002, were included in this denominator.15 Category specific rel- results from the direct effect of elec- study. Severe trauma was defined by ative risks were calculated as previ- tric current, conversion of electrical an electrical injury associated with a ously described.16 Ninety-five percent to thermal energy or mechanical ef- presenting Injury Severity Score confidence intervals (CIs) were cal- fects including shock waves, projec- (ISS) of 12 or greater or by death culated using exact statistics. tiles or violent muscle contractions. before the trauma team could make Although many people have experi- an assessment.14 This study received Results enced mild electrical shocks from do- approval of the ethics review board at mestic sources, severe trauma arising the University of Calgary and CHR. Incidence and demographic from electrocution is uncommon. In features the United States, estimates of fatali- Study protocol ties due to electrocution have been We identified 10 adults, residents of estimated at 5 per 1 million popula- We used a population-based surveil- the CHR, who suffered severe elec- tion annually.2–5 Death attributed lance cohort design. Study patients trical trauma over the 6-year study specifically to lightning has varied were identified by 3 methods. First, period, for an incidence of 2.4 per 1 considerably depending on geogra- the regional trauma services database million population annually (esti- phy, with the highest rates of ap- that prospectively collects informa- mated population at risk over 6 yr = proximately 1 per 1 million reported tion on all patients assessed at the 4 144 824). Six of the 10 patients from southern US.6,7 Male gender Foothills Medical Centre (FMC) was survived at least to the point of as- and an occupation requiring expo- studied. The FMC is the only tertiary sessment by the trauma team at the sure to electric current have been care trauma centre treating adults in FMC. Five cases were clustered dur- proposed as risk factors for severe southern Alberta and manages all ing the summer months of 1998, electrical injury, although the actual burns and nearly all traumatic in- but, with the exception of 2, they risks have been poorly quantified.4–8 juries occurring among adult resi- were epidemiologically unrelated. To define the burden of disease dents of the CHR. Second, the Because of the small number of cases and risk factors so as to identify poten- Health Information Services data- over the 6-year period, no clear tem- tial areas of prevention, population- base, which records administrative porally related trend in occurrence based studies are optimal. However, data on all hospitalized patients in could be established. existing studies in the literature have the CHR, was searched to identify Victims’ ages ranged from 19 to been either case reports or series or patients admitted to other acute care 52 years (mean and standard devia- population-based studies restricted to facilities in the region. Third, victims tion [SD] 38.2 [10.3] yr). All 10 vic- highly focused populations.9–12 Fur- of severe electrical trauma who died tims were male, for an annual sex- ther, no population-based studies before the trauma team could make specific incidence of 4.9 per 1 million have been reported from Canada. an assessment were identified by the population. Males aged 18–49 years Our objective was therefore to con- Medical Examiner’s Office of Alberta were nearly 3 times more likely to be duct a population-based surveillance database, which collects information victims of electrocution than older study to define the incidence, risk fac- on all unexpected deaths occurring men, although this finding was not tors and outcome of severe electrocu- in the CHR. Data from these 3 statistically significant (6.1 v. 2.1 per tion in a large Canadian region. sources were exported and merged in 1 million; relative risk 2.9; 95% CI Excel (Microsoft Corp., Redmond, 0.4–127.5). Although 8 of the 10 Methods Wash.). Patients who were not resi- victims resided within Calgary city dents of the CHR were excluded. A limits, 5 of the electrocutions were Patient population CHR resident was defined as any- documented to have occurred in body with a home address within the smaller communities or rural areas The Calgary Health Region (CHR) boundaries of the CHR.7 within the boundaries of the CHR. is a fully integrated, publicly funded health care system that provides vir- Statistical analysis Mechanism of injury tually all medical and surgical care to the residents of the cities of Calgary Analysis was performed using Stata All episodes of electrocution were acci- and Airdrie and approximately 20 version 7.0 (Stata Corp., College dental in nature, and 7 were occupa- nearby small towns, villages and Station, Tex.). Incidence and mortal- tionally related. The mechanism of in- hamlets.13 All adult (≥18 yr) residents ity were calculated by taking the jury and the outcome are summarized

290 J can chir, Vol. 48, N o 4, aoßt 2005 Severe electrocution in Calgary Health Region in Table 1. All 10 patients’ injuries trauma services, the mean (and SD) curring in a large Canadian popula- resulted from domestically generated ISS was 21.7 (5.6). In general, in- tion. We used the population-based electric current. All 6 patients who juries were characterized by head surveillance cohort design that is the were assessed by the trauma team suf- trauma and burns (Table 1). Two standard for observational epidemio- fered electrical shock related to power- patients presented with asystolic ar- logic investigation.16 As a result we line sources. Three patients touched rest at the accident scene. One of were able to establish population in- power lines with objects, and 2 of these patients was resuscitated with cidence and death rates for severe these fell from a height after electrocu- return of spontaneous circulation, al- electrical injuries. Although the inci- tion. Two other patients entered an though he died shortly after of com- dence of severe electrical trauma was energized field when they attempted plications from severe burns. Of the low, the case fatality rate was high to exit their vehicles after contacting 4 remaining patients assessed by the (60%). This study was adequately power lines. The remaining patient as- trauma service, all survived to hospi- methodologically designed to evalu- sessed by the trauma services suffered tal discharge (Table 1). Only limited ate and quantify category specific risk electrical shock and a flash burn while scene details were available for the 4 factors such as age and sex. How- working on a power transformer. Of patients who died before the trauma ever, since there was only a small the 4 patients who died before assess- service could make an assessment at number of cases during surveillance ment by the trauma service, 3 received FMC (Table 1). Overall, 6 adult pa- of more than 4 million patient-years a shock when ladders or scaffolding tients died for an annual death rate and none were female, information contacted active power lines. The re- due to severe electrical trauma of 1.4 on category specific risk factors was maining patient was electrocuted while per 1 million population. limited. Although not statistically using a homemade electrical apparatus significant, young men were at the on a wet lawn. Discussion highest risk for severe electrical trauma in the adult population and Management and outcomes This study provides important infor- represent a target group for future mation about the epidemiologic fea- preventive efforts. Of the 6 patients assessed by the tures of severe electrical trauma oc- A number of clinical observations

Table 1

Demographic and clinical summary of patients suffering severe electrocution in the Calgary Health Region, 1996–2002

Patient* Exposure Injuries Treatments Outcome 1 Accidentally touched live Closed head injury; second Two operations for wound Survived after 37-d hospital wire and fell while and third degree burns to debridement stay attempting to rescue animal abdomen, thorax and neck from height 2 Fall after ladder contacted Small subdural hematoma Admission for observation Discharged home after 1-d live power line admission 3 Tent pole touched active Asystolic arrest Cardiopulmonary Died in emergency power line resuscitation department 4 Shocked after exiting crane Epidural and subdural Craniotomy and wound Survived after 17-d hospital that contacted power line hematomas, burns to 1% body debridement stay surface area 5 Electrical flash burn while Second- and third-degree Multiple operations for Survived after 32-d hospital working on transformer burns to 41% body surface debridement and grafting stay area 6 Shocked after exiting vehicle Asystolic arrest at scene; burns Died 1 d into admission to that hit power line to 36% body surface area intensive care unit 7 Shocked while using Dead at scene homemade electrical apparatus on wet lawn 8 Scaffolding touched Dead at scene overhead wire 9 Scaffolding touched Severe burns Died shortly after admission to overhead wire (related to nontrauma hospital above patient) 10 Aluminum ladder touched Dead at scene 8000-V power line while victim was installing eavestrough *Mean age 38.2 (standard deviation 10.3) years

Can J Surg, Vol. 48, No. 4, August 2005 291 Laupland et al

in this study are noteworthy. First, all this is an unlikely source of bias be- 5. Wright RK, Davis JH. The investigation 10 electrocutions were caused by do- cause we used a comprehensive case of electrical deaths: a report of 220 fatalities. J Forensic Sci 1980;25:514-21. mestically generated electrical cur- identification protocol that involved rent; none were from lightning. 3 independent databases. One po- 6. Lightning-associated deaths — United Clearly, efforts aimed at reducing the tential for missed cases was if CHR States, 1980–1995. MMWR Morb Mortal Wkly Rep 1998;47:391-4. impact of serious electrocutions residents suffered electrical injuries should focus on this aspect. Second, outside the region and failed to sub- 7. Lightning-associated injuries and deaths 7 of the 10 cases were occupationally sequently receive health care services among military personnel — United States, 1998–2001. MMWR Morb Mortal related such that the workplace in the CHR. Given that Calgary is Wkly Rep 2002;51:859-62. should be recognized as an area on geographically isolated from other 8. Butler ED, Gant TD. Electrical injuries, which to focus preventive efforts. large metropolitan areas, with the with special reference to the upper extrem- However, the possibility does exist second nearest trauma centre being ities. A review of 182 cases. Am J Surg that there is no true association be- approximately 300 km away, we be- 1977;134:95-101. tween occupation and electrocution. lieve this bias is minor. Thus, the low 9. Lipscomb HJ, Dement JM, Rodriguez- This is because to convincingly es- rate of electrocution in our study Acosta R. Deaths from external causes of tablish an occupational link, knowl- likely reflects the true population in- injury among construction workers in edge of the baseline workplace expo- cidence. North Carolina, 1988–1994. Appl Occup sure of males in the base CHR Environ Hyg 2000;15:569-80. population would be required as de- Conclusions 10. Loomis D, Dufort V, Kleckner RC, Savitz nominator data for comparison. DA. Fatal occupational injuries among Third, all patients with spontaneous To our knowledge, this is the first electric power company workers. Am J Ind Med 1999;35:302-9. circulation documented at the scene population-based surveillance cohort survived and those in asystolic arrest study conducted in a nonselected 11. Rossignol M, Pineault M. Classification of fatal occupational electrocutions. Can J died. There are a number of case re- Canadian region, and it provides im- Public Health 1994;85:322-5. ports of electrocuted patients recov- portant information regarding the ering after prolonged resuscitation epidemiologic features of severe elec- 12. Taylor AJ, McGwin G Jr, Davis GG, Brissie RM, Rue LW 3rd. Occupational electrocu- attempts, and some experts recom- trical shock injuries. These injuries tions in Jefferson County, Alabama. Occup mend to continue efforts longer than occur rarely among residents of the Med (Lond) 2002;52:102-6. usual for other causes of asystolic CHR but are associated with high 13. Calgary Health Region Web site. Avail- arrest that often carry poor prog- death rates. Although the current able: www.calgaryhealthregion.ca/hecomm noses.3,17,18 Since the number of public health impact of these injuries /main/pdf/HecommBrochure.pdf (ac- patients in this study was small, our is relatively small, there exists an op- cessed 2005 May 4). data do not suggest that this practice portunity to prevent approximately 1 14. Baker S, O’Neill B, Haddon W Jr, Long should be otherwise. death per year among the residents WB. The injury severity score: a method There are few published data with of this region. for describing patients with multiple in- which to compare our incidence and juries and evaluating emergency care. J Acknowledgement: We thank MaryEllen Trauma 1974;14:187-96. mortality, but our rates appear to be Arnup, MSc, for her contribution to this article. lower than those observed in the US. 15. Alberta Health. Population estimates of US studies have suggested death Competing interests: None declared. Calgary Region by age group, 2000–2007. In: Calgary Health Region [Web site of rates due to electrocution of 5–25 References the Calgary Health Region] Available: per 1 million population annually, al- www.crha-health.ab.ca/hocr/influ/demo though these rates depend on the 1. Jain S, Bandi V. Electrical and lightning in- /popage.htm (accessed 2005 Jun 27). 3 specific populations studied. In ad- juries. Crit Care Clin 1999;15:319-31. 16. Laupland KB, Church DL, Mucenski M, dition, lightning injuries vary consid- Sutherland LR, Davies HD. Population- 2. Lederer W, Wiedermann FJ, Cerchiari E, based study of the epidemiology and the erably according to geography, with Baubin MA. Electricity-associated injuries I: risk factors for invasive Staphylococcus aureus southern and coastal regions at high- outdoor management of current-induced infections. J Infect Dis 2003;187:1452-9. est risk. Since the CHR is both land- casualties. Resuscitation 1999;43:69-77. locked and northern, it is not surpris- 17. Brindley PG, Markland DM, Mayers I, 3. Lederer W, Wiedermann FJ, Cerchiari E, ing that there were no cases Kutsogiannis DJ. Predictors of survival fol- Baubin MA. Electricity-associated injuries II: lowing in-hospital adult cardiopulmonary attributable to lightning in this outdoor management of lightning-induced resuscitation. CMAJ 2002;167:343-8. study. A reason why the electrocu- casualties. Resuscitation 2000;43:89-93. 18. Doig CJ, Boiteau PJ, Sandham JD. A tion rates were low in our study 4. Ore T, Casini V. Electrical fatalities among 2-year prospective cohort study of cardiac would be our failure to identify all U.S. construction workers. J Occup Envi- resuscitation in a major Canadian hospital. relevant cases. However, we believe ron Med 1996;38:587-92. Clin Invest Med 2000;23:132-43.

292 J can chir, Vol. 48, N o 4, aoßt 2005