ORIGINAL ARTICLE
Pediatric Poisonings in a Rural Ugandan Emergency Department Katherine L. Boyle, MD,*† Usha Periyanayagam, MD, MPH,†‡§ Kavita M. Babu, MD,|| Brian T. Rice, MD,§¶ and Mark Bisanzo, MD§#
Objective: This study aims to describe pediatric poisonings presenting to a rural Ugandan emergency department (ED), identifying demographic factors and causative agents. Methods: This retrospective study was conducted in the ED of a rural hospital in the Rukungiri District of Uganda. A prospectively collected quality assurance database of ED visits was queried for poisonings in patients under the age of 5 who were admitted to the hospital. Cases were included if the chief complaint or final diagnosis included anything referable to poisoning, ingestion, or intoxication, or if a toxicologic antidote was administered. The database was coded by a blinded investigator, and descriptive statistics were performed. Results: From November 9, 2009, to July 11, 2014, 3428 patients under the age of 5 were admitted to the hospital. A total of 123 cases (3.6%) met the inclusion criteria. Seventy-two patients were male (58.5%). The average age was 2.3 (SD, 0.97) years with 45 children (36.6%) under the age of 2 years. There were 19 cases (15.4%) lost to 3-day follow-up. The top 3 documented exposures responsible for pediatric poisonings were cow tick or organophosphates (36 cases, 29.2%), general poison or drug overdose (26 cases, 21.1%), and paraffin or hydrocarbon (24 cases, 19.5%). Of the admitted patients, 1 died in the ED and 2 died at 72-hour follow-up, for an overall 72-hour mortality of 2.4%. Patients who died were exposed to iron, cow tick, and rat poison. Conclusions: Pediatric poisoning affects patients in rural sub-Saharan Africa. The mortality rate at one rural Ugandan hospital was greater than 2%. Key Words: poisoning, rural, toxicology, Uganda (Pediatr Emer Care 2017;00: 00–00)
I
nitiatives to decrease pediatric mortality have resulted in reduction of the number of under-5 (U5) deaths to below 6 million worldwide in 2015.1 Despite initiatives to gather relevant data, only 2.7% of deaths in children U5 were medically certified in 2010.2 In addition, data indicate that there is variation in both causes of childhood mortality and mortality rates between rural and urban areas. There is increasing realization of the role that noncommunicable disease plays in mortality in sub-Saharan Africa. Pediatric poisonings in rural sub-Saharan Africa represent an infrequent but important cause of morbidity and mortality.3 In addition, pediatric poisonings are unintentional injuries that have the potential to be reduced with targeted and effective public health prevention strategies. Development and implementation of such From the *Beth Israel Deaconess Medical Center; †Harvard Medical School, ‡Brigham and Women's Hospital, Boston; §Global Emergency Care Collaborative, Shrewsbury; ||University of Massachusetts Medical School, Worcester, MA, ¶New York University School of Medicine, New York, NY, #University of Vermont College of Medicine, Burlington, VT. Disclosure: The authors declare no conflict of interest. Reprints: Katherine L. Boyle, MD, Department of Emergency Medicine, Beth Israel Deaconess Medical Center, One Deaconess Rd, Boston, MA 02215 (e‐mail: [email protected]). Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0749-5161
Pediatric Emergency Care • Volume 00, Number 00, Month 2017
strategies mandate a detailed understanding of the characteristics and scope of the problem in the area being targeted. To date, no epidemiologic data exist from rural areas of sub-Saharan Africa to guide public health initiatives aimed at decreasing the incidence of unintentional pediatric poisoning in those regions. This article addresses that gap in the literature by describing the incidence and characteristics of U5 poisonings presenting to a single emergency department (ED) in rural Uganda from 2009 to 2014.
METHODS The study is a retrospective observational study of pediatric toxicology cases that were admitted to Karoli Lwanga Hospital. This is a district hospital in the rural Rukungiri District of southwestern Uganda. Admitted patients were identified using a prospectively collected quality assurance (QA) database of all ED patient visits, including 3-day follow-up data on clinical status. The ED of the hospital is staffed by a group of nonphysician clinicians (emergency care practitioners) who completed an advanced training course in emergency care developed by the Global Emergency Care Collaborative.4 Each patient who presented to the ED was entered into the database (Microsoft Excel and Access) by a trained research assistant (RA). Demographic, clinical, and administrative information was recorded. The deidentified data were abstracted by a single researcher to produce the data set. All patients U5 who were admitted into the hospital were included in the data set. Cases were defined as poisoning if the patient had either a chief complaint or final diagnosis involving poisoning, ingestion, or intoxication, or if a patient received an antidote in the ED. The data abstracted included the following: patient demographic characteristics, medications administered in the ED, clinical outcomes, and condition at 3-day follow-up. All patients in the QA database were followed up at 3 days regardless of their disposition. For this study, analysis focused on mortality of in-hospital patients, so all patients that were “discharged” were excluded from analysis, and all others (including “admitted,” “dead on arrival,” “died in ED,” and “eloped”) were considered as possible cases. By local protocol, any poisoned patient with active symptoms or a concerning history is admitted for further monitoring. Overall, 68.8% of the 4985 pediatric U5 patients evaluated in the ED during this period were admitted.5 The cases were independently reviewed by 2 blinded United States emergency physicians (K.L. B. and M.B.). The final lists were then compared, and a K statistic was used to determine agreement. Discrepancies were reviewed by a third investigator who was also a United States emergency physician and toxicologist (K.M.B.). The database was subsequently coded by a blinded investigator, and descriptive statistics were performed. There were 4 gaps in follow-up data collection during the study period, when no RAs were available. These gaps include March 21 to July 21, 2010; September 25 to September 27, 2010; January 18 to February 2, 2011; and February 14 to February 26, 2011. These periods of missing data represented a www.pec-online.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
1
Pediatric Emergency Care • Volume 00, Number 00, Month 2017
Boyle et al
total of 153 days (9%) during the study period. All patients presenting outside of these dates were included in the database. This study was approved by the Mbarara University of Science and Technology Institutional Review Board, the Uganda National Council for Science and Technology, and the University of Massachusetts Medical School Institutional Review Board.
RESULTS From November 9, 2009, to July 11, 2014, a total of 3428 patients U5 were admitted to Karoli Lwanga Hospital. Of these, 123 cases (3.6%) met inclusion criteria as pediatric poisoning cases. Cases were reviewed for inclusion by 2 investigators; interrater reliability was 97.7%, with a K = 0.95. There were 72 boys (59%) and 51 girls (41%). The average patient age was 2.3 (SD, 0.97) years, with 45 children (36.6%) under the age of 2 years old. There were 19 cases (15.4%) of patients who were admitted and subsequently discharged who were lost to 3-day follow-up. The top 5 documented exposures responsible for pediatric poisonings requiring admission are listed in Table 1 and are organophosphates (OPs) or cow tick (36 cases, 29%), general poison or drug overdose (26 cases, 18%), paraffin/hydrocarbon (24 cases, 20%), rat poison (8 cases, 7%), and battery ingestion (5 cases, 4%). In this region, the term cow tick generally refers to OPs but may also be amitraz, which is a formamidine pesticide. Other exposures included the following: drug reaction leading to Stevens-Johnson syndrome, cassava ingestion (resulting in cyanide toxicity), promethazine, alcohol, foreign body, herbal medication, formaline, carvedilol, quinine, phenytoin, amoxicillin, calamine lotion, and phenobarbital. Of the pediatric patients admitted for poisoning, 119 received inpatient care, 3 eloped, and 1 died in the ED. At 72-hour follow-up, 2 more patients had died. Overall, children U5 admitted with poisoning had a 72-hour mortality rate of 2.4% (n = 3). The 3 fatal cases were exposed to iron, cow tick, and an unknown agent identified as rat poison. Overall, 25 patients received activated charcoal, 24 received atropine, 47 received intravenous saline, and 18 received antibiotics or anti-malarial agents.
DISCUSSION Our retrospective review of the QA database found that 3.6% of U5 admissions were due to poisoning (n = 123 of 3428). The mortality rate of children U5 admitted owing to poisoning was 2.4%(n = 3 of 123), which is lower than the previously reported overall mortality rate of 4.75% (n = 163 of 3428) for admitted patients in this age cohort at this center.5,6 Organophosphates and cow tick were the agents documented as responsible for most of the pediatric exposures presenting for care during our study period, which is consistent with top causes of poisoning worldwide.7 Organophosphates are widely available in the area surrounding the hospital and are easily obtained in town for low cost. As many families support themselves with agriculture or own animals, OPs can be found in many homes in the region, often stored in unlabeled plastic bottles. In this cohort, TABLE 1. Top 5 Documented Poisonings Requiring Admission Exposure
No. Cases
Cow tick/OPs General poison or drug overdose Paraffin/hydrocarbon Rat poison Battery ingestion
36 (29%) 26 (18%) 24 (20%) 8 (7%) 5 (4%)
2
www.pec-online.com
there was only 1 fatality in a patient exposed to an OP (1/36 cases, 2.7% mortality), which is significantly less than the mortality rate generally observed with OP toxicity.8 There are several possible reasons for this. First, the study population was comprised of only children with unintentional exposures to OPs, not adult patients with intentional ingestions and suicidal intent, which are known to be factors associated with increased mortality and included in the 15% mortality rate.8 In addition, the providers were trained to manage these patients aggressively with atropine, which may have helped reduce mortality. Finally, the terms OPs and cow tick are considered interchangeable in the provider documentation of the cases. However, cow tick may also refer to amitraz, a formamidine pesticide, which is a centrally acting α-agonist that is generally thought to be less toxic than OPs. Organophosphates work differently than formamidine pesticides by inhibiting acetylcholinesterase, leading to accumulation of acetylcholine in the body. Over 3 years in 1 hospital in Turkey, 11 children 6 years and under were treated for amitraz poisoning with no fatalities.9 Another group from Turkey reported 43 patients aged 13 and under from 1993 to 2000 with accidental amitraz poisoning with a 0% mortality rate.10 In comparison, during a study in Uganda, 40 hospitals reported seeing 71 adult and pediatric cases of OP poisonings over a 6-month period with a 27% mortality rate.11 The quantity of OP exposures observed in the Rukungiri ED highlights the difference in causative agents in pediatric poisonings in rural versus urban Uganda. In a paper published in the British Medical Journal in 1969, 130 children were observed to be admitted to an urban hospital for care after an accidental poisoning.12 Accidental poisoning accounted for 0.65% of all pediatric admissions, as a total of 20,061 children were admitted during the 6-year period. Seven of the 130 children admitted for accidental poisoning died (5.4% mortality), all of which occurred in children less than 4 years old.12 The seven fatalities were due to aspirin, chlorpromazine, chloroquine, codeine, dapsone, thiazine, and waragi (a local alcohol). However, OP exposure was rare with only 1 child presenting for care after exposure to OPs.12 Kerosene and waragi represented the toxins seen in the most unintentional poisonings.12 The decreased mortality rate observed in the present study may be reflective of the advances of care that have been made over the last decades. More recently, a survey of 20 Ugandan hospitals reported 38 pediatric deaths from poisoning (children up to 10 years old), with OPs being the second most common reported agent (8 cases, 21%) behind kerosene (21 cases, 55%).13 Differences in causative agents responsible for pediatric poisonings also vary across urban medical centers in Africa. Recent studies in Kenya have also cited kerosene and petroleum products as the most common toxicologic exposure in children requiring admission to the hospital. In 2008, Kilifi District Hospital in Kenya observed that 62% of pediatric poisoning cases admitted were due to kerosene exposure.14 In addition, from 2003 to 2007, petroleum products were responsible for the majority (30%) of hospital admissions in children up to 13 years old in Nairobi, Kenya, with only 6.7% of admissions being due to OPs.15 Similar trends were observed at an urban tertiary care center in South Africa where kerosene was the most common agent responsible for pediatric poisoning.16 However, a northern Africa urban Egyptian tertiary care center found the most common pediatric pharmaceutical exposure was to nonopioid analgesics, whereas the most frequent nonpharmaceutical exposures were corrosives in preschool children and pesticides in adolescents.17 Generally, poisoning in preschool children was unintentional and mostly due to nonpharmaceutical agents. The overall mortality rate at this center was 0.3%.17 The mortality rate was lower than what was observed in our cohort, but it included all patients that came to the ED, not © 2017 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care • Volume 00, Number 00, Month 2017
just admitted patients, and patients aged 0 to 18 years old. Patient fatalities were primarily from pesticides, including OPs and carbamates, followed by kerosene and corrosives. The variability in causative agents responsible for pediatric poisonings across different urban hospitals demonstrates the need for current epidemiological data reflective of the region where preventative efforts are to be deployed. This work does have a number of limitations. Cases of pediatric poisoning were identified through a search of an electronic database. Cases of poisoning that presented as altered mental status, seizures, respiratory failure, or death in which poisoning was not suspected by treating clinicians would not have been detected by this methodology. Because the study authors performed the analysis of the database, they were not blind to the aims of the study, potentially introducing bias. This study also is limited by its retrospective nature and reliance on the electronic database for information surrounding each poisoning. Incomplete or vague descriptions of events in the database could have led to misclassification. Also, confirmatory testing is not available at the hospital to identify the agents to which the patients were exposed. This mortality rate was determined using all patients who were admitted for care and does not include pediatric exposure patients who were discharged. Generally, most pediatric toxicology patients are admitted for observation owing to difficulty reaccessing care if symptoms develop. In addition, little literature exists on the treatment of patients in this setting to support an observation period in the ED with subsequent discharge to home, as is often seen in the ED in the United States. Thus, providers were trained to generally admit poisoned patients for observation, so we believe the number of pediatric toxicology patients that were discharged was low. In addition, 19 patients were lost to 3-day follow-up. Research assistants make multiple attempts to contact the patients to complete the 3-day follow-up; encounters take place on the ward for admitted patients or by phone for patients who are discharged from the ward before 3 days. Also, there were 4 gaps in follow-up data collection during the study period, when no RAs were available. Although these gaps totaled less than 6 months of cumulative time, we were potentially missing cases that would have otherwise met inclusion criteria. Finally, this study was conducted in a single rural Ugandan ED. This hospital is unusual in its novel structure and staffing model. However, the goal of this article is to describe the characteristics of the patients presenting with poisonings for care, which may potentially be generalizable to other settings with similar regional and patient characteristics. More research is needed to determine if these trends are consistent across rural Uganda and sub-Saharan Africa.
CONCLUSIONS Toxicologic illness affected a minority of pediatric patients presenting to the ED in rural sub-Saharan Africa; however, they represent a source of pediatric mortality that may be reduced with effective interventions. Organophosphates and hydrocarbons are a common cause of poisoning in this population, but the mortality is relatively low. The results of this study are from one site in rural
© 2017 Wolters Kluwer Health, Inc. All rights reserved.
Pediatric Poisonings in a Rural Ugandan ED
Uganda and may have limited generalizability. More data are needed from other sites to determine if these patterns are replicated throughout rural sub-Saharan Africa. Targeted prevention and treatment efforts should also be studied to determine local best prevention practices. REFERENCES 1. The Millennium Development Goals Report Summary. Available at: http:// www.un.org/millenniumgoals/2015_MDG_Report/pdf/MDG%202015% 20Summary%20web_english.pdf. Accessed March 23, 2016. 2. Liu L, Johnson HL, Cousens S, et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet. 2012;379:2151–2161. 3. Mutto M, Lawoko S, Nansamba C, et al. Unintentional childhood injury patterns, odds, and outcomes in Kampala City: an analysis of surveillance data from the National Pediatric Emergency Unit. J Inj Violence Res. 2011;3:13–18. 4. Hammerstedt H, Maling S, Kasyaba R, et al. World Health Assembly Resolution 60.22. [corrected]. Ann Emerg Med. 2014;64:461–468. 5. Rice B, Periyanayagam U, Chamberlain S, et al. Mortality in children under five receiving nonphysician clinician emergency care in Uganda. Pediatrics. 2016;137:e20153201. 6. Chamberlain S, Stolz U, Dreifuss B, et al. Mortality related to acute illness and injury in rural Uganda: task shifting to improve outcomes. PLoS One. 2015;10:e0122559. 7. Chen Y. Organophosphate-induced brain damage: mechanisms, neuropsychiatric and neurological consequences, and potential therapeutic strategies. NeuroToxicology. 2012;33:391–400. 8. Eddleston M, Buckley NA, Eyer P, et al. Management of acute organophosphorus pesticide poisoning. Lancet. 2008;371:597–607. 9. Yaramis A, Soker M, Bilici M. Amitraz poisoning in children. Hum Exp Toxicol. 2000;19:431–433. 10. Kalyoncu M, Dilber E, Okten A. Amitraz intoxication in children in the rural Black Sea region: analysis of forty-three patients. Hum Exp Toxicol. 2002;21:269–272. 11. Tibbutt D. Poisoning with organophosphates. South Sudan Medical Journal. 2013;6:13–16. 12. Bwibo NO. Accidental poisoning in children in Uganda. Br Med J. 1969;4:601–602. 13. Tibbutt D. Is poisoning a problem in South Sudan? South Sudan Medical Journal. 2011;4:90–91. 14. Lang T, Thuo N, Akech S. Accidental paraffin poisoning in Kenyan children. Trop Med Int Health. 2008;13:845–847. 15. Mutiso VM, Muoki AS, Kimeu MM. Patterns of poisoning among patients aged 0–13 years at a paediatric hospital in Nairobi. East Afr Med J. 2014;91:379–384. 16. Balme K, Roberts JC, Glasstone M, et al. The changing trends of childhood poisoning at a tertiary children's hospital in South Africa. S Afr Med J. 2012;102:142–146. 17. Azab SM, Hirshon JM, Hayes BD, et al. Epidemiology of acute poisoning in children presenting to the poisoning treatment center at Ain Shams University in Cairo, Egypt, 2009–2013. Clin Toxicol. 2016;54:20–26.
www.pec-online.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
3
Pediatric Poisonings in a Rural Ugandan Emergency Department Katherine L. Boyle, MD,*† Usha Periyanayagam, MD, MPH,†‡§ Kavita M. Babu, MD,|| Brian T. Rice, MD,§¶ and Mark Bisanzo, MD§#
Objective: This study aims to describe pediatric poisonings presenting to a rural Ugandan emergency department (ED), identifying demographic factors and causative agents. Methods: This retrospective study was conducted in the ED of a rural hospital in the Rukungiri District of Uganda. A prospectively collected quality assurance database of ED visits was queried for poisonings in patients under the age of 5 who were admitted to the hospital. Cases were included if the chief complaint or final diagnosis included anything referable to poisoning, ingestion, or intoxication, or if a toxicologic antidote was administered. The database was coded by a blinded investigator, and descriptive statistics were performed. Results: From November 9, 2009, to July 11, 2014, 3428 patients under the age of 5 were admitted to the hospital. A total of 123 cases (3.6%) met the inclusion criteria. Seventy-two patients were male (58.5%). The average age was 2.3 (SD, 0.97) years with 45 children (36.6%) under the age of 2 years. There were 19 cases (15.4%) lost to 3-day follow-up. The top 3 documented exposures responsible for pediatric poisonings were cow tick or organophosphates (36 cases, 29.2%), general poison or drug overdose (26 cases, 21.1%), and paraffin or hydrocarbon (24 cases, 19.5%). Of the admitted patients, 1 died in the ED and 2 died at 72-hour follow-up, for an overall 72-hour mortality of 2.4%. Patients who died were exposed to iron, cow tick, and rat poison. Conclusions: Pediatric poisoning affects patients in rural sub-Saharan Africa. The mortality rate at one rural Ugandan hospital was greater than 2%. Key Words: poisoning, rural, toxicology, Uganda (Pediatr Emer Care 2017;00: 00–00)
I
nitiatives to decrease pediatric mortality have resulted in reduction of the number of under-5 (U5) deaths to below 6 million worldwide in 2015.1 Despite initiatives to gather relevant data, only 2.7% of deaths in children U5 were medically certified in 2010.2 In addition, data indicate that there is variation in both causes of childhood mortality and mortality rates between rural and urban areas. There is increasing realization of the role that noncommunicable disease plays in mortality in sub-Saharan Africa. Pediatric poisonings in rural sub-Saharan Africa represent an infrequent but important cause of morbidity and mortality.3 In addition, pediatric poisonings are unintentional injuries that have the potential to be reduced with targeted and effective public health prevention strategies. Development and implementation of such From the *Beth Israel Deaconess Medical Center; †Harvard Medical School, ‡Brigham and Women's Hospital, Boston; §Global Emergency Care Collaborative, Shrewsbury; ||University of Massachusetts Medical School, Worcester, MA, ¶New York University School of Medicine, New York, NY, #University of Vermont College of Medicine, Burlington, VT. Disclosure: The authors declare no conflict of interest. Reprints: Katherine L. Boyle, MD, Department of Emergency Medicine, Beth Israel Deaconess Medical Center, One Deaconess Rd, Boston, MA 02215 (e‐mail: [email protected]). Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0749-5161
Pediatric Emergency Care • Volume 00, Number 00, Month 2017
strategies mandate a detailed understanding of the characteristics and scope of the problem in the area being targeted. To date, no epidemiologic data exist from rural areas of sub-Saharan Africa to guide public health initiatives aimed at decreasing the incidence of unintentional pediatric poisoning in those regions. This article addresses that gap in the literature by describing the incidence and characteristics of U5 poisonings presenting to a single emergency department (ED) in rural Uganda from 2009 to 2014.
METHODS The study is a retrospective observational study of pediatric toxicology cases that were admitted to Karoli Lwanga Hospital. This is a district hospital in the rural Rukungiri District of southwestern Uganda. Admitted patients were identified using a prospectively collected quality assurance (QA) database of all ED patient visits, including 3-day follow-up data on clinical status. The ED of the hospital is staffed by a group of nonphysician clinicians (emergency care practitioners) who completed an advanced training course in emergency care developed by the Global Emergency Care Collaborative.4 Each patient who presented to the ED was entered into the database (Microsoft Excel and Access) by a trained research assistant (RA). Demographic, clinical, and administrative information was recorded. The deidentified data were abstracted by a single researcher to produce the data set. All patients U5 who were admitted into the hospital were included in the data set. Cases were defined as poisoning if the patient had either a chief complaint or final diagnosis involving poisoning, ingestion, or intoxication, or if a patient received an antidote in the ED. The data abstracted included the following: patient demographic characteristics, medications administered in the ED, clinical outcomes, and condition at 3-day follow-up. All patients in the QA database were followed up at 3 days regardless of their disposition. For this study, analysis focused on mortality of in-hospital patients, so all patients that were “discharged” were excluded from analysis, and all others (including “admitted,” “dead on arrival,” “died in ED,” and “eloped”) were considered as possible cases. By local protocol, any poisoned patient with active symptoms or a concerning history is admitted for further monitoring. Overall, 68.8% of the 4985 pediatric U5 patients evaluated in the ED during this period were admitted.5 The cases were independently reviewed by 2 blinded United States emergency physicians (K.L. B. and M.B.). The final lists were then compared, and a K statistic was used to determine agreement. Discrepancies were reviewed by a third investigator who was also a United States emergency physician and toxicologist (K.M.B.). The database was subsequently coded by a blinded investigator, and descriptive statistics were performed. There were 4 gaps in follow-up data collection during the study period, when no RAs were available. These gaps include March 21 to July 21, 2010; September 25 to September 27, 2010; January 18 to February 2, 2011; and February 14 to February 26, 2011. These periods of missing data represented a www.pec-online.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
1
Pediatric Emergency Care • Volume 00, Number 00, Month 2017
Boyle et al
total of 153 days (9%) during the study period. All patients presenting outside of these dates were included in the database. This study was approved by the Mbarara University of Science and Technology Institutional Review Board, the Uganda National Council for Science and Technology, and the University of Massachusetts Medical School Institutional Review Board.
RESULTS From November 9, 2009, to July 11, 2014, a total of 3428 patients U5 were admitted to Karoli Lwanga Hospital. Of these, 123 cases (3.6%) met inclusion criteria as pediatric poisoning cases. Cases were reviewed for inclusion by 2 investigators; interrater reliability was 97.7%, with a K = 0.95. There were 72 boys (59%) and 51 girls (41%). The average patient age was 2.3 (SD, 0.97) years, with 45 children (36.6%) under the age of 2 years old. There were 19 cases (15.4%) of patients who were admitted and subsequently discharged who were lost to 3-day follow-up. The top 5 documented exposures responsible for pediatric poisonings requiring admission are listed in Table 1 and are organophosphates (OPs) or cow tick (36 cases, 29%), general poison or drug overdose (26 cases, 18%), paraffin/hydrocarbon (24 cases, 20%), rat poison (8 cases, 7%), and battery ingestion (5 cases, 4%). In this region, the term cow tick generally refers to OPs but may also be amitraz, which is a formamidine pesticide. Other exposures included the following: drug reaction leading to Stevens-Johnson syndrome, cassava ingestion (resulting in cyanide toxicity), promethazine, alcohol, foreign body, herbal medication, formaline, carvedilol, quinine, phenytoin, amoxicillin, calamine lotion, and phenobarbital. Of the pediatric patients admitted for poisoning, 119 received inpatient care, 3 eloped, and 1 died in the ED. At 72-hour follow-up, 2 more patients had died. Overall, children U5 admitted with poisoning had a 72-hour mortality rate of 2.4% (n = 3). The 3 fatal cases were exposed to iron, cow tick, and an unknown agent identified as rat poison. Overall, 25 patients received activated charcoal, 24 received atropine, 47 received intravenous saline, and 18 received antibiotics or anti-malarial agents.
DISCUSSION Our retrospective review of the QA database found that 3.6% of U5 admissions were due to poisoning (n = 123 of 3428). The mortality rate of children U5 admitted owing to poisoning was 2.4%(n = 3 of 123), which is lower than the previously reported overall mortality rate of 4.75% (n = 163 of 3428) for admitted patients in this age cohort at this center.5,6 Organophosphates and cow tick were the agents documented as responsible for most of the pediatric exposures presenting for care during our study period, which is consistent with top causes of poisoning worldwide.7 Organophosphates are widely available in the area surrounding the hospital and are easily obtained in town for low cost. As many families support themselves with agriculture or own animals, OPs can be found in many homes in the region, often stored in unlabeled plastic bottles. In this cohort, TABLE 1. Top 5 Documented Poisonings Requiring Admission Exposure
No. Cases
Cow tick/OPs General poison or drug overdose Paraffin/hydrocarbon Rat poison Battery ingestion
36 (29%) 26 (18%) 24 (20%) 8 (7%) 5 (4%)
2
www.pec-online.com
there was only 1 fatality in a patient exposed to an OP (1/36 cases, 2.7% mortality), which is significantly less than the mortality rate generally observed with OP toxicity.8 There are several possible reasons for this. First, the study population was comprised of only children with unintentional exposures to OPs, not adult patients with intentional ingestions and suicidal intent, which are known to be factors associated with increased mortality and included in the 15% mortality rate.8 In addition, the providers were trained to manage these patients aggressively with atropine, which may have helped reduce mortality. Finally, the terms OPs and cow tick are considered interchangeable in the provider documentation of the cases. However, cow tick may also refer to amitraz, a formamidine pesticide, which is a centrally acting α-agonist that is generally thought to be less toxic than OPs. Organophosphates work differently than formamidine pesticides by inhibiting acetylcholinesterase, leading to accumulation of acetylcholine in the body. Over 3 years in 1 hospital in Turkey, 11 children 6 years and under were treated for amitraz poisoning with no fatalities.9 Another group from Turkey reported 43 patients aged 13 and under from 1993 to 2000 with accidental amitraz poisoning with a 0% mortality rate.10 In comparison, during a study in Uganda, 40 hospitals reported seeing 71 adult and pediatric cases of OP poisonings over a 6-month period with a 27% mortality rate.11 The quantity of OP exposures observed in the Rukungiri ED highlights the difference in causative agents in pediatric poisonings in rural versus urban Uganda. In a paper published in the British Medical Journal in 1969, 130 children were observed to be admitted to an urban hospital for care after an accidental poisoning.12 Accidental poisoning accounted for 0.65% of all pediatric admissions, as a total of 20,061 children were admitted during the 6-year period. Seven of the 130 children admitted for accidental poisoning died (5.4% mortality), all of which occurred in children less than 4 years old.12 The seven fatalities were due to aspirin, chlorpromazine, chloroquine, codeine, dapsone, thiazine, and waragi (a local alcohol). However, OP exposure was rare with only 1 child presenting for care after exposure to OPs.12 Kerosene and waragi represented the toxins seen in the most unintentional poisonings.12 The decreased mortality rate observed in the present study may be reflective of the advances of care that have been made over the last decades. More recently, a survey of 20 Ugandan hospitals reported 38 pediatric deaths from poisoning (children up to 10 years old), with OPs being the second most common reported agent (8 cases, 21%) behind kerosene (21 cases, 55%).13 Differences in causative agents responsible for pediatric poisonings also vary across urban medical centers in Africa. Recent studies in Kenya have also cited kerosene and petroleum products as the most common toxicologic exposure in children requiring admission to the hospital. In 2008, Kilifi District Hospital in Kenya observed that 62% of pediatric poisoning cases admitted were due to kerosene exposure.14 In addition, from 2003 to 2007, petroleum products were responsible for the majority (30%) of hospital admissions in children up to 13 years old in Nairobi, Kenya, with only 6.7% of admissions being due to OPs.15 Similar trends were observed at an urban tertiary care center in South Africa where kerosene was the most common agent responsible for pediatric poisoning.16 However, a northern Africa urban Egyptian tertiary care center found the most common pediatric pharmaceutical exposure was to nonopioid analgesics, whereas the most frequent nonpharmaceutical exposures were corrosives in preschool children and pesticides in adolescents.17 Generally, poisoning in preschool children was unintentional and mostly due to nonpharmaceutical agents. The overall mortality rate at this center was 0.3%.17 The mortality rate was lower than what was observed in our cohort, but it included all patients that came to the ED, not © 2017 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care • Volume 00, Number 00, Month 2017
just admitted patients, and patients aged 0 to 18 years old. Patient fatalities were primarily from pesticides, including OPs and carbamates, followed by kerosene and corrosives. The variability in causative agents responsible for pediatric poisonings across different urban hospitals demonstrates the need for current epidemiological data reflective of the region where preventative efforts are to be deployed. This work does have a number of limitations. Cases of pediatric poisoning were identified through a search of an electronic database. Cases of poisoning that presented as altered mental status, seizures, respiratory failure, or death in which poisoning was not suspected by treating clinicians would not have been detected by this methodology. Because the study authors performed the analysis of the database, they were not blind to the aims of the study, potentially introducing bias. This study also is limited by its retrospective nature and reliance on the electronic database for information surrounding each poisoning. Incomplete or vague descriptions of events in the database could have led to misclassification. Also, confirmatory testing is not available at the hospital to identify the agents to which the patients were exposed. This mortality rate was determined using all patients who were admitted for care and does not include pediatric exposure patients who were discharged. Generally, most pediatric toxicology patients are admitted for observation owing to difficulty reaccessing care if symptoms develop. In addition, little literature exists on the treatment of patients in this setting to support an observation period in the ED with subsequent discharge to home, as is often seen in the ED in the United States. Thus, providers were trained to generally admit poisoned patients for observation, so we believe the number of pediatric toxicology patients that were discharged was low. In addition, 19 patients were lost to 3-day follow-up. Research assistants make multiple attempts to contact the patients to complete the 3-day follow-up; encounters take place on the ward for admitted patients or by phone for patients who are discharged from the ward before 3 days. Also, there were 4 gaps in follow-up data collection during the study period, when no RAs were available. Although these gaps totaled less than 6 months of cumulative time, we were potentially missing cases that would have otherwise met inclusion criteria. Finally, this study was conducted in a single rural Ugandan ED. This hospital is unusual in its novel structure and staffing model. However, the goal of this article is to describe the characteristics of the patients presenting with poisonings for care, which may potentially be generalizable to other settings with similar regional and patient characteristics. More research is needed to determine if these trends are consistent across rural Uganda and sub-Saharan Africa.
CONCLUSIONS Toxicologic illness affected a minority of pediatric patients presenting to the ED in rural sub-Saharan Africa; however, they represent a source of pediatric mortality that may be reduced with effective interventions. Organophosphates and hydrocarbons are a common cause of poisoning in this population, but the mortality is relatively low. The results of this study are from one site in rural
© 2017 Wolters Kluwer Health, Inc. All rights reserved.
Pediatric Poisonings in a Rural Ugandan ED
Uganda and may have limited generalizability. More data are needed from other sites to determine if these patterns are replicated throughout rural sub-Saharan Africa. Targeted prevention and treatment efforts should also be studied to determine local best prevention practices. REFERENCES 1. The Millennium Development Goals Report Summary. Available at: http:// www.un.org/millenniumgoals/2015_MDG_Report/pdf/MDG%202015% 20Summary%20web_english.pdf. Accessed March 23, 2016. 2. Liu L, Johnson HL, Cousens S, et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet. 2012;379:2151–2161. 3. Mutto M, Lawoko S, Nansamba C, et al. Unintentional childhood injury patterns, odds, and outcomes in Kampala City: an analysis of surveillance data from the National Pediatric Emergency Unit. J Inj Violence Res. 2011;3:13–18. 4. Hammerstedt H, Maling S, Kasyaba R, et al. World Health Assembly Resolution 60.22. [corrected]. Ann Emerg Med. 2014;64:461–468. 5. Rice B, Periyanayagam U, Chamberlain S, et al. Mortality in children under five receiving nonphysician clinician emergency care in Uganda. Pediatrics. 2016;137:e20153201. 6. Chamberlain S, Stolz U, Dreifuss B, et al. Mortality related to acute illness and injury in rural Uganda: task shifting to improve outcomes. PLoS One. 2015;10:e0122559. 7. Chen Y. Organophosphate-induced brain damage: mechanisms, neuropsychiatric and neurological consequences, and potential therapeutic strategies. NeuroToxicology. 2012;33:391–400. 8. Eddleston M, Buckley NA, Eyer P, et al. Management of acute organophosphorus pesticide poisoning. Lancet. 2008;371:597–607. 9. Yaramis A, Soker M, Bilici M. Amitraz poisoning in children. Hum Exp Toxicol. 2000;19:431–433. 10. Kalyoncu M, Dilber E, Okten A. Amitraz intoxication in children in the rural Black Sea region: analysis of forty-three patients. Hum Exp Toxicol. 2002;21:269–272. 11. Tibbutt D. Poisoning with organophosphates. South Sudan Medical Journal. 2013;6:13–16. 12. Bwibo NO. Accidental poisoning in children in Uganda. Br Med J. 1969;4:601–602. 13. Tibbutt D. Is poisoning a problem in South Sudan? South Sudan Medical Journal. 2011;4:90–91. 14. Lang T, Thuo N, Akech S. Accidental paraffin poisoning in Kenyan children. Trop Med Int Health. 2008;13:845–847. 15. Mutiso VM, Muoki AS, Kimeu MM. Patterns of poisoning among patients aged 0–13 years at a paediatric hospital in Nairobi. East Afr Med J. 2014;91:379–384. 16. Balme K, Roberts JC, Glasstone M, et al. The changing trends of childhood poisoning at a tertiary children's hospital in South Africa. S Afr Med J. 2012;102:142–146. 17. Azab SM, Hirshon JM, Hayes BD, et al. Epidemiology of acute poisoning in children presenting to the poisoning treatment center at Ain Shams University in Cairo, Egypt, 2009–2013. Clin Toxicol. 2016;54:20–26.
www.pec-online.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
3
