Background: Acute poisoning in children is a medical emergency and preventable cause of morbidity and mortality. Knowledge about the nature, magnitude, outcome and predictors of outcome is necessary for management and allocation of scant resources.
Methods: This is a retrospective study conducted in the Pediatric Intensive Care Unit (PICU) of an urban multi speciality teaching and referral hospital in North India from January 1993 to June 2008 to determine the epidemiology, clinical profile, outcome and predictors of outcome in children with acute poisoning. Data of 225 children with acute poisoning was retrieved from case records with respect to demographic profile, time to presentation, PRISM score, clinical features, investigations, therapeutic measures, complications and outcome in terms of survival or death. Survivors and non-survivors were compared to determine the predictors of mortality.
Results: Acute poisoning constituted 3.9% of total PICU admissions; almost all (96.9%) were accidental. The mean age of study patient's was 3.3 ± 3.1 (range 0.10-12) years with majority (61.3%) being toddlers (1-3 years). In the overall cohort, kerosene (27.1%) and prescription drugs (26.7%) were the most common causative agents followed by organophosphates (16.0%), corrosives (7.6%), carbamates (4.9%) and aluminum phosphide (4.9%). However the trends of the three 5-year interval (1993 till the end of 1997, 1998 till the end of 2002 and 2003 till the end of June 2008) revealed a significant decrease in kerosene, aluminum phosphide and iron with increase in organophosphate compound poisoning. Ninety nine (44%) patients required supplemental oxygen, of which nearly half (n = 42; 42.4%) needed mechanical ventilation. Twenty (8.9%) died; cause of death being iron poisoning in five; aluminum phosphide in four; organophosphates in three and one each because of kerosene, diesel, carbamate, corrosive, sewing machine lubricant, isoniazid, salicylate and maduramycin poisoning. There has been a significant decrease in the mortality over the years. The non-survivors were older, had a higher PRISM score and hypotension at admission and higher need for oxygen and ventilation. On multiple logistic regression analysis hypotension at admission was the most significant predictor of death (adjusted odds ratio: 5.59; 95% confidence interval: 1.38-22.63; p = 0.016).
Conclusion: Acute poisoning in children over the past 15 years has shown a changing trend with significant decrease in kerosene, iron and aluminum phosphide and an increase in organophosphate and prescription drugs. The overall mortality has decreased significantly. Hypotension at admission was the most significant predictor of death.