Deliberate drug poisonings admitted to an emergency department in Paris area - a descriptive study and assessment of risk factors for intensive care admission
Beaune, S.; Juvin, P.; Beauchet, A.; Casalino, E.; Megarbane, B.
European Review for Medical and Pharmacological Sciences 20(6): 1174-1179
2016
ISSN/ISBN: 1128-3602 PMID: 27049274 Document Number: 690278
Each year, approximately 165,000 poisonings are managed in the emergency departments (ED) in France. We performed a descriptive analysis of self-poisoned patients admitted to a university hospital ED in the Paris metropolitan area (France) aimed at investigating their outcome and the risk factors for transfer to the intensive care unit (ICU). We retrospectively reviewed patients' records and performed multivariate logistic regression analysis to identify risk factors for ICU admission. During 4 years, 882 self-poisoned patients (median age, 38 years [IQR, 26-47]; sex-ratio, 1M/3F) were admitted to the ED, representing 0.7% of all referred patients. Poisonings mainly resulted from multidrug exposures (53%), including benzodiazepines (78%), serotonin reuptake inhibitors (17%), acetaminophen (13%), antipsychotics (9.5%), imidazopyridines (9.5%), antihypertensive drugs (3%), and polycyclic antidepressants (1.3%). Ethanol was involved in 20% of the exposures. Patients were briefly (<24h) monitored in the ED (55%), transferred to the psychiatric department (30%), medical ward (2%) or ICU (6%), and took an irregular discharge (7%). Among the patients transferred to the ICU, 25% were mechanically ventilated and only one died. Risk factors for ICU admission included antihypertensive (Odds ratio (OR), 40.6; 95%-confidence interval (CI), 7.5-221.9) or antipsychotic drug ingestion (OR, 5.3; CI, 2.0-14.4), male gender (OR, 3.3; CI, 1.30-8.8), and consciousness impairment (OR, 2.1; CI, 1.8-2.5 per point lost in Glasgow coma score). Deliberate drug exposure represents a frequent cause of ED admission. Psychotropic drugs are most commonly involved. Transfer to the ICU is rare and predicted by male gender, drug class, and coma depth.