Summary

Patient characteristics and severity-scale prediction of adverse events in poisoning cases treated in an emergency department

Rebeca García Díaz¹, Ana Ramos Rodríguez², Rubén Pérez García², Francisco Martín Rodríguez^1,3,4, Raúl López Izquierdo^1,5,5, Beatriz Martín Pérez1^1,2

Affiliation of the authors

¹Universidad de Valladolid, Facultad de Medicina, Valladolid, Spain. ²Servicio de Urgencias, Hospital Universitario Río Hortega, Valladolid, Spain. ³Servicios Médicos de Emergencia (SACYL), Soporte Vital Avanzado, Valladolid, Spain. ⁴Grupo de Investigación del Sistema de Puntuación de Alerta Temprana Prehospitalaria, Valladolid, Spain. ⁵CIBER de Enfermedades Respiratorias, Instituto de Salud Carlos II, Madrid, Spain.

DOI

Download Article

Quote

García Díaz R, Ramos Rodríguez A, Pérez García R, Martín Rodríguez F, López Izquierdo R, Martín Pérez B. Patient characteristics and severity-scale prediction of adverse events in poisoning cases treated in an emergency department. Rev Esp Urg Emerg. 2025;4:70–5

Summary

OBJECTIVES. To analyze the usefulness of severity scales for assessing the development of adverse events in patients who attended our emergency department (ED) for acute poisoning.
MATERIAL AND METHODS. Retrospective descriptive study of patients over the age of 18 years who came to our ED for acute medication or street-drug poisoning in 2022. We collected demographic information, vital constants, and clinical and laboratory data. The severity scales used were the Poisoning Early Warning Score (PEWS), the National Early Warning Score (NEWS), and the Quick Sepsis-related Organ Failure Assessment (qSOFA). A composite adverse event outcome was the dependent variable. The composite included at least 1 of the following situations: admission to the intensive care unit (ICU), ED-ICU consultation, in-hospital
mortality, 30-day mortality, need for advanced airway management, and need for vasoactive drugs. Descriptive statistics were compiled for univariate and multivariate analyses. We calculated the area under the receiver operating characteristic curves (AUCs) of the various severity scales; 95% CIs were calculated for all statistics. Statistical significance was set at P < .05.
RESULTS. A total of 240 patients were included. The median (interquartile range) age was 40.8 years ( 29.7-52.6 years), and 58.3% were women. Medications were the most often used substances (in 77.1%), benzodiazepines being the medications most often used (57.5%). Two street drugs, cannabis (10%) and cocaine (8.8%), were the next most commonly used substances. Ten percent of the patients
developed adverse events. Medications were responsible for more adverse events (12.6%) than street drugs (1.6%) (P < .05). The AUCs for the 3 severity scales were, respectively: for the PEWS, 0.760 (95% CI, 0.614-0.907; P < .001); the NEWS, 0.681 (95% CI, 0.529-0.833; P < .05); and the qSOFA, 0.667 (95% CI, 0.533-0.802; P < .05). The PEWS had a sensitivity of 0.62% (95% CI, 0.38%-0.86%) and a
specificity of 0.89% (95% CI, 0.84%-0.94%). Odds ratios from the multivariate analysis were 13.993 (95% CI, 3.840-50.986; P < .001) for the PEWS; 1.385 (95% CI, 1.125-1.704); P < .05) for the NEWS; and 3.171 (95% CI, 1.563-6.434; P < .001) for the qSOFA.
CONCLUSIONS. The PEWS best predicts adverse events in cases of poisoning from medications or street-drugs.