A Time-Dependent Propensity Score Matching Approach to Assess Epinephrine Use on Patients Survival Within Out-of-Hospital Cardiac Arrest Care

• Published in: The Journal of emergency medicine Vol. 59; no. 4; pp. 542 - 552
• Main Authors: Baert, Valentine, Hubert, Hervé, Chouihed, Tahar, Claustre, Clément, Wiel, Éric, Escutnaire, Joséphine, Jaeger, Déborah, Vilhelm, Christian, Segal, Nicolas, Adnet, Frédéric, Gueugniaud, Pierre-Yves, Tazarourte, Karim, Mebazaa, Alexandre, Fraticelli, Laurie, El Khoury, Carlos
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.10.2020; Elsevier
• Subjects: Cardiology and cardiovascular system; epinephrine; Human health and pathology; Life Sciences; Mathematics; Methodology; out-of-hospital cardiac arrest; resuscitation; Santé publique et épidémiologie; Statistics; time-dependent propensity score; resuscitation; epinephrine; time-dependent propensity score; out-of-hospital cardiac arrest
• ISSN: 0736-4679; 2352-5029
• DOI: 10.1016/j.jemermed.2020.06.016

Epinephrine effectiveness and safety are still questioned. It is well known that the effect of epinephrine varies depending on patients’ rhythm and time to injection. We aimed to assess the association between epinephrine use during out-of-hospital cardiac arrest (OHCA) care and patient 30-day (D30) survival. Between 2011 and 2017, 27,008 OHCA patients were included from the French OHCA registry. We adjusted populations using a time-dependent propensity score matching. Analyses were stratified according to patient's first rhythm. After matching, 2837 pairs of patients with a shockable rhythm were created and 20,950 with a nonshockable rhythm. Whatever the patient's rhythm (shockable or nonshockable), epinephrine use was associated with less D30 survival (odds ratio [OR] 0.508; 95% confidence interval [CI] 0.440–0.586] and OR 0.645; 95% CI 0.549–0.759, respectively). In shockable rhythms, on all outcomes, epinephrine use was deleterious. In nonshockable rhythms, no difference was observed regarding return of spontaneous circulation and survival at hospital admission. However, epinephrine use was associated with worse neurological prognosis (OR 0.646; 95% CI 0.549–0.759). In shockable and nonshockable rhythms, epinephrine does not seem to have any benefit on D30 survival. These results underscore the need to perform further studies to define the optimal conditions for using epinephrine in patients with OHCA.