Midazolam versus morphine in acute cardiogenic pulmonary edema patients with and without atrial fibrillation: findings from the MIMO trial

Background and importance The MIMO clinical trial showed that patients with acute cardiogenic pulmonary edema (ACPE) treated with midazolam had fewer serious adverse events than those treated with morphine. Atrial fibrillation (AF) is a common comorbidity in heart failure and affects patient’s outcome. Objective The primary endpoint of this substudy is to know if AF modified the reduced risk of serious adverse events in the midazolam arm compared to morphine. The first secondary endpoint is to know if AF modified the reduced risk of serious adverse events or death at 30 days in the midazolam arm. The second secondary objective of this substudy is to analyze whether AF modified the reduced risk of midazolam against morphine on the total number of serious adverse events per patient. Design We conducted a secondary analysis of the MIMO trial. Patients more than 18 years old clinically diagnosed with ACPE and with dyspnea and anxiety were randomized (1:1) at emergency department arrival to receive either intravenous midazolam or morphine. Outcome measures and analysis In this post hoc analysis, we calculated the relative risk (RR) of serious adverse events in patients with and without AF. Calculating the Cochran-Mantel-Haenszel interaction test, we evaluated if AF modified the reduced risk of serious adverse events in the midazolam arm compared to morphine. Main results One hundred eleven patients (median = 78.9 years; IQR, 72.3–83.7; women, 52.2%) were randomized in the MIMO trial, 55 to receive midazolam and 56 to morphine. All randomized patients received the assigned drug and there were no losses to follow-up. Forty-four patients (39.6%) had AF. In the AF group, the RR for the incidence of serious adverse events in the midazolam versus morphine arm was 0.42 (95% CI, 0.14–1.3). In the group without AF, the RR was 0.46 (95% CI, 0.21–1). The presence of AF did not modify the reduced risk of serious adverse events in the midazolam arm compared with morphine (P for interaction = 0.88). Conclusion This post hoc analysis of the MIMO trial suggests that the reduced risk of serious adverse events in the midazolam group compared to morphine is similar in patients with and without AF.