Abstract
IMPORTANCE: Electroconvulsive therapy (ECT) practice during the COVID-19 pandemic necessitated restriction of bag-valve mask (BVM) ventilation to minimize disease transmission through aerosolization. To reduce BVM use, many sites decreased doses of anesthesia agents, but the associations of such dose reduction with clinical and tolerability outcomes are unknown. OBJECTIVES: To investigate the association of COVID-19 pandemic modifications in ECT, including reduction in anesthetic dosage and BVM use, with the rates of clinical complications and clinical outcomes. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study included data from patients receiving an index course of ECT for any clinical indication treated with methohexital and succinylcholine between March 16, 2017, and March 15, 2023, at the Centre for Addiction and Mental Health in Toronto, Ontario, Canada. Data analysis occurred from February to December 2024. EXPOSURE: Participants were split into 2 cohorts: pre-COVID-19 (treated between March 2017 and March 2020) and COVID-19 (treated between June 2020 and March 2023). MAIN OUTCOMES AND MEASURES: The primary outcomes were BVM use and rates of complications (specifically, aspiration, bradycardia, insufficient initial sedation dosage, postictal agitation [PIA], residual weakness, and prolonged seizures). Secondary outcomes were rates of response and cognitive impairment. Multivariable logistic regression was used to assess associations of COVID-19 pandemic modifications with outcomes. RESULTS: A total of 616 individuals (median [IQR] age, 45.0 [31.0-59.0] years; 342 female [55.5%]) were included in the analysis, with 362 in the pre-COVID-19 group and 254 in the COVID-19 group. Demographic and clinical characteristics were similar. The COVID-19 group received lower median (IQR) methohexital dosages (0.82 [0.72-0.96] mg/kg vs 0.87 [0.78-1.00] mg/kg; P = .007) and succinylcholine (0.33 [0.27-0.39] mg/kg vs 0.53 [0.45-0.59] mg/kg; P < .001) than the pre-COVID-19 group. The COVID-19 group had higher odds of insufficient initial sedation dosage (odds ratio, 2.16, 95% CI, 1.09-4.38) and PIA (odds ratio, 2.81; 95% CI, 1.76-4.56), with no difference for other complications. BVM use was reduced from standard use to 13.8% (35 individuals) during COVID-19. Groups were similar in the rates of response (pre-COVID-19, 246 individuals [68.0%]; COVID-19, 181 individuals [71.3%]; P = .38) and cognitive impairment (pre-COVID-19, 75 individuals [20.7%]; COVID-19, 49 individuals [19.3%]; P = .66). CONCLUSIONS AND RELEVANCE: In this retrospective cohort study, reduction of anesthesia agents was associated with reduced BVM use, with similar clinical outcomes and tolerability other than increased rates of insufficient initial sedation dosage and PIA, suggesting that management of PIA with this anesthesia dosing strategy warrants further investigation.