Use of Early Ketamine Sedation and Association With Clinical and Cost Outcomes Among Mechanically Ventilated Patients With COVID-19: A Retrospective Cohort Study

Abstract

OBJECTIVES: To describe the utilization of early ketamine use among patients mechanically ventilated for COVID-19, and examine associations with in-hospital mortality and other clinical outcomes. DESIGN: Retrospective cohort study. SETTING: Six hundred ten hospitals contributing data to the Premier Healthcare Database between April 2020 and June 2021. PATIENTS: Adults with COVID-19 and greater than or equal to 2 consecutive days of mechanical ventilation within 5 days of hospitalization. INTERVENTION: The exposures were early ketamine use initiated within 2 days of intubation and continued for greater than 1 day. MEASUREMENTS: Primary was hospital mortality. Secondary outcomes included length of stay (LOS) in the hospital and ICUs, ventilator days, vasopressor days, renal replacement therapy (RRT), and total hospital cost. The propensity score matching analysis was used to adjust for confounders. MAIN RESULTS: Among 42,954 patients, 1,423 (3.3%) were exposed to early ketamine use. After propensity score matching including 1,390 patients in each group, recipients of ketamine infusions were associated with higher hospital mortality (52.5% vs. 45.9%, risk ratio: 1.14, [1.06–1.23]), longer median ICU stay (13 vs. 12 d, mean ratio [MR]: 1.15 [1.08–1.23]), and longer ventilator days (12 vs. 11 d, MR: 1.19 [1.12–1.27]). There were no associations for hospital LOS (17 [10–27] vs. 17 [9–28], MR: 1.05 [0.99–1.12]), vasopressor days (4 vs. 4, MR: 1.04 [0.95–1.14]), and RRT (22.9% vs. 21.7%, RR: 1.05 [0.92–1.21]). Total hospital cost was higher (median $72,481 vs. $65,584, MR: 1.11 [1.05–1.19]). CONCLUSIONS: In a diverse sample of U.S. hospitals, about one in 30 patients mechanically ventilated with COVID-19 received ketamine infusions. Early ketamine may have an association with higher hospital mortality, increased total cost, ICU stay, and ventilator days, but no associations for hospital LOS, vasopressor days, and RRT. However, confounding by the severity of illness might occur due to higher extracorporeal membrane oxygenation and RRT use in the ketamine group. Further randomized trials are needed to better understand the role of ketamine infusions in the management of critically ill patients.