Association of dexmedetomidine with short-term outcome in patients with cardiogenic shock: a retrospective propensity score-matched cohort study from MIMIC-IV

Background Dexmedetomidine has been demonstrated to have cardioprotective effects in previous studies, prompting our investigation into its potential to improve survival outcomes in patients with cardiogenic shock (CS). Methods This retrospective cohort study analyzed data from the Medical Information Mart for Intensive Care (MIMIC) IV database, focusing on patients with CS. Exposure was defined as intravenous dexmedetomidine administration during intensive care unit (ICU) stay. The primary endpoints included 7-day and 30-day all-cause mortality. External validation was conducted using the eICU 2.0 database. Results The pre-matched and propensity score matched cohorts comprised 2,341 and 1,038 patients, respectively. Multivariable Cox regression analysis of the overall cohort revealed that dexmedetomidine administration was significantly associated with reduced risk of both 7-day (hazard ratio (HR) = 0.473, 95% confidence interval (CI): 0.359–0.624, p < 0.001) and 30-day all-cause mortality (HR = 0.606, 95% CI: 0.500–0.735, p < 0.001). This protective association persisted after propensity score matching (PSM) for 7-day (HR = 0.418, 95% CI: 0.317–0.552, p < 0.001) and 30-day mortality (HR = 0.579, 95% CI: 0.475–0.705, p < 0.001). Subgroup analyses demonstrated that patients older than 75 years, those with chronic pulmonary disease, or those with lower systolic blood pressure may not benefit from dexmedetomidine. External validation using 1411 CS patients from the eICU 2.0 database confirmed these findings, with PSM-adjusted analyses showing reduced in-hospital (HR = 0.597; 95% CI: 0.395–0.901; p = 0.014) and in-ICU mortality (HR = 0.425; 95% CI: 0.262–0.689; p < 0.001) among dexmedetomidine treated patients. Conclusion Dexmedetomidine administration was associated with reduced risk of 7-day and 30-day all-cause mortality in CS patients, though this protective effect may not be significant in patients over 75 years, those with chronic pulmonary disease, or those with lower systolic blood pressure. Prospective studies are required to validate these findings.