Association between vasoactive drug use and 28-day in-hospital mortality in patients with septic shock in the intensive care unit: A retrospective study based on the MIMIC-IV database.

Septic shock, a life-threatening intensive care unit condition characterized by persistent hypotension and organ dysfunction despite fluid resuscitation, is a leading cause of intensive care unit mortality. Vasoactive drugs are central to hemodynamic support in septic shock but their benefit-potential adverse event trade-off remains controversial; existing studies link high-dose vasopressors to increased mortality, though the association with 28-day mortality is unclear due to limitations like small sample sizes or single-center designs. This retrospective cohort study, based on the Medical Information Mart for Intensive Care-IV database (V3.1), included 2961 adult patients with septic shock who received vasoactive drugs and had hyperlactatemia between 2008 and 2022. The association between total vasoactive drug exposure and mortality was investigated using quartile grouping, Kaplan-Meier (K-M) survival analysis, restricted cubic spline regression, and subgroup analysis. Multivariable Cox proportional hazards models with incremental adjustments for demographics, comorbidities and laboratory markers were applied to evaluate hazard ratios. The overall in-hospital 28-day mortality rate was 31.37%. Mortality significantly increased across quartiles of vasoactive drug exposure, rising from 15.26% to 43.67% (χ² = 152.73, P < .001). Multivariable Cox models demonstrated consistently elevated mortality risks with higher exposure quartiles, showing significant dose-response trends. Restricted cubic spline analysis showed a nonlinear association between vasoactive drug exposure and 28-day mortality, with a threshold of 470.98 units: below it, lower exposure raised risk; above it, higher exposure reduced risk. Subgroup analyses found this association was more pronounced in males, patients ≤ 60 years old, those without cirrhosis, chronic kidney disease, cancer, and those with comorbid heart failure. The exposure dose of vasoactive drugs in patients with septic shock exhibits a non-linear association with 28-day mortality, with significant heterogeneity in risk characteristics across different dose ranges.