Assessing the Influence of Propofol Sedation Versus Propofol-Free Sedation on Recovery From Sepsis and Septic Shock in Mechanically Ventilated ICU Patients: A Retrospective, Propensity-Score Matched, Cohort Study.

BACKGROUND: Propofol-mediated immunosuppression has been postulated in preclinical research, yet translational clinical evidence is lacking. We, therefore, retrospectively evaluated the influence, if any, of propofol versus non-propofol sedation on critical care requirements, prognosis, and risk of re-infection among mechanically ventilated patients with sepsis or septic shock.

METHODS: This single-center, retrospective cohort analysis included adult patients admitted to any ICU at our institution from July 2017 through December 2022 with a diagnosis of sepsis or septic shock who required mechanical ventilation with sedation. Patients who received propofol sedation with or without other means of non-propofol sedation were compared to patients who received non-propofol sedation only. Univariate analysis was used to compare the primary outcome, ICU-free days, and secondary outcomes including intubation and mechanical ventilation-free days, infusion vasopressor support, and antimicrobial medications. Incidence of combined or hospital 30-day mortality and sepsis recurrence within six months were also compared. Time to infection clearance from blood was reported, as available. Total duration and infusion rate of propofol sedation were compared between patients who expired versus survived and between patients with versus without six-month sepsis recurrence. One-to-one propensity-score matching (PSM) was applied using sepsis diagnostic status, age, sex, BMI, and chronic comorbidity burden as covariates to manage confounding effects and identify comparable subpopulations.

RESULTS: A total of 615 patients were included, comprising 511 (83.1%) propofol-sedated patients and 104 (16.9%) non-propofol-sedated patients. Prior to PSM, propofol sedation was more common among younger patients (median (IQR) years, 62 (50-72) vs. 66 (58-75), p=0.021). Patients treated for septic shock comprised most of the dataset (n=498, 81.0%). The non-propofol group received fentanyl sedation more commonly (84 (80.8%) vs. 234 (45.8%), p<0.001), though the median duration of fentanyl sedation was longer in the propofol group (1.9 (0.9-4.2) vs. 1.5 (0.4-2.8) days, p=0.012). Median ICU-free days were similar between non-propofol and propofol-sedated groups (median (IQR) days, 6.3 (2.1-14.7) vs. 6.8 (3.3-13.3), p=0.303). The proportion of patients requiring three or four pressor infusions was significantly greater in the non-propofol group (32 (30.8%) vs. 110 (21.5%), p=0.041). The combined hospital or 30-day mortality rate was also higher in the non-propofol group (45 (43.3%) vs. 165 (32.3%), p=0.031). After PSM, fentanyl sedation remained more common in the non-propofol group. Median ICU-free days remained similar between groups, but there were no longer significant differences in extent of vasopressor support required or mortality rates.

CONCLUSIONS: Propofol-containing sedation was not associated with fewer ICU-free days, poorer prognosis, or greater rate of re-infection among mechanically ventilated patients with sepsis and septic shock. Residual confounding and limited precision preclude exclusion of clinically important benefit or harm. Randomized controlled research is necessary to elucidate the clinical impacts of propofol sedation among mechanically ventilated patients with sepsis and septic shock.