Impact of time of intensive care unit transfer and outcomes in patients with septic shock: An observational study

Abstract

AbstractObjectiveTo evaluate the association between time from ED presentation to intensive care unit (ICU) transfer on mortality in patients presenting with septic shock.MethodsAdult patients with suspected septic shock enrolled in the Australasian Resuscitation in Sepsis Evaluation trial were included. The primary outcome of this post‐hoc analysis was 90‐day mortality. ED‐to‐ICU time was analysed as both a continuous variable and a binary variable (≤vs>4 h). Analysis incorporated mixed effects regression, with ICU site as a random effect, time‐to‐event analysis and competing risks regression; all with and without inverse probability of treatment weighting to account for confounding baseline covariates.ResultsData from 1301 patients were included. Median (interquartile range [IQR]) ED‐to‐ICU time was 4.3 (3.1, 6.3) hours, with 588 patients (45%) transferred within 4 h. The ≤4‐h group were younger, 64 (51, 74)versus67 (52, 76) years (P = 0.04), with higher APACHE III scores, 50 (37, 65)versus47 (35, 62) (P = 0.002), and higher unadjusted 90‐day mortality, odds ratio (OR) 1.53 (95% confidence interval 1.15, 2.03),P = 0.01. After adjustment for pre‐specified confounders, the 90‐day mortality OR was 1.09 (0.83, 1.44),P = 0.52. Adjusted for death as a competing event and illness severity, hospital length of stay was similar between groups, whereas ICU duration remained longer for the ≤4‐h group.ConclusionIn patients presenting to the ED with septic shock, ED‐to‐ICU time less than 4 h was not associated with altered 90‐day mortality, although this should be interpreted with caution due to study limitations.