Long-term survival of patients with persistent critical illness in Australia and New Zealand; a retrospective registry-based cohort study

Objective To investigate whether increasing intensive care unit (ICU) length of stay (LOS) is associated with reduced long-term survival in patients with persistent critical illness (PerCI) (1,2) across Australian and New Zealand ICUs. Design Population-based retrospective cohort study. Setting All Australian and New Zealand ICUs reporting to the Australia and New Zealand Intensive Care Society’s Adult Patient Database. Participants Adult patients (age >16) with an index admission to ICU from July 2019 to June 2024. Main outcome measures All-cause mortality at 12, 24, and 60 months following ICU discharge in patients with PerCI. PerCI was defined as ICU length of stay (LOS) ≥10 d and was compared across five prespecified LOS categories: 10–13 d (reference group), 14–20, 21–29, 30–89, and ≥90 d. Results Of 985,594 ICU admissions, 35,573 (3.6%) met the PerCI criterion. Survival from ICU discharge was ∼70% at 12 m and varied between 47.5% and 61.4% at 60 m across LOS groups. After adjustment for illness severity, age, frailty, and comorbidity, LOS was not independently associated with survival in the ICU length of stay (ICU LOS) 21–29- and 30–89-day groups when compared with the 10–13-day group; however, the 14–20-day group (hazard ratio [HR] 1.06, 95% confidence interval [CI]:1.02–1.11, p = 0.005) and ≥90-d group (HR 1.31, 95% CI: 1.06–1.61, p = 0.011) had higher hazard of death. The strongest predictors of shorter survival were chronic liver disease (HR 1.76, 95% CI: 1.62–1.92, p < 0.001), frailty (HR 1.45, 95% CI: 1.38–1.51, p < 0.001), and dialysis-dependent renal failure (HR 1.34, 95% CI: 1.23–1.45, p < 0.001) Conclusions Most patients with PerCI in Australia and New Zealand had favourable long-term survival. After adjustment, the ≥90-day group emerged as a distinct exception, with a significantly higher hazard of death and five-year survival below 50%, despite its small size (N = 226). In contrast, ICU length of stay (ICU LOS) was not independently associated with survival across the 10–89-day range, indicating that for most patients, outcome is determined by underlying characteristics rather than time in the ICU. Future work should examine functional outcomes, frailty trajectory, quality of life, out of hospital resource use, and the characteristics distinguishing survivors from nonsurvivors.