Persistent critical illness in acute hypoxic respiratory failure: A retrospective cohort study

by Michael Kalinoski, Leah Bolden, Qi Wang, R. Adams Dudley, Nathan Mesfin, Nicholas E. Ingraham

Introduction

Persistent critical illness (PerCI) describes patients with prolonged intensive care unit (ICU) dependence after their acute phase of illness and affects 5–20% of the general ICU population. PerCI has been studied in the general ICU cohort, but prevalence and outcomes remain unclear among mechanically ventilated patients with acute hypoxic respiratory failure (AHRF), a high-risk cohort that is more prone to ICU-associated complications. The objective of this study is to quantify PerCI prevalence, identify risk factors for PerCI development, and describe one-year mortality in this population.

Methods

We conducted a retrospective study of patients with AHRF (PaO2/FiO2 ratio < 300) requiring mechanical ventilation within 72 hours of ICU admission across a 12-hospital health system (2011–2022). PerCI was defined as ICU length of stay ≥ 10 days. The primary outcome was one-year mortality, modeled using a Cox proportional hazards model with PerCI as a time-varying exposure. To identify risk factors for PerCI while accounting for competing early mortality, we used a dual-model approach: logistic regression comparing PerCI to early death (≤ 10-day mortality) and PerCI to early ICU discharge, with factors associated with both models considered robust predictors.

Results

Among 10,626 eligible patients, 3,107 (29.2%) met criteria for PerCI. One-year survival was 64.2% in the PerCI group compared to 73.8% in the non-PerCI group. After PerCI onset, mortality risk increased threefold (hazard ratio 3.01; 95% CI 2.72–3.33). Factors consistently associated with PerCI across both models included ARDS, pneumonia, aspiration pneumonitis, heart failure, septic shock, weight loss, surgical admission, and postprocedural respiratory and circulatory failure.

Conclusion

Nearly one in three mechanically ventilated AHRF patients developed PerCI, far exceeding rates in the general ICU populations and was associated with worse one-year survival. Identifiable clinical risk factors may support earlier goals-of-care conversations and post-ICU care planning in this high-risk population.