(1392) Late Reintubation Is a Distinct Subtype of Post-Procedural Respiratory Failure

Introduction: Post-procedural respiratory failure (PRF) is a major healthcare complication. Most definitions include prolonged intubation ≥24-hours (PI) and early < 7-day reintubation or non-invasive support (earlyRIRS). Several also include late ≥7-30 day reintubation and respiratory support (lateRIRS). There are knowledge gaps about differences between early versus late reintubation and respiratory support.

Methods: This was a retrospective cohort study of adult patients who underwent major invasive procedures between 2016–2023 in Bronx, NY. Exclusion criteria: (1) ASA-6; (2) subsequent procedure before 30-day reintubation or adverse discharge; (3) pre-existing respiratory failure or tracheostomy, and (4) DNI status. Primary outcome: adverse discharge defined as inpatient death or discharge to a nursing home. Study exposure: earlyRIRS, lateRIRS, or PI. Univariate association were tested using Chi-square analysis. Change in incidence was tested with univariable logistic regression. Association of all 3 exposures with the outcome was tested using multivariable logistic regression adjusted for confounders in a contemporary 2021-2023 data subset.

Results: 404,288 cases met study criteria. There was 5% annual decrease in incidence of PI (RR 0.95, 95% CI: 0.93-0.96; 0.4% in 2023). Similar incidence and change was noted for earlyRIRS (RR 0.97, 95% CI: 0.95-0.98; 0.4% in 2023). In contrast, there was no change in lateRIRS (2023 incidence 0.3%). Incidence of adverse discharge in 2023 was similar in all types of PRF (30-35%; P=0.21). Compared with earlyRIRS, lateRIRS was associated with less hypoxemia and more circulatory shock (lateRIRS: 28.9% and 58.6%, earlyRIRS: 55.6% and 36.7%; P< 0.001). Moreover, lateRIRS patients were older, more chronically sick, and more socially vulnerable. Patients with lateRIRS were also were more likely to have simple peripheral procedures (P < 0.05 for all). Compared to earlyRIRS, lateRIRS was associated with lower risk of adverse discharge (aRR 0.68, 95% CI: 0.51-0.92) whereas PI showed equal risk (P=0.219).

Conclusions: Early reintubation < 7 days after a major invasive procedure has similar morbidity and pathophysiology as prolonged intubation ≥24-hours. In contrast, late reintubation ≥7-days is a distinct complication with a lower mortality.