Abstract
BACKGROUND: In the intensive care unit (ICU), extubation failure has been associated with greater resource utilization and worsened clinical outcomes. Most recently, nighttime extubation (NTE) has been reported as a risk factor for increased ICU and hospital mortality. We hypothesized that, in a large, urban, university-affiliated hospital with multidisciplinary assessment for extubation, rigorously protocolized extubation algorithms, and expert airway managers available at all times of day for assessment of high-risk extubations, NTE would not confer additional risk of adverse clinical outcomes. METHODS: This was a retrospective cohort study of mechanically ventilated adults at a single university-affiliated hospital. NTE was defined as occurring between 7:00 pm and 6:59 am the following day. All data were extracted from the institution's electronic medical record. Multivariable regression analyses were used to assess associations between NTE and reintubation, ICU and hospital length of stay (LOS), and mortality with adjustments for demographic and clinical covariates defined a priori. Palliative, unplanned, and routine postoperative extubations were excluded in sensitivity analyses. RESULTS: Of 2241 patients, 204 of 2241 (9.1%) underwent NTE. The rates of reintubation (NTE 6.9% versus daytime extubation [DTE] 12.4%; adjusted odds ratio [95% confidence interval {CI}], 0.78 [0.43-1.41]; P =.41) and in-hospital mortality (NTE 3.4% versus DTE 5.9%; adjusted odds ratio [95% CI], 0.72 [0.28-1.84]; P =.49) were not found to differ. NTE, compared to DTE, was associated with shorter duration of mechanical ventilation (median [interquartile range], 1 [0-1] days vs 2 [1-4] days; adjusted ratio of geometric means [RGMs] [95% CI], 0.64 [0.54-0.70]; P <.001), ICU (2 [1-5] days vs 4 [2-10] days; adjusted RGMs [95% CI], 0.65 [0.57-0.75]; P <.001), and hospital LOS (6 [3-18] days vs 13 [6-25] days; adjusted RGMs [95% CI], 0.64 [0.56-0.74]; P <.001). These results were unchanged in sensitivity analyses. CONCLUSIONS: Patients who underwent NTE were not at increased risk of reintubation or in-hospital mortality. In addition, NTE was associated with a shortened duration of mechanical ventilation and hospital LOS. In health care systems with similar critical care delivery models, NTE may coincide with reduced resource utilization in appropriately selected patients. © 2020 Royal Society of Chemistry. All rights reserved.