Early Vs Delayed Tracheostomy In Mechanically Ventilated Patients With Copd

SESSION TITLE: Phenotyping and Management of COPD SESSION TYPE: Original Investigations PRESENTED ON: October 18-21, 2020 PURPOSE: Acute respiratory failure due to the acute exacerbation of chronic obstructive pulmonary disease (COPD) may need intensive care unit (ICU) admission and invasive ventilation that places COPD patients at a higher risk of prolonged ventilation. Tracheostomy procedures are widely used in the ICU to help wean patients who need prolonged ventilation. The optimal timing for tracheostomy remains controversial. We examined whether early tracheostomy in mechanically ventilated patients who have COPD was associated with lower 90-day all-cause in-hospital mortality. METHODS: Retrospective analysis of Health Facts, an electronic health record database, to study 72585 COPD patients who were hospitalized and mechanically ventilated between January 1, 2009, and June 29, 2018. The cohort was further limited to those receiving tracheostomy. Receiver operating characteristic (ROC) curve was used to define the most sensitive cut-off for the timing of tracheostomy for predicting 90-day mortality. The threshold was used to create the early and late tracheostomy groups. Multivariate logistic regression models, Generalized Linear Model, and Chi-square test were used to compare outcomes among patients who received an early tracheostomy and late tracheostomy. RESULTS: A total of 1728 patients met the inclusion criteria. The most sensitive cut-off for timing of tracheostomy was the 9th day of mechanical ventilation, which was the threshold used to create the early and late tracheostomy groups; 617 received an early tracheostomy and 1111 received a late tracheostomy. 90-day all-cause in-hospital mortality was significantly lower in the early group compared with a late group (15.24% vs 22.41%; Odds ratio, 0.632; 95% CI, 0.485-0.823) (P= 0.0006). Tracheostomy related complications were significantly higher in the early group (6.16% vs 4.5%; Odds ratio, 1.62; 95% CI, 1.041- 2.523) (P=0.0326). The risk of aspiration pneumonitis was not statistically different between both groups (Odds ratio, 0.771; 95% CI, 0.588-1.01) (P=0.059). The hospital length of stay was significantly lower in the early group compared with a late group (Median, 19 vs 29; Odds ratio, 0.709; 95% CI, 0.669-0.743) (P=<0.001). CONCLUSIONS: Early tracheostomy, before the 9th day, in mechanically ventilated COPD patients was associated with lower 90-day all-cause in-hospital mortality and hospital length of stay. However, they had a higher risk for tracheostomy complications. Further studies are required to explore factors that help clinicians predict which patients need prolonged ventilation. CLINICAL IMPLICATIONS: This study will advance knowledge regarding this controversial topic and provide information about the usual care provided in U.S hospitals for this subgroup of patients. The study results may guide physicians in making care decisions. DISCLOSURES: No relevant relationships by Julie Banderas, source=Web Response No relevant relationships by An-Lin Cheng, source=Web Response No relevant relationships by YAHIA MOHAMED, source=Web Response No relevant relationships by suman sahil, source=Web Response No relevant relationships by Gary Salzman, source=Web Response