Early Tracheostomy influences Length of Stay in patients requiring venovenous extracorporeal membrane oxygenation.

E. Tham¹, S. Campbell¹, P. Rothenberg¹, R. Strobel², I. Hasasna¹, J. Hunter. Mehaffey¹, A. Young³, N. Teman², P. Sappington¹, P. McCarthy¹, J. A.. Hayanga^1
1West Virginia University, Morgantown, West Virginia ²University of Virginia, Charlottesville, Virginia ³University of Virginia, Troy, Virginia

Purpose: Early tracheostomy can decrease length of stay and mortality.[1] However, there is a paucity of data regarding the benefits of early tracheostomy in patients requiring veno-venous extracorporeal membrane oxygenation (VV-ECMO). We sought to assess the outcomes of early tracheostomy in patients requiring VVECMO.

Methods: We queried a multi-institutional ECMO registry to evaluate patients requiring VV-ECMO for severe ARDS between 2017 -2021 and compared outcomes between early (ET – within 4 days of cannulation) and late tracheostomy (LT – after 4 days of cannulation) groups. Outcomes of interest included time to decannulation, extubation, discharge, survival, hospital, and ICU length of stay (LOS) in days. Other variables included in the analysis were age, gender, race, BMI, insurance type, pre-ECMO variables, comorbidities, and respiratory severity scores. T tests and chi-square tests were used in our univariate analysis for continuous and categorical variables where appropriate. Multi-state Cox proportional hazard models and generalized linear models were utilized for multivariable analysis.

Results: Our sample included 133 patients, 88 who underwent ET and 45 who underwent LT (Table). There were no differences between gender, race, insurance, creatinine, bilirubin, FiO2, SaO2, , or comorbidity scores,. . NNumber of days to tracheostomy was 1.65±1.26 days in the ET group and 13.6±10 in the LT group. Patients who underwent ET were younger (44.6±12.8 vs 49.4 ±13.1, p=0.046), had higher BMI (34.5±9.75 vs 31.7 ±6.18, p=0.046), higher RESP score (3.02±3.36 vs 0.19±3.45, p< 0.001) and higher Murray score (3.32±0.6 vs 3.02±0.5, p=0.0016). Univariate analysis demonstrated that ET patients had shorter interval to extubation (2.6± 1.6 vs 9.9±9.5, p< 0.001), and higher survival to extubation (83% vs 64.4%, p=0.03). There were no statistically significant differences in time to decannulation, discharge, or death.Multi-variable Cox proportional hazard models and generalized linear models demonstrated that ET patients had shorter duration to extubation (OR 3.39 [1.14-10.1], p=0.03), shorter time to discharge (OR 2.84 [1.24-6.47], p=0.013), shorter ICU LOS (17 [11-26.3] vs 33.2 [20.8-53.2], p=0.012) and shorter hospital LOS (22.1 [14.4-33.9] vs 37.2 [23.5-58.9], p=0.044). There was no significant difference in survival (aOR 1.14 [0.34-3.83], p=0.83) or time to decannulation (aOR 2.29 [0.8-6.5], p=0.12).

Conclusion: Extracorporeal support involves high intensity resource usage. Early tracheostomy in patients requiring VV-ECMO reduced time to extubation, discharge, hospital and ICU LOS, and may serve to decrease resource consumption.

Identify the source of the funding for this research project: Supported by NIH NHLBI # 2UM1 HL088925 12