ECMO support may be associated with improved survival in tuberculosis associated severe ARDS

BackgroundData describing outcome of extracorporeal membrane oxygenation (ECMO) support in Tuberculosis (Tbc)-associated acute respiratory distress syndrome (ARDS) remain sparce and are mostly confined to singular case reports. The aim of this case series was to analyze intensive care unit (ICU) survival in patients with Tbc-associated ARDS receiving veno-venous (vv-) ECMO support and to compare those to patients not receiving ECMO.Case presentationICU survival was analyzed retrospectively in 14 patients treated for Tbc-associated ARDS at three ECMO-referral university hospitals (Hannover Medical School, University Hospital Bonn (both Germany) and University Hospital Zurich (Switzerland)) during the last 14 years, of which eight patients received additional vv-ECMO support and six standard care only.ICU survival was significantly higher in patients receiving additional vv-ECMO support (62.5%, n = 5/8) compared to those that did not (16.7%, n = 1/6) (p = 0.021). ECMO support was associated with reduced ICU mortality (Hazard ratio adjusted for baseline SOFA score [adj. HR] 0.125 (95% confidence interval (CI): 0.023-0.689), p = 0.017). Median (IQR) time on ECMO and invasive ventilation in the vv-ECMO group were 20 (11-26) and 37 (27-53) days, respectively. Major bleeding defined as transfusion requirement of 4 units of blood or more or surgical and/or radiologic intervention occurred only in one patient, in whom pulmonary bleeding was fatal. Thromboembolic events occurred in none of the vv-ECMO patients.Case presentationICU survival was analyzed retrospectively in 14 patients treated for Tbc-associated ARDS at three ECMO-referral university hospitals (Hannover Medical School, University Hospital Bonn (both Germany) and University Hospital Zurich (Switzerland)) during the last 14 years, of which eight patients received additional vv-ECMO support and six standard care only.ICU survival was significantly higher in patients receiving additional vv-ECMO support (62.5%, n = 5/8) compared to those that did not (16.7%, n = 1/6) (p = 0.021). ECMO support was associated with reduced ICU mortality (Hazard ratio adjusted for baseline SOFA score [adj. HR] 0.125 (95% confidence interval (CI): 0.023-0.689), p = 0.017). Median (IQR) time on ECMO and invasive ventilation in the vv-ECMO group were 20 (11-26) and 37 (27-53) days, respectively. Major bleeding defined as transfusion requirement of 4 units of blood or more or surgical and/or radiologic intervention occurred only in one patient, in whom pulmonary bleeding was fatal. Thromboembolic events occurred in none of the vv-ECMO patients.Discussion and conclusionsThis retrospective analysis from three large ECMO centers with similar SOPs suggests vv-ECMO support as a feasible approach in patients with severe Tbc-associated ARDS. Although affiliated with extended runtimes, vv-ECMO might be associated with improved survival in those patients. Vv-ECMO support should thus be considered in Tbc-associated ARDS to enable lung protective strategies during prolonged lung recovery.