Outcomes Associated with Sustained Refractory Respiratory Failure After Recovery from Severe Cardiogenic Shock

P. Rothenberg¹, M. Platten¹, H. Rajjoub¹, N. Teman², B. Jalil¹, P. McCarthy¹, P. Sappington¹, V. Badhwar¹, A. Hayanga^1
1West Virginia University, Morgantown, West Virginia ²University of Virginia, Charlottesville, Virginia

Purpose: We compared outcomes in patients with persistent respiratory failure post-cardiogenic shock who required venovenous (VV) extracorporeal membrane oxygenation after venoarterial (VA) ECMO to patients requiring VA or VV ECMO only and hypothesized that VV ECMO may be a useful therapeutic option in sustained respiratory failure after cardiac recovery.

Methods: We used a multi-institutional ECMO registry to evaluate patients who developed sustained respiratory failure after recovery from cardiogenic shock and required conversion from VA to VV ECMO. We compared these outcomes to patients who were managed using single modality extracorporeal care, either VA or VV ECMO only. Outcomes of interest included length of stay, duration of ECMO, patient-reported outcomes, survival, and death. Chi-square test was used for categorical variables. T-test and one-way ANOVA were used for continuous variables. Propensity score analysis was performed using age, gender, race, BMI, and insurance as covariates. Inverse probability weighting was used for further adjustment.

Results: Of the total of 403 patients who were analyzed, 162 were supported using VA ECMO only while 218 patients were supported using VV ECMO only. Twenty-three patients were converted from VA to VV ECMO. Most patients were white (87.8%), male (62.5%) with median age of 49.5 ± 15.2 years, and BMI of 33.7 ±18.1 kg/m2. (Table) The indications for extracorporeal support in the conversion cohort included septic shock (n=5), cardiogenic shock (n=11), ARDS (n=3), respiratory failure (n=3), pulmonary embolus (n=3). Sequential organ failure assessment scores were higher in the conversion group (12.8 ± 3.26 vs 8.4± 3.26). Intensive care unit length of stay (LOS) was shorter in the VA group (14.1 vs. 20 days, p=0.02) and hospital LOS was shorter (17.8 vs. 27.7 days, p=0.003). The VV cohort had similar ICU LOS (20.8 days, p=0.952) and hospital LOS (27.4 days, p=0.996) to the conversion cohort. Both VA only and VV only cohorts had higher odds of death (VA: OR 3.39, p< 0.001; VV: OR 2.11, p< 0.001). (Figure) Duration of extracorporeal support was longer for VV only cohort (18 days, p< 0.001). Quality adjusted life years were higher in the conversion cohort (0.927 vs. VA: 0.815, p=0.008; VV: 0.849, p=0.008).

Conclusion: Patients requiring conversion from VA to VV ECMO had higher illness severity, but higher likelihood of survival compared to patients cannulated with single modality ECMO. Patient-reported outcomes were also better. Conversion from VA to VV ECMO may be a viable strategy in sustained refractory respiratory failure after cardiac recovery.

Identify the source of the funding for this research project:

Funding: Supported by NIH NHLBI # 2UM1 HL088925 12