Real-World Use of Extracorporeal CO2 Removal (ECCO2R) to Correct Respiratory Acidosis: Analysis of the Hemolung Registry

ABSTRACT

Hypercapnia and respiratory acidosis lead to increased morbidity and mortality in critically ill patients. Extracorporeal CO2 removal (ECCO2R) can rapidly correct pH and PaCO2 as a treatment for refractory, hypercapnic respiratory failure. Current clinical evidence for the benefits of ECCO2R is primarily limited to case series and single-center studies. The Hemolung (ALung Technologies, Inc., Pittsburgh, PA) is the only FDA cleared ECCO2R system and has been utilized to treat greater than 1,000 patients world-wide. The purpose of this study was to evaluate real-world evidence of the Hemolung ECCO2R system for the treatment of hypercapnic respiratory failure across a range of primary diagnoses. Methods: The Hemolung Registry was queried for patients with a baseline, pre-Hemolung pH < 7.35. Patients receiving either noninvasive or invasive ventilation were included in the analysis. Physiological benefits of Hemolung therapy were evaluated using a mixed model for repeated measures based on changes in pH and PaCO2 after 4-6 hrs and 16-35 hrs of Hemolung therapy compared to the baseline value. The model was used to calculate two-sided 95% confidence intervals and associated nominal p-values. Additional markers of clinical improvement included avoidance of intubation, survival to decannulation, and Hemolung CO2 removal rate and duration of use. Adverse events were also analyzed based on patient harm. Results: 176 Hemolung patients were included in the analysis. Multiple primary diagnoses were represented 31% ARDS, 22% COPD exacerbation, 32% COVID-19, and 15% Other. Median CO2 removal by the Hemolung during the first day of therapy was 88 mL/min and resulted in a concomitant correction of pH from a median of 7.20 to 7.35 (p<0.001) and median PaCO2 correction of 81.7 to 57.0 mmHg (p<0.001). Correction of respiratory acidosis was independent of primary diagnosis, age, and BMI. 69% (112/162) of patients survived to de-cannulation. 86% (19/22) of patients failing NIV avoided intubation. There were no unanticipated complications, and the majority of adverse events did not require medical intervention or discontinuation of Hemolung therapy. 3 deaths associated with Hemolung therapy occurred. Conclusion: These data represent the largest reported analysis of ECCO2R therapy to treat a diverse population of hypercapnic respiratory failure patients. The results demonstrate significant correction of pH and PaCO2 within the first day of Hemolung therapy without significant adverse events. Data from forthcoming RCTs will shed further light on whether these physiologic benefits translate to improved outcomes compared to current standard of care.