Hospital-level volume in extracorporeal membrane oxygenation cases and death or disability at 6 months

Ertugrul, A. D., Serpa Neto, A., Fulcher, B. J., Charles-Nelson, A., Bailey, M., Burrell, A. J. C., Anderson, S., Bernard, S., Board, J. V., Brodie, D., Buhr, H., Cooper, D. J., Dicker, C., Fan, E., Fraser, J. F., Gattas, D. J., Hopper, I. K., Huckson, S., Linke, N. J., Litton, E., McGuinness, S. P., Nair, P., Orford, N., Parke, R. L., Pellegrino, V. A., Pilcher, D. V., Stub, D., Udy, A. A., Reddi, B. A. J., Trapani, T. V., Jones, A., Higgins, A. M., Hodgson, C. L. Critical Care and Resuscitation DOI: 10.1016/j.ccrj.2024.08.006

Abstract:

Objective: Extracorporeal membrane oxygenation (ECMO) is a high-risk procedure with significant morbidity and mortality and there is an uncertain volume-outcome relationship, especially regarding long-term functional outcomes. The aim of this study was to examine the association between ECMO centre volume and long-term death and disability outcomes. Design, setting, and participants: This is a registry-embedded observational cohort study. Patients were included if they were enrolled in the binational ECMO registry (EXCEL). The exclusion criteria included patients on ECMO for heart/lung transplants. Data included demographics, clinical information on their first ECMO run, and six-month outcomes obtained by telephone interview. The primary outcome was death or new disability at six months. A multivariable analysis was conducted using hospitals' annual ECMO volume. High-volume centres were defined as having >30 ECMO cases annually, and analyses were run on ECMO subgroups of veno-venous (VV), veno-arterial (VA), and extracorporeal cardiopulmonary resuscitation (ECPR). Results: Of 1232 patients, 663 patients were cared for on ECMO at high-volume centres and 569 patients at low-volume centres. There was no difference in six-month death or new disability between high- and low-volume ECMO centres in VV-ECMO [OR: 1.09 (0.65e1.83), p ¼ 0.744], VA-ECMO [OR: 1.10 (0.66 e1.84), p ¼ 0.708], and ECPR-ECMO [OR: 1.38 (0.37e5.08), p ¼ 0.629]. This finding was persistent in all sensitivity analyses, including exclusion of patients who were transferred between high- and lowvolume centres. Conclusion: There was no difference in death or disability at six months between high- and low-volume centres in Australia and New Zealand, possibly due to the current model of coordinated care that includes patient transfers and training between high- and low-volume ECMO centres in our region.