PS52. Association of Intraoperative ECMO Cannulation Strategy with Primary Graft Dysfunction and Vascular Complications after Lung Transplantation
Presented During: Thoracic Poster Viewing Showcase
Baylor College of Medicine
Houston, TX
United States - Contact Me
Ethan D’Silva is a fourth-year medical student at Baylor College of Medicine. His interest in cardiothoracic surgery includes ex-vivo lung perfusion and mechanical circulatory support. His other academic interests include surgical education and applying artificial intelligence, ranging from machine learning to computer vision, to surgical care.
Saturday, May 6, 2023: 8:00 AM - Tuesday, May 9, 2023: 11:45 AM
Los Angeles Convention Center
Room: Outside of Room 408
Description
Objective: A recent report from the Extracorporeal Life Support in Lung Transplant (ECLS) Registry showed that lung transplantation (LT) using extracorporeal membrane oxygenation (ECMO) was associated with a greater incidence of primary graft dysfunction (PGD) compared to LT performed off-pump. We examined the association between cannulation strategy and the incidence of both PGD and cannulation-related vascular complications.
Methods: The ECLS registry includes entries from 7 US and 2 European centers each performing >40 LT per year. We retrospectively reviewed this registry to identify our study cohort, which included double LT performed on patients >13 years of age using intraoperative veno-arterial (VA) ECMO between January 1, 2016 and May 31, 2022. We excluded entries with multi-organ transplants and preoperative ECMO use. Our primary outcome was PGD defined as Grade 3 PGD at 48-72 hours after LT. Secondary outcomes included incidence of aortic dissection, femoral artery dissection, limb ischemia requiring surgical intervention, and stroke. We divided the study cohort into two groups based on intraoperative ECMO cannulation strategy: (1) central, defined as using aortic arterial inflow, and (2) peripheral, defined as using a peripheral arterial inflow. To evaluate the association between cannulation strategy and PGD, we used both a multivariate regression model, as well as optimal pair propensity score matching (PSM). We compared secondary outcomes using chi-square and Fisher's exact test.
Results: We identified 201 patients in the central group and 111 patients in the peripheral group. The incidence of PGD was 41 (20.4%) in the central group and 42 (37.8%) in the peripheral group (P<0.001). After adjusting for clinical differences (Table 1) using a multivariate regression model, the probability of developing PGD in the central group was 13% lower than in the peripheral group [odds ratio 0.87 (95% CI: 0.78-0.98, P=0.033)]. PSM analysis was consistent showing PGD incidence of 19 (20.7%) in the central group compared to 35 (38.0%) in the peripheral group (P=0.009). The peripheral group showed significantly greater incidence of femoral artery dissection and limb ischemia. There was no significant difference in aortic dissection and stroke incidence (Table 1).
Conclusions: This analysis of the ECLS registry suggests that a central aortic cannulation strategy may be preferred over peripheral arterial cannulation when VA ECMO is used during LT.
Methods: The ECLS registry includes entries from 7 US and 2 European centers each performing >40 LT per year. We retrospectively reviewed this registry to identify our study cohort, which included double LT performed on patients >13 years of age using intraoperative veno-arterial (VA) ECMO between January 1, 2016 and May 31, 2022. We excluded entries with multi-organ transplants and preoperative ECMO use. Our primary outcome was PGD defined as Grade 3 PGD at 48-72 hours after LT. Secondary outcomes included incidence of aortic dissection, femoral artery dissection, limb ischemia requiring surgical intervention, and stroke. We divided the study cohort into two groups based on intraoperative ECMO cannulation strategy: (1) central, defined as using aortic arterial inflow, and (2) peripheral, defined as using a peripheral arterial inflow. To evaluate the association between cannulation strategy and PGD, we used both a multivariate regression model, as well as optimal pair propensity score matching (PSM). We compared secondary outcomes using chi-square and Fisher's exact test.
Results: We identified 201 patients in the central group and 111 patients in the peripheral group. The incidence of PGD was 41 (20.4%) in the central group and 42 (37.8%) in the peripheral group (P<0.001). After adjusting for clinical differences (Table 1) using a multivariate regression model, the probability of developing PGD in the central group was 13% lower than in the peripheral group [odds ratio 0.87 (95% CI: 0.78-0.98, P=0.033)]. PSM analysis was consistent showing PGD incidence of 19 (20.7%) in the central group compared to 35 (38.0%) in the peripheral group (P=0.009). The peripheral group showed significantly greater incidence of femoral artery dissection and limb ischemia. There was no significant difference in aortic dissection and stroke incidence (Table 1).
Conclusions: This analysis of the ECLS registry suggests that a central aortic cannulation strategy may be preferred over peripheral arterial cannulation when VA ECMO is used during LT.