Effect of Time of Intraoperative Circulatory Support on Incidence of High-Grade Primary Graft Dysfunction (PGD): Multicenter Analysis on Use of Extracorporeal Life Support (ECLS) during Lung Transplantation
PURPOSE: Prior single center experiences suggest that use of ECLS during lung transplantation (LT) increases the risk of PGD. Here we analyze the effects of intraoperative ECLS time on PGD after LT in a multicenter ECLS registry. METHODS: The ECLS in LT registry includes data on bilateral LTs from 8 centers (2 from Europe, 6 from United States). The registry excludes single lung transplants and multi-organ transplants and defines primary graft dysfunction according to the ISHLT 2016 consensus statement. Herein, we refer to PGD as a single case developing PGD3 at time 48 and/or 72 hours. We explored the risk of intraoperative ECLS time on the incidence of PGD and death within 90 days using adjusted and unadjusted logistic regression and receiver operating curve (ROC) analysis. RESULTS: We identified 297 ECLS patients with intraoperative support times available between January 2016 and August of 2019. Breakdown of cases were as follows: 44% CPB, 66% ECMO (16% VV, 64%VA/VVA, 20% modified bypass-VA). Out of these, 35.6% developed PGD and 6% died within 90 days. Median time of ECLS support was 206 minutes (62-686 minutes). The median time of support was 200 minutes in patients without PGD and 206 minutes in cases with PGD. The median time of support was 202 minutes in patients who survived 90 days compared to 271 minutes in patients who did not survive 90 days. Our unadjusted and adjusted logistic regression analysis showed that time of ECLS was not associated with PGD irrespective of the mode of ECLS (CPB v ECMO). In addition, ROC analysis showed an area under the curve for ECLS time of 0.51 suggesting poor predictability for PGD. Our unadjusted analysis showed that increased time on ECLS was significantly associated with death within 90 days (P=0.014) although this was of borderline significance after adjusting for risk factors (P=0.067). Furthermore, this effect was of borderline significance for CPB (P=0.07) and not significant for ECMO (P=0.17). CONCLUSION: In an international multicenter registry on use of ECLS in LT, the time on ECLS did not alter the risk of PGD with either CPB or ECMO, although it was associated with perioperative death. Surgeons should limit the time spent on ECLS when feasible and consider ECMO for longer procedures.
Duke Scholars
Published In
DOI
EISSN
Publication Date
Volume
Issue
Start / End Page
Related Subject Headings
- Surgery
- 3202 Clinical sciences
- 3201 Cardiovascular medicine and haematology
- 1102 Cardiorespiratory Medicine and Haematology
Citation
Published In
DOI
EISSN
Publication Date
Volume
Issue
Start / End Page
Related Subject Headings
- Surgery
- 3202 Clinical sciences
- 3201 Cardiovascular medicine and haematology
- 1102 Cardiorespiratory Medicine and Haematology