Perioperative timing of surgically implanted microaxial flow pump support in high-risk coronary artery bypass grafting: a single-center Society of Thoracic Surgeons study.

BACKGROUND

Patients with ischemic cardiomyopathy undergoing coronary artery bypass grafting (CABG) represent a high-risk population when perioperative mechanical circulatory support is required. Evidence guiding the optimal timing of surgically implanted microaxial flow pump (mAFP) support in this setting remains limited. This study aimed to evaluate outcomes according to the timing of perioperative mAFP support in high-risk CABG patients.

METHODS

We conducted a single-center retrospective analysis of the Society of Thoracic Surgeons Adult Cardiac Surgery Database and electronic health records from August 2007 to September 2025. Patients undergoing isolated CABG with perioperative mAFP support were categorized as preoperative, planned intraoperative, or unplanned postoperative placement. The primary comparative analysis focused on preoperative versus planned intraoperative mAFP. Outcomes included mortality, survival, and postoperative complications. Survival was analyzed using Kaplan-Meier methods and Cox proportional hazards modeling adjusted for baseline left ventricular ejection fraction.

RESULTS

Among 34 patients, 11 received preoperative mAFP, 10 planned intraoperative mAFP, and 13 unplanned postoperative mAFP. In-hospital mortality was 0% in the preoperative group, 30% in the planned intraoperative group, and 61.5% in the unplanned postoperative group. In adjusted Cox analysis comparing preoperative versus planned intraoperative mAFP, mortality did not differ significantly [hazard ratio (HR) =2.08; 95% confidence interval (CI): 0.33-13.19; P=0.44]. Patients requiring unplanned postoperative mAFP experienced the longest intensive care unit (ICU) stays, ventilation times, and highest complication rates.

CONCLUSIONS

In high-risk CABG patients, preoperative mAFP support was associated with lower early mortality compared with planned intraoperative placement, although differences were not statistically significant. Patients requiring unplanned postoperative rescue support had markedly poor outcomes. These findings highlight the potential importance of early risk stratification and proactive consideration of perioperative mAFP support.