Does Aortic Manipulation Technique Influence Stroke Risk After CABG? - A Comparison of Observed to STS Predicted Outcomes

J. S.. Newman¹, A. Iribarne², S. Scheinerman³, A. Hartman⁴, D. Brinster⁵, H. Rahman¹, N. Patel^6
1Northwell Health, Manhasset, New York ²SIUH / Northwell Health, Staten Island, New York ³North well Health Lenox Hill Hospital, New York, New York ⁴North Shore Univ. Hospital, Dept of CV & Thoracic Surgery, Manhasset, New York ⁵Lenox Hill Hospital / Northwell Health, New York, New York ⁶Lenox Hill Hospital, Northwell Health, New York, New York

Purpose: Multiple techniques exist for performing proximal anastomoses during coronary artery bypass grafting (CABG), each with varying degrees of aortic manipulation. The goal of this analysis was to compare 5 techniques and assesses the association between proximal technique and observed to STS-predicted risk of stroke.

Methods: A retrospective review of a prospectively maintained multi-center registry was performed for 5,360 patients undergoing isolated CABG from July 2014 to June 2022. Aortic manipulation was assessed by proximal anastomosis clamp technique. Off-pump total anaortic bypass was employed in 2% (106) patients, off-pump with a clampless technique in 11% (588) patients, off-pump with a partial occlusion clamp in 14% (745) patients, on-pump single clamp technique in 26% (1,375) patients, and on-pump multiple clamp technique in 46% (2,487) patients. The primary outcome was postoperative stroke defined as per the STS as any confirmed neurological deficit of abrupt onset caused by a disturbance in blood supply to the brain that occurred during the index hospitalization. Postoperative stroke rates were compared to the STS predicted stroke rates as well as between groups. Secondary comparisons included post operative atrial fibrillation rates, utilization of epiaortic ultrasound, and number of grafts performed.

Results: Among the entire cohort, the observed stroke rate was 0.93% (50), which was within the STS predicted rate of 1.14% (p=0.081). Operations utilizing cardiopulmonary bypass had no difference in observed to expected (O:E) stroke rates in the multiple clamp (0.97% vs 1.08%, p=0.335) and single clamp groups (1.16% vs 1.15%, p=0.512). Off-pump operations had the highest predicted STS stroke risk (p < 0.0001), and significantly lower O:E risk for partial occlusion clamp (0.27% vs 1.04%, p=0.016) or an anastomotic assist device (0.51% vs 1.58%, p=0.017). Anaortic procedures did not demonstrate significantly reduced O:E rates (1.89% vs 0.94%, p=0.739). Compared with single clamp, off-pump partial occlusion clamp (1.16% vs 0.27%, p=0.176), on-pump multiple clamp (0.97%, p=1.00), and off-pump anastomotic assist devices (0.51%, p=0.860) trended towards reduced stroke rates. Intraoperative epiaortic ultrasound was used in 12% (623) patients, with 52.3% (326) patients with aortic disease. Significantly fewer strokes were observed in patients without aortic disease (0.00% vs 1.82%, p=0.032). Off-pump approach was employed in 93.6% (305) of patients with aortic disease, with 92.0% (300) undergoing a clampless operation.

Conclusion: Several techniques exist in performing proximal anastomoses during CABG, each with varying degrees of aortic manipulation and stroke risk. Significantly reduced O:E risk of stroke was observed among off-pump CABG with either a partial occlusion or anastomotic assist device which may be particularly advantageous in patients with calcified aortas.

Identify the source of the funding for this research project: No funding was received for this project.