Abstract
Background: Thrombocytopenia is a common post-operative finding in major surgeries including coronary artery bypass grafting (CABG). Thrombocytopenia is also frequently observed in intensive care unit patients supported by extracorporeal circuits such as extracorporeal membrane oxygenation (ECMO) and renal replacement therapy (RRT). CABG can be performed while the patient is on cardiopulmonary bypass (CBP) known as “on-pump,” or without CBP, known as “off-pump” or OPCAB. Purportedly, off-pump CABG is associated with less coagulopathy. However, there is limited long term benefit over on-pump CABG. Although thrombocytopenia is often reported in these populations, there is a paucity of data on the relative risk between modalities. Our aim was to compare the changes in platelet count among patients who underwent on- and off-pump CABG.
Methods: This is a retrospective cohort study including patients undergoing elective CABG at Mount Sinai Morningside Hospital from 2022 to 2025. Patients were excluded if they underwent a concomitant non-CABG procedure. The study exposure was use of CPB during CABG, dividing patients into on- and off-pump groups. Platelet counts (per microliter) were obtained from the most recent pre-operative complete blood count (CBC) and the first post-operative CBC. The differences in pre- and post-operative platelet counts were calculated and compared between on-pump and off-pump patients.
Results: The study included 235 patients with a median age of 65.6 years (IQR 59.6-72.7), including 194 (83%) males and 41 (17%) females. A total of 58 (25%) patients underwent on-pump CABG and 177 (75%) underwent off-pump CABG. The median pre-operative platelet count was 212 (IQR 177-269) for the on-pump group and 212 (IQR 182-257) for the off-pump group (p=0.996 for differences between the groups). The median drop in platelet count post-operatively was -71 (IQR -100 to -47) for the on-pump group and -57 (IQR -85 to -35) for the off-pump group. The Wilcoxon rank sum test with continuity correction revealed an estimated difference of -17 in platelet drops between groups (p=0.01; 95% CI -29 to -4).
Across the 58 patients who underwent on-pump CABG, the mean bypass time was 107 minutes (95% CI 100-115) and the mean aortic cross clamp time was 87 minutes (95% CI 81-93). Pearson's correlation coefficient for post-operative platelet count and bypass time was -0.09 (p=0.46, 95% CI -0.35 to 0.16) and -0.13 (p=0.35, 95% CI -.37 to 0.14) for post-operative platelet count and aortic cross clamp time.
Conclusions: We observed that patients who underwent CABG with the use of cardiopulmonary bypass had a significantly larger drop in platelet count compared to those who underwent CABG without cardiopulmonary bypass. We continue to research the clinical impact of this difference and future work includes gathering data on bleeding events and transfusion needs in this cohort.
