Optimizing Cardiopulmonary Bypass: Key Predictors and Strategies to Reduce Acute Kidney Injury in Cardiac Surgery

A. Arghami, K. Kashani, A. Reynolds, T. Dickinson, K. King, A. Lee, A. Pochettino, J. Crestanello, E. Wittwer
Mayo Clinic, Rochester, Minnesota

Purpose: Despite advancements in postoperative care, cardiac surgery-associated acute kidney injury (CSA-AKI) remains a relevant and significant complication with limited known preventive strategies. This research aims to identify modifiable cardiopulmonary bypass (CPB) parameters and their impact on the development of CSA-AKI to establish thresholds to reduce the risk of CSA-AKI.

Methods: In this historical cohort study, out of 2,384 screened cardiac surgery patients, we identified 70 patients who developed CSA-AKI. Using preoperative information related to the risk factors of AKI, we matched these patients to 251 patients without AKI using propensity scores. CPB parameters, including cardiac index (CI), oxygen delivery (DO2i), mean arterial pressure (MAP), lowest core temperature and ultrafiltration, and medication, were compared between the two groups. For hemodynamic parameters, the average time below the threshold, and area under the curve (AUC) below the threshold were calculated. AKI was defined based on STS definitions.

Results: The propensity-matched patients had similar baseline characteristics except for 16 minutes longer CPB time for the CSA-AKI group (p=0.03). No significant associations were found with ultrafiltration, lowest core temperature, or mean arterial pressure. Total duration and AUC of both CI and DO2i below threshold were strongly associated with CSA-AKI. Given similar predictive value of all models, cumulative time with CI below the threshold was selected for ease of use. Using different cutoff points, a CI of < 2 for over 12 minutes had the strongest association with CSA-AKI with a sensitivity of 0.61 and specificity of 0.70. To perform a sensitivity analysis, we used a lower creatinine value for the definition of AKI, and we still arrived at similar results indicating a dose-response effect. Epinephrine at any time, vasopressin after CPB, and norepinephrine in the ICU were all associated with CSA-AKI, but not phenylephrine, or intraoperative norepinephrine dose. Red blood cells and platelet transfusion were associated with CSA-AKI but not plasma. The time to peak creatinine was 4.9 days for CSA-AKI patients as compared to 2.2 days for those without.

Conclusion: In a propensity-matched cohort study, we demonstrated that blood flow during CPB is an important predictor of CSA-AKI. A cardiac index < 2.0 for more than 12 (cumulative) minutes is the simplest yet strong predictor of CSA-AKI. An automated pump alarm systems warning the perfusionist and surgical team of this critical threshold, can potentially reduce postoperative AKI.

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