Objectives: The recently published PROTECTION trial (Intravenous Amino Acid Therapy for Kidney Protection in Cardiac Surgery) demonstrated that, in adult cardiac surgery patients, preoperative amino acid (AA) administration has a protective effect on renal function. However, large differences were observed within the participating centers. We aimed to investigate whether such center effects would remove the impact of AA on the prevention of cardiac surgery-associated acute kidney injury (CSA-AKI). Design: A post hoc analysis of data from the PROTECTION trial was performed. Setting: Multi-institutional data from centers that participated in the PROTECTION trial. Participants: Adult patients undergoing cardiac surgery enrolled in the PROTECTION trial. Interventions: Two centers showed a significantly lower rate of CSA-AKI with respect to the pooled rate in all centers (low-rate centers) and 3 centers had a significantly higher rate (high-rate centers [HR-C]). These centers were compared for preoperative and intraoperative variables. Results: Patients in the HR-C were significantly (p = 0.001) older, with lower left ventricular ejection fraction and hemoglobin values, a higher rate of class III and IV New York Heart Association functional class, arterial hypertension, previous myocardial infarction, diabetes, peripheral vascular disease, and more frequently received myocardial revascularization. During surgery, patients in the HR-C group had a longer aortic cross-clamp time, higher temperature on cardiopulmonary bypass, and received diuretics and hemofiltration at a lower rate. Additionally, a greater number of patients in the HR-C group required norepinephrine. However, once corrected for such a center effect, AA remained significantly independently associated with a reduction in CSA-AKI (relative risk, 0.79). Conclusions: HR-Cs treated patients with greater severity (unmodifiable risk factors) and received different operative and perioperative management. Taking into account such center effects, however, AA therapy remained independently associated with CSA-AKI prevention.
