ABSTRACT
Increased procalcitonin concentration (PCT) is known to be reliable for the identification of infections even in the presence of the non-specific systemic inflammatory response seen after cardiopulmonary bypass (CPB), whereas increased C-reactive protein concentration (CRP) is not. The present work explored the ability of neonate PCT measured early after cardiac surgery to identify postoperative infections. This was a retrospective case-control study, where PCT was matched between patients with and without infections according to the patient's age, the CPB length, the use of deep hypothermic circulatory arrest (DHCA), and the postoperative day (POD). The accuracy in the prediction of infections was ascertained and cutoff thresholds were identified. 144 neonates were eligible, and 89 pairs of measurements from 94 patients were analyzed. PCT was a good predictor of infections within POD4, and was a better predictor when compared with CRP at POD1 and POD2. The sum of PCT (pg mL-1) and CRP (mg L-1) > 33 on POD1 or POD2 predicted infections with a 0.68 sensitivity and a 0.82 specificity, and a sum > 49.36 on POD3 or POD4 predicted infections with a 0.82 sensitivity and a 0.93 specificity. In patients with DHCA, PCT was higher than in those without DHCA, and was not predictive of infections. The accuracy of PCT to identify infections after neonatal cardiac surgery is better than that of CRP when measured within 48 h of surgery. The sum of the two markers measured early after surgery is an excellent predictor of postoperative infections.
