Prognostic nutritional index value in the prognosis of Kawasaki disease with coronary artery lesions.

Abstract

The prognostic nutritional index (PNI) is a purported predictor of intravenous immunoglobulin (IVIG) resistance and coronary artery aneurysm (CAA) development in patients with Kawasaki disease (KD). However, limited data exist on CAA regression. This study aimed to confirm whether the PNI is a predictor for CAA persistency in patients with KD. This retrospective study grouped 341 patients with KD based on the coronary artery status and time of aneurysm persistence. The clinical and laboratory parameters were compared, and multivariate logistic regression analysis was performed to identify the independent risk factors for persistent CAA. The receiver operating characteristic (ROC) curve was further used to assess the predictive values of the PNI in persistent CAA. Among the study patients, 80 (23.5%) presented with CAA, including CAA persisting for 2 years in 17 patients (5.0%). Patients with CAA were more frequently treated with corticosteroids (p < 0.016). No statistically significant differences were found in the nutritional status and PNI among patients with or without coronary artery lesions, regardless of injury severity. Patients in the persistent CAA group presented with higher rates of overnutrition and showed lower PNI values and a higher incidence of thrombosis than those in the normal group (p < 0.05). The PNI and the maximum Z-score at 1 month of onset were significantly associated with CAA persisting for 2 years and may be used as predictors of persistent CAA. The area under the ROC curve was 0.708 (95% confidence interval, 0.569-0.847), and a 40.2 PNI cutoff yielded a sensitivity and specificity of 41 and 92%, respectively, for predicting CAA persisting for 2 years. Kaplan-Meier survival analysis revealed that the estimated median time of aneurysm persistence was significantly higher in patients with PNI values of ≤40 than in those with PNI values of >40 (hazard ratio, 2.958; 95% confidence interval, 1.601-5.464; p = 0.007). After sampling-time stratification, the PNI differed significantly between patients with and without persistent CAA when sampled on the second (p = 0.040), third (p = 0.028), and fourth days (p = 0.041) following disease onset. A lower PNI value is an independent risk factor for CAA persisting for 2 years in patients with KD, besides the maximum Z-score at 1 month after onset. Furthermore, the PNI obtained within 4 days from fever onset may possess greater predictive power for patients with persistent CAA.