Intravenous immunoglobulin (IVIG) serves as the first-line therapy for Kawasaki disease (KD); however, a significant minority of patients exhibit resistance to this treatment. Currently, there is a lack of easily accessible and reliable predictors for IVIG resistance. This study systematically evaluates the utility of baseline peripheral blood cell counts and their ratios in predicting IVIG non-response. We assembled a retrospective cohort from the Children’s Hospital of Chongqing Medical University, including all KD patients admitted between January 1, 2023, and June 30, 2025. Patients were classified as either IVIG-responsive or IVIG-resistant based on the resolution of fever following the initial IVIG treatment. We compared baseline demographics, laboratory data, and coronary artery lesion (CAL) status. Univariable analysis was conducted to identify factors associated with IVIG resistance, and receiver-operating characteristic (ROC) curves were utilized to quantify predictive performance. Among the 210 children diagnosed with KD, 15 (7.1%) were classified as IVIG-resistant, while 195 (92.9%) were responsive to treatment. No significant differences were observed in sex, age at onset, or the frequency of CAL between the two groups; however, the hospital length of stay was significantly longer in the resistant cases (P < 0.001). Prior to IVIG treatment, the resistant group exhibited elevated neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio (MLR), and systemic immune-inflammation index (SII) (P = 0.003, 0.001, 0.013, 0.029, respectively), alongside a lower absolute lymphocyte count (P < 0.001). ROC analysis revealed that NLR (AUC 0.733; cut-off 1.07; sensitivity 100%, specificity 84.6%) and PLR (AUC 0.766; cut-off 64.43; sensitivity 100%, specificity 84.1%) were the most robust pre-treatment predictors of IVIG resistance. Following IVIG administration, responsive patients demonstrated significant alterations in multiple cellular parameters, whereas resistant patients exhibited only minimal changes. The extent of change was markedly different between the groups, particularly for PLR: PLR decreased in resistant patients but increased in responders. The difference in PLR before and after infusion (ΔPLR) provided excellent discrimination (AUC 0.840; cut-off 81.77; sensitivity 93.3%, specificity 90.3%). Elevated baseline NLR, PLR, MLR, and SII are identified as risk factors for IVIG resistance in KD. NLR values exceeding 1.07 and PLR values greater than 64.43 demonstrate high predictive value. A decline in post-treatment PLR of ≥ 81.77 serves as an additional strong indicator of non-response to IVIG.
Tang et al. (Thu,) studied this question.