Platelet-related hematologic markers and genetic associations of aspirin resistance in kawasaki disease

Kawasaki disease (KD) is the leading cause of acquired cardiovascular disease in children and is characterized by intense immune activation and platelet dysfunction. High platelet reactivity (HPR) is increasingly recognized as a biological basis of aspirin resistance (AR), which may increase the risk of adverse coronary outcomes, including coronary artery aneurysms (CAA). However, the hematologic dynamics and genetic determinants underlying AR in KD remain unclear. In this study, the association between AR and CAA was assessed using chi-square analysis. We compared platelet parameters between KD aspirin-resistant (KD-AR) and KD non-aspirin-resistant (KD-NAR) patients across different disease phases using linear mixed-effects models (LMM). Baseline complete blood count (CBC) derived inflammatory indices, including the systemic immune-inflammation index (SII), platelet-to-lymphocyte ratio (PLR), and neutrophil-to-lymphocyte ratio (NLR), were evaluated using restricted cubic spline (RCS) and receiver operating characteristic (ROC) analyses. Integrated transcriptomic and expression quantitative trait loci (eQTL) analyses were performed to identify candidate genetic factors associated with the KD-AR phenotype. The results showed that AR was significantly associated with CAA formation. LMM showed significant phase-dependent changes in platelet parameters, with distinct longitudinal trajectories between KD-AR and KD-NAR patients. Between-group differences were mainly observed during the subacute phase (D7–14), when KD-AR patients showed lower platelet count (PLT) and plateletcrit (PCT), but higher platelet distribution width (PDW) and platelet large cell ratio (PLCR). Baseline SII, PLR, and NLR were significantly elevated in KD-AR patients (all P < 0.001). RCS analyses demonstrated significant overall associations between these indices and AR risk (all P overall < 0.001). ROC analyses showed moderate discrimination for SII (AUC = 0.702) and NLR (AUC = 0.722), whereas PLR showed lower performance (AUC = 0.626). MBP was consistently upregulated in HPR-associated samples, and eQTL integration identified MBP/rs8090438 as a candidate variant linked to KD-AR. These findings suggest that AR in KD represents a multifactorial phenotype involving immune-driven platelet dysregulation and genetic susceptibility. Baseline inflammatory indices, particularly NLR and SII, may assist in early identification of KD patients at increased likelihood of AR.