Demographic data and clinical features of 66 Pediatric TAK patients
Sixty-six pediatric TAK patients (65.2% female) were selected. The median age of onset was 85.0 months (one month–15 years), and 16, eight, seven, and 35 patients were aged <12 months, 1–3 years, 3–7 years, and ≥7 years, respectively. The clinical characteristics of the pediatric TAK patients in this study are summarized in Table 1. Fever (57.6%) and hypertension (45.5%) were the most common constitutional symptoms, and Numano classification was mostly type 3 (50.0 %) or 1 (36.4 %). In addition, 66 (100 %) patients had active disease based on the National Institutes of Health (NIH) criteria. The median CRP, AESR, PLT, TNF-α,IL-6, and FS values were higher than the reference ranges. None of the patients had a history of smoking, drinking, diabetes, or a family history of hereditary hypercholesterolemia or hypertriglyceridemia. Serum total cholesterol, LDL-C, HDL-C, and TG levels were within normal ranges.
Clinical characteristics and coronary artery involvement in the CAL group
As shown in Table 2, there were 26 (39.4 %) cases in the CAL group; 19 (73.1%) cases started within 36 months, of which two (7.7%) started at one month. Eighteen (69.2%) patients had a fever, eight (30.8%) had hypertension, and eight (30.8%) had rash. 12 (46.2%) patients were misdiagnosed with atypical KD due to CAL. These patients were under three years of age and did not meet the diagnosis of typical KD. Among them, nine children received intravenous immunoglobulin (IVIG) and oral high-dose aspirin (ASP), but the inflammatory reaction was not controlled. The temperature of the three patients was normal after IVIG combined with oral prednisone (PRE), but fever recurred when PRE was stopped.
None of the patients had symptoms of angina or ischemia on electrocardiogram (ECG), and all CALs, which were dilated, were detected using coronary ultrasound. The median Z values were between 3.0 and 5.3, indicating that most CALs were small or middle coronary artery aneurysms. Five giant coronary aneurysmal dilations were formed with a maximum Z-value of 15.37. Patient (P) 1 showed beaded changes, and patients P8 and P20 showed mural thrombosis (Fig. 2). The most frequently affected coronary arteries were left main coronary artery (LMCA), right coronary artery (RCA), and left circumflex artery (CX), with the affected rates of 84.6%,73.1%,65.4% respectively. In eight cases (30.7%), four coronary arteries were all involved (Table 3). Cardiac findings showed heart failure, pericardial effusion, and pulmonary hypertension in four, one, and two patients, respectively.
All 26 patients were treated with corticosteroids precombined with infliximab (IFX) and methotrexate (MTX), tocilizumab (TCZ) and MTX, and cyclophosphamide (CP) in 17 (65.4%), two (7.7%), and five (19.2%) cases, respectively. All the patients received ASP and/or warfarin as anticoagulant therapy. Four patients were administered anti-heart failure treatment simultaneously, and two patients received bosentan orally to reduce pulmonary artery pressure. The median follow-up time was four years, and 24 patients were stable. Nine patients with coronary artery dilatation recovered in > 1 year, and 14 regressed in more than 0.5 years. None of the patients developed MI or angina. Two patients died during follow-up. Patient(P) 20 died because of irregular treatment with PRE and CP, and refused IFX therapy. In 1.5 years, a thrombus in the coronary artery, moderate pulmonary hypertension, and a large amount of effusion in the pericardium were observed; the patient’s parents discontinued treatment after left heart failure. P21 received chemotherapy due to chronic active Epstein-Barr virus infection, but after nine months of regular treatment, hemophagocytic syndrome occurred, the parents discontinued treatment, and the patient died.
Comparison of clinical characteristics and laboratory findings between pediatric TAK patients with and without CAL
Differences in clinical characteristics and laboratory findings between pediatric TAK patients with and without CAL are listed in Table 4. The age of onset and disease duration in TAK patients with CAL were lower than those in TAK patients without CAL [12.0 (3.1–48.7) vs. 115.6 (82.1–146.0), P < 0.001; 11.5 (7.0–20.0) vs. 28.0 (14.8–60.0), P = 0.01]. Laboratory findings showed that TAK patients with CAL had significantly higher CRP,WBC, PLT,TNF-α, and IL-2R levels than those in other patients [64.5 (22.2–139.8) vs. 25.0 (3.2–82.5), P =0.04; 13.9 (9.9–18.8) vs. 8.7 (7.3–12.1), P = 0.002; 481.5 (310.8–596.5) vs. 346.5 (253.2–428.8), P = 0.01; 14.4 (10.6–18.6) vs. 10.0 (7.9–16.0), P = 0.01;527.0 (365.5–1547.2) vs. 380.0 (292.2–566.5), P = 0.03]; HGB was significantly lower in patients with CAL than in other patients (99.4±19.8 vs. 111.4±21.1, P = 0.01). Among patients without CAL, the number of patients with renal artery stenosis (RAS) was significantly higher than that in the CAL group (P = 0.009). However, no statistical differences were found in the clinical manifestations, NIH score, disease activity, sex, fibrinogen (FIB), sodium (Na), and albumin.
The risk factors of pediatric TAK involving the coronary artery
Univariate logistic regression analysis of all baseline variables showed that early disease onset, CRP, WBC, HGB, PLT, IL-2R, and RAS may be associated with pediatric TAK combined with CALs (P < 0.05) (Table 5). Factors with P < 0.1 in the univariate logistic regression analysis were included in the multivariate logistic regression analysis. Finally, the age of onset and RAS were predictive factors associated with pediatric TAK combined with CAL (P < 0.05) (Table 6). According to the receiver operating characteristic (ROC) curve results (Fig. 3), compared with model 2 (including RAS), models 1 (including age of onset) and 3 (including age of onset and RAS) had better values in predicting pediatric TAK combined with CAL (all P < 0.05). A cutoff value of 54.75 months maximized the diagnostic efficacy compared with CAL, with a sensitivity/specificity of 82.50%/76.90% (Table 7).
Survival analysis of CAL and non-CAL groups:
After treatment, there was no statistically significant difference in the survival rate between the CAL and non-CAL groups (P > 0.05) (Fig.4).