Study Population

A total of 309 pediatric patients who underwent MPI at Children’s Hospital of Fudan University between July 2019 and February 2023 were included. 152 cases were excluded from the study due to the absence of a definitive diagnosis of KD or presence of other anomalies such as myocardial bridge, history of coronary artery revascularization, and lack of concurrent CAG during the same hospitalization. Finally, a total of 177 children were included for analysis in this study (Fig. 1). The study was approved by the Institutional Review Board of Children’s Hospital of Fudan University (Approval No. 2022-391), in accordance with the Declaration of Helsinki. All clinical data was de-identified.

Fig. 1

Study Flowchart. MPI 99mTc-MIBI myocardial perfusion imaging, CAG coronary artery angiography

The diagnosis of KD followed the 2017 American Heart Association (AHA) guidelines [1] and the Japanese diagnostic guideline (6th revised edition) [16]. Coronary artery aneurysm (CAA) inner diameter was measured by CAG, and the Z-score of coronary arteries was calculated by the online calculator (https://www.pedz.de/de/pedz/mmode.html). According to AHA guidelines, CAA was classified as small (2.5 ≤ Z-score < 5), medium (5 ≤ Z-score < 10), and giant (Z-score ≥ 10 or absolute inner diameter ≥ 8 mm) [17].

Imaging Examination Methods and Diagnostic CriteriaCAG

All patients underwent CAG using the Siemens (AXIOM Artis dBA Detector System)/ Canon (INFX-9000 V) DSA machine through femoral punctures. Aortic root angiography and selective CAG were performed to visualize the left main coronary artery (LMCA), left anterior descending branch (LAD), left circumflex branch (LCX), and right coronary artery (RCA) from at least two projections. Significant stenosis of coronary artery on CAG was defined as either 50% luminal narrowing of the LMCA or 70% luminal narrowing of the remaining of coronary arteries [18]. The CAG results were analyzed and confirmed by at least two blinded pediatric cardiologists.

99mTc-MIBI MPI

99mTc-MIBI MPI were performed by the dual-head Siemens E.CAM SPECT system. Patients would regularly undergo both adenosine triphosphate (ATP)-induced MPI and resting MPI. If a patient presents with evidence of significant myocardial ischemia such as history of myocardial infarction or decreased systolic function in echocardiogram, only a resting MPI will be performed.

The protocol of ATP-induced 99mTc-MIBI MPI is as follows: (1) Prior to the test, intake of dipyridamole and theophylline was discontinued for 24 h, refrained from consuming food for 4–6 h, and beverages containing caffeine were avoided. (2) In a supine position, bilateral intravenous access was established and intravenous drip of ATP at a rate of 0.14 mg/kg/min (equal to 0.84 mg/kg) was initiated for 6 min. 3 min after the start of the drip, the contrast agent was injected into the contralateral vein. (3) Continuous monitoring of the electrocardiogram (ECG) was performed throughout the ATP infusion. Heart rate, blood pressure, ECG performance, and any symptoms at three time points were recorded: 3 min after the start of the drip, the end of the drip, and 5 min after completing the drip. (4) After the injection of contrast agent, high-fat foods such as cake or milk were consumed within 30 min, followed by MPI within 60 min. (5) Tomographic images of the left ventricular short axis, vertical axis, and horizontal long axis at a zoom level of 1.45, matrix size of 128*128, and 30 s per frame were obtained. (6) If the stress test yielded a positive result, resting MPI was performed 24 h later.

The presence of sparse or faulty distribution of the radiographic agent across two or more levels was considered indicative of myocardial ischemia [19]. If a radioactive defect was observed in the stress image but was filled in the rest image, it suggested reversible myocardial ischemia.

CMRI

CMRI was performed by the Siemens MR Avanto 1.5 T equipment (Germany). Examination techniques included Trufi sagittal, coronal, transverse, 2-chamber, 4-chamber, short-axis, short-axis T1WI, T2WI, T2WI SPAIR, myocardial perfusion, delayed enhancement, short-axis + 4-chamber cine sequence, thin 3D layer. The presence of delayed T2WI elevation, under-perfusion, or signal anomalies was considered indicative of myocardial ischemia in CMRI [20].

Echocardiogram

The presence of any of the following conditions was considered indicative of myocardial ischemia in echocardiogram: (1) decreased left ventricular ejection fraction (LVEF, < 50%) or fraction shortening (LVFS, < 25%); (2) decreased or absence of ventricular wall motion; (3) paradoxical motion or systole asynchrony [21] evaluated by at least two pediatric cardiologists. Different types of ventricular wall motion were defined as follows: (1) normal motion, characterized by a ventricular wall wave amplitude > 5 mm, (2) decreased motion, characterized by a ventricular wall wave amplitude 2–5 mm, and (3) loss of motion, characterized by a ventricular wall wave amplitude < 2 mm.

Comprehensive Electrocardiogram-Related Examinations

CEEs include 12-Lead electrocardiogram (ECG), 24-h Holter monitor (Holter), and exercise treadmill stress test (ETST). The presence of any of the following conditions was considered indicative of myocardial ischemia. (1) ECG: a 0.1 mV deviation in the resting ST segment from baseline and/or T-wave abnormalities (T-wave inversion or T-wave amplitude less than 1/10 of the R-wave in the same leads) [22, 23]. (2) Holter: horizontal or downward-sloping ST-segment depression 0.1mv for > 1 min [24, 25]. (3) ETST: ETST examination followed the modified Bruce-1 protocol for children. A positive ETST diagnosis was defined as the occurrence of the typical angina pectoris, ST-segment horizontal or downward-sloping depression 0.1 mV (60–80 ms after the J-point) lasting > 1 min, or severe arrhythmia during the test [26].

Statistical Analysis

Statistical analysis was performed using SPSS (version 26, IBM SPSS Statistics for Mac). Continuous data were presented as mean ± SD. The student’s t-test and Mann–Whitney U test were used for analysis. Categorical data were compared using the chi-squared test and presented as numbers (n) and percentages (%). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the diagnostic values of each examination were determined and displayed in a 2*2 table. Agreement between different tests was estimated using the Kappa statistic: Kappa < 0.40 was regarded as poor consistency, 0.40 ≤ Kappa < 0.75 was regarded as fair consistency, Kappa ≥ 0.75 was regarded as good consistency. Statistics were considered as significant when P < 0.05.