Abnormal Myocardial Perfusion in Kawasaki Disease Convalescence

Kawasaki disease (KD) is a generalized systemic vasculitis, although the coronary artery system is typically involved. Coronary artery lesions (CAL) develop in 25% of children during the acute stage of KD and may lead to infarction, sudden death, or chronic coronary artery insufficiency (1). In about 50% of cases, complete, spontaneous resolution occurs within 1 year after onset (2).

The purpose of this study was to evaluate myocardial perfusion reserve with the use of perfusion cardiac magnetic resonance (CMR) in children with a previous history of KD and coronary involvement and correlate it with coronary morphologic abnormalities.

Fourteen asymptomatic patients with history of KD and coronary involvement (male patients, n = 8; mean age 10.2 ± 7.2 years) were prospectively examined with CMR using a 1.5-T magnetic resonance (MR) unit (Intera, Philips Healthcare, Best, the Netherlands). Electrocardiography-gated 2-dimensional steady-state free precession images were acquired for function assessment. Stress/rest first-pass perfusion was assessed using a turbo fast low-angle shot sequence (in-plane spatial resolution: 2.5 × 2.5 × 10.0 mm; 3 short-axis slices during intravenous contrast medium infusion (gadobutrol 0.1 mmol/kg body weight) after administration of 140 μg/kg/min adenosine for 4 min and at rest, with a 15-min interval. To assess fibrosis, late gadolinium enhancement (LGE) images were acquired 10 min after (inversion recovery turbo fast low-angle shot). A 3-dimensional steadystate free precession sequence with T₂ and fat saturation pre-pulses was used for MR angiography. Imaging was performed under anesthesia, if necessary, with continuous intravenous infusion of remifentanil and controlled ventilation.

Mean and segmental myocardial perfusion reserve index (MPRI), as defined by the ratio of stress to rest myocardial signal intensity relative upslope, were measured quantitatively (QMass MR 7.5, Medis, Leiden, the Netherlands). Left ventricular endocardial and epicardial boundaries of the left ventricle were automatically outlined and manually edited for through-plane motion on a 16-segment model, in time series of short-axis cine MR images. An abnormal coronary artery was defined according to criteria established by the Japanese Ministry of Health. A 5-SD threshold above the mean remote myocardial signal was used to study LGE images.

Comparisons of continuous variables with a normal distribution were performed using the independent-sample Student t test. Correlation analysis was assessed using Pearson correlation. The coefficient of variation was calculated to study the variability of the measurements. Data are expressed as mean ± SD unless otherwise specified. Values of p < 0.05 were considered statistically significant.

In 8 of 14 patients, CMR was performed under anesthesia. Persisting CAL were identified with MR angiography in 5 patients. Inducible perfusion defect by visual assessment was detected in 1 patient. LGE identified myocardial scar in 1 patient. Mean MPRI was significantly impaired in all patients, compared with historical pediatric control subjects (0.86 ± 0.256 vs. 2.46 ± 0.3, p < 0.001, 1 sample Student t test) (3). No significant difference in mean MPRI was identified between patients with regressed CAL (9 of 14) and persistent CAL (5 of 14) (1.0 ± 0.3 vs. 0.79 ± 0.225; p = 0.31). In patients with persisting CAL, no differences in MPRI were demonstrated between segments subtended by arteries with regressed and persistent CAL. Patients' clinical and CMR characteristics are presented in Table 1.

Table 1. Clinical and CMR Characteristics of KD Patients (N = 14).

Age at CMR examination, yrs	10.2 ± 7.2
Male	8/14 (57)
Age at KD onset, yrs	5.0 ± 3.5
Interval from acute event, yrs	4.9 ± 5.6
Treatment
IVIG	14/14 (100)
Aspirin	14/14 (100)
Steroids	0/0 (0)
Exam under anesthesia	8/14
Coronary artery status by MRA
Transient CAL	9/14 (64)
Persistent CAL	5/14 (36)
Dilation of RCA origin/dilation of LAD-LCX bifurcation mm	3/4
Dilation of LMCA, mm	4
Dilation of LAD origin, mm	4
Dilation of LM/LAD, mm	4
Giant aneurysm of LMCA/LAD aneurysm, mm	17/4
Function
LVEF, %	60.0 ± 4.7
Normal	67.0 ± 5.0
LV EDVI, ml/m2	73.0 ± 22.4
Normal	90.0 ± 11.0
LV ESV, ml/m2	29.0 ± 10.1
Normal	30.0 ± 7.0
Myocardial perfusion
Inducible visual defect	1/14
Basal to apical lateral subendocardial defect, related to LMCA giant aneurysm
Mean MPRI, all patients	0.86 ± 0.256
Mean MPRI, transient CAL	1.0 ± 0.30
Mean MPRI, persistent CAL	0.79 ± 0.225
LGE+	1/14
Basal inferior subendocardial scar

Values are mean ± SD, n/N (%), n/N, or n.

+ = positive; CAL = coronary artery lesion; CMR = cardiac magnetic resonance; EDVI = end-diastolic volume indexed; ESV = end-systolic function; IVIG = intravenous immunoglobulin; KD = Kawasaki disease; LAD = left anterior descending artery; LCX = left circumflex artery; LGE = late gadolinium enhancement; LMCA = left main coronary artery; LVEF = left ventricular ejection fraction; MRA = magnetic resonance angiography; MPRI = myocardial perfusion reserve index; RCA = right coronary artery.

Previous histologic studies in the acute and chronic phase of KD have confirmed inflammatory cell infiltration of the arterial wall, with subsequent reparative elastic fibers replacement by fibrous tissue and chronic myofibroblastic intimal proliferation (4). A limited number of studies with nuclear medicine tracers in KD patients identified reduced hyperemic flow and flow reserve, despite resolution of epicardial vessel disease, as indicative of functional microvascular abnormalities (5).

In our study, we demonstrated with the use of CMR that MPRI was markedly decreased in all myocardial segments regardless of the epicardial disease status. These findings are in agreement with previous studies, suggesting that microvascular dysfunction is a substantial feature of the underlying diseased myocardium.

Only quantitative perfusion analysis was able to detect the abnormalities due to the lack of regional differences in perfusion. Quantitative CMR perfusion imaging influences substantially the appreciation of myocardial perfusion pattern in KD and provides further insight into its pathophysiological substrate.

The limitations of the study were the small number of patients and the lack of a pediatric normal refer-ence group, due to the ethical obstacles in applying a stress exam in a healthy pediatric population.

In conclusion, quantitative perfusion CMR identifies abnormal perfusion reserve in KD convalescent patients irrespective of the coronary artery status, which is suggestive of significant coronary microvascular dysfunction. The clinical implications of these findings need further assessment.