Normal myocardial perfusion in kawasaki disease convalescence: an age, body-surface-area and sex-matched analysis using fully automated quantitative stress perfusion cardiac MRI

Background Kawasaki disease (KD) is a childhood systemic inflammatory illness associated with the development of coronary artery aneurysms (CAAs) in up to 25% of children. A minority of CAAs persist into adulthood and may predispose patients to adverse cardiac events in the convalescent phase. Previous semi-quantitative studies1 suggest KD confers a sustained impairment of coronary microvascular perfusion, independent of epicardial coronary artery morphology. This has not been evaluated using fully automated inline quantitative perfusion cardiac MRI (CMR). Purpose To assess the impact of childhood CAAs on fully automated quantitative myocardial stress perfusion in adult patients with convalescent KD, compared to an age-, body surface area (BSA)- and sex-matched control cohort. Methods A retrospective single centre cohort analysis of 30 adults with convalescent KD was performed. Patients with known childhood CAA (Z score >2.5) and recent automated quantitative perfusion CMR were selected for further analysis. Those with prior myocardial infarction, coronary revascularisation or late gadolinium enhancement were excluded. An age-, BSA-, and sex-matched group with normal CMRs was used for comparison. Resting myocardial blood flow (MBF) was normalised to rate pressure product. Automated quantification of MBF and myocardial perfusion reserve (MPR) was obtained during rest and stress (adenosine 140-210ug/kg/min). A per-vessel analysis was performed using a standardised 16-segment AHA myocardial segmentation. Pairwise comparisons were conducted using a Student’s t-test and correlations were assessed using Pearson’s correlation. Results Eighteen patients were included (9 Kawasaki; 9 Controls). Patients were well-matched to age, BSA and sex (KD: 25±5.6 years; 1.9±0.25m2; 6 males vs. Controls: 25±5.3 years; 1.9±0.25m2; 6 males). All KD patients had coronary anatomy assessed by CT (89%) or invasive (11%) angiography. 54 vessels were analysed in total. In KD patients, no significant difference was observed in MPR, stress MBF (sMBF) or endocardial: epicardial (Endo:Epi) perfusion ratio, between myocardium subtended by unaffected, regressed aneurysmal or persistently aneurysmal vessels, or controls. Childhood CAAs had regressed in 70% of vessels. Persistently aneurysmal vessels had significantly higher childhood Z scores than regressed CAAs (10.9±5.0 vs 5.05±1.7, p = 0.0004) however no significant correlation was identified between childhood Z scores and MPR, sMBF or Endo:Epi perfusion ratio in adulthood. Conclusions Epicardial morphology did not contribute to a significant impairment of myocardial perfusion indices in adult patients with convalescent Kawasaki disease. Furthermore, no association was found between childhood CAA size and myocardial perfusion. Whilst hypothesis-generating, these results suggest KD may not result in sustained coronary microvascular dysfunction in this population. Additional studies are required to validate these initial findings.