Prognostic Value of Stress Perfusion CMR in Kawasaki Disease with Coronary Artery Aneurysms

Presenting Author(s)

• 

  Alexander Kiener, MD, MPH

  Post-Doctoral Clinical Fellow
  Baylor College of Medicine / Texas Children's Hospital
  Houston, Texas, United States

Primary Author(s)

• 

  Alexander Kiener, MD, MPH

  Post-Doctoral Clinical Fellow
  Baylor College of Medicine / Texas Children's Hospital
  Houston, Texas, United States

Co-Author(s)

• 

  Tam T. Doan, MD

  Assistant Professor
  Baylor College of Medicine / Texas Children's Hospital
  Houston, Texas, United States
• CA

  Carolyn Altman, MD

  Professor
  Baylor College of Medicine
  Houston, Texas, United States
• KS

  Kristen Sexson, MD, PhD

  Assistant Professor
  Baylor College of Medicine
  Houston, Texas, United States
• AL

  Aimee Liou, MD

  Cardiologist
  Baylor College of Medicine, United States

Background: Kawasaki disease (KD) complicated by coronary artery aneurysms (CAA) confers long-term risk for myocardial ischemia and major cardiac events (MCE). Surveillance imaging is recommended, but the prognostic role of stress perfusion cardiac magnetic resonance imaging (spCMR) has not been established. We aimed to evaluate whether findings of inducible perfusion defects from spCMR predict future MCE in KD patients with CAA.

Methods: We performed a retrospective cohort study of KD patients with AHA risk level 3-5 who underwent vasodilator spCMR between 2014-2024 at a tertiary pediatric referral center. Inducible perfusion defects, ventricular function, and late gadolinium enhancement (LGE) were assessed. The primary outcome was MCE (myocardial infarction, ventricular dysfunction, arrhythmia, coronary revascularization, transplant, or death). Event-free survival was evaluated with Kaplan-Meier analysis, and Cox regression was used to assess risk.

Results: Sixty-two patients underwent 107 spCMR, including 18 patients (29%) with AHA risk level 3-4 and 44 patients (71%) with AHA risk level 5. Their median age at KD diagnosis was 1.9 years (IQR 0.4-5.1) with a median follow-up of 8.6 years (IQR 5.9-12.1) from KD diagnosis. The initial spCMR was performed at 9.9 years (IQR 7.7–14.2). Overall, 18 patients (29%) experienced at least one MCE, all within the AHA 5 group. Nine patients (14.5%) had inducible perfusion defects on their first spCMR; these patients had significantly reduced event-free survival compared with those without perfusion abnormalities (HR 4.2, 95% CI 1.1–15.6; p=0.03, Figure 1) over a median follow-up time from initial spCMR of 3.4 years (IQR 0.8-4.9). Among AHA 5 patients, 44% with a positive first spCMR developed MCE compared with 14% with a negative first spCMR (p=0.045). Serial spCMR was performed in 21 patients during the study duration and identified new inducible defects in patients who subsequently experienced MCE. Post-revascularization spCMR confirmed resolution or improvement of ischemia in all six cases and guided return-to-activity decisions.

Conclusion: In KD patients with CAA, inducible perfusion defects on the first spCMR independently predict increased risk of MCE. MCE occurred only in patients with large or giant CAA. Serial spCMR was valuable for identifying evolving ischemia and assessing post-revascularization outcomes. These findings support the integration of spCMR into longitudinal surveillance of high-risk KD patients and provide rationale for multicenter validation.

Figure 1. Kaplan-Meier survival curves showing freedom from new MCEs among patients with KD and CAA following initial spCMR. Patients with inducible perfusion defects (red) had significantly lower event-free survival than those without perfusion abnormalities (blue) (HR 4.2, 95% CI 1.1–15.6; p=0.03). Shaded areas represent 95% confidence intervals. "Number at risk" tables show the number of patients remaining under observation at each time point. Time 0 represents time of initial spCMR. CAA, coronary artery aneurysm; KD, Kawasaki disease; MCE, major cardiac events; spCMR, stress perfusion cardiac magnetic resonance imaging