Introduction: Children with Kawasaki disease (KD) and coronary artery aneurysms (CAA), particularly giant aneurysms, may develop coronary artery calcification (CAC) detectable on coronary computed tomography angiography (CTA). We sought to describe CAC findings and technical considerations in a pediatric cohort with giant CAA.
Methods: We performed a retrospective review of patients with KD and giant CAA who underwent ECG-gated coronary CTA at a single center over a 10-year period. All CTAs were reviewed by a single pediatric radiologist with specialized cardiac training. CAC scoring was ideally determined from true non-contrast (TNC) series or a virtual non-contrast (VNC) series when available. Mann–Whitney U and Fisher’s exact tests were used for group comparisons.
Results: Fifty-one coronary CTAs were identified. Thirty-one(61%) lacked a TNC or artifact free VNC series. 20 CTAs in 16 patients were adequate for CAC classification, 75% were male(n=12) with median age at KD diagnosis of 3.25 years(Q1–Q3 1.6–6.5). All 16 patients had giant LAD aneurysms(max Z-score 10.2–42.4), and 10 also had giant right CAA(max Z-score 10.1-38) at diagnosis. Time from KD diagnosis to CTA was 0.2–18 years (median 5.3 years, Q1–Q3 1.55–11.95). At the time of CTA, persistent giant aneurysms were uncommon (2/16 giant LAD; 1/10 giant right).
CAC was present in 12/20 (60%), with CAC scores 2 to 1196. Time from KD diagnosis to CTA did not differ between CAC-positive and CAC-negative patients (7.0±5.9 vs 6.6±5.8 years, p=0.9), nor did age at CTA (10.8±6.0 vs 11.7±6.6 years, p=0.9). CAC prevalence did not differ by time since diagnosis (0–<5 years: 5/8 [63%]; 5–<10 years: 3/6 [50%]; >10 years: 4/6 [67%]; p=1.0). All 4 patients with a subsequent CTA which was positive for CAC had a prior negative study. Notably, CAC was detected as early as ~2 months after diagnosis (n=1) and within the first year (n=2).
Discussion: Among patients with KD and giant CAA, CAC was common (60%) and was observed even early after KD diagnosis. In this cohort, VNC-based CAC assessment was often limited by image artifacts; therefore, a TNC acquisition may be indicated to improve confidence in CAC detection and quantification.