2.1. Patients
This retrospective study enrolled KD patients who developed a CAA (inner coronary artery diameter ≥4 mm) on UCG at1 month of disease onset. CAAs were defined according to the 1983 Japanese Ministry of Welfare criteria13) and confirmed by CAG, which was performed repeatedly in accordance with KD guidelines14). In follow-up CAG performed between January 2012 and March 2018, OCT was performed concurrently in patients with bodyweight ≥30 kg. The study was approved by the Wakayama Medical University Ethical Review Board (No. 2916) and conforms with the ethical guidelines of the 1975 Declaration of Helsinki. Written informed consent was obtained from all participants.
2.2. CAG
Bilateral selective CAG was performed. After a 6 Fr sheath was placed in the right radial artery, 5 Fr catheters were advanced to the coronary ostia bilaterally for contrast injection.
2.3. OCT procedure
For OCT, a 6 Fr guiding coronary artery catheter was placed at the origin of each coronary artery to advance a guidewire. A C7 Dragonfly Intravascular Imaging Catheter (0.036-inch outer diameter; St. Jude Medical, Inc. St. Paul, MN, USA) was inserted into the distal coronary artery. The guiding catheter was continuously flushed with contrast medium to remove blood cells from the observation region during OCT scanning. OCT was imaged using an automatic pullback device (20 mm/s) at a frame rate of 100 frames/s and recorded on a C7-XR OCT Intravascular Imaging System (St. Jude Medical, Inc.).
2.4. Evaluation of coronary artery wall structure changes
In coronary arteries in which OCT could be applied, we focused on branches with no dilatation or stenosis on the most recent CAG. Branches were classified into two groups based on CAG findings. Branches that developed CAA in the acute phase that later regressed in the convalescent phase (“normalization”) were classified as group R. Those with no abnormal findings in either the acute or convalescent phases were classified as group N. We excluded branches with residual CAA on the most recent CAG from this study, as it is difficult to accurately evaluate the entire circumference of the CAA in these branches on OCT because of insufficient blood cell washout with contrast medium and/or inadequate observational range.
In group R, OCT images of a 10-mm segment of the region that included the acute-phase CAA were extracted at 1-mm intervals. In group N, images were extracted from a 10-mm segment of the origin of each branch, where CAA had never developed. We excluded branches that had imaging artifacts or included a side branch that comprised >25% of the observed region.
2.5. Analysis of OCT images
All OCT imaging data were digitized, transferred to ImageJ (US National Institutes of Health, Bethesda, MD, USA), and analyzed in two-dimensions by 2 observers who were blinded to CAG findings. OCT imaging of the three laminar areas was acquired in a concentric pattern from the vascular lumen to the outside to show the intima, media, and adventitia. Maximum intimal diameter was measured. Intimal cross-sectional area was calculated by subtracting the medial area (i.e., vascular lumen cross-sectional area) from the circumferential medial cross-sectional area; the mean area of the target cross sections was determined (Figure 1).
In previous studies, coronary artery intimal thickness <300 µm was considered normal7,15,16,17), so we defined thickness >400 µm as abnormal. In addition to intimal thickness and cross-sectional area, disruption of the media, calcification (sharply delineated borders with heterogeneous poor signal composition), macrophage colonization (signal-rich, distinct, and greatly attenuated OCT light), and thrombus (mass attached to the luminal surface or floating within the lumen) were evaluated and compared between groups R and N. All evaluations were made in accordance with the consensus standards for acquisition, measurement, and reporting of OCT studies18).
2.6. Statistical analysis
Quantitative variables are expressed as medians with range and categorical variables as numbers with percentage. Statistical analyses were performed using the Mann-Whitney U test and the chi-square test in JMP Pro 13 software (SAS Institute Japan, Tokyo, Japan). P <0.05 was considered significant.