Coronary artery ectasia (CAE) has been defined as an ectatic artery diameter more than 1.5 times greater than the diameters of the adjacent normal segments (1). Its prevalence is 0.3–4.9% in patients who have undergone coronary angiography (CAG) (1). Its incidence increases with age, and the average diagnosis occurs at 55 ± 10 years old (2). The current understanding of CAE is still relatively limited. More than 80% of CAE patients experience coronary artery disease (CAD) (1, 3); their main symptoms include angina, myocardial infarction, arrhythmia, and sudden death, and most are treated under the principle of current guidelines for CAD (4, 5). However, CAE is also characterized by obvious slow blood flow in dilated coronary arteries (6, 7), increased inflammatory markers in peripheral blood (8), anomalies in other blood vessels (9–11), and decreased left ventricular diastolic function (12). Pathologically, CAE is characterized by the extensive destruction of the musculoelastic elements in the coronary wall (3, 4). Possible etiologies include enzyme destruction (2), vascular endothelial dysfunction and oxidative stress (13), and genetic factors (14). It was unclear whether CAE was progressive disease until recently, and in clinical practice one of the major concerns is that CAE patients might be at risk of continuous rapid expansion even the coronary rupture like giant coronary aneurysm (another major concern was thrombotic event) (15), and stent implant in the coronary artery should be avoided if ectatic changes progress. This self-controlled retrospective cohort study evaluated patient medical records and images from two separate CAG sessions to evaluate dynamic changes of CAE over time.

Table 1 presents the basic CAE patient characteristics at the time of the first and second CAGs. All of the CAE patients in both groups exhibited atherosclerosis, and most presented with acute myocardial infarction (AMI). The blood pressure and lipid concentration measurements indicated decreasing tendencies in both groups. However, statistical significance was observed in only the 5–16-year follow-up group. The treatment information for all study subjects is provided. A subset of patients underwent revascularization with stent implantation (Supplementary Material).

The data in Table 2 show no obvious changes in the inflammation indicators for either group. CAG evaluations revealed increases in the left atrium dimensions and fractional shortening percentages, but no other characteristics were different over time.

Table 3 shows the coronary imaging characteristics of the enrolled subjects. The lack of change in the extent of ectasia was indicated by blood vessel involvement, Markis type, ectatic diameter, ectatic length, and corrected thrombolysis in the myocardial infarction frame count. The Gensini scores and blood vessel involvement assessment revealed increases in the extent of stenosis over time.

Figure 1 shows minimal changes in ectatic extent as indicated by the ectatic diameter and length. Coronary blood flow indicated by corrected thrombolysis in the myocardial infarction frame count showed an increasing tendency, and atherosclerotic progression was observed in all CAE patients. The detailed measurements of the first and second CAGs for all patients are listed in Table 3.

Table 4 shows the multiple linear regression results regarding the fold change of Gensini scores between the first and second CAGs (fold changes, equal to ratio of the second Gensini score to the first Gensini score). The results suggest that longer follow-up times, lower baseline Gensini score, and higher hypersensitive C-reactive protein concentration contributed to the increased Gensini scores.

Table 5 shows the relationship between ectasia and stenosis. The right coronary artery (RCA) was most frequently involved in cases with a combination of ectasia and stenosis. On the second CAG, there were increases in the percentage of patients with both stenosis and ectasia in the RCA (all 99 affected) and in the ratio of vessels with both ectasia and stenosis to all abnormal vessels (dilation and/or atherosclerosis).

Figures 2A,B show the Pearson correlation analysis of the relationship between ectasia diameter and Gensini scores for the first and t second CAGs. Gensini scores had a significant relationship with ectasia diameter at both time points (Pearson correlation coefficients and p-values were: 0.227, 0.024, 0.214, and 0.033, respectively).

In Figure 3, typical CAG images from two CAE patients showed there were minimal changes along with time for ectasia extent.

This self-controlled retrospective cohort study mainly explored whether CAE was a progressive disease by comparing clinical features and imaging data between patients' first and second CAG examinations. The results revealed that: (1) All CAE patients had evidence of atherosclerosis, and a majority presented with AMI. Most of the basic characteristics increased over time except blood pressure and lipid concentrations, which had decreasing tendencies. (2) There were no significant changes in inflammation indicators, but there were small changes in heart function including increases in left atrium dimensions and fractional shortening percentages. (3) Atherosclerosis increased significantly as indicated by the distribution of stenosis in coronary arteries and Gensini scores. However, the ectasia extent showed minimal change as reflected by blood vessel involvement, Markis type, ectatic diameter, ectatic length, and coronary blood flow. (4) Multilinear regression analysis revealed that the underlying factors for stenosis changes (as indicated by the fold change of Gensini score) were longer follow-up time, lower baseline Gensini scores, and higher hypersensitive C-reactive protein concentrations (5). The RCA was most frequently involved in the ectatic changes. The percentage of patients with both stenosis and ectasia in the RCA was higher after the second CAG. The ectatic diameter was related to the extent of stenosis, as indicated by the Gensini score at both CAG time points. The self-comparisons for subgroups including CAE with AMI and without AMI, as well as CAE with revascularization and without revascularization, were also applied and there were no changes for ectatic extent while the extent of stenosis progressed (Supplementary Material).

The first major finding was that there were no significant changes in ectatic extent and coronary blood flow in either follow-up group of 1–5 years (time interval, 3.04 [2.48–4.17] years) and 5–16 years (time interval, 8.15 [6.52–11.48] years). These results are clinically significant for several reasons. (1) CAE patients might have a longer natural history than presumed if their condition progressed very slowly. They might have experienced coronary ectatic changes at younger ages, although the average age at diagnosis was 55 ± 10 years (2). (2) CAE seemed to have a more intimate relationship with genetic abnormalities in patients who presented with coronary dilation at a younger age and did not have Kawasaki disease or other pathogenic factors (e.g., infections, percutaneous coronary intervention, etc.) (18). (3) Because of the slow progression of ectasia, one of major dangerous outcome might not be continuous rapid expansion even the coronary rupture at the ectatic sites like giant coronary artery aneurysm, which is currently considered an important issue by some researchers (15, 19). CAE might be different from aneurysms in the brain and aortic artery, which are prone to rupture and thus need to be remedied as soon as possible. Rather, the greatest damage might be thrombosis and slow blood flow in the coronary artery, which would result in AMI and unstable angina pectoris (20). (4) Percutaneous coronary intervention, especially stent implantation in the ectatic site, is seldom considered because of concerns about stent migration due to continuous dilation in patients with coronary ectasia (21). Our results suggest that stent implantation might be acceptable in CAE patients if the coronary was coexisted with ectasia and CAD.

The second important finding was significant progression of coronary atherosclerotic changes in CAE patients. Part of the patients were received stent implant outside of ectatic site after first CAG (Supplementary Material), their atherosclerotic changes also progressed while ectatic changes did not (Supplementary Material)… An important question is why CAE was not diagnosed at a younger age if it is a chronic progressive disease. A possible reason is that the clinical manifestations of CAE (e.g., angina, AMI, arrhythmia, and sudden death) might originate from not just ectasia but also other pathological changes such as stenosis progression. It is likely that there were no obvious clinical manifestations at the early stage because the stenosis was slight or absent. However, with atherosclerotic progression, the patients eventually developed the above-mentioned symptoms and were then diagnosed by CAG or other imaging techniques. Thus, the prevention and treatment of atherosclerotic changes might have more clinical significance than the prevention and treatment of ectatic changes. Indeed, there are no universally accepted treatment strategies for CAE. Most proposals are based on the treatments for CAD (22, 23). Antithrombotic agents, statins and anti-ischemic therapies are the main treatments because most CAE cases are associated with CAD, and approximately 50% of patients experience angina symptoms (24–26). In the present study, all CAE patients exhibited atherosclerosis, 48.48% (48/99) had experienced AMI, and 45.45% (45/99) had unstable angina pectoris. The AMI rate in CAE population was higher than in CAD patients (20), in those patients atherosclerotic changes also progressed while ectatic changes did not (Supplementary Material), there is no consensus which was the best anti-thrombotic strategy as the single antiplatelet agent, dual antiplatelet therapy (DAPT) or anti-coagulant therapy currently, CAE patient with larger coronary diameter and recurrent thrombotic event might require anti-coagulant therapy (27). The study results thus provide additional evidence to support the current CAD-based treatment strategy for CAE patients.

The relationship between CAE and CAD remains unclear. We observed that most ectatic changes were anatomically accompanied by stenosis, especially in the RCA. There was a significant relationship between ectasia and stenosis. There are two possible explanations for the relationship between CAE and CAD. The first is that CAE is a variation of CAD (14), with more than 80% of CAE patients also presenting with CAD (3, 28). CAE patients experience atherosclerotic changes in the intimal–media interface of the coronary artery. The vessel wall is infiltrated by several types of inflammatory cells (neutrophils, lymphocytes, plasma cells, eosinophils, and mononuclear cells) (8). It has therefore been suggested that the two diseases might share a pathological process and that CAE is a variation of CAD. The other standpoint, CAE is a risk factor for CAD for several reasons. (1) Pathologically, CAE is characterized by significant damage in the smooth muscle and elastic fiber layers of the coronary artery, while CAD is not (4). (2) Clinical epidemiology data demonstrated that most CAE patients experience CAD while few CAD patients exhibit CAE. In a large population-based survey with 10,524 cases, 87% of the CAE patients had CAD, but only 3% of CAD patients exhibited CAE (20). (3) The two diseases also have different progression rates. We found that ectatic changes occurred over a longer period in patient with CAE compared to those with CAD. Therefore, we prefer that CAE might present before CAD and is more likely to be an underlying risk factor for CAD progression. This hypothesis requires further exploration.

The data analyzed in this study is subject to the following licenses/restrictions: extra data is available by emailing Siwen Liang. Requests to access these datasets should be directed to Siwen Liang, 278713478@qq.com.

This study was approved by the Ethics Committee of Beijing Friendship Hospital Affiliated to Capital Medical University and was in accord with the Declaration of Helsinki (P2-013-02). Informed consent was not obtained from all the study participants because it was a retrospective research and no extra medical interventions were introduced for all subjects.

RL and SL: conceived the study and its design, conceptualization, formal analysis, resources, and software. SL: data curation. RL, XG, HZ, and SL: investigation and validation. XG: methodology. RL, HZ, and SL: project administration. All authors contributed to the article and approved the submitted version.

This study was sponsored by the National Natural Science Foundation of China (Grant No. 81600276).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.