Coronary Artery Ectasia Predicts Future Cardiac Events in Patients With Acute Myocardial InfarctionHighlights
Objective—Coronary artery ectasia (CAE) is an infrequently observed vascular phenotype characterized by abnormal vessel dilatation and disturbed coronary flow, which potentially promote thrombogenicity and inflammatory reactions. However, whether or not CAE influences cardiovascular outcomes remains unknown.
Approach and Results—We investigated major adverse cardiac events (MACE; defined as cardiac death and nonfatal myocardial infarction [MI]) in 1698 patients with acute MI. The occurrence of MACE was compared in patients with and without CAE. CAE was identified in 3.0% of study subjects. During the 49-month observation period, CAE was associated with 3.25-, 2.71-, and 4.92-fold greater likelihoods of experiencing MACE (95% confidence interval [CI], 1.88–5.66; P<0.001), cardiac death (95% CI, 1.37–5.37; P=0.004), and nonfatal MI (95% CI, 2.20–11.0; P<0.001), respectively. These cardiac risks of CAE were consistently observed in a multivariate Cox proportional hazards model (MACE: hazard ratio, 4.94; 95% CI, 2.36–10.4; P<0.001) and in a propensity score–matched cohort (MACE: hazard ratio, 8.98; 95% CI, 1.14–71.0; P=0.03). Despite having a higher risk of CAE-related cardiac events, patients with CAE receiving anticoagulation therapy who achieved an optimal percent time in target therapeutic range, defined as ≥60%, did not experience the occurrence of MACE (P=0.03 versus patients with percent time in target therapeutic range <60% or without anticoagulation therapy).
Conclusions—The presence of CAE predicted future cardiac events in patients with acute MI. Our findings suggest that acute MI patients with CAE are a high-risk subset who might benefit from a pharmacological approach to controlling the coagulation cascade.
Coronary artery ectasia (CAE) is a vascular phenotype that is infrequently observed in patients who undergo coronary angiography.1–6 It is characterized by abnormal vessel dilatation associated with disturbed coronary flow.7,8 Although CAE has been shown to be associated with enhanced thrombogenicity and inflammatory reactions, such as activation of tumor necrosis factor-α and interleukin-1β,8 the clinical significance of CAE has not been fully elucidated yet.
Although CAE is uncommonly seen on coronary angiography, several case reports showed acute occlusion at the ectatic coronary segment within infarct-related arteries in patients with acute myocardial infarction (AMI).9,10 In these cases, thrombectomy catheters retrieved substantial amounts of thrombus around ectatic lesions.10 Postmortem pathological studies also demonstrated spontaneous occlusion of dilated coronary arteries because of the presence of massive thrombi.1,5 Because dilatation of the coronary arteries disturbs coronary flow and thereby increases blood viscosity and activates coagulation,11 CAE may be a high-risk lesion causing acute coronary events. Furthermore, the enhanced thrombogenicity in CAE suggests a potential benefit of pharmacological agents for modulating the coagulation cascade to prevent CAE-related coronary events.12 However, the association of CAE with cardiac events and the clinical efficacy of anticoagulation therapy remain uncertain. Therefore, the current study sought to (1) investigate clinical outcomes in AMI patients with CAE, and (2) evaluate the efficacy of anticoagulation therapy on major adverse cardiac events (MACE).
Materials and Methods
Materials and Methods are available in the online-only Data Supplement.
Prevalence and Angiographic Characteristics of CAE
Patient disposition is shown in Figure 1. In the current study, CAE was observed in 89 coronary arteries of 51 patients (3.0% of patients; 95% confidence interval [CI], 2.1–4.3). Representative cases of classical and alternative definition of CAE is illustrated in Figure 2. The conventional definition of CAE was applied in 67% (n=34) of CAE subjects, and the alternate definition was applied in the remaining cases (33%, n=17). Overall, the CAEs were located in 29 infarct-related arteries and 60 noninfarct related arteries, respectively. One patient each in CAE subjects had polycystic kidney disease and Ehlers–Danlos syndrome. None of patients with CAE had any angiographic coronary dissection and fibromuscular dysplasia. Table 1 summarizes the clinical characteristics of CAE. CAE was more frequently observed within the right coronary artery (39 of 51; 76%), followed by the left circumflex artery (28 of 51; 55%), left anterior descending artery (22 of 51; 43%), and left main trunk (10 of 51; 20%). Fifty-nine percent of CAE subjects showed multivessel CAE involvement. With regard to angiographic features of CAE, oscillatory and static flow were observed in 47% (24 of 51) and 27% (14 of 51) of patients with CAE, respectively. Of particular interest was that 37% (19 of 51) of patients with CAE exhibited concomitant dilatation of other vessels, including thoracic aorta (8%), abdominal aorta (8%), iliac artery (4%), and intracranial arteries (8%). One patient (2%) had both iliac and intracranial artery aneurysms. Dissecting aortic aneurysm was identified in 4 patients (8%).
Baseline Clinical Demographics in Patients With CAE
Baseline clinical demographics in patients with and without CAE are shown in Table 2. Patients with CAE, compared with those without, were more likely to be younger (63±13 versus 68±12 years; P=0.005), men (84% versus 71%; P=0.04), more obese (body mass index ≥30 kg/m2; 12% versus 4%; P=0.008), and smokers (86% versus 71%; P=0.02) and to exhibit an increased number of coronary risk factors (2.9±1.1 versus 2.6±1.2; P=0.04) and higher left ventricular ejection fraction (49±8% versus 45±10%; P=0.003). Warfarin was more frequently used in patients with CAE (37% versus 13%; P<0.001), but these patients were less likely to receive dual antiplatelet therapy (22% versus 34%; P=0.04). The use of other established medical therapies was comparable between the 2 groups.
Lesion Characteristics and Percutaneous Coronary Intervention Procedures
Lesion characteristics and details on percutaneous coronary intervention procedures are summarized in Table 2 and Table II in the online-only Data Supplement. There were no significant differences in the numbers of diseased coronary vessels (1.8±0.7 versus 1.7±0.8; P=0.30) and the prevalence of triple vessel disease (20% versus 22%; P=0.64) between patients with and without CAE. The Gensini score in patients with CAE was significantly lower compared with those without CAE (48 [27–81] versus 58 [40–87]; P=0.03). Primary percutaneous coronary intervention was conducted in 82% and 74% of patients with and without CAE, respectively (P=0.17). Lesion characteristics and number of treated vessels were comparable between the 2 groups. In patients receiving primary percutaneous coronary intervention, the frequency of stent use was lower in individuals with CAE than those without (62% versus 91%; P<0.001), whereas those with CAE were more likely to be treated with an intra-aortic balloon pump (29% versus 14%; P=0.01). In patients not treated with stents, balloon angioplasty without stent use (31% versus 7%; P<0.001) and thrombectomy (31% versus 2%; P<0.001) were more frequently used in the CAE group. As a consequence, the proportion of patients who achieved Thrombolysis in Acute Myocardial Infarction flow grade 3 was significantly lower in CAE subjects (45% versus 88%; P<0.001; Figure 1).
Long-Term Outcomes in Patients With CAE
Figure 3, Tables 3 and 4, and Tables IV and V in the online-only Data Supplement compare the occurrence of MACE, cardiac death, and nonfatal myocardial infarction (MI) in the 2 groups. During the observation period (median, 49 months; interquartile range, 19–93 months), significantly higher incidences of MACE (hazard ratio [HR], 3.25; 95% CI, 1.88–5.66; P<0.001), cardiac death (HR, 2.71; 95% CI, 1.37–5.37; P=0.004), and nonfatal MI (HR, 4.92; 95% CI, 2.20–11.0; P<0.001) were observed in patients with CAE (Table 4; Table IV and Table V in the online-only Data Supplement). Of note, 5 of 7 nonfatal MI events in patients with CAE occurred at the segment with CAE in infracted-related artery (Table 3; Table III in the online-only Data Supplement). In patients with CAE, 2 patients occurred nonfatal MI at culprit lesion (29%=2/7), whereas 13 patients (33%=13/40) occurred in patients without CAE (29% versus 33%; P=0.84). After adjusting for covariates, CAE was still an independent predictor of MACE (HR, 4.94; 95% CI, 2.36–10.4; P<0.001), cardiac death (HR, 7.97; 95% CI, 2.85–22.3; P<0.001), and nonfatal MI (HR, 3.90; 95% CI, 1.61–9.46; P=0.003; Table 4; Tables IV and V in the online-only Data Supplement). These associations were further analyzed using propensity score–matching analysis, which selected 49 patients in each of the CAE and non-CAE groups. In this model, there were no significant differences in clinical characteristics between the 2 groups (Table VI in the online-only Data Supplement). Even in this propensity score–matched cohort, the frequency of MACE was still significantly higher in patients with CAE (HR, 8.98; 95% CI, 1.14–71.0; P=0.03).
Anticoagulation Therapy and MACE in Subjects With CAE
Of the 51 patients with CAE, 19 were treated with warfarin at the time of discharge. Of these, 8 achieved the percent time in the target therapeutic range (%TTR ≥60%). During the observation period, the occurrence of MACE was not observed in these patients. By contrast, the incidence of MACE was considerably higher in patients with %TTR <60% or in those who did not take warfarin (14 of 43; 33%; P=0.03; Figure 4).
Although CAE is associated with abnormal vessel dilatation, disturbed coronary flow, and enhanced thrombogenicity, its association with cardiac events has not been fully characterized. Our study demonstrated that (1) 3% of patients with AMI harbored CAE, and (2) patients with CAE more frequently developed nonfatal MI and cardiac death. These findings suggest that CAE may be an important morphological characteristic of coronary artery causing cardiac events.
CAE has generally been defined according to the ratio of the size of ectatic segment to that of the adjacent normal reference segment.4,13 However, as shown in Figure 2, this method is not applicable to patients with diffuse CAE because of the absence of normal reference segments. Krüger et al14 developed another definition of CAE that used the mean size of each coronary segment in an age- and sex-matched cohort without heart disease as a reference value. In our study, the conventional definition of CAE was inapplicable to 17 patients (33%), who were, therefore, diagnosed using Kruger’s definition. Consequently, we identified CAE in 3% of the study population, which was within the range of the previously reported prevalence (0.9%–5.3%).4,6,13,15,16 As such, patients with diffuse CAE might be underdiagnosed when the conventional definition alone is used. The use of both definitions is more rigorous but still practical for properly identifying ectatic segments within coronary arteries.
As reported previously, patients with CAE in this study exhibited distinct clinical characteristics, such as more frequent involvement of the right coronary artery,6,13,16,17 a lower frequency of stent implantation, and final Thrombolysis in Acute Myocardial Infarction flow grade 3.16,17 Moreover, enlargement of other vessels was identified in 37% of patients with CAE. Mechanistically, both atherosclerosis and nonatherosclerotic pathogeneses18–21 have been considered to cause CAE, and these potentially affect other systemic arteries as well. Although our CAE subjects were more likely to show clustering of coronary risks, 2 CAE cases had polycystic kidney disease or a connective tissue disorder accompanied by dilated intracranial arteries or descending thoracic aorta. As such, the influence of CAE-related pathogeneses on noncoronary vessels underscores the importance of meticulous screening of systemic arteries in CAE subjects.
The current study analyzing 1698 patients with AMI demonstrated that the presence of CAE was significantly associated with an increased risk of MACE during the 49-month observation period. Moreover, >70% of nonfatal MI in CAE subjects occurred at nonobstructive ectatic coronary segments. Mechanistically, dilatation of vessel size has been shown to slow coronary flow velocity, which increases blood viscosity and promotes coagulation.12 This pathophysiological element might result in thrombotic occlusion of ectatic coronary arteries. Another possible mechanism is an increased level of inflammatory cytokines, such as tumor necrosis factor-α and interleukin-1β, in CAE subjects.8 Because these cytokines stimulate the synthesis of tissue factor, which activates the coagulation cascade, a more inflamed substrate in CAE subjects might account for an increased risk of acute coronary events.
Our findings are different from previous studies which did not identified association between CAE and cardiac events.4,13,16,17,22–24 This might be partly because of analyzing patients with stable coronary artery disease in other published studies. Because risk of cardiac events in stable coronary artery disease subjects is lower compared with those with AMI, it may be difficult to detect significant differences in clinical outcomes between stable coronary artery disease subjects with and without CAE. Longer observational period in our study might have enabled to elucidate worse outcomes in patients with CAE.
In this study, subjects who achieved %TTR ≥60% with warfarin therapy exhibited a lower occurrence of MACE compared with those with %TTR <60% or without anticoagulation therapy. Given a significant reduction of thrombotic events in patients with optimal control of %TTR with warfarin,9 this observation might underscore the importance of carefully monitoring the international normalization ratio in CAE subjects. Further investigation is warranted to evaluate the clinical efficacy of anticoagulation therapy in CAE subjects.
Several caveats should be noted. This was an observational study at a single center that included relatively small numbers of patients with CAE and cardiac events. Despite these limitations, a multivariate Cox proportional hazards model and propensity score–matched analyses consistently showed a significant relationship between CAE and cardiac events. The frequency of statin use was only 55% because guidelines before 2007 did not recommend early statin therapy in patients with AMI. Management based on current therapeutic guidelines might affect the clinical outcomes in our study. Anticoagulation therapy was initiated according to each physician’s discretion, a procedure that is susceptible to selection bias.
In conclusion, the presence of CAE predicted the future occurrence of nonfatal MI and cardiac death in the setting of AMI. Our findings highlight CAE as a high-risk vascular phenotype requiring additional pharmacological approaches to optimally control the coagulation cascade.
We thank Hiromi Maeda of the National Cerebral and Cardiovascular Center for assistance with data acquisition and Yoko Masukata-Nakao for advice on statistical analysis.
Sources of Funding
This manuscript was sent to Robert Hegele, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.117.309683/-/DC1.
- Nonstandard Abbreviations and Acronyms
- acute myocardial infarction
- coronary artery ectasia
- confidence interval
- hazard ratio
- major adverse cardiac events
- percent time in target therapeutic range
- Received May 17, 2017.
- Accepted September 25, 2017.
- © 2017 American Heart Association, Inc.
- Hartnell GG,
- Parnell BM,
- Pridie RB
- Swaye PS,
- Fisher LD,
- Litwin P,
- Vignola PA,
- Judkins MP,
- Kemp HG,
- Mudd JG,
- Gosselin AJ
- Demopoulos VP,
- Olympios CD,
- Fakiolas CN,
- Pissimissis EG,
- Economides NM,
- Adamopoulou E,
- Foussas SG,
- Cokkinos DV
- Krüger D,
- Stierle U,
- Herrmann G,
- Simon R,
- Sheikhzadeh A
- Fujii T,
- Sakai K,
- Kimura M,
- Nakano M,
- Ohno Y,
- Nakazawa G,
- Shinozaki N,
- Matsukage T,
- Yoshimachi F,
- Ikari Y
- Boles U,
- Zhao Y,
- Rakhit R,
- Shiu MF,
- Papachristidis A,
- David S,
- Koganti S,
- Gilbert T,
- Henein MY
- Campanile A,
- Sozzi FB,
- Consonni D,
- Piscione F,
- Sganzerla P,
- Indolfi C,
- Stabile A,
- Migliorini A,
- Antoniucci D,
- Ferraresi R,
- Boccuzzi G,
- Danzi GB
- Zhang Y,
- Huang QJ,
- Li XL,
- Guo YL,
- Zhu CG,
- Wang XW,
- Xu B,
- Gao RL,
- Li JJ
Coronary artery ectasia (CAE) is characterized as abnormal vessel dilatation associated with disturbed coronary flow. Enhanced thrombogenicity and inflammatory reactions at ectatic coronary segment suggest that the presence of CAE may cause future coronary events.
In the current analysis, CAE was observed in 3.0% of patients with acute myocardial infarction.
The presence of CAE was associated with a higher occurrence of cardiac death and nonfatal myocardial infarction. This association was consistently observed by multivariate Cox proportional hazards model and a propensity score–matched cohort analysis.
Despite a higher risk of cardiac events in patients with CAE, the use of anticoagulation therapy which achieved an optimal percent time in target therapeutic range, defined as ≥60%, did not experience any occurrence of major adverse cardiac events.
Evaluation of CAE may enable to identify high-risk subjects who are more likely to develop cardiac death and nonfatal myocardial infarction. The efficacy of pharmacological intervention, such as anticoagulation therapy on cardiovascular outcomes, in patients with CAE should be investigated in future prospective randomized trial.