Forget Ischemia: It’s All About the Plaque
The introduction of invasive coronary angiography in the 1960s was a major step forward in the understanding and treatment of coronary artery disease. However, high initial complication rates and limited availability led to the widespread development of ischemia testing techniques as safer, noninvasive surrogate methods of identifying obstructive coronary artery disease. Subsequent studies demonstrated the powerful prognostic value of ischemia testing. These advances have shaped clinical cardiology practice over the past 50 years.
The prognostic benefits of coronary revascularization have been widely tested and debated. The aptly named COURAGE trial (Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation) caused major controversy when it reported that percutaneous coronary intervention had no prognostic benefits in the prevention of cardiovascular death or myocardial infarction (MI) in patients with stable coronary artery disease. Many observers criticized the trial for the inclusion of patients with mild disease and suggested that differences would have been observed in those with more widespread ischemia and perhaps with the use of coronary artery bypass graft surgery. A subsequent substudy tentatively suggested that coronary revascularization may have some benefit in those with a high ischemic burden. Thus, ISCHEMIA (International Study of Comparative Health Effectiveness With Medical and Invasive Approaches) was conceived and undertaken. The main findings of ISCHEMIA appear to recapitulate the findings of the COURAGE trial: no prognostic benefit was found from coronary revascularization in patients with stable coronary artery disease and myocardial ischemia after a median of 3 years of follow-up.
In this issue of Circulation, 2 important substudies of ISCHEMIA are presented by Reynolds et al.[4,5] In one analysis, ISCHEMIA investigators examined both the prognostic predictive power and the potential trial treatment interactions according to the degree of ischemia demonstrated by stress testing (predominantly nuclear perfusion imaging) and the anatomic severity of coronary artery disease (predominantly the number of obstructive stenoses on coronary computed tomography angiography). Their findings are provocative. They demonstrate that the extent of ischemia is a poor discriminator of risk for most clinical end points, although a small hazard can be documented for nonfatal MI. In contrast, irrespective of ischemic burden, the severity of coronary artery disease was a strong predictor of outcome across nearly all clinical end points, including all-cause death, cardiovascular death, and MI. The superiority of anatomic assessments over functional ischemia testing has been increasingly demonstrated in other trials, including the SCOT-HEART trial (Scottish Computed Tomography of the Heart) and PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain).[6,7] In these 2 trials, similar differential effects and marked superiority of coronary computed tomography angiography in predicting events have been well documented.[8,9] From first principles, this should not be a surprise. Spontaneous MI is caused by rupture of either nonobstructive or obstructive plaques. Therefore, why would ischemia testing be superior to an anatomic test given that it is plaque instability, not reduced perfusion, which causes MI? As again demonstrated by Reynolds and colleagues, ischemia and coronary artery disease severity are correlated and inextricably linked, but ischemic burden is an imprecise and insensitive surrogate marker of coronary plaque burden. These study findings are especially surprising given that less than half of the study population had an interpretable assessment of the severity of coronary artery disease, whereas ischemic burden was available in 99% of the study population. So even with less than half the power, measures of coronary artery disease severity remained dominant. However, it should be acknowledged that ischemia was an entry criterion for the trial and therefore the full range of ischemia was not represented and the prognostic effect of ischemia may be underappreciated. This was also true for the severity of coronary artery disease because the entry criteria included the requirement for obstructive coronary artery disease. Because of these factors, the magnitude of the hazard ratios for both ischemia burden and severity of coronary artery disease were attenuated compared with broader, less restricted populations.[8,9]
When it comes to treatment effects within the trial, it is perhaps unsurprising that ischemia did not demonstrate an interaction with the trial intervention, given its poor predictive performance. This essentially disproved the previous COURAGE ischemia substudy findings. However, in patients with severe coronary artery disease on computed tomography, coronary revascularization reduced cardiovascular death or MI by more than one-third, with similar trends seen for the primary end point, but no effect on overall mortality. This is particularly impressive given that several factors would have hindered the detection of a treatment effect. First, less than half the study population contributed to this analysis, perhaps contributing to the lack of a statistically significant treatment interaction. Second, the assessment of severity of coronary artery disease was based on the rather outdated use of stenosis severity. Although this is arguably more relevant to a trial of ischemia and coronary revascularization, plaque burden and plaque type are much stronger independent predictors of outcome,[10,11] especially given that most MIs are caused by nonobstructive plaques or those with high-risk plaque features. Perhaps greater differential effects would have been seen if more modern and sensitive measures of anatomic disease severity were used. Third, the length of follow-up is relatively short and spontaneous plaque rupture events occur slowly. Last, we should remember that coronary revascularization was achieved by percutaneous coronary intervention in three-quarters of cases. This does little to protect the coronary circulation from plaque rupture and merely relieves the hemodynamic flow obstruction. As the COURAGE trial amply demonstrated, percutaneous coronary intervention does not reduce rates of MI in patients with stable disease. In contrast, coronary artery bypass graft surgery does reduce the long-term rates of MI and all-cause mortality.This is because long segments of proximal coronary artery plaque are bypassed by the grafts and most plaque rupture events upstream of the bypass anastomosis are unlikely to lead to MI. It will be interesting to see whether ISCHEMIA confirms these benefits in the longer term and whether this could be attributable to coronary artery “plaque” bypass surgery.
In their second analysis, Reynolds and colleagues report the findings of a registry of patients who were excluded from ISCHEMIA because they had evidence of ischemia but no obstructive coronary artery disease. The method of ischemia testing in this substudy was stress echocardiography and all images were reviewed and scored by a blinded core laboratory. There were a number of notable observations. Exercise-limiting angina was only present in 20% of participants on formal stress testing and most participants had relief of their symptoms by 1 year. Moreover, the stress-induced wall motion abnormalities resolved at follow-up in most patients, with very few participants having evidence of worsening abnormalities. Last, the participants had a much better prognosis than did those with obstructive coronary artery disease.
What do we make of these findings? As with any test, a false-positive stress echocardiogram is an important concern. There is also clear evidence of regression to the mean, which is common to all studies that have entry criteria for the subject under investigation. However, for a modest percentage of patients, there were continuing debilitating symptoms associated with objective evidence of persistent stress-induced cardiac dysfunction. Many theories abound, such as microvascular angina, and yet we still have little understanding regarding the underlying pathophysiology of these phenomena or how best to treat them. This is important given the substantial morbidity associated with severe, persistent cases. However, for the majority, this condition is self-limiting and associated with a relatively good prognosis. This analysis by Reynolds and colleagues highlights that systematic evaluation is important to document the presence of ischemia and nonobstructive plaque. The former is needed to provide a clear diagnosis and rational approach to the symptomatic management of those with persistent symptoms and, for the latter, to ensure that nonobstructive coronary artery disease is treated with appropriate preventative therapies. For those with documented ischemia and persistent symptoms, antianginal therapy using vasodilator agents remains an imperfect empirical choice, which lacks definitive demonstrable efficacy but aims to improve the patient’s quality of life in the absence of other proven interventions.
These ISCHEMIA substudies reinforce the growing consensus that ischemia testing is a suboptimal surrogate measure of coronary artery disease severity and that anatomic imaging, preferably with noninvasive coronary computed tomography angiography, provides superior diagnostic information and risk stratification in patients with atherosclerotic coronary artery disease. It is highly likely that measures of plaque burden and type will further enhance risk prediction and may be a future method of determining the intensity of therapies and interventions for the prevention of MI. For ischemia testing, there remains an important role for the objective assessment of persistent symptoms and their management, especially where anatomy or presence of ischemia is uncertain. We look forward to the long-term outcomes of ISCHEMIA. If further long-term benefits are confirmed, this likely will be attributable to the application of coronary artery “plaque” bypass surgery, especially in patients with 3-vessel coronary artery disease. However, the plaque stabilizing effects of guideline-directed medical therapy are hard to beat and the primary role of ischemia testing and coronary revascularization is to provide symptom relief when other antianginal therapies and interventions have failed.
Sources of Funding
Drs Newby, Williams, and Dweck are supported by the British Heart Foundation (FS/ICRF/20/26002, FS/14/78/31020, CH/09/002, RG/16/10/32375, RE/18/5/34216, and FS/SCRF/21/32010). Dr Newby is the recipient of a Wellcome Trust Senior Investigator Award (WT103782AIA). Dr Williams is supported by The Chief Scientist Office of the Scottish Government Health and Social Care Directorates (PCL/17/04). Dr Dweck is the recipient of the Sir Jules Thorn Award for Biomedical Research 2015.
Circulation. 2021;144(13):1039-1041. © 2021 American Heart Association, Inc.