Persisting chest pain in nonobstructive coronary artery disease

This editorial refers to ‘Natural history of patients with insignificant coronary artery disease’, by R. Tavella et al., on page 117. Chest pain is the most common symptom of coronary artery disease (CAD) prompting subjects to seek attention from physicians. Angina is an important predictor of outcomes and to a large extent determines health-related quality of life (HRQoL) even in the overall apparently health population.1 Furthermore, physical disability caused by angina has an impact on prognosis as well as on HRQoL.2 Early reports have demonstrated the value of supplementing the clinical presentation of angina with angiographic information.3 Angiography does not miss significant life-threatening disease. The severity and the extent of obstructive coronary lesions demonstrated by angiography are powerful predictors of death. Clinical presentation is also a strong predictor of mortality, as type and severity of chest pain act indirectly as predictors because of their association with the severity of the coronary stenosis. For this reason, studies designed to evaluate the influence of medical or interventional therapy on survival and/or HRQoL should utilize angiographic findings in the selection of groups of patients. The results of the study by Tavella et al.4 are at variance with these ‘rules of thumb’. These authors looked at patients with stable CAD and reported 1% of recurrence of myocardial infarction in a group of 253 patients with chest pain and non-obstructive CAD who were followed for 12 months. Similar outcomes were found in those patients labelled as obstructive CAD who, indeed, had clinically negligible differences in prognosis: 1% death and 1% myocardial infarction during the same time span. Similarity was even more when evaluating HRQoL status in the two groups of patients. These findings raise several important issues. The population reported by Tavella et al.4 was highly selected in that the 253 patients with non-obstructive CAD and their counterpart with obstructive CAD (n = 505) were drawn from 1146 patients referred for evaluation of their chest pain from April 2003 to May 2007. Patients with non-obstructive CAD were young (average 58 years); there was no large sex differences in prevalence of disease (57% of women); many patients had survived prior manifestations of acute cardiovascular disease (9% cerebrovascular disease, 4% myocardial infarction, 8% acute chest pain admission, and 12% prior angiography) and many had reached a stable stage of chronic angina (54% with Canadian Cardiovascular class II-IV). Previous large studies on this issue have been done either in patients with acute coronary syndrome5,6 or in patients with suspected angina.7 The latter have mainly focused on women with the exclusion or under-representation of men. The extrapolation of the data of the current study to populations of dissimilar characteristics with regard to selection or composition is not possible. The results of the HRQoL analysis from the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial in patients with stable angina and obstructive CAD are in somewhere different, but also in somewhere consistent with the message from this study. Optimal medical therapy either with or without routine early percutaneus intervention (PCI) rapidly improves the patients' Seattle Angina Questionnaire (SAQ) scores for physical limitation, angina frequency, and overall HRQoL. The gains with routine PCI are significantly greater than those with optimal medical therapy only throughout the first year of follow-up.8 Later PCI advantages disappear shortly and no significant differences in HRQoL scores are seen between the two treatment strategies. In the study Tavella et al.,4 the proportion of patients still with symptoms after the first year of follow-up was high in both the non-obstructive and obstructive CAD patients (48 and 59%, respectively). Many patients with obstructive CAD (40%) had a past cardiac history of revascularization, and presumably have repeated interventions during follow-up. These data leave open the question of whether a PCI-first strategy is justified, but also leave open two further questions: what the reasons of chest pain are in these patients? And what the best therapeutic medical strategy is in both obstructive and non-obstructive CAD? Angina is not a disease; it is a symptom of an underlying heart problem. Persistence of angina symptoms is not normal and is inevitably associated with an impairment of HRQoL. Common causes of recurrent chest pain in obstructive CAD can be summarized as causes linked to the anatomy of the vessels (in-stent restenosis, in-stent thrombosis, incomplete revascularization, rapid progression of coronary atherosclerosis in non-culprit lesions) or to the function of the vessels (microvascular dysfunction, diffuse epicardial coronary vasoconstriction, or focal segmental epicardial coronary spasm), or a combination of both. Reports of studies from our institution have shown that relief of a significant stenosis by PCI in stable angina globally improves regional and global myocardial blood flow at rest.9 Flow improvement, however, was not apparent at the time of revascularization, but at 6 months' follow-up. As coronary artery diameters of the non-target vessels did not significantly differ before and after PCI and at 6-month follow-up, late upturn of the microcirculation was assumed to be the reason for delayed recovery of myocardial perfusion. Abnormal flow recovery following PCI can also be demonstrated by single-photon emission computed tomography. Yet after successful PCI, perfusion defects still persist in the absence of restenosis on an average of 6 months, and can be reversed by statin treatment.10 It follows that functional abnormality of the coronary vessels is the major culprit in chronic stable angina either in patients presenting with or without obstructive lesions. Indeed, many of these patients exhibit a disorder of the entire coronary arterial tree. Non-obstructive coronary artery disease may overlap with other clinical conditions known as ‘syndrome X’,11 and ‘vasotonic angina’.12 These patients continue to have recurrent chest pain often despite maximal anti-ischaemic treatment, thus they are substantially limited in everyday life and consume a great deal of healthcare resources.7 Angina, despite non-obstructive CAD, has traditionally been thought of as a female affliction. A remarkable finding from the study by Tavella et al.4 is demonstrating that it equally affects women and men. Because most studies have either included only women, or had a small sample size of men, it is currently unclear whether there are sex differences in the HRQoL status of patients with non-obstructive CAD. The expected consequence is the assumption that whatever are the findings for women, the results would also hold true for men. The study by Tavella et al.4 shed some light on this issue. Backward elimination of baseline predictors for ongoing chest pain at 12 months, removed all variables except female gender. Therefore, it seems quite apparent that women who have chest pain and native or post PCI normal or near normal coronary arteries do not have a benign prognosis at least in terms of persistence of symptoms and HRQoL status. Significant coronary plaque burden can be present despite an apparent normal or near normal coronary lumen. Women are much more likely than men to develop endothelial dysfunction, which may provide a link between a seemingly benign normal coronary angiogram and an unanticipated associated increased risk of subsequent episodes of angina and perhaps major future cardiac events. Finally, the data by Tavella et al.4 suggest that medical compliance, particularly with multiple medications, is considerably less in the ‘real world’ than that it has been observed in the COURAGE trial. At 1 year follow-up, 90% or more of the medically treated cohort in COURAGE was compliant with aspirin, statin, and beta-blocker therapies. This level of protocol-driven therapeutic compliance stands in contrast with the registry data of the study by Tavella et al. In this registry, 26–35% of patients with non-obstructive and obstructive CAD, respectively, were compliant with beta-blockers, 40–69% with statins, and 65–85% with antiplatelet agents. No information on the use of multiple concomitant medications is available. In this context, several important questions arise. First, and foremost, are the results of medical therapy as achieved in the COURAGE trial applicable to general clinical practice? ‘Failure’ of medical therapy to meet the hypothesized remission of symptoms is neither surprising nor new if medications are underused. In this registry data, medical therapy was not as intensive as currently possible. Second, was the choice of medical therapy ‘optimal’ in this registry? The time of enrolment of the registry is from 2002 to 2007. Medical therapy and PCI have each improved from 2007. Medical therapy has changed the most in terms of intensive use of ancillary antiplatelet and anticoagulant drugs. Little data exist from the study of Tavella et al.4 to guide decision-making in future stable angina patients. Finally, should medical therapy for obstructive and non-obstructive CAD be similar? Unfortunately, there are few studies on patients with non-obstructive CAD. Such patients have atherosclerosis and, therefore, would presumably benefit from treatment with all secondary prevention drugs. Beta-blockers have been shown to treat effectively ischaemic chest pain.13 Available data seems to show that nitrates are ineffective in this population.14 The best therapeutic option remains largely speculative, even though recent available data on ACE inhibitors suggest this medication to be used as first-line therapy.15 Clearly, the relationship between ischaemia, treatment approach, and symptoms is worthy of further investigation. Inconsistent information on therapy limits reliability of conclusions on lack of change in HRQoL. Conflict of interest: none declared.