Mechanisms and Diagnostic Evaluation of Persistent or Recurrent Angina Following Percutaneous Coronary Revascularization

Filippo Crea; Cathleen Noel Bairey Merz; John F. Beltrame; Colin Berry; Paolo G. Camici; Juan Carlos Kaski; Peter Ong; Carl J. Pepine; Udo Sechtem; Hiroaki Shimokawa; On behalf of the Coronary Vasomotion Disorders International Study Group (COVADIS)


Eur Heart J. 2019;40(29):2455-2462. 

In This Article

Mechanisms of Persistent or Recurrent Angina After Percutaneous Coronary Intervention

Elective PCI in patients with stable angina is routinely successful and procedure-related problems are uncommon. Procedural success reflects effective education and training, the availability of state-of-the-art stent technologies, use of adjunctive intravascular imaging and physiology techniques, and quality assurance.[21]

Despite procedural success, angina may persist or recur in a large proportion of patients, which is frustrating for patients and clinicians. In contemporary practice, clinicians are focused on identifying a 'flow-limiting' stenosis amenable to PCI with stents when sometimes other potential causes are relevant. This issue could be considered as 'detection bias'.

Yet, the pathophysiology of persisting or recurring angina after successful PCI is heterogeneous. Aside from non-cardiac causes of angina, structural and functional alterations in the coronary circulation may be implicated (Table 1 and Figure 1).

Figure 1.

Structural and functional alterations of coronary circulation responsible for persistence or recurrence of angina after percutaneous coronary intervention.

Flow Limiting Epicardial Obstructions

Stent thrombosis and in-stent restenosis are infrequent causes of recurrent angina and ischaemia after PCI in contemporary clinical practice with rates of stent thrombosis <1% at 1 year and 0.2–0.4% per year thereafter, and rates of clinically relevant in-stent restenosis of 5% at 1-year follow-up.[22] Recurrence rate of major adverse cardiac events due native vessel disease progression is <5% at 1-year follow-up,[23] although disease progression and cardiac events are higher in patients with diabetes.[24]

Persistence of angina caused by incomplete coronary revascularization may occur in up to 30% in the current era, although definitions of incomplete revascularization are heterogeneous.[25]

Other potential causes of angina post-PCI include diffuse atherosclerosis without a focal stenosis leading in turn to an insidious pressure gradient along the length of the coronary artery, coronary dissection, myocardial bridging, and thromboembolism, all of which may have been overlooked during the index procedure.

Coronary Vasomotion Disorders of Epicardial Arteries and Microcirculation

Functional causes of persistent or recurrent angina following a technically successful PCI include vasomotion disorders of epicardial coronary arteries, as well as coronary microvascular dysfunction (CMD). Accordingly, current European Society of Cardiology guidelines on stable CAD have acknowledged the role of coronary vascular dysfunction in causing angina post-PCI.[1] If vasomotor function has not been specifically assessed in a patient with angina before undergoing PCI, it is impossible to know the time-course of these functional alterations. Of note, stent implantation may cause or enhance coronary vascular dysfunction increasing the propensity of epicardial coronary vasospasm and/or CMD. Overall, these functional mechanisms may be causally implicated in about half of patients with angina post-PCI although, again, prospective studies are lacking.[3]

Significant constriction of epicardial coronary arteries at or distal to the PCI site is a potential cause of recurrent angina.[26] Ong et al.[27] documented epicardial spasm (>75% narrowing), associated with reproduction of patient's symptoms, in response to intracoronary acetylcholine (ACh) at increasing doses; up to 200 μg for the left coronary artery or with 80 μg for the right coronary artery (Figure 2). Significant constriction occurred in about half of the patients undergoing coronary angiography for post-PCI recurrent angina without haemodynamically significant coronary stenoses. These observations support a mechanistic role for epicardial coronary artery constriction in patients with angina persistence or recurrence post-PCI. Heightened activation of the Rho-kinase pathway, a central molecular mechanism for vascular smooth muscle constriction, might also be involved in the pathogenesis of DES-related enhanced epicardial artery constriction.[28]

Figure 2.

(A and B) Epicardial coronary spasm: representative images of a 55-year-old male patient with several previous stent implantations in the right coronary artery (red lines). The patient-reported recurrent resting angina and underwent invasive coronary angiography. The right coronary artery showed mild in-stent proliferation. Intracoronary acetylcholine testing revealed spasm of the right coronary artery distal to the stents at 80 μg with reproduction of the reported symptoms (A). After intracoronary nitroglycerine injection (400 μg) the spasm and the symptoms resolved (B). (C and D) Coronary microvascular spasm: representative images of a 64-year-old female patient with previous stent implantation in the left anterior descending artery (red lines). The patient-reported recurrent angina at rest as well as during exertion and underwent invasive coronary angiography. The left anterior descending showed mild in-stent proliferation. Intracoronary acetylcholine testing revealed no epicardial spasm at 200 μg, but reproduction of the reported symptoms together with ischaemic ECG shifts on the simultaneously recorded 12-lead ECG (C). After intracoronary nitroglycerine injection (200 μg) the ischaemic ECG changes and the symptoms resolved (D).

Coronary microvascular dysfunction in post-PCI angina, Type 4 by the classification of Camici and Crea,[29] can be caused by impaired microvascular dilation resulting in a reduction of coronary flow reserve (CFR), similar to that caused by an epicardial stenosis, or by microvascular spasm (Figure 2). In a study of patients with post-PCI recurrent angina compared with matched patients without angina post-PCI, Li et al.[30] observed a more profound reduction in coronary blood flow (CBF) in association with an increase in the index of microvascular resistance (IMR) in response to intravenous adenosine (140 μg/kg/min). The impairment of hyperaemic CBF and increase in IMR was even more relevant in those with abnormal exercise stress test result and persisted at 6-month and 12-month follow-up. These findings are consistent with studies using transthoracic Doppler echocardiography by Milo et al.[31] In their study, hyperaemic coronary artery blood flow velocity in the left anterior descending (LAD) was impaired at 1 day, 3 months, and 6 months after successful PCI compared with controls. Importantly, similar impairment of the CBF response to cold pressor testing was also observed, suggesting impaired endothelium-dependent vasodilator function of the coronary microcirculation. Furthermore, a greater impairment of the CBF response predicted restenosis in the LAD coronary artery during long-term follow-up.[32] Among patients who underwent PCI with second-generation DES, Hokimoto et al.[33] found impaired CBF response to both ACh (an endothelium-dependent stimulus) and adenosine (mostly an endothelium-independent stimulus). Overall, in this study, evidence of CMD was found in 59% of patients with previous PCI. Finally, Ong et al.[27] suggested coronary microvascular spasm might contribute to post-PCI recurrent angina in approximately one in five affected patients. Specifically, following intra-coronary infusion of ACh, affected patients experienced angina and ischaemic electrocardiographic (ECG) changes without evident epicardial constriction, suggesting that microvascular spasm rather than epicardial constriction was causal for the angina.

Mechanisms underlying CMD post-PCI are poorly understood, heterogeneous, and potentially more than one problem may be operative in any given patient.[34] First, chronic coronary microcirculation adaptation to reduced perfusion pressure distal to a stenosis or to an occlusion would be expected to negatively influence microvascular remodelling and its capacity to maximally dilate after restoration of a more physiologic perfusion pressure. Furthermore, the time required for the coronary microcirculation to recover to baseline vasomotor reactivity is likely to be variable among patients.[35]

Second, PCI might contribute to CMD by causing microembolization by debris material, which may obstruct small coronary arteries, arterioles, and/or capillaries to cause perivascular inflammation and even capillary obliteration and cardiomyocyte injury.[36] Furthermore, the DES per se causes vasoconstriction in the adjacent and downstream coronary macro- and microcirculation, potentially due to elution of the active drug and polymer constituents downstream. Although these detrimental effects of DES might be mediated by endothelial dysfunction,[37] more recent studies suggest that they are mainly mediated by activation of Rho-kinase in smooth muscle cells and, accordingly, are prevented by Rho-kinase inhibitors.[38,39]

Third, CMD may have pre-dated the index PCI procedure, type 3 by the classification of Camici and Crea,[29] which was 'unmasked' when the epicardial obstruction was relieved by PCI. In particular, CMD with microvascular rarefaction is implicated in heart failure with preserved ejection fraction.[40,41]

Last but not least, coronary microcirculation regulates myocardial blood flow, and therefore, shear stress in large epicardial arteries which in turn directs vessel remodelling and plaque formation.