Prognostic association of plasma NT-proBNP levels in patients with microvascular angina -A report from the international cohort study by COVADIS-

1. Introduction

Up to half of patients undergoing diagnostic coronary angiography for typical chest pain have angiographically normal coronary arteries or non-obstructive coronary artery disease (CAD). These patients with signs and symptoms suggestive of ischemic heart disease with no obstructive coronary arteries (INOCA) are increasingly recognized.

In INOCA patients, coronary functional abnormalities could be involved, including epicardial spasm responsible for vasospastic angina (VSA) and/or coronary microvascular dysfunction (CMD) responsible for microvascular angina (MVA). MVA is typically defined as enhanced contraction (microvascular spasm) and/or impaired vasodilatation of coronary microvessels, leading to inadequate increase in blood flow in response to stress with resultant myocardial ischemia. Thus, CMD can cause myocardial ischemia in a sizeable proportion of angina.

Recently, several studies with either invasive or non-invasive techniques demonstrated that patients with MVA have significantly higher rates of cardiovascular events, as compared with non-anginal control populations, indicating the importance of their identification. As the COronary VAsomotor Disorders International Study (COVADIS) group, we have proposed the diagnostic criteria of MVA [10] and demonstrated the clinical characteristics and prognosis of MVA patients in our international prospective cohort study. Briefly, the diagnosis of MVA is based upon symptoms suggestive of myocardial ischemia in the absence of obstructive CAD (<50 % diameter reduction and/or FFR > 0.80) associated with objective evidence of myocardial ischemia and impaired coronary microvascular function defined by one of the following 4 findings, reduced coronary flow reserve (CFR), microvascular spasm, increased microvascular resistance, and coronary “slow flow phenomenon”. Our international MVA cohort study has demonstrated substantial risk of major adverse cardiac events (MACE), especially hospitalization for unstable angina, with hypertension and previous history of CAD being independent clinical predictors of MACE; however, prognostic biomarkers for MVA remain unclear.

We have previously suggested that CMD may play an important role in patients with heart failure with preserved ejection fraction (HFpEF) in the mechanism of ventricular hypertrophy and fibrosis, contributing to diastolic dysfunction and that natriuretic peptides modulates the effect of CMD leading to ventricular hypertrophy and fibrosis. B-type natriuretic peptide (BNP) is a cardiac neurohormone specifically secreted from the ventricles in response to volume expansion and pressure overload. Thus, plasma levels of BNP have been shown to be diagnostic and/or prognostic biomarker in patients with heart failure. On the other hand, inactive N-terminal fragment of pro-brain natriuretic peptide (NT-proBNP) has been reported to be a marker of long-term mortality in stable angina patients with obstructive coronary atherosclerosis. However, a possible relationship between plasma NT-proBNP levels and prognosis of MVA patients has not been investigated. Thus, in the present study, we aimed to determine whether plasma NT-pro BNP levels are a novel prognostic biomarker in patients with MVA.

2. Methods

2.1. International and prospective cohort study on MVA by COVADIS

Details of the international and prospective cohort study of MVA patients have been previously described. Briefly, this cohort study is a multinational, multicenter, multiethnic, prospective, observational, and longitudinal cohort study. We enrolled 686 eligible patients fulfilling the COVADIS diagnostic criteria for MVA as follows; (1) signs and/or symptoms of myocardial ischemia, (2) absence of obstructive CAD, (3) objective evidence of myocardial ischemia, and (4) evidence of impaired coronary microvascular function, as determined by the enrolling site (Supplemental methods, Supplemental Table S1).

Patients with obstructive CAD were excluded, which was defined as the presence of any coronary stenosis > 50 % on invasive angiography or computed tomography angiography. Evidence of myocardial ischemia was obtained by rest/stress ECG and/or non-invasive imaging by assessing either myocardial perfusion with single photon emission computed tomography (SPECT), positron emission tomography (PET), cardiac magnetic resonance (CMR), or left ventricular wall motion abnormality with stress echocardiography. Coronary microvascular function was assessed invasively by using coronary functional testing, including measurements of CFR and/or microvascular resistance and/or acetylcholine provocation testing for coronary microvascular spasm.

During the period from July 1, 2015 to December 31, 2018, the participating centers prospectively enrolled patients with MVA. All patients underwent clinical assessments and received usual medical care as determined by attending physicians. Follow-up of each patient was conducted at least once from study entry to the end of December 2019 either by a telephone call or a site visit, depending on the approach considered most practical and effective (Supplemental methods).

The ethics committee of Tohoku University Graduate School of Medicine approved the study protocol (No. 2015-1-188) followed by the ethics committee and/or sponsors at each participating institute, in compliance with the Declaration of Helsinki (UMIN000035177) (Supplemental methods).

2.2. Study population

Of the 686 patients registered in the COVADIS MVA cohort study, we finally included 226 consecutive patients who had both baseline plasma NT-proBNP levels and echocardiographic data including LV ejection fraction (LVEF) and E/e’ (Fig. 1). At enrollment, we also obtained clinical details, including patient demographic profiles, cardiovascular risk factors, past history of CAD including acute coronary syndrome and stable angina pectoris, non-invasive markers of myocardial ischemia, invasive assessment of microvascular function, initial treatment after diagnosis and assessment of health status by the Seattle Angina Questionnaire (SAQ).

Fig. 1. Patient enrollment and follow-up.

2.3. Study endpoints

The primary endpoint was the composite of major cardiovascular events (MACE), including cardiovascular death, non-fatal myocardial infarction (MI), non-fatal stroke, and hospitalization due to heart failure or unstable angina (UA), which were determined by the institutional investigators at each site or an independent clinical event committee. Definition of MI was based on the Third Universal Definition, and that of UA in the presence of ischemic chest pain and hospitalization within 24 h of most recent symptoms, without elevation in cardiac biomarkers but with ischemic ECG changes. Stroke was defined as neurological deficit due to an ischemic or hemorrhagic central nervous system event with residual systems > 24 h after onset or leading to death. For each patient, a MACE was defined as the first occurrence of one of these events during follow-up period. The associations between plasma NT-proBNP levels, echocardiographic data, and MACE were evaluated.

2.4. Statistical method

Continuous variables are presented as mean ± SD or medians and interquartile range, depending on the distribution of the data that was tested by Shapiro-Wilk normality test. Categorical variables are presented as counts and percentages. We used the Wilcoxon rank-sum test to compare continuous variables and the Pearson chi-square test to compare categorical variables. Events were analyzed as time from enrolment to first occurrence of any event from the composite endpoint. We used the Kaplan-Meier method to provide survival estimates, which were assessed with a log-rank test. C-statistics, which equal to the area under the Receiver Operating Characteristic curve (AUC), was used to summarize the performance of the predicted probability of the outcomes for discrimination. Additional supportive analyses included time to first occurrence of each component of the composite endpoint individually. Event rate of the composite endpoint and that of each of endpoint are reported separately at 1, 2, and 3 years since enrolment. To examine the association between plasma NT-proBNP levels and incidence of primary endpoint, we used multivariable logistic regression model. A P-value < 0.05 was taken as statistically significant.