The Institutional Review Board of Korea University Guro Hospital specifically approved this entire study and all the consent procedures (No. KUGH10045). The authors of this manuscript certify that the information contained herein is true and correct as reflected in the records of the Institutional Review Board.

A total of 10,177 consecutive patients with typical or atypical chest pain were enrolled between November 2004 and May 2015 and underwent coronary angiography (CAG) at the Cardiovascular Center of Korea University Guro Hospital, Seoul, Republic of Korea. Among these, 6,430 patients with typical or atypical chest pain without significant coronary artery disease (defined as a stenosis diameter of less than 70% on quantitative coronary angiography) underwent an acetylcholine (ACH) provocation test. Ineligible patients were excluded, as presented in Fig. 1, if they had conditions that could be major causes of adverse cardiovascular events and could bias the results. In total, 3,072 CAS patients were enrolled in the study and divided into two groups based on whether they took LDA: the LDA group (n=338) and non-LDA group (n=2,734) (Fig. 1).

Significant CAS was defined as having more than 70% of luminal narrowing of the artery during the ACH provocation test regardless of the presence of ischemic changes on electrocardiography (ECG) changes or chest pain. Deaths were regarded as being of cardiac causes unless a noncardiac death could be confirmed. Repeated CAG (mostly due to recurrent angina) was performed in patients who complained of recurrent angina despite having received adequate anti-anginal medication for at least 6 months since the onset of first CAG. In these cases, the physician assumed that CAS may have progressed or there may have been newly developing atherosclerotic coronary artery disease [5]. Major adverse cardiovascular events (MACE) were defined as the composite of total death, myocardial infarction (MI), and revascularization (including percutaneous coronary intervention [PCI] and coronary artery bypass graft [CABG]) [11]. Major adverse cardiovascular and cerebrovascular events (MACCE) 1 were defined as the composite of MACE and stroke events [11]. MACCE 2 were defined as the composite of MACCE 1 and recurrent angina.

The design of the ACH test has been described before [4–8]. In brief, CAS induction was tested by an intracoronary injection of ACH immediately after diagnostic angiography by either a transradial or transfemoral approach. ACH was injected by incremental doses of 20 (A1), 50 (A2), and 100 (A3) µg/min into the left coronary artery over a 1-minute period with 5-minute intervals up to the maximum tolerated dose under continuous monitoring by ECG and measurements of blood pressure. Routine provocation tests of the right coronary artery were not done due to safety issues regarding the higher prevalence of advanced atrioventricular block, which needs a temporary pacemaker for maintaining an adequate ACH infusion rate and cost-effectiveness for the diagnosis and management of significant CAS. Angiography was repeated after each ACH dose until a significant focal or diffuse narrowing of more than 70% was observed. An intracoronary injection of 0.2 mg of nitroglycerine was administered after completing the ACH provocation test, followed by CAG 2 minutes later. If significant focal or diffuse vasoconstriction (>70%) of the coronary arteries was induced at any dose, the ACH infusion was stopped. End-systolic images for each segment of the left coronary artery were chosen according to the corresponding points on the ECG trace (QRS onset or end of T wave) and analyzed using the quantitative coronary arteriography system of the catheterization laboratory (FD-20; Phillips, Amsterdam, Netherlands).

For continuous variables, differences between the two groups were evaluated by the unpaired t-test or Mann-Whitney rank test. Data were expressed as mean±standard deviation. For discrete variables, differences between the two groups were expressed as counts and percentages and analyzed with the chi-square or Fisher exact test. To adjust for any potential confounders, propensity score matching (PSM) was performed using a logistic regression model. We tested all available variables that could be of potential relevance: age, sex, cardiovascular risk factors (hypertension, diabetes, dyslipidemia, current smokers, and current alcohol drinkers), angiographic and clinical parameters (myocardial bridge, ACH dose [20, 50, and 100 µg/min], CAS site [left arterial descending, left circumflex], number of CAS vessels, CAS length, ECG changes, chest pain, and atrioventricular block), and medical treatment (renin-angiotensin system inhibitors, calcium channel blockers, diltiazem, nitrate, trimetazidine, molsidomine, beta-blockers, diuretics, aspirin, and statins). The propensity score was estimated using the C-statistic in the logistic regression model, and the propensity score for the two groups was 0.827. Matching was performed using a 1:1 matching protocol without replacements (nearest neighbor matching algorithm), with a caliper width equal to 0.05 of the standard deviation of the logit of the propensity score. Various clinical outcomes were estimated with the Kaplan-Meier method, and differences between the groups were compared with the log-rank test before and after PSM. A proportional-hazard model was used to assess the hazard ratio of the LDA group compared with the non-LDA group in the PSM population. For all analyses, a two-sided P<0.05 was considered to indicate statistical significance. All data were processed with IBM SPSS ver. 20.0 (IBM Corp., Armonk, NY, USA).

After diagnosis, CAS patients were prescribed anti-anginal medication such as nitrate, calcium channel blockers (including diltiazem), and/or nicorandil for at least 6 months depending on the physician’s discretion.

The primary endpoint was the cumulative incidence of major clinical endpoints such as various composite events of total death, MI, PCI, CABG, stroke, and repeat CAG. The secondary endpoint was recurrent angina. In this study, the mean follow-up period was 1,216±597 days (after PSM, 1,318±561 days), and we could follow up the clinical data of all enrolled patients through face-to-face interviews at regular outpatient clinic visits, medical chart reviews, and telephone contacts.

In this study, a total of 3,072 CAS patients were enrolled, among whom 11.0% received LDA (Fig. 1). The patients’ baseline clinical characteristics are shown in Table 1, and their angiographic characteristics are presented in Table 2. In the overall population, there was a considerable imbalance between the LDA group and non-LDA group in baseline clinical characteristics such as sex, age, body mass index, left ventricular ejection fraction, and history of hypertension, diabetes, and dyslipidemia. However, there were no significant differences between both groups in angiographic characteristics. After a matched analysis, the baseline clinical and angiographic characteristics of the two PSM groups (313 pairs, n=626 total) were balanced in all measured criteria (Tables 1, 2).

In the overall population, there was a considerable imbalance between the LDA and non-LDA groups in the use of medications such as calcium channel blockers, beta-blockers, diuretics, renin-angiotensin-aldosterone system inhibitors (e.g., angiotensin receptor blockers or angiotensin-converting enzyme inhibitors), and statins. However, in the matched analysis, the medical treatments were balanced between the two groups (Table 3).

In the 5-year clinical follow-up of the entire population, in comparison with the non-LDA group, the LDA group showed significantly higher rates of MACCE1, MACCE2, and recurrent angina. However, after PSM analysis to adjust for baseline confounders, the two groups were undifferentiated regarding all follow-up events (Fig. 2). Multiple Cox-regression analysis after PSM showed that LDA use did not affect any follow-up events (Table 4).