Angina with non-obstructive coronary artery disease (ANOCA) is a prevalent condition that has recently received increased attention [1, 2]. The leading causes of ANOCA are vasospastic angina (VSA), coronary microvascular dysfunction (CMD), or both [1, 2]. The coexistence of both underlying causes has a poor prognosis [3]. Treatment is superior for improving subjective symptoms in ANOCA with a known cause treated with pharmacological therapy [4]. Consequently, VSA is recommended to be identified using spasm provocation testing (SPT) and CMD with the coronary microvascular function test (CMFT).

SPT and CMFT are widely recognized for testing [1, 2]. However, which test should be performed first remains unestablished and also varies based on factors such as subjective symptoms, institutional experience, and policies. CMFT requires maximally dilated coronary arteries; thus, nitroglycerin (NTG) preadministration is mandatory. However, NTG preadministration may have a significant effect on SPT results [5]. Furthermore, patients with CMD (especially those with reduced coronary flow reserve [CFR]) demonstrate a poor prognosis [6], but VSA also causes cardiac arrest [7, 8]. Further, beta-blockers (the main treatment for CMD) [1] are restricted for treating VSA when administered alone [9]. Therefore, VSA must be diagnosed before beta-blocker administration. Hence, SPT is frequently administered before CMFT in Japan, including our institution.

Conversely, one of the complications of SPT with acetylcholine (ACh) is paroxysmal atrial fibrillation (PAF) [10, 11]. Clinical characteristics that cause PAF during SPT remain unclear. Furthermore, atrial fibrillation (AF) may affect CMFT testing, especially in the CFR [12, 13, 14, 15]. Furthermore, the effect of PAF on CMFT outcomes needs to be identified. We investigated the clinical characteristics of patients who underwent SPT and CMFT to investigate the etiology of ANOCA, the clinical characteristics of patients who developed PAF, and the association of PAF occurrence with CMFT outcomes.

This retrospective observational study investigated the clinical characteristics of patients with ANOCA who developed AF when first treated with SPT with ACh, followed by CMFT treatment. Approximately 25% of the patients developed PAF, but our study could not determine any clinical characteristics that would predispose patients to PAF. Additionally, we examined the effect of AF on subsequent CMFT. Our results revealed that patients with persistent AF demonstrated little effect on IMR despite a reduced CFR caused by increased coronary blood flow velocity at rest. Conversely, CFR and IMR may be affected in patients who developed PAF but recovered to SR, indicating an improved coronary microcirculation. Moreover, caution should be exercised when interpreting SPT results using ACh preceded by CMFT because AF complications may affect CMFT measurements.

PAF occurred in approximately 8%–17% of the patients in the SPT using the ACh group [10, 11]. This was more predominant in RCA provocations [10, 11]. This could be caused by the differences in the distribution of muscarinic receptors in the coronary arteries [22]. The present study revealed PAF in 23% of the patients receiving RCA and in 6% administered with LCA, indicating that PAF was more prevalent in the RCA provocation. Our study observed provocation was more common than previously reported [10, 11], which may be caused by the definition of AF used and because we included patients with PAF of short durations (such as those that were brief and quickly restored to SR). Additionally, the definition of low, moderate, and high doses of ACh differed from study to study, and these definitions must be cautiously interpreted, but it did not occur at higher ACh doses as reported by Sueda et al. [23]. Few studies have examined which patients are more likely to develop PAF during SPT with ACh [11]. Saito et al. [11] revealed that PAF is more likely to occur in patients with a PAF history and low body mass index (BMI). The present study revealed that patients with PAF reported no PAF history, and BMI did not differ between patients with or without PAF. The differences in the groups studied may explain the differences in patients with ANOCA. Saito et al. [11] conducted a univariate analysis revealing that men were less likely to have PAF, but the differences were not significant in the multivariate analysis. The present study indicated no such significant differences but observed a trend toward a higher PAF incidence in women. Further research is warranted to understand the sex-related differences in PAF complications during the SPT in a multicenter registry. The present study revealed a trend toward lower NT-proBNP and echocardiographic E/e’ in the PAF group, although the differences were not statistically significant. These results indicated that patients with less severity of cardiac organic abnormalities may be more prone to PAF caused by ACh provocation. Additionally, MB also tended to be more prevalent in the PAF group. An association between MB and PAF during the operation was observed in patients with hypertrophic cardiomyopathy [24]. More research is required to clarify the associations of NT-proBNP, E/e’, and MB with PAF during ACh provocation. Additionally, no significant association was observed between the occurrence of AF and coronary spasms, based on previous reports and our research results [10, 11]. Our study revealed that diffuse spasm was less common in AF vessels than in SR-1 vessels among the endotypes of coronary spasm. The number of cases in a multicenter registry or other methods to investigate this issue needs to be increased because of the small number of cases of AF vessels.

Several reports [12, 13, 14, 15] on the effects of AF on coronary microcirculation have reported that increased heart rate with AF results in a faster coronary blood flow velocity at baseline, causing a lower CFR. The present study revealed no difference in the maximal coronary blood flow velocity under ATP infusion although the resting coronary blood flow velocity increased with increased heart rate. Additionally, no significant changes were observed in blood pressure at baseline (data not shown) and during ATP infusion. Thus, an increased heart rate was the only significant factor responsible for a reduced CFR on logistic regression analysis. Our data were consistent with the results of previous studies [12, 13, 14, 15]. However, the results were somewhat different when the correlations between heart rate and Tmm at baseline and CFR were investigated separately for SR-1 and AF vessels. Significant negative correlations between heart rate and Tmm at baseline and CFR were observed in SR-1 vessels, whereas no significant correlations between them were found in AF vessels. Factors other than simply increased heart rate, such as irregular rhythm, may be involved in the shortening of the Tmm at baseline and the decrease in CFR although the small number of AF vessels may contribute to this result. In any case, the number of cases of AF vessels is small and should be increased in the future. Myocardial ischemia has been reported in animal models of AF complicated by significant coronary artery stenosis [13]. However, we considered that no apparent myocardial ischemia was induced during AF because the patients in our study demonstrated no significant stenosis in the coronary arteries and no significant decreases in intracoronary pressure (Pd) or FFR during ATP infusion. Increased heart rate (although not AF) increases CFR variability, indicating that it could be a reliable IMR indicator in such cases [25]. The present study revealed that IMR was unaffected by the heart rate and did not significantly differ between the AF and SR vessels. Many cases revealed that an increase in IMR does not always coincide with a decrease in CFR in clinical practice [16, 26]. Therefore, IMR alone cannot diagnose CMD, but it is recommended to refer to the IMR value when the CFR decreases during the AF duration.

The coronary microcirculation in vessels that recovered quickly from PAF to SR was interesting. CFR was preserved in these vessels and IMR appeared lower, although this was not statistically significant. Consequently, a lower rate of having the two components of CMD was observed with CFR of 2.0 and IMR of $\ge$25. The vessels with preserved CMF can be quickly restored to SR after an initial AF. However, no differences in patient background between the three groups were observed, making it unlikely that CMF was preserved in such patients. Hence, a compensatory improvement in the coronary microcirculation during the stages of PAF and immediate recovery. Reportedly, coronary microcirculation does not restore immediately after electrical cardioversion in a clinical setting [15]. Additionally, whether CMF improves immediately after PAF is restored to SR is speculative, but a compensatory effect may be possible if the AF duration is extremely short. The number of such vessels was not large, the PAF duration was not measured and varied, and we cannot deduce how long the improvement in coronary microcirculation would last because this was a cross-sectional study. These issues could be resolved by increasing the number of patients in a multicenter registry or by investigating CMF in patients with AF-SR. In any case, coronary microcirculation should be carefully assessed even in vessels in which AF is restored to SR.

Clinically, this study implies that SPT with ACh first followed by CMFT was effective in improving the rate of coronary spasm identification, but concomitant AF may affect the results of subsequent CMFT. IMR values are advisable to be referred for CMD diagnosis because CFR decreases with increased heart rate in cases of persistent AF. Diagnosing CMD may be difficult if PAF is restored to SR immediately after it occurs. In any case, the CMFT should be interpreted with caution when AF occurs. AF frequently occurs during ACh provocation of the RCA. Performing SPTs and CMFTs in the LCA and RCA may increase the diagnostic yield of VSA and CMD [27], but limiting them to the LCA could be a countermeasure [26].

This study has some limitations. First, this was a single-center, retrospective study and the number of patients with persistent AF or with AF restoring to SR was quite small. Additionally, many cases underwent CMFT in only one coronary artery at the attending physician’s discretion. The application of these results is unclear internationally. Second, recording during CMFT measurements with intracoronary pressure and Tmn data missing in some cases was insufficient. Third, none of the patients in this study reported a PAF history, as far as we could identify from their medical records and interviews. However, the method of confirmation may not be sufficient because Holter ECGs were not performed frequently enough. Fourth, LAD and RCA were analyzed together because the number of AF cases was not large in this analysis. However, RCAs demonstrated generally higher Tmn and higher IMR than LADs [28]. Our study revealed no differences in the frequency of LAD and RCA measurements between the AF and SR groups, which did not influence the present results, but analyzing LAD and RCA separately for each vessel as they should seem correct. This is another issue that should be considered with an increased number of cases in a multicenter registry or other studies. Finally, the data were obtained from a cross-sectional study, and the CMFT data evolution in patients with AF was not observed. A method is used to measure CMFT after electrical cardioversion or administration of antiarrhythmic drugs by restoring AF to SR, but restoration of AF to SR is only performed after completing the examination due to the effects of these procedures and drugs on coronary microcirculation and because of time constraints.

We focused on AF which occurs when SPT with ACh is utilized for chest pain screening in patients with ANOCA in the present study. PAF occurred in 25% of the patients in our protocol. Our study did not reveal any particular clinical features in the background of patients that would predispose these patients to PAF development. We then examined whether AF would affect the results of the subsequent CMFT and the results indicated that CFR decreased with increasing heart rate in cases of persistent AF, but IMR was not related to heart rate. CMD with reference to IMR is recommended to be evaluated when AF is persistent and CFR declines. Conversely, CFR was preserved and IMR was low in vessels where AF was quickly restored to SR. This indicates retained coronary microcirculation. Subsequent interpretation of CMFTs should be done cautiously when AF occurs during SPT with ACh, whether persistent or not. The results of this study warrant validation in a multicenter registry.

No AI was used in the preparation of this paper.

ACh, acetylcholine; AF, atrial fibrillation; ANOCA, angina with non-obstructive coronary artery disease; ATP, adenosine triphosphate; CAD, coronary artery disease; CAG, coronary angiography; CFR, coronary flow reserve; CKD, chronic kidney disease; CMD, coronary microvascular dysfunction; CMF, coronary microvascular function; CMFT, coronary microvascular function test; DM, diabetes mellitus; eGFR, estimated glomerular filtration ratio; EM, methylergometrine maleate; FFR, fractional flow reserve; H, high dose; IMR, index of microcirculatory resistance; L, low dose; LAD, left anterior descending coronary artery; LCA, left coronary artery; LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index; M, moderate dose; MB, myocardial bridging; MVS, microvascular spasm; n, number; ND, not diagnosed; NTG, nitroglycerin; No., number; NT-proBNP, N Terminal-pro brain natriuretic peptide; Pa, aortic pressure; PAF, paroxysmal atrial fibrillation; PCI, percutaneous coronary intervention; Pd, distal pressure; RAS, renin–angiotensin system, RCA, right coronary artery; SPT, spasm provocation test; SR, sinus rhythm; Tmn, transit mean time; UCG, ultrasound cardiography; VSA, vasospastic angina.

Data and materials inquiries can be directed to the corresponding author.

HT, CO, YH and SN were involved in the patient enrolment and the collection of clinical data. HT drafted the manuscript and CO, YH and SN checked the manuscript before submission. All authors have read and agreed to the published version of the manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

This study was conducted in accordance with the guidelines of the Declaration of Helsinki and was approved by the Institutional Review Board of JR Hiroshima Hospital (protocol code: 2024-16; date of approval: 15 August 2024). Informed consent was obtained from all the patients who underwent the SPT and CMFT before the procedure. The opt-out method was adopted because of the retrospective study design.

This research received no external funding.

The authors declare no conflict of interest. Hiroki Teragawa is serving as Guest Editor of this journal. We declare that Hiroki Teragawa had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to Francesco Pelliccia.