Ischemia with nonobstructive coronary artery disease (CAD) has recently garnered significant recognition as a crucial subset of ischemic heart disease, among which vasospastic angina (VSA) is a major etiology [1]. In patients with suspected VSA, intracoronary acetylcholine (ACh) provocation testing is the gold standard invasive procedure [1, 2]. Previous studies have identified several clinical factors associated with VSA, including genetic variants, current smoking habits, and others [3, 4, 5]. From a mechanistic perspective, coronary vasospasm may be associated with the presence of perivascular adipose tissue (PVAT) and adventitial vasa vasorum as sources of inflammatory conditions [6]. Japanese researchers have confirmed that patients with VSA have significantly increased fluorodeoxyglucose uptake on positron emission tomography (PET) and PVAT volume on computed tomography (CT) surrounding the coronary arteries compared to those without VSA [6]. Recently, coronary artery inflammation has been evaluated by coronary CT angiography (CTA) using the fat attenuation index (FAI), a novel imaging biomarker that maps spatial changes in perivascular fat attenuation. This biomarker is reportedly associated with future cardiovascular outcomes in patients with obstructive and non-obstructive coronary diseases [7, 8]. Some previous studies have indicated a potential relationship between such findings on coronary CTA and vasospasm, however, most of these studies focused on patients with VSA (i.e., patient-level analysis) rather than coronary arteries with vasospasm (i.e., vessel-level analysis) [9, 10]. Thus, we aimed to assess the characteristics of PVAT on CTA of coronary arteries with vasospasms in patients undergoing ACh provocation tests.

Of the 51 patients, 24 (47.1%) yielded a positive ACh provocation test (VSA), with 40 of the 125 (32.0%) vessels having significant ACh-induced vasospasm (Fig. 1). Obstructive epicardial CAD was observed in 12 (9.6%) vessels. The baseline characteristics did not differ significantly between patients with or without VSA, except for BSA (Table 1). The findings of the ACh provocation tests are presented in Table 2. In patients with vasospasms in the LAD and RCA in the negative ACh test group (Table 2), neither chest symptoms nor ECG changes were observed.

Overall, volumes of EAT and PVAT were 84.9 $\pm$ 41.2 and 28.3 $\pm$ 17.2 cm³, respectively, while perivascular FAI was –79.8 $\pm$ 8.4. In the vessel-level analysis, the BSA-indexed volume of the PVAT was not significantly different between the vessels with (n = 40) and without (n = 85) ACh-induced vasospasms (Fig. 3). Similarly, no significant difference was identified in FAI between the two groups (Fig. 3). In each major coronary artery, the presence of ACh-induced vasospasm was not significantly associated with the PVAT volume or FAI (Table 3). Conversely, vessels with epicardial obstructive CAD had a significantly higher FAI than those without (Fig. 4). In the patient-level analysis, the BSA-indexed volumes of the PVAT and FAI did not differ significantly between patients with and without positive ACh provocation test results in the LAD, LCX, and RCA (Table 4).

Fig. 3.

Perivascular adipose tissue (PVAT) volume and fat attenuation index (FAI) in vessels with and without acetylcholine-induced vasospasm. The PVAT volume is indexed to the body surface area. HU, Hounsfield unit.

Fig. 4.

Perivascular adipose tissue (PVAT) volume and fat attenuation index (FAI) in vessels with and without obstructive coronary artery disease (CAD). The PVAT volume is indexed to the body surface area. HU, Hounsfield unit.

The present study demonstrated that PVAT analysis using CTA was feasible in 125 of 144 (86.8%) coronary vessels and that ACh-induced vasospasm was not significantly associated with increased PVAT volume and FAI at the vessel level. The presence of obstructive epicardial CAD was associated with elevated FAI. The present study findings suggest that the PVAT, but not the VSA, was involved in the potential pathogenesis of obstructive CAD.

Previous studies have identified several underlying mechanisms of coronary artery spasms, including inflammation, endothelial dysfunction, and smooth muscle cell hypercontractivity [20]. In this context, PVAT may play a crucial role as a source of inflammation. The PVAT functions as an active endocrine and paracrine organ that secretes inflammatory cytokines [21]. In an experimental study involving pigs, inflammatory changes in PVAT on PET images were associated with coronary hypercontractivity [22]. Additionally, previous clinical studies have demonstrated that fluorodeoxyglucose uptake on PET, a surrogate of inflammation, was significantly increased in patients with positive ACh provocation tests compared to the uptake in those without negative test results [6]. Consequently, an association between coronary spasm and inflammation of the coronary adventitia and PVAT was suggested. Furthermore, a correlation between increased PET/CT uptake and elevated CT attenuation values around the coronary arteries has been reported [23]. Previous studies have investigated the clinical relevance of the pericoronary adipose tissue in a broad spectrum of patient populations [24, 25]. CTA is another non-invasive modality used to evaluate PVAT and inflammation. Although PVAT assessed by CTA has been associated with future cardiovascular risk in previous studies [17, 26], its prognostic significance remains to be established. In the CRISP-CT study, another CT measure of PVAT, namely FAI, was evaluated, which enhanced cardiac risk stratification in patients with mild-to-moderate epicardial CAD [7]. The subsequent ORFAN study demonstrated the clinical utility of FAI on coronary CTA for cardiovascular risk estimation in patients without obstructive CAD [8]. Although some previous studies have investigated the potential increase in PVAT and FAI on coronary CTA in patients with VSA [9, 10], the results were preliminary and focused only on patient-level analysis, preventing evaluation of a direct link between ACh-induced epicardial vasospasm and PVAT. Thus, in the present study, we primarily focused on a vessel-level analysis of the relationship between CTA metrics and vasospasm.

In this context, the present study investigated the impact of PVAT volume on FAI. In a previous study in Japan by Ohyama et al. [10], which included 66 patients who underwent ACh provocation tests and coronary CTA, the BSA-indexed PVAT volume ranged from approximately 7 to 20 cm³/m² in each major coronary artery. These findings potentially align with our results [10]. As reported in the Japanese study, the PVAT volume was reduced in the LCX compared to the LAD and RCA in the present study. Ohyama et al. [10] demonstrated that the PVAT volume in the LAD on coronary CTA was significantly increased in patients with VSA compared to that in the control group. In the landmark CRISP-CT study, the mean FAI reportedly ranged from –75.1 to –77.0 HU in their derivation and validation cohorts [7], which also aligned with our study results. Another Japanese study indicated that FAI in the RCA could predict positive ACh provocation tests [9, 10]. Collectively, previous studies suggest that PVAT volume and FAI assessed on coronary CTA may be associated with positive results in ACh provocation tests. However, the key coronary arteries differed between the two metrics (LAD vs. RCA). In the present study, the vessel-level analysis did not demonstrate a significant association between ACh-induced vasospasm and PVAT volume or FAI. Although the underlying mechanisms of these insignificant results remain unclear, we believe that this association is neither simple nor robust. Importantly, the FAI was significantly higher in vessels with obstructive CAD in the present study, reinforcing the idea that inflammation enhances coronary atherosclerosis [27]. Further studies are needed to refine the methodology of CTA analysis and to identify patients with VSA that are highly likely to be attributable to PVAT and inflammation. Moreover, PET may be a superior diagnostic modality for assessing inflammation surrounding the coronary arteries [6].

Our study had some limitations that should be considered when interpreting the results. As this was a single-center retrospective study with a relatively small sample size, external generalizability may be limited, and potential selection bias should be acknowledged. Thus, our results should be considered hypothesis-generating. Although the CT images were analyzed following previously reported methods [22], these have not yet been fully established. In the present study, intracoronary ACh provocation testing and coronary CTA were performed within 6 months, during which the characteristics of PVAT possibly changed due to lifestyle modifications and medications. In the negative ACh test group, some patients presented with chest symptoms and ECG changes during provocation testing (Table 2), suggesting a potential diagnosis of microvascular spasm. However, the small sample size precluded further analyses of this patient population.

In patients undergoing intracoronary ACh provocation tests and coronary CTA, no significant association was noted between ACh-induced coronary vasospasm and PVAT volume or FAI at the vessel level, although FAI significantly increased in vessels with epicardial obstructive CAD. Future prospective, multicenter, well-designed investigations with large sample sizes are warranted to clarify the impact of PVAT quantity and quality on coronary vasospasm.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

KA, conceptualization, data curation, investigation, formal analysis, and writing draft; YS, conceptualization, investigation, review and writing, and supervision; HT, investigation, review and writing, and supervision; HK, investigation and review and writing; YK, conceptualization, review and writing, supervision, and project administration. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The present study was performed in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Chiba University Hospital. Ethics Committee of Chiba University Graduate School of Medicine (approval no. 3983). Informed consent for this study was obtained in an opt-out manner.

This research received no external funding.

The authors declare no conflict of interest.