Variant Angina is Associated with Myocarditis

Introduction

Vasospastic angina (VSA), formerly known as variant angina or Prinzmetal angina, is mainly caused by transient narrowing due to vasoconstriction of the coronary artery, leading to clinical symptoms that present as chest pain and are usually accompanied by ST-segment elevation or ST-segment depression on the ECG.¹ It is classified as a focal or diffuse coronary artery spasm (CAS). Clinically, severe vasospastic angina causing total coronary occlusion can induce a wide range of myocardial infarctions and is life-threatening.² It was reported that the prevalence rate of VSA ranged from 33.4% to 57.6% in western countries and 40% to 79% in Asian countries. Importantly, patients with VSA have relatively poor clinical outcomes, even if current therapies with calcium channel blockers and nitrates can effectively relax spastic coronary arteries and improve angina symptoms in these patients.³

The exact etiology of VSA remains unknown, and spastic coronary artery segments have no stenosis in most patients. Therefore, VSA may differ from atherosclerotic lesions. Furthermore, statins have been used worldwide to reduce adverse events in patients with coronary artery disease (CAD), and antiplatelet therapy using aspirin or aspirin plus clopidogrel is used for the primary and secondary prevention of CAD. However, the application of these treatments has not reduced adverse events as expected in patients with VSA.^4,5 These studies suggest that hypercholesterolemia and platelet activation are not pathogenic factors of VSA.

An early study found that platelet-activating factor (PAF) can induce coronary spasm, inflammatory cell-derived interleukin 1 (IL-1) is a potent inducer of PAF,^6,7 and mast- -and eosinophil-released histamine (HA) induces coronary artery contraction, which suggests that inflammation and inflammatory cytokines may be an important cause of VSA. One example is that allergic reactions (usually severe reactions) can cause VSA by releasing HA and other inflammatory mediators (Kounis syndrome).⁸

It is well known that inflammatory processes are involved in various vascular diseases. Previous observations have shown that coronary inflammatory changes may be more important in the pathogenesis of CAS than coronary atherosclerotic changes.⁹ Shimokawa et al found that chronic treatment with IL-1β, a pro-inflammatory cytokine, induced intimal thickening and CAS in vivo in animal experiments.⁹ An interesting study in 15 patients with VSA found that coronary adventitial and perivascular adipose tissue inflammation are associated with CAS using 18-fluorine fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography/computed tomography (PET/CT).¹⁰ This primary exploration suggests that VSA is associated with inflammation. However, more evidence is needed to establish the direct association between inflammation, especially in the heart, and VSA.

In the present observational study, we investigated whether the increased plasma inflammatory cytokines levels are associated with recurrent CAS using multivariate logistic regression to establish a predictive model to evaluate whether the levels of cytokines serve as important biomarkers to identify VSA. We further evaluated whether myocarditis and coronary adventitial and perivascular adipose tissue inflammation exist in the hearts of patients with VSA.

Methods

Study Participants and Blood Sample Collection

Patients who visited the Department of Cardiovascular Medicine, Tongji Hospital (Wuhan, China) donated blood samples for this study. Informed consent was obtained from the participants. This study was approved by the Ethics Committee of Tongji Hospital and the trial conformed to the principles outlined in the Declaration of Helsinki. The process of patient screening and selection is documented in detail in a flowchart (Supplementary Figure 1). Between March 2012 and January 2021, 137 patients were clinically diagnosed with VSA at Tongji Hospital (Wuhan, China). Electrocardiography (ECG), Holter ECG, and coronary angiography were performed. Six patients with suspected VSA who did not meet the criteria were excluded according to the latest diagnostic criteria for VSA. We excluded 22 patients with VSA because blood samples were not obtained from these patients during hospitalization. The exclusion criteria included non-ST-elevation myocardial infarction, stable angina, and unstable angina. Finally, 109 patients with VSA who met the diagnostic criteria, from whom blood samples were obtained, were successfully recruited for the study. We enrolled 61 normal controls (CON) and 61 patients with ST-segment elevation myocardial infarction (ST) determined by the latest diagnostic criteria as negative and positive controls to reflect the clinical situation more accurately. Reports of this research adhered to the STROBE checklist for cohort studies (Supplementary material).

Peripheral blood samples were collected in ethylenediamine tetra-acetic acid-coated tubes and centrifuged at 2800 rpm for 8 min, followed by the collection of plasma into cryovials. Plasma was stored at −80°C until analysis. In addition, plasma samples needed to be thawed overnight at 4°C before testing.

Cytokine Measurement of Plasma

Cryopreserved plasma cytokine concentrations were quantified using the Human Immunotherapy Magnetic Luminex Performance Assay 24-plex panel (R&D Systems, USA) according to the manufacturer’s instructions: granulocyte-macrophage colony-stimulating factor, granzyme B, interferon-α (IFN-α), IFN-γ, IL-1α, IL1-β, IL-1ra, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, IL-15, IL-17A, IL-33, IFN inducible protein 10, macrophage inflammatory protein-1α (MIP-1α), MIP-1β, macrophage chemoattractant protein-1 (MCP-1), programmed death ligand 1 (PD-L1), tumor necrosis factor-α (TNF-α), and CD40L. A two-fold dilution with a calibrator was used for all samples, based on pre-experimental results. The plate was read using a Luminex magnetic analyzer (R&D Systems, USA).

Enzyme-Linked Immunosorbent Assay (ELISA)

The protein expression of endothelin-1 (ET-1) was detected using human ET-1 ELISA kits (Immunoway, USA) following the manufacturer’s instructions. The absorbance at 450 nm was measured using a microplate reader.

Metabolomic Analysis

The native standards, acetylcholine (ACh), HA, and its deuterated internal (d9-ACh and d4-HA), were procured from Sigma-Aldrich (St. Louis, MO, USA). Methanol and acetonitrile were procured from Merck Millipore (Billerica, MA, USA). Ammonium hydroxide (25%, w/v) solution and formic acid were procured from Fisher Scientific. All the solvents and chemicals used were of high-performance liquid chromatography grade or higher.

Ach and HA were quantified using previously optimized methods.¹¹ Briefly, plasma samples (20 µL) from each patient, 10 µL of IS solution (d9-Ach concentration 50 ng/mL, d4-HA concentration 2 ng/mL), and 170 µL of methanol were vortex-mixed for 2 min, followed by centrifugation at 12,000 × g at 4°C for 15 min. The supernatant (100 µL) was then transferred to a chromatographic vial for further measurement. Five microliters of plasma from each patient were prepared in a pooled sample to define the quality control.

Metabolites were obtained using a UPLC system interfaced with a mass spectrometer (Waters Xevo™ TQXS system). The injection volumes of 5 µL were employed in a column (ACQUITY UPLC BEH Amide, 100 mm × 2.1 mm, 1.7 μm, Waters, MA, USA) and maintained at 40°C at a flow rate of 0.5 mL/min. The mobile phases comprised (A) 0.05% ammonium hydroxide, 20 mM ammonium formate in water and (B) 0.05% ammonium hydroxide, 20 mM ammonium formate in 90% acetonitrile. A gradient elution was optimized as following: 0–3.0 min, 95–45% B; 3.0–6 min, 50% B; 6–6.1 min, 95% B; 6.1–9 min, 95% B. The optimized parameters for mass spectrometry were: capillary voltage, 4.5 kV; cone voltage, 50 V; source temperature, 150°C; desolvation temperature, 500°C; cone gas flow, 150 L/h; desolvation gas flow, 800 L/h; collision gas flow, 0.5 mL/min. Measurements were performed using multiple reaction monitoring in positive mode. A quality control sample was acquired for every 10 samples to assess the stability of the experiment. Quantification was performed against the area of the internal standards using Masslynx (version 4.2; Waters, USA). The concentration is presented in ng/mL for each compound. (ClinicalTrials.gov Identifier: NCT05282511. Registered 16 March 2022, retrospective, https://www.clinicaltrials.gov/ct2/show/NCT05282511)

Statistical Analysis

Cytokines were transformed using log2 to prevent the skewing of results by outliers. The data are presented as scatter plots with each circle representing a single individual. All data were analyzed using GraphPad Prism 9 (version 9.0), R software (version 4.1.1), and R Studio (version 1.4.1717). Continuous variables are expressed as mean ± standard deviation or median (interquartile range), according to their distribution, and categorical variables as frequencies and percentages. The difference in cytokine levels in the plasma samples was assessed using the two-tailed Mann–Whitney U and Kruskal–Wallis tests. The three groups were compared using Dunn’s post hoc test to correct for multiple comparisons. Generalized linear modeling-logistic regression (GLM-LR) was used to assess the ability of cytokines with significant differences to classify VSA, and stepwise regression was used to screen the optimal model. Model performance was evaluated based on the area under the receiver operating characteristic curve using pROC (package version 1.18.0). The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the model were determined using standard definitions. Statistical significance was set at p<0.05.

Results

Baseline Characteristic of the Study Population

Plasma samples from 224 participants, comprising 60 normal participants, 56 patients with ST, and 108 patients with VSA, were included in this study. The clinical characteristics are summarized in Table 1. There was no difference in the prevalence of coronary risk factors (age, sex, current smoking status, alcohol consumption, and blood pressure) among the three groups, except for diabetes mellitus (p=0.008) and dyslipidemia (p=0.020). Similarly, no statistically significant difference was observed in the history of asthma (p=0.646) or allergy (p=0.596) among the participants in the various groups. Among the cardiac markers, erythrocyte sedimentation rate levels were comparable among the three groups (p=0.164), whereas the levels of high-sensitivity C-reactive protein (hs-CRP) increased significantly in the VSA and ST groups compared with the CON group (p=0.002). Regarding medical treatment, there was no difference in nitrate (0.078) and corticosteroids (0.137) among the various groups, whereas the use of angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers and calcium channel blockers increased significantly in the VSA and ST groups compared with the CON group (p<0.001 and p<0.001, respectively). Similarly, the use of statins and beta-blockers was significantly higher in the CON and ST groups than that in the VSA group (p<0.001 and p<0.001, respectively).

Table 1 Baseline Clinical Characteristics and Treatments of the CON Group, ST Group and VSA Group

Plasma Levels of Cytokines and Chemokines were Elevated in the VSA Group Compared with the both CON and ST Groups

Plasma levels of all 24 cytokine were measured properly and 17 of them had significantly different between VSA, ST and normal control (Table 2). Interestingly, plasma levels of the pro-inflammatory cytokines IL-6, IL-12p70, and IL-15 (Figure 1A, B and E) and the anti-inflammatory cytokines IL-13 and IL-10 (Figure 1C and D) were elevated in the VSA group compared with the CON and ST groups in this study. In addition, plasma chemokine levels of PD-L1, MIP-1α and MIP-1β (Figure 1F–H) were higher in the VSA group than in the CON and ST groups. In addition, the levels of these cytokines were not significantly different between the CON and ST groups, except for IL-15.

Table 2 Cytokines with Significant Difference in CON Group, ST Group and VSA Group

Figure 1 (A–H) Plasma levels of cytokines and chemokines are elevated in VSA group compared to CON group and ST group. Abbreviations: FC, fold change; CON, controls; ST, ST segment elevation myocardial infarction.

Plasma Levels of Cytokines Were Elevated in the VSA Group Compared with the ST Group

It is noteworthy that the levels of the pro-inflammatory cytokines IL-4, IFN-α, IL1-α, IL-2 and chemokine MCP-1 (Figure 2A–E) were significantly higher in the VSA group than in the ST group. This indicates that these cytokines can distinguish between the VSA and ST groups.

Figure 2 (A–E) Plasma levels of cytokines are elevated in VSA group compared to ST group.

Plasma Levels of IL-8, IFN-γ, and Anti-Inflammatory Cytokine IL-1ra are Elevated in Patients with VSA Compared with the Normal Participants

The levels of the pro-inflammatory cytokines IL-8 and IFN-γ (Figure 3A, C) and anti-inflammatory cytokine IL-1ra (Figure 3B) in the VSA group increased compared with the CON group. This also shows that these three cytokines could distinguish between the VSA and CON groups. In addition, the plasma levels of IFN-γ and IL-1ra in patients with ST were higher than those in patients with VSA. Other significantly different cytokines are shown in the Supplementary Material.

Figure 3 (A–C) Plasma levels of cytokines are elevated in VSA group compared to CON group.

Plasma Vasoconstrictor Metabolites Increased in the VSA and ST Groups

We found that the levels of plasma vasoconstrictor metabolites, HA and ET-1 (Figure 4A and B) were significantly higher in the VSA and ST groups compared with those in the CON group. These vasoconstrictor metabolites showed a higher trend in the VSA group than in the ST group, although the difference was not significant.

Figure 4 (A and B) Plasma vasoconstrictor metabolites are increased in both VSA and ST group.

Plasma Levels of Cytokines Serve as Biomarkers of VSA to Distinguish Between the VSA and ST and CON groups

The GLM-LR model was used to evaluate the predictive power of cytokines with significant differences. The area under the curve (AUC) was 0.84 (95% confidence interval [CI] 0.78–0.90) between the VSA and CON groups, and the sensitivity and specificity of developing VSA were 71.56% and 86.88%, respectively (Figure 5A). The AUC was 0.94 (95% CI 0.91–0.98) between the VSA and ST groups, and the sensitivity and specificity of developing VSA were 87.16% and 88.53%, respectively (Figure 5C). We also assessed the predictive power of other clinical markers of inflammation, such as hs-CRP and common inflammatory cytokines (including IL-1β, IL-6, IL-8, IL-10, TNF-α.). The AUC of the two predictors were 0.55 and 0.78 between the VSA and CON groups (Figure 5B). The AUC of the two predictors were 0.76 and 0.59 between the VSA and ST groups (Figure 5D). This showed that the predictive power of these cytokines was better than that of hs-CRP and inflammatory cytokines.

Figure 5 (A–D). ROC analysis of cytokines with significant differences distinguishing VSA group from ST and CON group. Abbreviations: AUC, area under curve; hs-CRP, high-sensitivity C-reactive protein; ROC, receiver operating characteristic. Notes: (B) Cytokine biomarkers (IL10, IL-12p70, IL-1ra) differentiate VSA and CON groups. (D) Cytokine biomarkers (MCP-1, IFN-α, IFN-γ, IL-13, IL-15, IL-1ra, IL-2, IL-4) differentiate VSA and ST groups.

Case Presentation

We performed ¹⁸F-FDG PET/CT in four patients and found that all patients had a significantly increased ¹⁸F-FDG uptake in the left ventricle and also showed a marked increase in the ¹⁸F-FDG uptake in the proximal segment of the right coronary artery. We also tried to treat these four patients with VSA using glucocorticoids, and long-term remission was achieved during the follow-up period of 3 and 6 months. Below are the reports of two typical cases.

Patient 1:

A 51-year-old man had a history of recurrent allergic asthma. The patient was admitted for recurrent chest pain for 7 years, accompanied by chest pain-related syncope five times in 3 months. The chest pain was not related to effort but to asthma occurrence, aggravation, or sleep, and lasted for 5–30 min. Three months ago, he experienced chest pain again after getting a cold, and when it was severe, he lost consciousness for approximately 2 min. He previously underwent ECG and Holter ECG showing recurrent ST-segment elevation in leads II and III, as well as augmented vector foot and chest leads and intermittent cardioplegia for up to 9 s. On admission, ECG demonstrated ST-segment elevation and Q-wave myocardial infarction and grade III atrioventricular block (Supplementary Figure 2). Coronary angiography revealed severe local stenosis of the right coronary artery, which resolved after nitroglycerine injection (Supplementary Figure 3). Additionally, the left anterior descending artery and left circumflex artery narrowed down by approximately 20%, and the right coronary artery narrowed down by approximately 30% locally. Serum cardiac troponin was significantly elevated (3000.0 pg/mL, reference <34 pg/mL), and N-terminal pro-brain natriuretic peptide was mildly elevated (300 pg/mL). It was interesting that the blood test revealed elevated white blood cell and eosinophil levels of 15.1 × 10⁹/L and 0.84 × 10⁹/L while he was symptomatic. Plasma cytokine assays showed that concentrations of IL-8, TNF-α, and soluble suppression of tumorigenesis-2 (sST2) were significantly elevated (0.15 ng/mL, 0.03 ng/mL, and 80 ng/mL, respectively). ¹⁸F-FDG PET/CT revealed a diffuse FDG uptake increase in the coronary artery and the left ventricle (Supplementary Figure 4). Based on clinical symptoms and angiography and ECG results, the patient was diagnosed with VSA caused by the inflammatory response of the coronary artery. He subsequently underwent immunomodulatory treatments with intravenous injection of methylprednisolone (80 mg/day) and immunoglobulin (10 g/day), as well as oral sustained-release isosorbitol dinitrate and diltiazem (90 mg/day). After treatment for 1 week, he noted significant improvement in his symptoms, and IL-8, TNF-α, and sST2 levels returned to normal. During the 6-month follow-up period, there was no onset of chest pain.

Patient 2:

A 64-year-old woman was hospitalized for appendicitis and experienced sudden chest pain 2 days after admission. The ECG revealed inferior T-wave inversion and ST-segment flat, but T wave disappear in all leads when she has mild chest pain, which is different from normal condition (no chest pain) during Holter ECG monitoring (Supplementary Figure 5). Her troponin I level was 11.02 ng/mL. Moreover, her blood tests revealed an increased neutrophil ratio and elevated hs-CRP (76.3% and 7.8 mg/L, respectively). On day 4 of hospitalization, she had ventricular fibrillation during coronary angiography of the right coronary artery which showed diffuse spasm (narrowed >85%), and the left coronary artery also showed mild diffuse spasm. After defibrillation and vasodilator therapy (intracoronary nitroglycerine injection, 100 µg), the coronary spasm improved. Spontaneous circulation returned (Supplementary Figure 6). On day 5 of hospitalization, the blood test revealed an elevated neutrophil count (89.7%), and the ¹⁸F-FDG PET/CT revealed that FDG uptake in the right coronary artery was elevated and diffusely increased in the left ventricle (Supplementary Figure 7). Therefore, we believe that pericoronary inflammation causes coronary spasms. She was treated with an intravenous injection of methylprednisolone (40 mg/day) and immunoglobulin (5 g/day) to inhibit the inflammatory response. Eventually, her plasma inflammatory cytokine levels returned to normal, and she was asymptomatic. After 2 months of follow-up, the patient had no chest pain.

Discussion

This study is the first to investigate the relationship between the levels of multiple plasma inflammatory cytokines and VSA in a Chinese cohort and included patients with ST and normal participants as controls. We found that (1) of the tested 24 inflammatory factors, eight were significantly elevated in patients with VSA than in patients with ST and normal controls (p<0.001). IL-4, IFN-α, IL1-α, IL-2, and MCP-1 were lower in patients with ST than in patients with VSA. Plasma levels of IL-8, IFN-γ, and IL-1ra were significantly elevated in patients with VSA and ST compared with the controls, which suggests that VSA is an inflammation-related disease and it is different from acute myocardial infarction (AMI); (2) substances that caused VSA, namely HA and ET-1^12,13 were also elevated; (3) further, we observed that patients with VSA had pericoronary inflammation as well as myocarditis, as shown by ¹⁸F-FDG PET/CT. Our findings reveal that VSA is associated with inflammation, particularly myocarditis. In addition, we found that a panel of inflammatory cytokines might help diagnose VSA.

Previous case reports^14–17have indicated that CAS is associated with inflammation and myocarditis, as confirmed by endomyocardial biopsy. ¹⁸F-FDG uptake can effectively reflect tissue metabolism; therefore, ¹⁸F-FDG PET/CT can assess tissues, including the cardiac inflammatory status, by ¹⁸F-FDG uptake.¹⁸ A small cohort study using ¹⁸F-FDG PET/CT found that coronary perivascular FDG uptake significantly increased in the VSA group but not in controls, suggesting coronary adventitial and perivascular adipose tissue inflammation in patients with VSA.¹⁰ The present study demonstrates that the plasma levels of multiple inflammatory cytokines in patients with VSA are markedly elevated, which is different from AMI. These findings reveal that VSA is an inflammatory-associated disease.

The mechanisms of coronary spasms are important to understand this special recurrent angina pectoris. Clinically, coronary spasms can be induced by ACh or ergoline. Nihei et al found that Rho-kinase activity in circulating leukocytes in patients with VSA increased, and during the 3-year follow-up period, cardiac events occurred in 10 patients (5.7%) with VSA but in none of the patients without VSA, while VSA patients with higher Rho-kinase activity had a significantly worse prognosis than other comparable patients.¹⁹ Increased Rho-kinase activity represents an inflammatory condition and can activate the mitogen-activated protein kinase and nuclear factor kappa B signaling pathways, as well as the 20-kDa regulatory myosin light chain, thus inducing vessel contraction.²⁰ Increased cytokine levels from neutrophils and macrophages, such as IL-6, can enhance coronary contractions by activating Rho-associated kinase.^21,22 In addition, IL-1 induces the synthesis of inflammatory mediators, prostacyclin, and platelet-activating factor, which directly induce CAS.^6,23

Hypersensitivity has been recognized to be related to VSA, which is referred to as Kounis syndrome.⁸ HA, a key factor in inflammation, especially mast cells, is an important mediator of inflammation and vasospasm. Therefore, Kounis syndrome represents a subset of patients with VSA due to inflammation. Shimokawa et al found that the local coronary segment exposed to inflammatory stimuli (IL-1β) was linked to vascular smooth muscle cell hyperreactivity in a porcine model with significantly elevated levels of serotonin, HA, and prostaglandin F2.⁹ Our observation found that plasma HA and ET-1 levels in patients with VSA increased, as proven by Figeuras et al.²⁴ Two of our VSA patients with refractory angina pectoris achieved long-term remission after receiving treatment with glucocorticoids and immunoglobulin. The detection of inflammation may provide a new approach for diagnosing recurrent coronary spasms. Furthermore, our case report not only contributes to the diagnosis of CAS but also paves the way for alternative therapy in patients with VSA. This is also our first attempt to use glucocorticoids combined with immunoglobulin to treat patients with coronary spasms caused by inflammation. Although we have only observed a small number of clinical cases, the therapy has proven to be effective in practice.

In summary, the levels of multiple inflammatory cytokines in patients with VSA were generally increased, different from that in patients with AMI. Therefore, we were able to set a multiple inflammatory cytokine biomarker to help differentiate VSA from AMI or regular angina and provide some insight into mechanical therapy. A previous study reported that plasma levels of hs-CRP are higher in patients with coronary spasms than in those without, and elevated CRP is independently and significantly associated with coronary spasm.²⁵ To further explore the relationship between inflammatory cytokines and other inflammatory markers, our multivariate logistic regression showed that the predictive power of these cytokines was significantly better than that of hs-CRP and commonly inflammatory cytokines, including IL-1β, IL-6, IL-8, IL-10, and TNF-α.

The next question concerns the origin of the inflammatory cytokines. Combining the literature and this study, we suggest that VSA is a complication of myocarditis in most patients. Thus, for patients in the acute phase of VSA or recurrent attacks of VSA, anti-inflammatory therapy such as using glucocorticoids is reasonable.

A limitation of this study is its relatively small sample size. In addition, for most patients in this cohort, we only provided plasma levels of multiple inflammatory cytokines and baseline data; however, ¹⁸F-FDG PET/CT and positive responses to glucocorticoid therapy were available for only a few patients. Finally, regarding the predictive efficacy of inflammatory factors, we still lack a validation population for further confirmation. Therefore, further studies, including therapeutic clinical trials and multivariate cytokine panel validation, are required.

The present study demonstrates that various inflammatory cytokines in the plasma are increased and that elevated inflammatory cytokines may be useful for diagnosis and serve as a guide for anti-inflammatory treatment in patients with VSA.