Coronary Microvascular Dysfunction in Systemic Sclerosis: Insights from PET-Derived Myocardial Flow Reserve

Abstract

Objectives:

Cardiovascular death is the second leading cause in systemic sclerosis (SSc), with coronary microvascular dysfunction (CMVD) playing a key role. PET-derived myocardial blood flow (MBF) and flow reserve (MFR) offer a validated, noninvasive way to assess CMVD. We studied the prevalence of impaired MFR in SSc and its association with vasodilator or immunomodulatory therapy.

Methods:

SSc patients who underwent dynamic 82-Rubidium PET myocardial perfusion imaging (MPI) at Yale New Haven Hospital (July 2016 to April 2025) were studied. Patients were matched 3:1 with controls without autoimmune rheumatologic disease based on age, sex, body mass index, and cardiovascular comorbidities. Abnormal MFR was defined as <2.0.

Results:

The cohort included 67 SSc patients (87% female, age:61±11 years, 21% diffuse cutaneous SSc, SSc duration:12±10 years) and 201 controls. Rest MBF was higher in SSc (1.14 [IQR 0.91–1.39] ml/g/min) vs. controls (1.00 [IQR 0.78–1.25]; p=0.01), while MFR was lower (2.17 [IQR 1.84–2.57] vs. 2.44 [IQR 1.96–2.96]; p=0.01). Stress MBF was similar (2.47 [IQR 1.99–2.82] vs. 2.43 [IQR 1.91–3.04]; p=0.85). Multivariable analysis showed mycophenolate mofetil (MMF) use linked to lower odds of abnormal MFR (OR 0.09 [95% CI 0.01–0.56]; p=0.017), while calcium-channel blockers (OR 7.77 [1.93–42.53]; p=0.008) and statins (OR 7.14 [1.86–36.87]; p=0.008) increased odds.

Conclusions:

PET MPI reveals reduced MFR in SSc. Treatment associations, including MMF, should be interpreted cautiously given the retrospective design. PET-derived MFR may serve as a noninvasive marker of vascular involvement, warranting prospective validation.

Introduction

Systemic sclerosis (SSc) is characterized by dysregulated inflammation, fibrosis and obliterative vasculopathy with cardiovascular disease representing a major contributor to morbidity and mortality [1–4]. Microvascular injury is an early and fundamental SSc feature, yet its impact on the coronary circulation remains incompletely understood. Coronary microvascular dysfunction (CMVD) as assessed by transthoracic Doppler echocardiography [5] or cardiac magnetic resonance imaging [6] has been implicated in SSc-related cardiac complications, including ischemia, arrhythmia, heart failure, and sudden cardiac death, but its prevalence and clinical correlates remain poorly defined.

Rubidium (⁸²Rb) positron emission tomography (PET) myocardial perfusion imaging (MPI) enables quantification of absolute myocardial blood flow (MBF) and myocardial flow reserve (MFR/ defined as stress over rest MBF), providing a well-validated noninvasive assessment of coronary microvascular function [7, 8]. Reduced MFR is associated with adverse cardiovascular outcomes in multiple populations [9–11], but data in SSc are limited, and the potential influence of disease-modifying therapies on coronary microvascular function has not been systematically evaluated.

We aimed to determine the prevalence and severity of impaired MFR in patients with SSc compared with matched controls, and to examine the associations between vasodilator and immunomodulatory therapies and microvascular function. Clarifying these relationships may help identify therapeutic strategies that preserve vascular health and improve cardiovascular risk stratification in SSc.

Methods

Study Design and Population

We conducted a retrospective, single-center study at Yale New Haven Hospital (New Haven, CT) that was approved by the Yale Institutional Review Board (HIC # 2000025019). The data used in this research are available from the corresponding author upon reasonable request. 67 patients with SSc, diagnosed by a rheumatologist, who underwent rest and stress ⁸²Rb PET/CT MPI for the evaluation of clinically suspected coronary artery disease and/or coronary microvascular disease between July 2016 and April 2025, were included. Electronic records were manually reviewed to confirm SSc diagnosis, disease subtype (diffuse vs. limited cutaneous SSc), and medication history (mycophenolate mofetil/MMF, hydroxychloroquine, methotrexate, prednisone, calcium channel blockers [CCBs], angiotensin receptor blocker, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, phosphodiesterase [PDE]-5 inhibitors, beta blockers, aspirin and statins). Overlap disease was defined as the coexistence of SSc with features of other autoimmune diseases, such as rheumatoid arthritis or systemic lupus erythematosus.

Additionally, a matched cohort of 201 patients without a history of autoimmune rheumatologic disease (ARD) who had undergone dynamic rest-stress 82-Rubidium PET was matched to SSc patients in a 3:1 ratio. Electronic health records of controls were manually reviewed to adjudicate the absence of ARD diagnosis at PET. Matching was performed for age, sex, body mass index, smoking history, and clinical diagnosis of hypertension, hyperlipidemia, diabetes, heart failure (HF), transient ischemic attack or stroke, peripheral artery disease, history of myocardial infarction (MI), coronary bypass surgery, percutaneous coronary intervention, chronic kidney disease, and presenting symptoms. Matching was performed using a nearest-neighbor method in the R environment (R version 3.4; RStudio version 1.1.453; MatchIt package version 3.0.4).

PET Imaging Protocol

Dynamic rest-stress 82-Rubidium PET was performed on a hybrid 64-slice PET/CT scanner (Discovery 690, GE Healthcare). Patients abstained from caffeine for ≥12 hours and continued routine medications. Following intravenous injection of ~23 ± 4 mCi ⁸²Rb, rest dynamic images were acquired for 7 minutes. Pharmacologic stress was induced using peripherally administered IV regadenoson (0.4 mg bolus), adenosine (140 μg/kg/min), or dobutamine (up to 40 μg/kg/min) selected based on comorbidity profile. At peak stress (45 seconds after regadenoson administration or under steady infusion of adenosine or dobutamine), a second dose of 82-Rubidium was administered, and dynamic acquisition was repeated. Low-dose chest CT was performed for attenuation correction to correct for attenuation distortions in the reconstructed images, improving the overall diagnostic quality of the scans. Heart rate, blood pressure, and 12-lead ECG were recorded throughout.

Image Analysis

PET images were reconstructed with attenuation correction and analyzed using Invia Corridor 4DM software (v2017, Ann Arbor, MI). Global rest and stress MBF (ml/g/min) were calculated using a validated one-compartment kinetic model as previously described [12]. In the analysis, values uncorrected for rate pressure product (RPP, heart rate x systolic blood pressure) were utilized, however to account for hemodynamic variabilities, RPP corrected stress, rest MBF and MFR values were also calculated as previously described [10]. MFR was calculated as the ratio of stress to rest MBF (thus the value is unitless). MFR <2.0 and stress MBF <1.8 mL/g/min were considered abnormal [13]. Verification of quantitative values and visual estimation of the presence of perfusion defects and coronary calcification were performed by board certified nuclear cardiologists. The presence or absence and the extent of CAC was evaluated based on CT acquired for attenuation correction (CTAC) findings as previously described [14]. The visual CAC score was determined by individually assessing calcification in each coronary artery (left main, left anterior descending, left circumflex, and right coronary) and assigning a score of 0 (absent), 1 (mild), 2 (moderate), or 3 (severe) [14]. These scores were then summed to produce a total score ranging from 0 to 12 for each patient. Rest and stress left ventricular ejection fractions were derived from ECG gated images.

Laboratory and Clinical Variables

In both cohorts, the history of presenting symptoms or the indication for PET and current medications were recorded. Among SSc patients, we collected inflammatory markers such as C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), as well as disease duration defined as the duration from the onset of the first non-Raynaud symptom attributed to SSc (e.g., puffy fingers or telangiectasias) to PET. We also recorded SSc subtype (limited vs diffuse), ILD and PAH status, and immunosuppressive medication use (e.g., MMF, prednisone).

Statistical Analysis

Continuous variables were evaluated for normality using the Shapiro–Wilk test. Variables following a normal distribution were summarized as mean ± standard deviation and compared with the Student’s t-test; non-normal variables were shown as median [interquartile range (IQR)] and compared using the Wilcoxon rank-sum test. Categorical variables were analyzed with the Chi-square or Fisher’s exact test, as appropriate. Spearman correlation coefficients were calculated between MFR and inflammatory markers (CRP, ESR) and SSc duration. To identify factors associated with abnormal MFR (<2.0), univariable and multivariable logistic regression models were developed. Variables with p < 0.05 in univariable analysis were included in the multivariable model. All analyses were conducted with GraphPad Prism (version 10.6.1) and R version 4.4.2. A two-sided p < 0.05 was considered statistically significant.

Results

Study Population

Sixty-seven SSc patients (87% female, median age 61 years, SSc duration 12 ± 10 years) and 201 matched controls (84% female, median age 62 years) participated in the study. The baseline clinical characteristics of the participants are shown in Table 1 – there were no significant differences in age, sex, BMI, or traditional cardiovascular risk factors. In the study cohort, 14 patients (21%) had diffuse cutaneous SSc, 21 patients (31%) had overlap disease, and 20 patients (30%) had a history or current digital ulceration. Additionally, 12 SSc patients (18%) had PAH, and 34 (51%) had ILD.

Table 1.

Clinical characteristics of the study participants

Baseline characteristics	SSc patients n=67	Control participants n=201	p-value
Age, years, mean ± SD	62 ± 9.84	62 ± 14.55	0.94
Body mass index, kg/m2, mean ± SD	29.61 ± 8.06	29.51 ± 9.18	0.94
Biological sex female, n (%)	58 (86.6)	169 (84.1)	0.769
Cardiovascular comorbidities, n (%)
Essential hypertension	34 (50.7)	112 (55.7)	0.571
Dyslipidemia	26 (38.8)	81 (40.3)	0.943
Diabetes	5 (7.5)	19 (9.5)	0.805
Ever smoking	3 (4.5)	6 (3.0)	0.845
Myocardial infarction	2 (3.0)	5 (2.5)	1.000
Coronary artery bypass surgery	0 (0)	0 (0)	NA
Percutaneous coronary intervention	3 (4.5)	12 (6.0)	0.878
Heart failure	6 (9.0)	29 (14.4)	0.346
Cerebrovascular accident	1 (1.5)	4 (2.0)	1.000
Chronic kidney disease	1 (1.5)	5 (2.5)	1.000
Peripheral artery disease	5 (7.5)	12 (6.0)	0.885
Diffuse cutaneous SSc	14 (20.9)		N/A
Overlap disease	21 (31.3)		N/A
Prior/current digital ulceration	20 (29.8)		N/A
Pulmonary artery hypertension	12 (17.9)		N/A
Interstitial lung disease	34 (50.7)		N/A
ESR, median [IQR]	20 [10–36]		N/A
CRP, median [IQR]	3 [1.55–7.35]		N/A
Medications, n (%)
Calcium channel blocker	40 (59.7)	46 (22.9)	<0.001
Angiotensin receptor blocker	14 (20.9)	33 (16.4)	0.458
Statin	26 (29.9)	94 (46.8)	0.32
Aspirin	16 (23.9)	89 (44.3)	0.004
Beta blocker	16 (23.9)	91 (45.3)	0.002
ACE inhibitors	7 (10.4)	60 (29.9)	0.103
Methotrexate	3 (4.5)	0 (0%)	0.0167
Mycophenolic mofetil	24 (35.8)	0 (0%)	<0.001
Glucocorticoids	16 (23.9)	0 (0%)	<0.001
PAH-specific therapy	16 (23.9)	0 (0%)	<0.001
Presenting symptoms, n (%)			0.953
Abnormal ECG	5 (7.5)	18 (9.0)
Dyspnea	29 (43.3)	90 (44.8)
Heart failure	1 (1.5)	5 (2.5)
Chest pain	25 (37.3)	63 (31.3)
Pre-operative evaluation	4 (6.0)	15 (7.5)
Other	3 (4.5)	10 (5.0)

Table 1 presents demographic characteristics, SSc disease features, and concurrent medication use in patients with systemic sclerosis and control participants. Continuous variables are summarized as mean ± standard deviation or median (interquartile range), depending on distribution, and categorical variables as number (percentage). Group comparisons were conducted using the unpaired Mann–Whitney U test for continuous variables and Fisher’s exact test for categorical variables.

Medication profiles varied significantly between SSc patients and controls. Calcium channel blocker use was notably higher in SSc (60% vs 23%; p < 0.001), consistent with their use in Raynaud phenomenon. MMF (36%), glucocorticoids (24%), and PAH–targeted therapy (24%) were used exclusively in the SSc group (all p < 0.001). Methotrexate was used by 4.5% of SSc patients (p = 0.017). In contrast, aspirin (24% vs 44%, p = 0.004) and beta-blockers (24% vs 45%, p = 0.002) were less frequently prescribed in the SSc group. The rates of statin therapy (30% vs 47%, p = 0.320) and use of renin–angiotensin blockers (ACE inhibitors or ARBs) did not differ significantly.

Both groups presented with similar symptoms with dyspnea being the most common at 43% and 45%, followed by chest pain at 37% and 31% for the SSc and control groups, respectively (collective p = 0.953). The prevalence of previous MI, HF, or chronic kidney disease was comparable between the cohorts.

Myocardial Blood Flow and Flow Reserve

Median stress MBF was similar between SSc and controls (2.47 [IQR 1.97–2.84] vs 2.43 [IQR 1.90–3.05] ml/g/min, p = 0.85, Figure 1). Rest MBF was significantly higher in SSc compared to controls (1.14 [IQR 0.91–1.39] vs 1.00 [IQR 0.78–1.25], p = 0.01), resulting in a lower MFR in the SSc group (2.17 [IQR 1.84–2.58] vs 2.44 [IQR 1.96–2.96], p = 0.01). Overall, 39% of SSc patients had abnormal MFR compared to 28% of controls. SSc patients were significantly more likely than controls to fall into the quadrant characterized by preserved stress MBF but reduced MFR due to the abnormal rest MBF (high stress MBF/low MFR), with 33% of SSc patients in this category compared with 15% of controls (p=0.006; Figure 2). After RPP correction, the difference between SSc and controls in resting MBF (corrected rest MBF SSc 1.22 [IQR 1.07–1.45] vs. control 1.13 [IQR 0.95–1.42], p = 0.05), and in MFR (corrected MFR SSc 1.60 [IQR 1.41–1.90] vs. control 1.75 [IQR 1.45–2.06], p = 0.11) no longer persisted, suggesting that hemodynamic factors at least partially responsible for the elevated resting MBF supported by the elevated resting heart rate and elevated diastolic blood pressure in the SSc group. Rates of abnormal perfusion, ischemic ECG changes, and stress LVEF were similar between groups (Table 2). Hemodynamic responses were consistent with vasodilator stress testing (predominantly regadenoson), which typically increases heart rate without a corresponding rise in systolic blood pressure and may slightly lower blood pressure.

Figure 1. Myocardial blood flow (MBF) distribution.

Violin plots depict positron emission tomography–derived resting (Panel A) and stress (Panel B) myocardial blood flow and myocardial flow reserve (MFR, Panel C) in patients with systemic sclerosis (SSc) and control participants.

Figure 2. Stress myocardial blood flow and myocardial flow reserve distribution.

Scatter plot showing positron emission tomography–derived stress myocardial blood flow (MBF) plotted against global myocardial flow reserve (MFR) in patients with systemic sclerosis (SSc) and control participants. Quadrant lines indicate predefined abnormal thresholds for stress myocardial blood flow and myocardial flow reserve. Points are color-coded by study group.

Table 2.

Positron emission tomography (PET) myocardial perfusion imaging indications, stress protocol, hemodynamics, and quantitative results

Imaging Characteristics	SSc n=67	Control n=201	p
Hemodynamics, median [IQR]
Rest HR, bpm	75 [66 – 85]	71 [63 – 80]	0.03
Rest SBP, mmHg	124 [112 – 140]	124 [111 – 139]	0.61
Rest DBP, mmHg	73 [62 – 76]	66 [61 – 74]	0.04
Rest MAP, mmHg	88 [78 – 97]	85 [79 – 94]	0.35
Rest RPP, mmHg*bpm	9126 [7866 – 10540]	8646 [7424 – 10345]	0.15
Stress HR, bpm	96 [88 – 112]	99 [87 – 111]	0.91
Stress SBP, mm Hg	123 [109 – 138]	120 [108 – 138]	0.65
Stress DBP, mmHg	70 [59 – 76]	65 [58 – 71]	0.03
Stress MAP, mmHg	86 [78 – 97]	84 [76 – 93]	0.13
Stress RPP, mmHg*bpm	12483 [9888 – 14933]	11952 [10060 – 14018]	0.52
Stressor agent, n (%)			0.70
Regadenoson	64 (95.5)	195 (97.0)
Dobutamine	3 (4.5)	6 (3.0)
Study results, n (%) or median [IQR]
Ischemic stress ECG response	0 (0.0)	7 (3.0)	0.198
Stress LVEF, %	67.08 ± 9.99	63.59 ± 13.21	0.16
Abnormal perfusion	11 (16)	45 (22)	0.30
Coronary calcification	41 (61.2)	97 (48.3)	0.69
Visual coronary calcium score	1.8 ± 2.2	2.0 ± 2.9	0.37
MBF uncorrected, stress, mL/g/min	2.47 [1.97–2.84]	2.43 [1.90 – 3.05]	0.85
MBF uncorrected, rest, mL/g/min	1.14 [0.91 – 1.39]	1.00 [0.78 – 1.25]	0.01
MFR uncorrected	2.17 [1.84 – 2.58]	2.44 [1.96 – 2.96]	0.01
MBF RPP corrected, stress, mL/g/min	1.98 [1.67 – 2.36]	1.97 [1.57 – 2.46]	0.83
MBF RPP corrected, rest, mL/g/min	1.22 [1.07 – 1.45]	1.13 [0.95 – 1.42]	0.05
MFR RPP corrected	1.60 [1.41 – 1.90]	1.75 [1.45 – 2.06]	0.11

Table 2 summarizes indications for positron emission tomography myocardial perfusion imaging, pharmacologic stressor agents, resting and stress hemodynamic parameters, and quantitative PET results, including myocardial blood flow (MBF) and myocardial flow reserve (MFR), in patients with systemic sclerosis and control participants. Continuous variables are presented as mean ± standard deviation or median (interquartile range), as appropriate, and categorical variables as number (percentage). Between-group comparisons were performed using the unpaired Mann–Whitney U test and Fisher’s exact test.

Factors Associated with Abnormal MFR

On univariable analysis, CCB use (p = 0.01), statin use (p = 0.02), and longer SSc duration (p = 0.03) but not subtype were associated with impaired MFR (Figure 3). MMF therapy (mean ± SD duration of therapy: 3.5 ± 2.8 years) was associated with preserved MFR (p = 0.02). While MFR was numerically higher among patients on PDE-5 inhibitors than among those not receiving pulmonary vasodilators, the difference was not statistically significant (2.35 ± 0.58 vs 2.18 ± 0.46; p = 0.36). In multivariable analysis, MMF remained independently protective (OR 0.09 [95 % CI 0.01–0.56]; p = 0.017), while CCB (OR 7.77 [1.93–42.53]; p = 0.008) and statin use (OR 7.14 [1.86–36.87]; p = 0.008) were independently associated with abnormal MFR (Figure 3). When compared to patients not on statin therapy, patients on statin therapy had no significant difference in the rate of abnormal perfusion (23 vs 12%, respectively, p=0.32), coronary calcifications (69 vs. 56%, respectively, p=0.32) and had similar visual CAC score (2.4 ± 2.7 vs 1.4 ± 1.8, respectively, p=0.16). Calcium channel blocker use was not different in patients with or without digital ulceration history (77 vs 56%, p=0.21). There were no significant correlations between MFR and SSc duration (r=−0.15, p=0.23), ESR (r=−0.20, p=0.46) or CRP (r=−0.09, p=0.13) (Supplementary Figure 1). In addition, MFR was comparable between patients with or without overlap syndrome (2.21 ± 0.37 versus 2.27 ± 0.53 respectively, p=0.10).

Figure 3. Factors associated with reduced myocardial flow reserve.

Forest plot of odds ratios (OR) for clinical factors associated with reduced myocardial flow reserve (MFR <2.0) in univariate (Panel A) and multivariate (Panel B) regression models. CI: confidence interval, dcSSc: diffuse cutaneous systemic sclerosis, PAH: pulmonary artery hypertension, ACE: angiotensin converting enzyme.

Discussion

Our study demonstrated significantly lower PET-derived MFR in patients with SSc compared with matched controls, supporting the presence of diffuse coronary microvascular dysfunction in SSc. Additionally, there was a significantly higher resting MBF in SSc patients, despite similar stress MBF, which suggests baseline microvascular vasodilation and/or increased myocardial oxygen demand at rest. The underlying mechanisms could include disease-related capillary loss, dysregulated vasomotor tone, and abnormal sympathetic innervation; hallmarks of SSc-related vasculopathy. Consequently, decreased MFR in SSc may act as a functional marker of vascular remodeling that increases the risk of ischemic injury, fibrosis, and cardiac dysfunction.

Our findings add to the growing evidence that CMVD is a systemic vascular manifestation of autoimmune rheumatic diseases [15]. We also provide the first evidence that disease-modifying therapy, especially MMF, may help preserve microvascular function in SSc.

Mechanistic Considerations

Microvascular dysfunction is a key pathogenic feature of SSc. Endothelial injury, oxidative stress, and dysregulated nitric oxide signaling drive early vascular remodeling, leading to intimal proliferation, luminal narrowing, and capillary loss. These changes mirror the histopathologic features of digital ischemia and pulmonary arteriopathy seen in advanced disease. SSc CMVD results from structural and functional abnormalities in both the microvasculature and surrounding myocardium [8]. Similar to Raynaud phenomenon, the heart's microvasculature experiences widespread concentric proliferation of the coronary intima, causing lumen narrowing and endothelial dysfunction, even without obstructive epicardial CAD [16–18]. Moreover, fibrosis and inflammation of the surrounding myocardium also contribute to CMVD [7, 19].

Our study’s finding of elevated resting MBF with preserved (or relatively normal) stress MBF is informative when interpreted alongside MFR, which is calculated as stress MBF divided by resting MBF and reflects the coronary microcirculation’s vasodilator reserve. Thus, a pattern of high resting MBF with low MFR does not necessarily indicate impaired hyperemic flow; rather, it suggests that baseline microvascular tone is already relatively vasodilated and/or resting myocardial oxygen demand is increased, leaving less “reserve” to augment flow under stress. Elevated resting flow may indicate tonic vasodilation due to impaired autoregulation potentially related to dysautonomia [20] or increased resting myocardial oxygen demand reflected by increased elevated resting heart rate and elevated diastolic blood pressure in our cohort. In other autoimmune rheumatic diseases, our group has reported similarly abnormal resting MBF and reduced MFR, which correlate with inflammatory burden and long-term outcomes [11]. Collectively, these findings suggest that CMVD in SSc reflects a widespread, systemic vascular process rather than localized ischemia.

Comparison with Prior Imaging Studies

Previous imaging studies have demonstrated myocardial perfusion abnormalities and replacement fibrosis in SSc using modalities such as cardiac magnetic resonance imaging (CMR) and echocardiography [6, 21–23]. Dumitru et al. identified subendocardial perfusion deficits and diffuse myocardial fibrosis on stress CMR in patients with SSc without epicardial coronary disease, supporting a microvascular cause [6]. Weber et al. expanded this understanding across various autoimmune diseases, finding that impaired PET MFR independently predicted mortality and major adverse cardiovascular events, even after accounting for traditional risk factors [9]. In a large PET registry, Feher et al. showed that patients with autoimmune rheumatic diseases had significantly lower MFR than matched controls, and that reduced MFR was associated with a twofold increased risk of cardiac death [11]. Our findings align with and build upon these studies by specifically examining SSc and incorporating pharmacologic exposure into the analysis. They highlight that PET MFR is a sensitive, reproducible indicator of CMVD in SSc, as well as present the first evidence suggesting that immunosuppressive therapy may modulate microvascular function. Previous studies have demonstrated that CMR is an invaluable tool for the assessment of replacement fibrosis [6, 21, 24, 25], myocardial inflammation [21, 24, 26], and subclinical systolic dysfunction in SSc [27]. Building on these findings, the integration of PET with CMR in hybrid PET/MRI systems offers a potential diagnostic advantage. This combined approach enables comprehensive evaluation of cardiac involvement in SSc by integrating detailed tissue characterization with sensitive assessment of microvascular dysfunction and inflammatory activity, thereby enhancing diagnostic accuracy and providing a more holistic understanding of disease pathology.

Therapeutic Associations

Calcium Channel Blockers and Statins

The link between CCB or statin use and abnormal MFR probably stems from confounding by indication and the physiological effects of these therapies (see Figure 3). CCBs are often prescribed for patients with Raynaud phenomenon or vasospastic angina, which are features of a vascular-predominant SSc phenotype that may not be fully evident from digital ulcer history alone. Additionally, CCBs can increase resting MBF, potentially lowering the MFR ratio without implying coronary microvasculature structural disease. Consequently, lower MFR in patients on CCBs might reflect both the hemodynamic effects of vasodilator treatment and a greater burden of microvascular dysfunction. Similarly, statins provide limited anti-inflammatory and endothelial benefits and are mainly used for dyslipidemia or macrovascular disease prevention, not to improve microvascular health.

Mycophenolate Mofetil

In our study, MMF use was independently associated with preserved MFR (Figure 3). Although this suggests that immunosuppressive therapy might relate to better microvascular profiles in SSc, it should be interpreted cautiously. MMF is often prescribed for patients with active inflammatory or fibrotic disease, but treatment decisions may also depend on provider preference, comorbidity profiles, or unmeasured clinical factors that could affect vascular function independently of MMF use. MMF works by inhibiting inosine monophosphate dehydrogenase, which reduces lymphocyte proliferation and cytokine release and decreases oxidative stress and vascular inflammation. Preclinical research shows that MMF increases nitric oxide (NO) availability, decreases endothelial cell apoptosis, and reduces fibroblast activation [28], mechanisms that could plausibly support microvascular health. However, our cross-sectional design cannot determine whether MMF directly improves coronary microvascular function or if patients on MMF simply represent a subgroup with different microvascular trajectories. Prospective longitudinal studies with serial PET imaging will be necessary to clarify whether MMF has a disease-modifying effect on CMVD in SSc.

Limitations

Our study has several limitations. It was a retrospective analysis conducted at a single center with a relatively small sample size, which limits the generalizability of the results. Medication use was not randomized and might reflect underlying disease severity, potentially confounding the outcomes. The lack of longitudinal follow-up data prevented analysis of the temporal relationship between therapy and microvascular function. Although matching helped control for common cardiovascular confounders, unmeasured factors like autonomic dysfunction, microinfarction load, or inflammatory processes might have influenced the results. Additionally, invasive coronary physiology measurements were not available to validate PET-derived MFR, but previous research has shown a strong correlation between PET and invasive flow assessments [29].

Future Directions

Future studies are needed to verify these results and determine whether immunosuppressive treatments can improve coronary microvascular function over time. Serial PET scans might show how MFR changes dynamically with MMF therapy or dose adjustments. Combining PET results with blood biomarkers of endothelial damage, such as endothelin-1, VEGF, adhesion molecules, and CMR tissue characterization [30], could enhance understanding of how inflammation, fibrosis, and vascular health are interconnected.

Conclusion

In summary, patients with SSc show a high prevalence of CMVD compared to controls with similar co-morbidities, indicated by increased resting MBF and decreased MFR on PET perfusion imaging. These results may reflect systemic endothelial damage and impaired autoregulation commonly seen in SSc. The association between MMF use and preserved MFR suggests that immunosuppressive treatments might provide vascular protection, while the use of CCB and statins could indicate more advanced disease though no definitive conclusions can be drawn. Nonetheless, our results demonstrate that PET imaging is a quantitative, noninvasive method for detecting and monitoring CMVD in SSc, potentially serving as a biomarker for assessing CV risk and treatment response.

Key messages:

• SSc patients show reduced MFR compared to controls.

• PET-derived MFR could be a noninvasive biomarker for vascular involvement and treatment response in SSc.

• Associations between MMF use and MFR warrant cautious interpretation and prospective validation.

Data availability statement:

The data underlying this article will be shared on reasonable request to the corresponding author.