Combined diagnostic value of the systemic immune-inflammation index and LDL for coronary microvascular disease: a retrospective cohort study

Abstract

Background:

Coronary microvascular disease (CMVD) is an important contributor to ischemic heart disease. Symptomatic patients without obstructive coronary lesions may still have microvascular ischemia, which is associated with an increased risk of major adverse cardiovascular events. The diagnosis of CMVD relies on coronary function testing or noninvasive imaging techniques, which are not universally available and may limit routine clinical implementation.

Objectives:

Growing evidence supports the notion that biological processes such as inflammation and oxidative stress drive vascular aging and contribute to microvascular dysfunction. The systemic immune-inflammation index (SII) is a marker used to assess immune-inflammatory and thrombotic status and has shown prognostic value in chronic heart failure, and acute myocardial infarction. However, its role in CMVD remains underexplored.

Design:

To address this gap, we retrospectively analyzed 180 patients with angina who underwent myocardial contrast stress echocardiography between January and November 2022 at the First Affiliated Hospital of Harbin Medical University.

Methods:

Based on coronary microcirculatory function, patients were divided into two groups: 91 with normal function and 89 with microcirculatory disorders. Clinical data and lab indicators were compared.

Results:

SII was significantly higher in the CMVD group (722.96 ± 354.10 vs 618.51 ± 324.50; p < 0.05). Logistic regression identified SII, LDL cholesterol as independent CMVD predictors. Pearson correlation demonstrated an inverse relationship between SII and coronary flow reserve, supporting its link with impaired microvascular function. Receiver operating characteristic analysis further showed that the combined SII and LDL model provided improved discriminatory performance compared with either marker alone, achieving an area under the curve of 0.76 with a sensitivity of 77.5% and specificity of 69.2%.

Conclusion:

SII is independently associated with CMVD and improved discriminatory performance when combined with LDL.

Plain Language Summary

Can Combining Inflammation and Cholesterol Blood Markers Improve the Detection of Coronary Microvascular Disease?

Why was the study done?

Coronary microvascular disease (CMVD) is a heart condition that affects the small blood vessels of the heart. It can cause chest pain and other symptoms even when the major coronary arteries appear normal. Detecting CMVD can be challenging, and doctors are looking for simple blood tests that may help identify patients at risk. The Systemic Immune-Inflammation Index (SII) is a blood marker that reflects inflammation and clotting activity in the body. It has already been shown to predict outcomes in cancer, neurological diseases, heart failure, and heart attacks. However, its role in CMVD has not been well studied.

What did the researchers do?

In this study, we reviewed medical records from 180 patients who underwent a specialized heart ultrasound test (myocardial contrast stress echocardiography) in 2022. Based on their coronary microvascular function, patients were divided into two groups: those with normal small-vessel function and those with CMVD.

What did the researchers find?

We found that SII was independently associated with CMVD in patients with angina. Among the evaluated inflammatory indices, SII showed improved discriminatory performance compared with NLR, PLR, while the combined SII and LDL model further improved diagnostic performance over either biomarker alone. In addition, modest but significant negative correlation between SII, LDL, and CFR.

What do the findings mean?

The findings suggested a potential relationship between inflammatory and lipid-related pathways and impaired coronary microvascular function. To the best of our knowledge, this is among the first studies to evaluate the diagnostic value of SII for CMVD assessment in patients with angina. Given their low cost and routine availability in clinical practice, the combination of SII and LDL may represent a practical approach for identifying patients at risk of CMVD; however, prospective multicenter studies are needed to validate its clinical utility.

Introduction

Coronary microvascular disease (CMVD) is increasingly recognized as a significant contributor to the development of ischemic heart disease. Patients with symptoms but without detectable obstructive plaques may still have significant nonobstructive atherosclerosis and microvascular ischemia, which are associated with higher rates of major adverse cardiovascular events. ¹ A recent systematic review and meta-analysis revealed that, among patients with angina or ischemia but no obstructive coronary artery disease (ANOCA/INOCA), approximately half have underlying conditions such as CMVD or coronary vasospasm. ² Currently, the diagnosis of CMVD can only be confirmed through invasive cardiac catheterization, such as IMR (index of microcirculatory resistance) or noninvasive imaging techniques. ³ Therefore, it is crucial for physicians to remain vigilant for ischemia even in the absence of obstructive coronary arteries, to ensure accurate diagnosis and provide personalized patient management.

A meta-analysis found that CMVD is highly prevalent in both sexes; however, women are more likely to be affected. ⁴ Microembolization caused by atherosclerotic debris, platelet–neutrophil recruitment, and fibrin deposition can influence the degree of microvascular occlusion. ⁵ Additionally, growing evidence supports the notion that biological processes such as inflammation and oxidative stress drive vascular aging and contribute to microvascular dysfunction as a result of exposure to cardiovascular risk factors such as hypertension, diabetes, smoking, and elevated low-density lipoprotein (LDL) levels.^6,7 Recent evidence suggests a positive association between inflammation, CMVD, and epicardial adipose tissue, a metabolically active visceral fat depot located between the myocardium and visceral pericardium, that may modulate coronary microvascular function through the secretion of pro-inflammatory mediators. ⁸ The IL-1β/TNF-α/IL-6/CRP pro-inflammatory pathway, along with epigenetic changes in the promoter region of TNF-α^9,10 has been significantly associated with patients suffering from angina pectoris in combination with cardiovascular disease (CVD), further supporting the role of systemic inflammation in CMVD.^11–13

The systemic immune-inflammation index (SII) is a novel inflammatory marker that has demonstrated prognostic value in infections, chronic inflammatory diseases, and various cancers.^14–16 SII has also been studied in the context of CVD and identified as a potential biomarker for isolated coronary artery ectasia, saphenous vein graft disease.^17–20 A retrospective cohort study showed that, among patients with heart valve disease, the incidence of poor outcomes, 30-day mortality, and 30-day readmission was significantly higher in the high SII group compared to the low SII group. ²¹ Another study demonstrated that SII is an independent predictor of mortality in patients with advanced chronic heart failure and renal dysfunction. ²² However, no study has specifically investigated its predictive value in CMVD.

Thus, in this cohort study, we investigated the predictive value of SII in symptomatic patients with CMVD and examined its correlation with coronary flow reserve (CFR), a key parameter used in the diagnosis of CMVD.

Subjects and methods

Data collection

This study was conducted as a single-center, retrospective cohort analysis. From January 2022 to November 2022, 180 patients with angina who underwent myocardial contrast echocardiography at First Affiliated Hospital of Harbin Medical University were consecutively included in this retrospective cohort study. The inclusion criteria were: (1) patients with symptoms of myocardial ischemia (exertional or resting angina); (2) evidence of myocardial ischemia on a resting electrocardiogram (at least two contiguous leads with ST-segment depression ⩾0.1 mV) or a positive treadmill exercise test; (3) coronary angiography or coronary computed tomographic angiography (CCTA) showing <50% stenosis in the affected coronary vessels; (4) availability of complete clinical data. The exclusion criteria included: (1) patients with a history of percutaneous coronary intervention or coronary artery bypass grafting; (2) patients with a history of myocardial infarction; (3) patients with an ejection fraction <50%, valvular heart disease, or cardiomyopathy; (4) patients with malignancies or autoimmune diseases; (5) patients with severe liver or kidney dysfunction, or those allergic to adenosine triphosphate; (6) patients with acute or chronic infections. Selection bias was minimized through predefined inclusion and exclusion criteria and age- and sex-matched group selection from the hospital’s electronic health records. Information bias was reduced by using objective laboratory, imaging, and pharmacy data extracted from medical records rather than self-reported information. No formal a priori sample size calculation was performed due to the retrospective study design; however, the final sample size was considered adequate for receiver operating characteristic-based analysis assuming an expected area under the curve (AUC) of approximately 0.75 and a CMVD prevalence of 40%. ² Ethical approval for the use of human subjects was obtained from the Research Ethics Committee of the First Affiliated Hospital of Harbin Medical University. All patient information and medical records were de-identified and/or anonymized prior to analysis, and the requirement for informed consent was waived. The reporting of this study conforms to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Supplemental Materials). ²³

Clinical and laboratory measurements

All participants completed a questionnaire regarding medical history and tobacco use. Body mass index (BMI) was calculated as body weight (in kilograms) divided by height (in meters squared). Hypertension was defined as a systolic blood pressure >140 mmHg or a diastolic blood pressure >90 mmHg on three separate occasions. Diabetes mellitus was defined as a fasting blood glucose level >126 mg/dL (7 mmol/L) on two separate tests. Hyperlipidemia was defined as total cholesterol >200 mg/dL (5.2 mmol/L) or triglycerides >150 mg/dL (1.7 mmol/L). High-density lipoprotein (HDL) cholesterol was considered low if <40 mg/dL (1.0 mmol/L) in men or <50 mg/dL (1.3 mmol/L) in women ²⁴ ; LDL cholesterol was considered elevated if >100 mg/dL (2.59 mmol/L). Additionally, all participants underwent a complete blood count, cardiac enzyme testing, a lipoprotein panel, and blood tests for kidney function during hospitalization.

Coronary computed tomographic angiography

Coronary computed tomographic angiography (CCTA) was performed using a 256-slice scanner (Brilliance iCT; Philips Healthcare, Cleveland, USA), which enables comprehensive imaging of the heart within a short time frame and with low radiation exposure. This provides enhanced visual detail of the heart’s function and structure. The images were reconstructed in 3D format to assess heartbeat strength and detect plaque deposits within the coronary arteries. Procedures were followed as previously described. ²⁵ Oral and/or intravenous β-blockers were administered to reduce heart rate when it exceeded 70 beats/min. The CCTA results were evaluated by two imaging specialists.

Invasive coronary angiography

Invasive coronary angiographic images were obtained using the Allura Xper FD 20/10 system (Philips Medical Systems), with guidance from CCTA performed prior to the procedure. Cardiac ultrasound, coagulation tests, and biochemical blood tests were conducted and evaluated before angiographic imaging. Detailed analysis of the angiographic images was performed by two experienced interventional cardiologists.

Myocardial contrast stress echocardiography

Echocardiographic images were obtained using the Philips CX50 POC echocardiography system (Philips Ultrasound, Bothell, USA) and stored for subsequent analysis. Myocardial contrast stress echocardiography (MCSE) was performed in the apical 4-, 2-, and 3-chamber views at rest and during pharmacological stress at 3, 5, and 6 min following intravenous administration of ATP at a maximum dose of 140 μg/kg/min over 6 min.^26,27 SonoVue was initially infused at a rate of 0.8–0.9 mL/min using a dedicated Vueject pump (Bracco, Geneva, Switzerland) to achieve optimal myocardial opacification with minimal attenuation. Patients were continuously monitored via blood pressure and electrocardiography before, during, and for 20 min after ATP infusion. CFR was calculated as the ratio of total myocardial blood flow during stress to that at rest. To prevent interaction, a 3- to 5-day interval was maintained between coronary angiography and MCSE.

Systemic immune-inflammation index

The SII was calculated as follows: SII = P × N/L, where P, N, and L represent the preoperative peripheral platelet, neutrophil, and lymphocyte counts, respectively.

Statistical analysis

Participants were divided into two groups: those with CMVD and those without. Clinical characteristics and investigation results were compared between the groups. Statistical analyses were performed using statistical packages in R version 4.1.1 (R Core Team, 2021). Chi-square analysis was used to compare categorical variables, which are presented as percentages. The Kolmogorov–Smirnov test was employed to assess the normality of distribution for continuous variables. For variables following a normal distribution, the t test was used, with results expressed as mean ± standard deviation. For non-normally distributed variables, the Mann–Whitney U test was applied. Both univariate and multivariate logistic regression analyses were performed to identify independent risk factors for CMVD. Hosmer–Lemeshow test was used to evaluate the model’s calibration and goodness-of-fit. Receiver operating characteristic (ROC) analysis was used to evaluate discrimination, sensitivity and specificity and optimal cutoff values. Incremental value was assessed by comparing AUCs between nested models. Internal validation was performed using 10-fold cross-validation. A two-tailed p value <0.05 was considered statistically significant. Pearson correlation analysis was performed to assess the linear relationship between two quantitative variables.

Results

Baseline clinical characteristics of participants

Based on CFR values—the standard for diagnosing CMVD ²⁸ —all participants with angina were categorized into the CMVD group if CFR <2 (n = 89), and into the control group if CFR ⩾ 2 (n = 91). ²⁹ The characteristics of the two groups are presented in Table 1. The median age of all participants was 57 years, and 66.1% were female. No significant differences were observed between the groups in terms of BMI, blood pressure, left ventricular ejection fraction, or prior medication use.

Table 1.

Baseline characteristics of the study population.

Characteristics	Control (n = 91)	CMVD (n = 89)	p Value
Age (year)	57.98 ± 9.67	59.17 ±10.29	0.5
Sex, n (%)			0.6
Male	29 (31.87%)	32 (35.96%)
Female	62 (68.13%)	57 (64.04%)
BMI (kg/m2)	21.70 ± 2.28	21.17 ± 2.13	0.078
Systolic blood pressure (mmHg)	136.10 ± 19.97	138.27 ±20.91	0.6
Diastolic blood pressure (mmHg)	85.46 ± 11.95	83.15 ±12.05	0.2
Heart rate (bpm)	77.86 ± 12.12	74.58 ± 12.40	0.070
Diabetes mellitus (n, %)	12.00 (13.19%)	13.00 (14.61%)	0.8
Smoking (n, %)	5.00 (5.49%)	12.00 (13.48%)	0.067
Hypertension (n, %)	32.00 (35.16%)	38.00 (42.70%)	0.3
History of hyperlipidemia (n, %)	34.00 (37.36%)	36.00 (40.45%)	0.7
Left ventricular ejection fraction (n, %)	65.36 ± 5.80	64.70 ± 5.37	>0.9
Previous drug used
β-blocker (n, %)	11.00 (12.09%)	15.00 (16.85%)	0.4
Statins (n, %)	34.00 (37.36%)	36.00 (40.45%)	0.7
Clopidogrel (n, %)	4.00 (4.40%)	4.00 (4.49%)	>0.9
ACEI/ARB (n, %)	32.00 (35.16%)	27.00 (30.34%)	0.5

CMVD, coronary microvascular disease.

Laboratory markers, including neutrophil count, platelet count, cardiac troponin, creatine kinase-MB, ³⁰ and glomerular filtration rate, were significantly higher in the CMVD group compared to the control group (p < 0.001). Total cholesterol levels were significantly lower, whereas LDL levels were significant higher, in the CMVD group (total cholesterol: 4.43 ± 0.96 mmol/L; LDL: 3.25 ± 0.8 mmol/L) than in the control group (total cholesterol: 4.66 ± 0.98 mmol/L; LDL: 2.97 ± 0.85 mmol/L; p < 0.001; Table 2). Moreover, higher platelet-to-lymphocyte ratio (PLR) (p = 0.012), neutrophil-to-lymphocyte ratio (NLR) (p < 0.001), and monocyte-to-lymphocyte ratio (MLR) (p < 0.001) values were observed in the CMVD group. A particularly notable finding was the significantly elevated SII in the CMVD group (722.96 ± 354.10) compared to the control group (618.51 ± 324.50; p < 0.001; Figure 1). In contrast, there were no significant differences between the two groups in other laboratory parameters, including white blood cell count, monocyte count, lymphocyte count, serum glucose, B-type natriuretic peptide (BNP), creatinine, and uric acid levels (p > 0.05).

Table 2.

Laboratory markers and analysis results of myocardial stress echocardiography.

Characteristics	Control	CMVD	p Value
White blood cell count (×109/L)	6.44 ± 1.86	6.41 ± 1.89	0.4
Monocyte count (×109/L)	0.36 ± 0.12	0.41 ± 0.19	0.079
Neutrophil count (×109/L)	3.97 ± 1.58	4.82 ± 1.64	<0.001*
Lymphocyte count (×109/L)	2.01 ± 0.60	1.91 ± 0.66	0.11
Platelet count (×109/L)	245.20 ± 59.64	270.17 ± 79.89	0.012*
Hemoglobin (g/dL)	140.37 ± 13.66	143.70 ± 15.25	0.4
Serum glucose (mmol/L)	5.49 ± 1.67	5.76 ± 1.95	0.5
Cardiac troponin (ng/mL)	8.30 ± 1.22)	10.42 ± 1.13	<0.001*
Creatine kinase-MB (U/L)	2.55 ± 0.41	2.77 ± 0.53	<0.001*
B-type natriuretic peptide (pg/mL)	23.26 ± 28.01	23.95 ± 26.60	0.2
Total cholesterol (mmol/L)	4.66 ± 0.98	4.43 ± 0.96	<0.001*
Triglyceride (mmol/L)	1.70 ± 1.02	1.73 ± 0.96	0.6
High-density lipoprotein (mmol/L)	1.13 ± 0.29	1.11 ± 0.29	0.3
Low-density lipoprotein (mol/L)	2.97 ± 0.85	3.25 ± 0.80	<0.001*
Serum creatinine (mmol/L)	68.77 ± 18.07	69.11 ± 18.09	0.7
Glomerular filtration rate (mL/min/1.73 m2)	110.09 ± 6.35	112.66 ± 5.7	<0.001*
Uric acid (µmol/L)	326.6 ± 85.81	323.63 ± 95.72	0.3
SII	618.51 ± 324.5	722.96 ± 354.10	<0.001*
Platelet-to-lymphocyte ratio	130.58 ± 49.78	150.87 ± 55.79	0.012*
Neutrophil-to-lymphocyte ratio	2.14 ± 1.12	2.72 ± 1.19	<0.001*
Monocyte-to-lymphocyte ratio	0.19 ± 0.07	0.22 ± 0.09	0.009*
Monocyte-to-high-density lipoprotein ratio	0.33 ± 0.15	0.41 ± 0.24	0.058
CFR	2.11 ± 0.54	1.68 ± 0.24	<0.001*

CFR, coronary flow reserve; SII, systemic immune-inflammation index.

^*p < 0.05.

Figure 1.

Comparison of mean SII levels between CMVD and control groups.

CMVD, coronary microvascular disease; SII, systemic immune-inflammation index.

SII is independently associated with CMVD

Univariate and multivariate logistic regression analyses were performed to evaluate the associations between inflammatory biomarkers, LDL, and CMVD. To avoid multicollinearity among inflammatory indices, SII, NLR, PLR, and MLR were evaluated in separate multivariable models. Among these variables, after adjustment for age, sex, and medications, SII demonstrated the strongest association with CMVD (OR = 3.71, p = 0.016), performing similarly to LDL (OR = 3.77, p < 0.001). This association was notably stronger than those of PLR (OR = 2.37, p = 0.007), NLR (OR = 1.97, p = 0.03), and MLR (OR = 1.94, p < 0.033), all of which were also independently associated with CMVD (Table 3).

Table 3.

Binary logistic regression analysis.

Characteristics	Univariate analysis			Multivariate analysis			Calibration p value*
	OR	95% CI	p Value	OR	95% CI	p Value
SII	3.86	1.45–12.2	0.011	3.71	1.36–11.9	0.016	0.6314
PLR	2.36	1.30–4.32	0.005	2.37	1.27–4.49	0.007	0.8705
NLR	2.15	1.19–3.94	0.012	1.97	1.06–3.7	0.033	0.7032
MLR	2.05	1.14–3.74	0.018	1.94	1.06–3.61	0.033	0.4953
LDL	3.45	1.88–6.43	<0.001	3.77	2.01–7.25	<0.001	0.9762

SII, systemic immune-inflammation index.

^*Hosmer and Lemeshow goodness-of-fit test was used. The calibration p value >0.05 indicates that the logistic regression model has an excellent fit and is well-calibrated.

Incremental diagnostic value of combined SII and LDL for CMVD

ROC curve analysis was subsequently performed using model-derived predicted probabilities to evaluate and compare the diagnostic performance of SII, and LDL for CMVD (Figure 2; Table 4). Other inflammatory indices (PLR, NLR, and MLR) were not included in ROC analysis, as they were not the primary focus of diagnostic model construction in this study. The results showed that SII (AUC = 0.691, 95% CI = 0.614–0.768) and LDL (AUC = 0.691, 95% CI = 0.613–0.768) demonstrated moderate discriminatory ability for CMVD. The combined model incorporating SII and LDL significantly improved diagnostic performance (AUC = 0.759, 95% CI = 0.688–0.83), with better sensitivity compared with individual biomarkers. Consistently, DeLong test confirmed that the combined SII + LDL model significantly outperformed both SII alone (Z = −2.06, p = 0.040) and LDL alone (Z = −2.35, p = 0.019) in terms of discriminatory ability, indicating incremental diagnostic value from integrating inflammatory and lipid-related biomarkers.

Figure 2.

ROC curves for SII, LDL, and SII + LDL.

AUC, area under the curve; ROC, receiver operating characteristic; SII, systemic immune-inflammation index.

Table 4.

Receiver operating characteristic (ROC) curve analysis.

ROC							DeLong test*
Variables	AUC	95% CI	Sensitivity	Specificity	Youden index	Threshold	Z score	p Value
SII	0.691	0.614–0.768	0.438	0.846	0.284	706.375	−2.0565	0.04
LDL	0.691	0.613–0.768	0.573	0.769	0.342	3.14	−2.3531	0.019
SII + LDL	0.759	0.688–0.83	0.775	0.692	0.468	N/A	N/A	N/A

ROC, receiver operating characteristic; SII, systemic immune-inflammation index.

^*DeLong Test compared AUCs for SII or LDL model alone with SII and LDL combined model.

SII negatively correlates with CFR

Furthermore, Pearson’s correlation analysis demonstrated but statistically significant negative correlation between SII and CFR (R = −0.233, p = 0.002; Figure 3). Similar inverse associations were observed in both male (R = −0.303, p = 0.017) and female (R = −0.254, p = 0.005) subgroups (Table 5). LDL also showed a modest negative correlation with CFR (R = −0.306, p = 0.001), which remained significant only in the female subgroup (R = −0.348, p < 0.001).

Figure 3.

Pearson analysis show a negative correlation between SII scores and the value of CFR.

CFR, coronary flow reserve; SII, systemic immune-inflammation index.

Table 5.

Correlations between SII, LDL, PLR, NLR, and the value of CFR.

Variables	All patients		Male		Female
	R	p Value	R	p Value	R	p Value
SII	−0.233	0.002*	−0.303	0.017*	−0.254	0.005*
LDL	−0.306	0.001*	−0.213	0.099	−0.348	<0.001*
PLR	−0.182	0.046*	−0.201	0.121	−0.173	0.059
NLR	−0.211	0.021*	−0.213	0.099	−0.184	0.045*

CFR, coronary flow reserve; SII, systemic immune-inflammation index.

^*p < 0.05.

Discussion

This study demonstrated that SII was independently associated with CMVD in patients with angina. Among the evaluated inflammatory indices, SII showed improved discriminatory performance compared with NLR, PLR, while the combined SII and LDL model further improved diagnostic performance over either biomarker alone. In addition, modest but significant negative correlation between SII, LDL, and CFR, suggesting a potential relationship between inflammatory and lipid-related pathways and impaired coronary microvascular function. To the best of our knowledge, this is among the first studies to evaluate the diagnostic value of SII for CMVD assessment in patients with angina. Given that SII and LDL are derived from routine laboratory parameters, they may represent a simple, cost-effective, and easily accessible biomarkers with potential clinical utility for CMVD risk stratification.

However, the overall discriminatory ability of these biomarkers remained modest, with AUC values below 0.80. This indicates that, although statistically significant, their capacity to distinguish CMVD from non-CMVD cases is limited, likely reflecting the complex and multifactorial nature of coronary microvascular impairment. CMVD involves heterogeneous mechanisms, including functional dysregulation, structural remodeling, and vasomotor abnormalities, which cannot be fully captured by a single circulating biomarker. Therefore, SII and LDL should be considered adjunctive indicators rather than standalone diagnostic tools.

Lipid metabolism and inflammation are closely intertwined in the pathogenesis of atherosclerosis. Extensive evidence has established that LDL plays a causal role in the development and progression of atherosclerotic CVD, making LDL measurement a key component of risk assessment in all major international guidelines.^31,32 Specifically, LDL has also been identified as an independent predictor of CMVD. It is negatively correlated with CFR ³³ and positively correlated with the index of microvascular resistance. ³⁴ Clinical evidence suggests that targeting either lipid or inflammatory pathways alone leaves substantial residual cardiovascular risk, whereas simultaneous regulation of both pathways provides greater clinical benefit. ³⁵ Furthermore, recent studies have linked increased vascular-specific epicardial adipose tissue and peri-coronary adipose tissue volumes to coronary artery spasm ³⁶ and CMVD. ⁸ In obesity and coronary artery disease (CAD), the secretome of these adipose depots shifts toward a pro-inflammatory phenotype, potentially contributing to coronary vasomotor dysfunction. These observations underscore the interconnected roles of dyslipidemia and inflammation in the pathogenesis of CMVD. Therefore, integrating lipid-related and inflammatory biomarkers may improve the diagnostic discrimination in patients with CMVD. ³⁷

Inflammation plays a pivotal role in the development, pathogenesis, and prognosis of CVDs. Anti-inflammatory therapy targeting the interleukin-1β pathway with canakinumab has been shown to significantly reduce the rate of recurrent cardiovascular events, including myocardial infarction, stroke, cardiovascular death, and hospitalization for heart failure.^38,39 Eighteen pro-inflammatory biomarkers, including those related to the TNF-α/IL-6 pathway, have been reported to be significantly associated with CMVD in females. As an emerging inflammatory marker, SII has demonstrated value in predicting adverse clinical outcomes in CVD and is significantly associated with the SYNTAX score, which is used to assess the severity of CVD.^17,40 A 20-year follow-up cohort study reported that individuals with SII levels >655.56 had higher all-cause and cardiovascular mortality compared to those with SII levels <335.36 in the general population. ⁴¹ The present study demonstrated that, in patients with symptoms but without detectable obstructive plaques, an SII value ⩾706.38 was associated with a higher probability of having CMVD. This finding highlights the potential clinical utility of SII in this specific patient population.

Calculated based on platelet, neutrophil, and lymphocyte counts, SII was found to be independently associated with CMVD. Meanwhile, we also observed that platelet and neutrophil counts were significantly higher in CMVD patients compared to controls, consistent with previous cohort studies. ⁴² State-of-the-art RNA sequencing of whole blood from patients with INOCA and CAD has identified neutrophil activation as a key mechanism potentially contributing to impaired cardiac microvascular function, whereas transcriptomic changes in CAD are mainly associated with T lymphocytes, particularly a reduction in regulatory T-cell populations. ⁴³ In addition, activated platelets secrete more than 300 proteins and molecules, including thromboxane A2, serotonin, and ADP, which directly cause coronary vasoconstriction.^44,45 Women with ANOCA were found to have significantly higher thrombin-induced platelet–fibrin clot strength (TIP–FCS, mm), clotting index, and fibrinogen activity than men, as measured by thromboelastography. ⁴⁶ Together, neutrophils and platelets form complexes that further activate both cell types, amplifying information and coagulation responses. This cascade contributes to microvascular remodeling, endothelial dysfunction, and reduced CFR, all of which are central features of CMVD. ⁴⁷

In addition, lymphocyte homeostasis and cross-talk between platelets and lymphocytes have been discussed for decades due to their vital importance in endothelial dysfunction and atherothrombosis.^48,49 Significant direct correlations were found between peripheral CD4 subpopulations Th1 and Th17 and microvascular oxidative stress. In contrast, an inverse correlation was observed between Tregs and microvascular damage. ⁵⁰ These findings highlight the significant involvement of lymphocytes in CMVD. However, changes in subsets or balance of these cells explain why the overall peripheral lymphocyte count did not differ between the CMVD and control groups in our patients.

Symptomatic women are less likely than men with similar symptoms to have obstructive CAD; instead, they are more prone to CMVD, plaque erosion, and thrombus formation.^24,30,51 This occurs despite symptomatic women being, on average, over 10 years older than men at presentation and having more risk factors. ⁵¹ The postmenopausal period may further accentuate this sex-related difference through increased adipose-associated inflammatory cytokine production. ⁵² In this study, we found that approximately 48% of women presenting with chest pain had a CFR <2. However, no significant difference in the prevalence of CMVD was observed between genders, although the average age of women was approximately 3 years higher than that of men in our cohort. A negative correlation between the SII and CFR was observed in both men and women, though the correlation was slightly weaker in women. Considering the modest correlation due to the relatively small sample size, its association with CFR warrants further investigation. Further research should explore how this differs by gender. Additionally, our data showed that plasma LDL levels were negatively correlated with CFR values, but this correlation was observed only in the female group when analyzed by gender. LDL has also been identified negatively correlated with CFR in asymptomatic men at high risk for CAD ³³ and positively correlated with the index of microvascular resistance (IMR), another important metric for assessing microvascular function of coronary arteries. ³⁴ However, both studies show a modest correlation due to their relatively small sample size. To our knowledge, no gender-specific differences in the association between LDL and CMVD have been previously reported. Further investigation is needed to clarify this potential disparity.

CMVD encompasses structural and functional abnormalities of the coronary microcirculation responsible for myocardial ischemia. Although CFR is widely used for physiologic assessment of CMVD, CFR alone may not fully characterize all endotypes of coronary vasomotor dysfunction, particularly vasospastic mechanisms. ⁸ Notably, coronary vasospasm may occur despite preserved CFR and normal microvascular resistance indices. Therefore, additional vasoreactivity assessment, including acetylcholine provocation testing, may provide complementary diagnostic information in selected patients within the broader spectrum of ANOCA/INOCA.^1,53–55 Despite the considerable number of noninvasive and invasive diagnostic tools for disease characterization and risk stratification, research on emerging therapeutic strategies and clinically accessible biomarkers for CMVD remains limited. In this context, our findings—which combine SII and LDL to reflect both inflammatory and lipid-related pathways—display improved discriminatory performance for CMVD and may provide complementary information for CMVD risk assessment.

Several limitations should be acknowledged. First, the retrospective and single-center design may introduce selection bias and limit the generalizability of the findings, and the absence of external validation further limits the robustness and transportability of the model. Second, the SII was calculated at only one time point; therefore, temporal variations in inflammatory status and their relationship with symptom onset and disease progression could not be evaluated. Accordingly, larger prospective multicenter studies with longitudinal measurements are needed to further validate and extend these findings. Another limitation of this study is that CMVD was defined based on CFR. Consequently, different INOCA endotypes, including epicardial vasospasm, microvascular spasm, endothelial dysfunction, and mixed phenotypes, could not be characterized within the current study design. Future studies incorporating comprehensive coronary function testing may help further clarify the relationship between inflammatory markers and specific mechanisms of coronary vasomotor dysfunction.

Conclusion

To our knowledge, this is the first report to demonstrate the predictive value of the SII for assessing the risk of CMVD in patients with angina. Our findings suggest that SII is a novel, independent marker associated with CMVD and has a stronger association than other inflammatory indices, such as NLR, PLR, and MLR, for identifying CMVD in symptomatic patients. SII and LDL provide improved discriminatory performance compare with either marker alone. Given their low cost and routine availability in clinical practice, the combination of SII and LDL may represent a practical approach for identifying patients at risk of CMVD; however, prospective multicenter studies are needed to validate its clinical utility.