Introduction

Coronary microvascular disease (CMVD) as a subtype of ischemia with non-obstructive coronary artery disease (INOCA) is a non-atherosclerotic disease leading to heart ischemia1. Some studies have shown that patients with have a significantly increased risk of developing heart failure with preserved ejection fraction (HFpEF)2. The GUSTO IIb study found that non-obstructive chest pain was present in 4.2–30.5% of 12,142 patients with ST-segment elevation myocardial infarction (STEMI), non-ST-segment elevation myocardial infarction (NSTEMI) or unstable angina3. Moreover, a study using PET as a diagnostic tool to explore the incidence of CMVD found that myocardial perfusion abnormalities were present in 51% of male and 54% of female patients on the prevalence of CMVD using PET as a tool4. In recent years, the pathophysiology of CMVD has been attracted physician to extensive attention, and various hypotheses have emerged, platelet activation, capillary thinning, and inflammatory response5,6.

The coronary endothelium cells play a pivotal role in maintaining microcirculatory hemodynamics. Some risk factors such as diabetes, hyperlipidemia and insulin resistance (IR) can contribute to endothelial damage and consequently arteriolar vasomotor dysfunction. The TG/HDL-C ratio was initially regard as an indicator of atherosclerosis7 and the Triglyceride to high density lipoprotein (TG/HDL) was more valuable than other single lipid indices, due to strength of reflecting the complex interaction between lipoprotein metabolism, and better predicting the development of atherosclerosis8. Moreover, TG/HDL is considered a biomarker of IR, closely related to metabolic syndrome9. Elevated TG/HDL is not only strongly associated with IR, but is also often a concomitant phenomenon of centripetal obesity, both of which are key drivers of escalating cardiovascular disease risk10. IR can affect multiple systems throughout the body, including lipid metabolism, which in turn promotes the development of atherosclerosis. Recent studies have shown that inflammation not only directly contributes to vascular endothelial damage and promotes plaque formation and destabilization, but also has an intricate interaction with dyslipidemia11. Disturbances in lipid metabolism, especially the oxidation of low-density lipoprotein cholesterol (LDL-C) and the accumulation of triglycerides, can trigger the inflammatory response, and ultimately damage to endothelial cells12.

Emerging evidence underscores the pivotal role of inflammatory mechanisms in cardiovascular pathogenesis, notably in acute coronary syndromes (ACS) and heart failure progression13,14. Proinflammatory mediators - particularly C-reactive protein (CRP), tumor necrosis factor-alpha (TNF-α), and interleukin-6 (IL-6) - induce coronary endothelial dysfunction through synergistic interactions between oxidative stress and structural endothelial damage15. Elevated CRP levels, reflecting systemic inflammation, correlate with impaired endothelial homeostasis in coronary microvascular disease (CMVD) patients16. Recent advances propose novel composite biomarkers - C-reactive protein/lymphocyte ratio (CLR) and inflammatory burden index (IBI) - as superior predictors of cumulative inflammatory load and adverse cardiovascular outcomes compared to single-parameter indices17,18.

The aim of this study is to investigate the association between lipid metabolism, especially changes in TG/HDL ratio, and CMVD, and to further analyze whether this association is mediated through the key pathway of inflammatory response. By elucidating these mechanisms, we can provide a scientific basis for clinical prevention and treatment.

Methods

Data source

This is a prospective study to explore the correlation of lipid metabolism and CMVD mediated by inflammatory indicators. All clinical data of patients included in the study were sourced from inpatients admitted to the Cardiovascular Department of China Academy of Chinese Medical Sciences Xiyuan Hospital, China Academy of Traditional Chinese Medicine, between January 2022 and September 2024. The patient screening process was shown in Fig. 1 (Flowchart of study participants). The study passed the ethics committee of Xiyuan Hospital. Ethical approval was granted by the hospital’s ethics committee. Furthermore, all methods were performed in accordance with the relevant guidelines and regulations.

Fig. 1
figure 1

Flowchart of study participants.

Full size image

Diagnostic criteria for CMVD

Diagnostic criteria for CMVD adhered to the Chinese Multidisciplinary Expert Consensus on the Diagnosis and Treatment of Microvascular Diseases (2020)19, encompassing myocardial ischemia symptoms, objective evidence of myocardial ischemia, non-obstructive coronary artery disease (< 50% stenosis on CTA/CAG), and impaired coronary microcirculatory function, involving impaired coronary flow reserve (CFR) < 2.5 or < 2.0, or the slow coronary flow phenomenon (TIMI frames > 25), the index of microvascular resistance index (IMR) > 25.

Inclusion criteria

CMVD group: (i) no significant stenosis in 3 coronary arteries including left anterior descending artery (LAD), left circumflex artery (LCX), and right coronary artery (RCA) by coronary angiography (CAG) (stenosis < 50%); (ii) 3 coronary arteries with normal quantitative flow ratio (QFR) > 0.8, and the angiographic microvascular resistance (AMR) > 250 mmHg*s/m20; Control group (non-CMVD): (i) no significant stenosis of coronary artery shown by CAG; (ii) 3 coronary arteries measured by QFR, QFR > 0.8, AMR < 250 mmHg*s/m; Furthermore, all recruited patients aged over 20, regardless of gender.

Exclusion criteria

(i) Undergoing coronary revascularization within 1 month before this CAG, including Percutaneous Coronary Intervention (PCI), Percutaneous Transluminal Coronary Angioplasty (PTCA) and coronary artery bypass grafting (CABG); (ii) organic heart disease, such as severe heart failure (HF), malignant arrhythmia, valvular disease, etc.; (iii) acute and chronic inflammation, immune diseases or malignant tumors; (iv) severe renal insufficiency: glomerular filtration rate < 45 mL/min; (v) severe coagulation disorders or bleeding; (vi) allergy to contrast media. In addition, the exclusion criteria of CAG included poor visualization during or overlapping of the contrasted vessels.

AMR examination

Patients who meet the inclusion criteria underwent QFR examination. Real-time CAG images were conducted to the QFR instrument (Pulse Medical, Shanghai) and analyzed under a single angle of a single vessel using the software AngioPlus Core (version V2). The process is performed by a trained cardiologist, who performed the analysis. After selecting the optimal contrast view with minimal vessel overlap, the software automatically outlined the detected vessel lumen. The contrast flow rate is derived by dividing the vessel length by the contrast fill time, which is then converted to the filling flow rate9. Subsequently, frames in which the lumen contour was fully exposed were selected as analysis frames to determine the main and side branches of the analyzed vessels. According to Murray’s bifurcation-principle, the reference vessel diameter needs to be reconstructed21. Finally, according to fluid dynamics, AMR was calculated following the blow equations (Supplemental Fig. 1)22,23,24,25,26.

$$\user2{AMR} = \frac{{{\mathbf{Pd}}}}{{{\mathbf{Velocity}}_{{{\mathbf{hyp}}}} }} = \frac{{\user2{Pa} \times \user2{uQFR}}}{{{\mathbf{Velocity}}_{{{\mathbf{hyp}}}} }}$$
(1)

Note

Pa is aortic pressure at baseline; Pd is distal coronary artery pressure; Velocityhyp indicates coronary blood flow velocity in the hyperemic status.

Definition of inflammatory indexes and lipid metabolism

The inflammatory index involving CRP, CLR and IBI were calculated using the below formulas:

  • The CLR is defined as C-reaction protein/lymphocyte count.

  • The IBI is formulated as C-reaction protein × neutrophil count/lymphocyte count27.

  • TG/HDL is used to characterize lipid metabolism levels, and it is expressed as serum triglycerides/ serum high-density lipoprotein28.

Statistical analysis

All data were analyzed via R statistical (version 4.2.2). According to the normal distribution, all continuous variables do not follow normal distribution, and thus were presented in the form of medians and interquartile ranges. Whereas, categorical variables were expressed as frequencies and percentages. To clarify the differences in variables among different TG/HDL levels, TG/HDL were categorized into quartile groups, ranging from the lowest quartile (maximally anti-inflammatory) to the highest quartile (maximally pro-inflammatory). The Chi-squared test or Kruskal-Wallis H test was used to analyze various DII four categories. One-way ANOVA test, Kruskal-Wallis H test or Chi-squared test were applied to compare continuous or categorical variables in different TG/HDL quartile groups29. A statistically significant result was determined as a two-sided P value < 0.05.

Multivariate logistics regression models were employed to estimate the associations of TG/HD, inflammatory indexes and CMVD, and to calculate hazard ratios (HRs) and 95% confidence interval (CI). Model 1 was unadjusted. Model 2 was adjusted for gender, age, smoking, and drinking; Model 3 was additionally modified to account for gender, age, smoking, drinking, hypertension, hyperlipidemia, glucose, cholesterol and triglyceride. Next, we performed the subgroup analyses to explore potential differences among specific populations including gender, smoking, drinking, hypertension, hyperlipidemia and diabetes. The restricted cubic spline (RCS) model was applied to investigate the non-linear relationship between TG/HDL and CMVD. Furthermore, to clarify the correlation of inflammatory indicators, TG/HDL and CMVD, we conducted a linear analysis and And the violin plots was applied to characterize the differences in these indicators between the CMVD and non-CMVD group. Afterwards, the receiver operating characteristic (ROC) curve with the area under the curve (AUC) was used to explore the cutoff value of TG/HDL, CRP, CLR and IBI to identify CMVD. Eventually, the presence of a mediating effect was defined as existing a significant indirect effect, direct effect and total effect30.

Results

Baseline characteristic

Table ​1 presents the baseline characteristics of enrolled participants in this study with different TG/HDL quartiles. Within all included participants, the median TG/HDL values were 1.20, and ranging from 0.60 in Q1, 1.01 in Q2, 1.41 in Q3 and 2.45 in Q4. The median age of the 242 enrolled patients was 66, and about 57.44% sample were male. In comparison with those in the low-level TG/HDL group, participants with elevated levels of TG/HDL had an increasing proportion of male, drinking alcohol, inflammatory indexes (including CRP, CLR and IBI). Additionally, regarding laboratory indexes, participants in higher quartiles had a relatively higher level of creatine kinase isoenzyme-MB, brain natriuretic peptide, alanine aminotransferase, uric acid, triglyceride, HDL and AMR of RCA.

Associations of TG/HDL with CMVD

In this study, the incidence of CMVD was 38.02% among these participants with angina pectoris. Table 2 demonstrates the associations of TG/HDL with CMVD. TG/HDL was significantly associated with an increased risk of CMVD (OR = 1.62, 95%CI = 1.15–2.28) in the crude model. After multivariable adjustment, the results remained robust and statistically significant, with model 2 (OR = 1.53, 95%CI = 1.09–2.17), but the results showed no significance with model 3 (OR = 1.71, 95%CI = 0.69–4.25). Compared to the first quartile of TG/HDL, ORs for participants in the second, third and fourth quartile tend to be higher regardless of whether to adjust variables.

Table 1 Characteristics of the study population.
Full size table
Table 2 The associations of TG/HDL with CMVD.
Full size table
Table 3 The associations between TG/HDL and inflammatory indexes.
Full size table

Results of nonlinear of TG/HDL and CMVD

By the RCS models with adjustment for above mentioned confounders (model 3), the results showed that there was the L-shaped association between TG/HDL and CMVD (Fig. 2 and the cut-off value for CMVD was 1.22.

Fig. 2
figure 2

Restricted cubic spline regression curve of the association between TG/HDL and CMVD. CMVD Coronary microvascular disease, TG/HDL triglycerides to high-density lipoprotein cholesterol ratio.

Full size image

Subgroup analysis

We performed subgroup analyses to explore the relationships of TG/HDL with CMVD and the detailed results were showed in Fig. 3 Most analyses showed no differences within groups, for example the association between TG/HDL and CMVD was irrelevant in the participants who were smoking, drinking alcohol, hypertension and hyperlipidemia.

Fig. 3
figure 3

Subgroup analysis of associations between TG/HDL and CMVD. CMVD Coronary microvascular disease, TG/HDL triglycerides to high-density lipoprotein cholesterol ratio.

Full size image

Associations of inflammation with TG/HDL and CMVD

Table ​3 displays the associations of TG/HDL and inflammatory indexes after multivariate logistic regression. As identical adjusted variables method mentioned above, DII was positively associated with CPR (OR = 1.25, 95% CI = 1.03–1.51), CLR (OR = 1.35, 95% CI = 1.04–1.76), and IBI (OR = 1.04, 95% CI = 0.98–1.11).

Furthermore, logistics regression results of inflammatory indexes with CMVD are demonstrated in Table 4. The results showed that all indicators were positively related to CMVD except IBI in model 3.

Table 4 The associations of inflammatory indexes with CMVD.
Full size table

Evaluation of the impact of inflammatory parameters and TG/HDL on CMVD

A ROC analysis to determine whether TG/HDL, CRP, CLR and IBI can be a diagnostic and prognostic marker in CMVD patients revealed TG/HDL (AUC: 0.634, 95% CI = 0.56–0.71, P < 0.01), CRP (AUC: 0.714, 95% CI = 0.65–0.78, P < 0.01), CLR (AUC: 0.698, 95% CI = 0.63–0.76, P < 0.01) and IBI (AUC: 0.696, 95% CI = 0.63–0.76, P < 0.01) with cutoff valvule of 0.93, 1.59, 1.23 and 6.00, respectively (Fig. 4.

Fig. 4
figure 4

Receiver operating characteristic (ROC) curve of the TG/HDL-C, C-reaction protein, C-reaction protein to lymphocyte ratio and inflammatory burden index as high-risk indicators of CMVD. CMVD Coronary microvascular disease, CRP C-reaction protein, CLR C-reaction protein to lymphocyte ratio, IBI inflammatory burden index, TG/HDL triglycerides to high-density lipoprotein cholesterol ratio.

Full size image

Linear correlation of inflammatory parameters and TG/HDL with AMR

The Spearman correlation analysis showed that TG/HDL, CRP, CLR and IBI have positive correlation with AMR (R = 0.17, P = 0.007; r = 0.24, P < 0.001; r = 0.22, P < 0.001; r = 0.19, P < 0.001; respectively) in the whole study population (Supplemental Figure).

Distribution differences of inflammatory parameters and TG/HDL

We conducted violin plot analysis to compare the expression of inflammatory parameters and TG/HDL between CMVD and non-CMVD group. The result showed the level of these indices in the CMVD group (Supplemental Figure).

Mediating role of inflammatory indexes

Figure 5 shows that CRP mediated 26.37% of the association between TG/HDL and CMVD, the CLR mediated 16.89% of the association between TG/HDL and CMVD, and the IBI mediated 10.45% of the association between TG/HDL and CMVD. (Fig. 5.

Fig. 5
figure 5

Analysis of the mediation by C-reaction protein (a), C-reaction protein to lymphocyte ratio (b) and inflammatory burden index (c) of the associations of TG/HDL with CMVD. CMVD Coronary microvascular disease, TG/HDL triglycerides to high-density lipoprotein cholesterol ratio.

Full size image

Discussions

In this study, we found a positive association of TG/HDL with CMVD among the angina pectoris population, and after the comprehensively adjusted model, these results remained significantly. Given that the positive correlations of inflammatory indexes with TG/HDL and CMVD were uncovered by logistic regression, we performed the mediation analysis to identify the significant roles of CRP, CLR, and IBI in linking TG/HDL with CMVD, further expounding inflammation as a potential underlying mechanism to trigger CMVD. Therefore, the level of blood lipid metabolism may be a potent approach for the onset of CMVD and this process may be mediated by inflammation.

Although CMVD is a non-atherosclerotic disease, the WISE sub-study utilized intravascular ultrasound (IVUS) to detect female patients with non-obstructive CAD and demonstrated that the increased risk in patients with CMVD is strongly associated with atherosclerosis-related risk factors, such as hypertension31, diabetes mellitus32,33, and hyperlipidemia34. It was found that TG levels were higher in the CMVD group in comparison. In binary logistic regression, TC promoted CMVD (OR:2.19, 95%CI = 1.12–4.25) 35. Hyperlipidemia is regard as one of the main risk factors for the development of atherosclerosis and epidemiologic studies suggest that elevated triglyceride levels are a biomarker of cardiovascular disease36. Michael Miller suggested impact of triglyceride levels beyond low-density lipoprotein cholesterol after acute coronary syndrome in the PROVE IT-TIMI 22 trial37. Notably, among the various lipoprotein classes, however, HDL is inversely associated with the incidence of CVD38,39. Because they can exert a series of atheroprotective functions40. Currently, many scholars have proposed that the ratio of the two is more reflective of lipid metabolism levels and is also more closely linked to the development of IR. RCSCD-TCM study reported TG/HDL is positively risk of prediabetes and Type 2 diabetes in patients with CAD in China (OR = 1.19, 95%CI = 1.16–1.23)41. A cross-sectional study suggested high TG/HDL with association of a higher risk of metabolic syndrome (OR = 3.07, 95% CI = 2.402 to 3.924, P < 0.001)42. The Jichi Medical School Cohort Study suggested increased TG/HDL ratio correlated with a significant increase in stroke risk in the healthy body mass index (BMI) participants43. Some studies also indicated that the TG/HDL is a marker of CAD44,45. Moreover, TG/HDL is an independent risk factor for the occurrence of CMVD. Li ping Liao, etc. indicated that the proportion of females, the incidence of hypertension and diabetes, the level of TG, and C-reactive protein, and the ratio of TG/HDL-C were higher in the CMVD group46. And they defined CMVD based on lactic acid levels. For this study CMVD is diagnosed based on AMR without pressure guidewires and adenosine injection to reach the maximum hyperemic state given that the accuracy of diagnosis has been clearly reported before, whereas our conclusions are similar in that TG/HDL promotes CMVD. Interestingly, there were more male patients in our study. CMVD is an abnormality of microvascular diastolic function due to endothelial damage caused by various injurious factors47. Hyperlipidemia and IR has been reported to directly exacerbate vascular endothelial cell injury48. A study concerning prevalence of CMVD showed hyperlipidemia (69.48% vs. 55.22%) was relatively more prevalent in CMVD group49. Moreover, the content of vasodilator factors decreases in response to insulin resistance may cause endothelial-dependent functional impairment50. IR mainly involve endothelium-independent mechanisms in which stimulation of both adenosine A1 and A2 receptors and the opening of ATP-sensitive potassium channels on smooth muscle cells play a role51. On the other hand, IR in cardiomyocytes results in decreased glucose uptake and glucose oxidation52 and activates the inflammatory pathway, in which the proinflammatory TNF-α/IL-6/C-reactive protein (CRP) pathway was significantly associated with CMVD53. Emerging evidence indicates that hypovitaminosis D in patients with microvascular dysfunction may exacerbate endothelial pathology through multifaceted mechanisms, including (1) potentiation of pro-inflammatory cytokine cascades (e.g., IL-1β, TNF-α), (2) dysregulation of immunomodulatory pathways (reduced Treg cell activity), and (3) facilitation of atherosclerotic plaque formation via lipid peroxidation and foam cell accumulation54,55. Several of the mechanisms of endothelial dysfunction for IR are shared with some signaling pathways, including formation of advanced glycation end products, activation of protein kinase C (PKC). With increased AGE level, the level of AGE receptors accordingly increases with expression on inflammatory T cells, leading to increased cytokine production and ROS generation. Meanwhile, AGE/RAGE signaling is involved in increased expression of TNF-α and oxidative stress markers including NOX-256. Moreover, IR also contributes to CMVD, particularly via impairment of the phosphatidylinositol-3-kinase pathway15.

CMVD as a result of both endothelium-dependent and non-endothelium-dependent injury, and endothelial injury in response to inflammation has been recognized as a major pathogenesis. A study showed a significantly higher level of IMR in the high CRP group (P = 0.02)57. Our research also showed that CRP, and CLR are positively correlated with CMVD (OR = 1.76,95% CI 1.29 to 2.41and OR = 2.48, 95%CI 1.50 to 4.09, respectively). The damage of CRP on endothelium-dependent dilation is mediated by activation of NADPH oxidase, as a catalyst for free radical production, and p38 kinase, an upstream activator of NADPH oxidase58. Karasu, M. showed that IL-34 is an independent predictor of slow coronary flow (OR: 1.044, 95% CI: 1.006–1.084)59. In addition, some studies have shown that in patients with CMVD, pro-inflammatory factors predispose to endothelial dysfunction, which is due to the presence of reactive oxygen species, of which the intermediary cytokines IL-1, IL-6, and TNF-α are key mediators of the inflammatory cascade response with reduced eNOS expression60.

This prospective cohort study yields three principal findings: (1) It provides novel epidemiological evidence supporting the positive correlation between triglyceride-to-high-density lipoprotein cholesterol ratio (TG/HDL-C) and coronary microvascular dysfunction (CMVD); (2) Mediation analysis confirms the intermediary role of systemic inflammation (CRP-mediated proportion: 28.6%, P < 0.01) in the TG/HDL-C–CMVD pathway; (3) Longitudinal monitoring of lipid ratios may serve as a cost-effective strategy for early CMVD risk stratification. Notably, the observational design precludes causal inference, and the single-center recruitment limits generalizability of findings.

Conclusions

This study provided evidence for the positive associations of TG/HDL with CMVD, while also indicating the significant mediating role of inflammatory indices in this association.