The relationship between triglyceride/high-density lipoprotein cholesterol ratio and coronary slow-flow phenomenon

In this study, we aimed to investigate the relationship between high triglyceride (TG)/high-density lipoprotein cholesterol (HDL-C) ratio and coronary slow flow phenomenon (CSFP) in patients undergoing elective coronary angiography for suspected coronary artery disease. This prospective study included a total of 84 CSFP patients and 83 controls with normal coronary flow, as evidenced by coronary angiography. The Thrombolysis in Myocardial Infarction (TIMI) frame count (TFC) was used to measure the coronary blood flow velocity. The lipid profiles were analyzed and TG/HDL-C ratio were calculated dividing absolute TG levels by absolute HDL-C levels in peripheral blood. The median TG/HDL-C ratio was higher in the CSFP group than the control group (3.4 [2.6 to 4.9] vs. 2.3 [1.8 to 3], respectively; p < 0.001). The multivariate logistic regression analysis revealed that TG/HDL-C ratio was an independent predictor of CSFP (odds ratio [OR] 1.78, 95% confidence interval [CI] 1.59–2.32; p = 0.001) and TG/HDL-C ratio was positively correlated with the TFC in the CSFP group (r = 0.311, p < 0.001). The area under the receiver operating characteristic curve of TG/HDL-C for the diagnosis of CSFP was 0.73 (95% CI 0.65–0.81; p < 0.001). If a cut-off value of 2.75 was used, higher levels of TG/HDL-C ratio could predict the presence of CSFP with 72% sensitivity and 71% specificity. Our study results suggest that TG/HDL-C ratio is associated with CSFP and may be a useful biomarker for predicting CSFP and its severity.


Introduction
The coronary slow-flow phenomenon (CSFP) is a coronary artery disease, characterized by delayed vessel opacification in the absence of obstructive coronary artery disease [1]. It is defined angiographically and its incidence varies from 1 to 7% among the patients undergoing coronary angiography for suspected coronary artery disease (CAD) [2]. Although the exact mechanisms of CSFP are still unclear, small vessel disease, endothelial dysfunction, diffuse atherosclerosis, microvascular vasomotor dysfunction, and inflammation are implicated in its pathophysiology [3][4][5][6]. In addition, CSFP may cause transient myocardial hypoperfusion in patients with both normal coronary arteries and angina, thereby, leading to increasing the risk of CAD and worsening of the prognosis [7]. There is no definitive treatment of CSFP and traditional anti-anginal drugs are often used in the treatment. A particular attention has been paid to control hypertension and dyslipidemia in CSFP cases [8]. In some studies, 1 3 statins are used to regulate cholesterol levels and take vascular inflammation under control [9].
In recent years, intimal thickening in the coronary arteries, diffuse calcification, lumen irregularity, and non-occlusive atheroma plaques have been identified in the majority of cases with CSFP [3,10,11]. In the light of recent findings, CSFP should be evaluated as CAD, which was earlier considered a subgroup of cardiac syndrome X [2]. Several studies have suggested that established cardiovascular risk factors may play a role in the pathogenesis of microvascular angina in healthy individuals, result in coronary microvascular dysfunction [8,12].
Dyslipidemia is one of the risk factors of cardiovascular disease (CVD) [13]. It is defined as a disorder of lipoprotein metabolism, rather than elevations in total cholesterol (TC) or low-density lipoprotein cholesterol (LDL-C) alone [14]. Triglyceride (TG), LDL-C, and high-density lipoprotein cholesterol (HDL-C) are important components of the lipid fraction of the human body [14,15]. However, epidemiological researches have demonstrated that lipid-related ratios such as TC/HDL-C, TG/HDL-C, or LDL-C/HDL-C may be probably better predictors of CVD risk than any other single lipid marker [16,17]. Among these, the TG/HDL-C ratio was first proposed by Gaziano et al. [18] as an atherogenic index. Several studies have shown that TG/HDL-C ratio is a strong predictor of CAD such as myocardial infarction (MI), LDL phenotype B, and atherogenic risk [19,20]. In addition, the TG/HDL-C ratio appears to be more valuable than any other single lipid marker, since it has an ability to reflect the complex interactions between the lipoprotein metabolism and to better predict plasma atherogenicity [17,20]. Besides its simplicity and practicality, there is a growing number of evidences supporting the predictive value of TG/HDL-C ratio in cardiovascular events and may be of clinical relevance [21].
In the light of the literature data, we, in the present study, aimed to investigate the relationship between the TG/HDL-C ratio and CSFP in patients undergoing elective coronary angiography for suspected CAD and to identify whether TG/HDL-C ratio was a feasible biomarker in distinguishing CSFP cases from healthy individuals.

Study design and study population
This single-center, prospective study was conducted at Department of Cardiology of a tertiary care center between October 2017 and February 2020. A total of 84 CSFP patients who were admitted with chest pain and had myocardial ischemia as assessed by myocardial perfusion scintigraphy with positive treadmill test results and age-and sex-matched 83 controls with normal coronary flow were included. The CSFP was confirmed by coronary angiography and the Thrombolysis in Myocardial Infarction (TIMI) frame count (TFC) was used to measure the coronary blood flow velocity. Those undergoing angiography previously and undergoing surgery or mechanical revascularization for significant CAD were excluded from the study. Patients with left ventricular dysfunction (ejection fraction < 50%), peripheral artery disease, congenital heart disease, chronic renal or hepatic insufficiency, chronic obstructive pulmonary disease, acute and chronic inflammatory diseases, or autoimmune disorders, and those receiving drugs affecting the lipid metabolism were also excluded. All participants were informed about the nature of the study and a written informed consent was obtained. The study protocol was approved by the institutional Ethics Committee with the Approval No. 0447. The study was conducted in accordance with the principles of the Declaration of Helsinki.

Data collection and definitions
Baseline demographic and clinical characteristics of all participants including age, sex, body mass index (BMI; calculated as weight in kg divided by height in meters squared), and comorbidities were recorded. Hypertension was defined as a systolic blood pressure of ≥ 140 mmHg and diastolic blood pressure of ≥ 90 mmHg or the current use of antihypertensive. Diabetes mellitus was defined as a fasting blood glucose of ≥ 126 mg/dL in two consecutive measurements or having a history of diabetes and using insulin or oral antidiabetics. Smoking was defined as current daily smoking.

Coronary angiography-based CSFP assessment
Coronary angiography was percutaneously performed using the Judkins technique. The coronary arteries were visualized at right and left oblique and caudal planes at 30 frames per second (fps). A contrast agent was injected manually during coronary angiography (6 to 10 mL of contrast agent at each position using right and left, and cranial and caudal angulations). Angiographic images were stored in 34 runs in a CD in accordance with the Digital Imaging and Communications in Medicine (DICOM) standards and the flow velocity was measured. All angiographic measurements were carried out by two cardiologists who were Şblind to the clinical status of the patients. The flow in coronary arteries was measured using the TFC as described by Gibson et al. [22]. The TFC was calculated from the difference between the first and last frames. The first frame was when nearly complete lumen opacification with antegrade filling, while the final frame was accepted when the contrast dye reached distal landmarks. The following distal landmarks were used for the analysis: the distal bifurcation of the left anterior descending artery (LAD) (i.e., the "mustache") ( Fig. 1a) and the distal bifurcation of the segment with the longest total distance in the left circumflex artery (LCX) and the first branch of the posterolateral artery in the right coronary artery (RCA) (Fig. 1b) [22]. As the LAD artery is often longer than the other main coronary arteries, the LAD TFC is usually high. The longer LAD frame counts were corrected dividing by 1.7 to calculate the corrected TFC [22]. The mean TFC was accepted as 36 ± 2.5 for LAD, 22 ± 4.1 for LCX, and 20.4 ± 3.1 for RCA. The corrected cut-off value for the LAD artery was 21.1(± 1.5) frames. Any TFC values with more than 2 standard deviations (SDs) were considered CSFP [21] (Fig. 2). The mean TFC for each subject was calculated by summing the TFC for LAD, LCX, and RCA and, then, dividing this sum into 3 [22].

Biochemical analyses
Venous blood samples were drawn from each participant after an overnight (12-h) fasting. Hematological analyses were carried out using the BC-6800 Hematology Analyzer (Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, China). All biochemical parameters including fasting blood glucose, creatinine, urea, uric acid, aspartate aminotransferase (AST), alanine aminotransferase (ALT), TC, TG, HDL-C, and LDL-C were analyzed on the day of sample collection using a commercial autoanalyzer (c8000i; Abbott Diagnostics GmbH, Germany). For samples with a TG level of < 400 mg/dL, the LDL-C was calculated using the Friedewald formula [(LDL-C = TC -HDL-C -(TG/5)] [23]. Non-HDL-C was calculated as: TC-HDL-C. TG/ HDL-C ratio was calculated dividing absolute TG levels by absolute HDL-C levels in peripheral blood.

Statistical analysis
Statistical analysis was performed using the SPSS version 26.0 software (IBM Corp., Armonk, NY, USA). Continuous data were expressed in mean ± SD or median (interquartile range [IQR]), while categorical data were expressed in number and percentage. The normality assumption was checked using the Kolmogorov-Smirnov test. One-way analysis of variance (ANOVA) with post-hoc Tukey test, independent samples t-test, Kruskal-Wallis and Mann-Whitney U tests were used to examine independent quantitative variables. The paired samples t-test and Wilcoxon test were performed to analyze dependent quantitative variables. Independent qualitative variables were examined using the chi-square test or Fisher's exact test, when the chi-square assumptions were not met. The Pearson or Spearman correlation analysis was carried out to identify the relationship between the continuous variables. The correlation between the TG/HDL-C ratio and TFC was examined using the Spearman correlation coefficient. Backward stepwise univariate and multivariate logistic regression analyses were performed to determine independent predictors of CSFP. The receiver operating characteristic (ROC) curve was used to identify sensitivity and specificity of independent predictors of CSFP. The optimal cut-off value was calculated from the point of maximal sensitivity and specificity, as described by Youden [24]. The DeLong test was used to compare predictive performance of independent predictors of CSFP. A p value of < 0.05 was considered statistically significant at 95% confidence interval.

Results
Of the 84 CSFP patients, 52 were males and 32 were females with a mean age of 54 ± 8.9 (range 39 to 67) years. Of the 83 control individuals, 41 were males and 42 were females with a mean age of 55.2 ± 8.5 (range 33 to 68) years. There was no statistically significant difference in the demographic characteristics such as age, sex, BMI and risk factors for CAD such as diabetes mellitus, hypertension, or smoking between the two groups (p > 0.05 for all). The use of antiplatelet and angiotensin-converting enzyme inhibitors (ACEI)/angiotensin receptor blockers (ARBs) was more frequent in the control group. Baseline demographic and clinical characteristics of both groups are summarized in Table 1.
Laboratory test results of the CSFP and control groups are presented in Table 2. Accordingly, there was no statistically significant difference in the mean fasting blood glucose, urea, AST, ALT, white blood cell (WBC) count, hemoglobin, and platelets between the groups (p > 0.05 for all). However, the mean creatinine and uric acid levels were significantly higher in the CSFP group than the control group (p = 0.002 and p = 0.004, respectively). The mean HDL-C level was significantly lower in the CSFP patients (p = 0.001) than control group. CSFP patients had higher TG levels (p < 0.001), higher Non-HDL-C (p < 0.033) and higher TG/HDL-C ratio (p < 0.001) than control group. On the other hand, there was no statistically significant difference in the TC and LDL-C levels 3, respectively; p < 0.001) were significantly higher in the CSFP group than the control group. In addition, the mean TFC was significantly higher in these patients (34.8 ± 3 vs. 19 ± 1.6, respectively; p < 0.001). The correlation analysis between lipid parameters (LDL-C, TC, HDL-C, non-HDL-C, TG), TG/HDL ratio and mean TFC are shown in Table 3. Accordingly, there was a positive and significant correlation between the TFC and TG/HDL-C ratio (r = 0.311, p < 0.01; Fig. 3) and TG levels (r = 0.222, p = 0.06) and a negative and significant correlation between the mean TFC and HDL-C levels (r = 0.275, p = 0.001). No statistically significant correlation was found between mean TFC and LDL-C (r = 0.002, p = 0.809), non-HDL-C (0.147, p = 0.069) or TC (r = 0.003, p = 0.975). Backward stepwise univariate and multivariate logistic regression analyses were performed to determine independent predictors of CSFP. The relation of CSFP with age, sex, diabetes mellitus, smoking, creatinine, uric acid, and TG/ HDL-C ratio was examined. Given the fact that TG/HDL-C ratio, TG, and HDL-C show a strong correlation with each other, these variables were excluded from the multivariate analysis. In the multivariate regression analysis, the TG/ HDL-C ratio was found to be a significant independent  Table 4). The ROC curve was used to evaluate the ability of TG, HDL-C, and TG/HDL-C ratio to discriminate CSFP and healthy individuals (Fig. 4). The ROC curve showed an area under the curve (AUC) value of 0.72 (%95 CI 0.65-0.81; p < 0.001). The sensitivity and specificity of TG/HDL-C ratio for predicting CSFP were 72% and 71%, respectively. The AUC value of TG was 0.69 (95% CI 0.61-0.76; p < 0.001) with 64% sensitivity and 63% specificity at values above 126.5 mg/dL. The AUC value of HDL-C was 0.66 (95% CI 0.58-0.73; p = 0.001) with 59% sensitivity and 60% specificity at values below 45.5 mg/dL. The TG/ HDL-C ratio, TG, and HDL-C were compared in terms of their predictive performance using the DeLong method. The predictive performance of TG/HDL-C ratio was significantly superior to that of TG, but it was borderline significant compared to HDL-C, indicating no statistical significance.

Discussion
In the present study, we investigated the possible relationship between the TG/HDL-C ratio and CSFP in patients undergoing elective coronary angiography. Our study results showed that the mean TG/HDL-C ratio was significantly higher in the CSFP patients than those with normal coronary flow. In addition, the mean TCF was positively and significantly correlated with the TG/HDL-C ratio. The ROC curve analysis showed that the cut-off value of TG/HDL-C ratio was sufficient to discriminate CSFP and non-CSFP individuals with a high sensitivity and specificity.
The CSFP is an angiographic evidence and should be considered as a distinct entity with specific characteristics, pathogenic mechanisms, and specified diagnostic criteria [1,2]. Although the exact pathophysiology of CSFP is still unclear, endothelial dysfunction and atherosclerosis have been thought to be involved [3,6]. Abnormally slow flow in the coronary arteries has been proposed to be an indicator of diffuse atherosclerosis related to the endothelial damage before angiographically visible coronary lesion, indicating an early manifestation involving both the microvascular system and epicardial coronary arteries [3,11]. In a study, Avsar et al. [25]. Examined the possible link between CSFP and carotid artery intima-media thickness (CIMT) and found the increased CIMT in CSFP patients. In this study, the CIMT and corrected TCF were also significantly  correlated, implying that CSFP might be an early marker of both endothelial dysfunction and subclinical atherosclerosis. In another study investigating the epicardial coronary morphology and intracoronary pressure in CSFP patients using intravascular ultrasonography (IVUS) and flow rate measurements, Cin et al. [3] reported diffuse intimal thickening, widespread calcification through the vessel wall, and non-obstructive atheroma. Consistent with these findings, Pekdemir et al. [11] described longitudinally extended massive calcification along the epicardial arteries as evidenced by IVUS examination as the most common finding in CSFP patients. Furthermore, Camsari et al. [26] found the CIMT to be associated with IVUS indices in CSFP patients. Taken together, these findings support the notion that CSFP is an indicator of microvascular diseases and diffuse, non-obstructive atherosclerotic disease of the epicardial arteries. Endothelial dysfunction is an early indicator of atherosclerosis [27]. Endothelial-dependent, flow-mediated dilatation (FMD) of the brachial artery has been shown to reduce in CSFP patients, indicating that CSFP may play a role in the etiology of endothelial dysfunction [5]. In addition, plasma homocysteine and endothelin levels increase in CSFP patients, both which have detrimental effects on endothelial function [4,10]. In the light of existing data, CSFP seems to be a form of early diffuse coronary atherosclerosis, predominantly characterized by microvascular endothelial dysfunction.
It has been well established that lipid abnormalities contribute to the development of atherosclerosis [13]. An extensive number of studies have demonstrated that high TC and LDL-C and low HDL-C are strongly associated with ischemic heart disease, stroke, peripheral vascular disease, and atherosclerosis [13][14][15]. Elevated LDL-C levels with high TG and low HDL-C have been shown to increase the risk of CAD than elevated LDL-C levels alone [15,28]. Review of the literature reveals that different subfractions of the lipid particles are involved in the atherogenesis. Small dense LDL particles are more atherogenic than larger particles [15,28]. High TG and low HDL-C levels are strongly associated with small dense LDL-C levels and RLP-C levels [18,28]. TG levels are elevated in the setting of decreased lipoprotein lipase activity. This leads to higher remnant-like particle cholesterol and lower HDL levels (18). RLP-C is TG rich lipoproteins products of partially metabolized chylomicrons and very Low-Density Lipoproteins (VLDL) and associated with increased risk for cardiovascular disease [15,18,29,30]. Consistent with previous studies showing a link between low HDL-C and high TG levels and CSFP [12,31,32], we found high TG levels low HDL-C levels in the CSFP group in our study.
A higher TG/HDL-C ratio was associated with an increasingly atherogenic lipid phenotype, characterized by primarily higher RLP-C and LDL density and a superior predictor of coronary heart disease than conventional lipid parameters [15,19]. This ratio has been also considered an independent predictor of CVD and all-cause mortality [19,21]. Review of the literature reveals only one retrospective study investigating the relationship between proportional serum lipid parameters and CSFP. In this study including 54 CSFP patients and 39 controls, Kalayci et al. [32] found higher TG, TC/HDL-C, TG/HDL-C, LDL-C/HDL-C, and non-HDL-C levels to be associated with CSFP. In our prospective study, we also found a significant correlation of the high TG, TG/ HDL-C ratio, and low HDL-C and mean TFC, but non-HDL-C levels wasn't associated with CSFP. However high TG/HDL-C ratio is an independent predictor of CSFP. High TG/HDL ratio may be a valuable marker for atherogenic lipid profile and abnormal TG metabolism in CSFP patients. In the literature, several studies have shown the link between LDL-C levels and CSFP [33][34][35]. In a study investigating the clinical and angiographic properties of CSFP, Yilmaz et al. [33] found LDL-C levels to be correlated with CSFP. On the other hand, no significant correlation between these two variables was reported in some other studies [34,35]. Similarly, in our study, LDL-C levels were similar between the two groups, and no statistically significant correlation was found with mean TFC. Based on these findings, it can be speculated that TG and HDL-C levels are more valuable indicators of CSFP than LDL-C alone and TG/HDL-C ratio shows a higher correlation with the mean TFC, than TG and HDL-C levels.
Unfortunately, there is no established therapy for the treatment of CSF, mainly due to the lack of homogenous large, randomized clinical trials. Based on the small number of studies have reported benefit in the treatment of CSF patients with dipyridamole, statins, angiotensin-converting enzyme inhibitors and, α-blockers [36,37]. In our study, there was no difference in statin, calcium channel blocker or beta-blocker use between the two groups. But there was excess use of aspirin and angiotensin-converting enzyme inhibitors/ angiotensin receptor blockers in the control group. These medications might affect study results. Other limitations of this study it was an observational single center study with a small sample size and other pharmacological therapy, dietary or exercise not evaluated. All of them may affect lipid parameters.
We were unable to evaluate the atherosclerotic alterations in the coronary arteries via sophisticated imaging modalities such as IVUS or optical coherence tomography precluding the evaluation of coexisting non-obstructive CAD in isolated CSFP patients. Of note, IVUS is not recommended in the routine daily practice for the evaluation of CSFP patients and the diagnosis is usually confirmed based on visible angiographic findings. İnflammatory biomarkers such as the high-sensitivity C-reactive protein, interleukin-6, and tumor necrosis factor-alpha were unable to be analyzed. It was another limitation of our study. Further large-scale, prospective, randomized-controlled studies are needed to confirm these findings.

Conclusions
In conclusion, the CSFP is a poorly recognized clinical entity. Our study results suggest that TG/HDL-C ratio is strongly associated with CSFP and is a significant independent predictor of CSFP. Therefore, it may be a useful biomarker for predicting CSFP and its severity. Beyond its predictive value, it may be of value as a therapeutic target owing to its simplicity and as it allows longitudinal followup in the clinical setting. Further studies are warranted to gain a better understanding of the pathogenesis of CSFP and the diagnostic and therapeutic value of TG/HDL-C ratio in this group of patients. Funding This study received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Conflict of interest
The authors declare they have no potential conflict of interest regarding the investigation, authorship, and/or publication of this article.