Predictors and clinical outcomes of slow flow phenomenon in diabetic patients with chronic coronary syndrome

Abstract

Background

Coronary slow flow (CSF) is characterized by late distal coronary perfusion of coronary arteries at the time of angiography despite the vessels appearing normal. The importance of CSF is still debatable. Therefore, this study aimed to investigate CSF’s predictors and clinical outcomes in diabetic patients with chronic coronary syndrome (CCS).

Patient and methods

This retrospective study included 250 diabetic patients diagnosed with chronic stable angina and referred for coronary angiography (CAG), showing normal coronaries with CSF (Group I) and 240 diabetic patients with normal coronaries and normal flow (Group II). The patients in both groups were followed up for one year to evaluate clinical outcomes.

Results

The incidence of major adverse cardiac events (MACE) was higher in Group I than in Group II, but the difference was not statistically significant except when the composite endpoints of STEMI, NSTEMI, and unstable angina were combined under the term ACS. The independent predictors of CSF, as detected by multivariate regression analysis, were body mass index (BMI) (OR = 0.694, 95% CI = 0.295–0.842, P = 0.010), blood glucose during catheterization (OR = 0.647, 95% CI = 0.298–0.874, P = 0.008), serum triglycerides (OR = 0.574, 95% CI = 0.289–0.746, P = 0.010), and the neutrophil/lymphocyte ratio (NLR) (OR = 0.618, 95% CI = 0.479–0.892, P = 0.001).

Conclusion

Serum triglyceride levels, BMI, NLR, and high blood glucose levels at the time of catheterization were independent predictors of CSF in diabetic patients. MACE levels were higher in diabetic patients with CSF.

Introduction

Microvascular disease and endothelial dysfunction are closely related to diabetes mellitus (DM), which is a significant risk factor for coronary artery disease (CAD). Even in patients with angiographically normal coronary arteries, the diabetic population tends to have an increased number of cardiovascular events [1]. Approximately 1–7% of coronary angiographies performed to assess patients with stable chronic coronary syndrome (CCS) reveal the coronary slow-flow phenomenon (CSFP) [2]. Small artery disease and coronary endothelial dysfunction are related to the development of CSFP, yet the specific pathophysiologic pathways involved are unknown [3]. The quality of life is markedly reduced in patients with slow flow, and in certain cases, it has been linked to increased morbidity and mortality [4].

Increased blood uric acid and C-reactive protein (CRP) levels, along with reduced serum albumin levels, have all been shown to be independent predictors of cardiovascular events, coronary artery disease (CAD), and CSFP [5].

A few studies have reviewed coronary blood flow in normal coronary arteries without categorizing diabetes separately, and the results of these studies were largely contradictory. Furthermore, none of these studies addressed why not all diabetic patients exhibit slow coronary flow. This led us to conduct the present study, which aimed to assess the predictors and clinical outcomes of CSFP in diabetic patients with CCS.

Methods

In the current study, 5400 patients underwent coronary angiography in our cardiovascular department from July 2020 to February 2022. Of these, 3240 were diagnosed with CCS, and 490 diabetic patients (15.1%) were included in the study, as shown in the flow chart (Fig. 1). Patients were divided into two groups: Group I included 250 diabetic patients with normal coronary arteries and slow flow, and Group II included 240 diabetic patients with normal coronary arteries and normal flow, serving as the control group. Informed consent was obtained from all patients, and the Ethical Committee approved the study. Exclusion criteria included impaired left ventricular function (ejection fraction less than 40%), coronary arteries with any luminal stenosis, coronary artery ectasia, acute coronary syndrome, and age less than 18 years (Fig. 1).

Fig. 1

A flow chart of the patients included in the study

All patients underwent a detailed history discussion regarding age, sex, and risk factors for CAD (hypertension, DM, smoking, dyslipidemia, and family history of CAD), as well as a general examination, including blood pressure assessment and BMI. Clinical examination, resting 12-lead ECG, and a review of patient files were performed. Investigations included urea and creatinine levels, estimated glomerular filtration rate (e-GFR), complete blood count and hematocrit value, complete lipid profile, fasting blood glucose and HbA1c, serum fibrinogen, high-sensitive CRP (Hs-CRP), neutrophil/lymphocyte ratio (NLR), and blood glucose level during catheterization. Echocardiographic measurements were performed according to American Society of Echocardiography guidelines [6] using an M5S phased array transducer (2.5–5.0 MHz) and a Vivid E9 ultrasound machine (GE Vingmed Ultrasound, Horten, Norway). Measurements included left atrial diameter, valvular affection, end-diastolic and end-systolic left ventricular volumes, and ejection fraction, which were determined using the Simpson technique.

Coronary angiography

The standard technique for left cardiac catheterization was employed, utilizing both caudal and cranial angulations to obtain angiographic images. Each patient received sublingual tablets or an intracoronary injection containing 200 µg of nitroglycerin throughout the procedure. Two skilled interventionists reviewed angiography for each participant, and TIMI frame counts were documented for each coronary artery. The clinical features of the participants were not known to the two investigators. Briefly, the first frame was defined as the frame that showed > 70% opacification of the lumen with antegrade filling. For each vessel, the last frame was identified when dye appeared at a specific distal landmark, with the distal bifurcation being utilized for the left anterior descending artery (LAD) [7].

The distal landmark for the left circumflex artery (LCX) was the furthest bifurcation of the obtuse marginal branch, which was farthest from the coronary ostium. The right coronary artery (RCA) was served by the first branch of the posterolateral artery [8]. Images were acquired at 15 frames per second, and the numbers were multiplied by two. Any number of frames exceeding 27 was deemed unusual and suggestive of slow flow [9]. Cardioactive medications, antiplatelet drugs, and anticoagulants were the most commonly used medications during coronary angiography.

Follow-up was conducted for one year in both groups to assess chest pain recurrence, acute coronary syndrome (STEMI, NSTEMI, and unstable angina), arrhythmias (including atrial fibrillation, other atrial and ventricular arrhythmias), hospitalization, stroke, and cardiovascular mortality.

Statistical analysis

IBM SPSS version 23 (Armonk, New York) was used for the statistical analysis. Quantitative data were expressed as the mean ± SD, while categorical data were expressed as percentages and absolute numbers. The significance of differences between the two groups was examined using Student’s t-test for numerical data and the chi-square test for categorical variables. A P-value of < 0.05 indicated statistical significance. Multivariate regression analysis using binary logistic regression was performed to identify the independent variables influencing CSFP.

Results

The study was conducted on 250 diabetic patients diagnosed with chronic stable angina and referred for coronary angiography, which showed normal coronaries with slow flow (Group I) and 240 diabetic patients with normal coronaries and normal flow (Group II). There was a significant increase in BMI in Group I (31.56 ± 1.90 kg/m²) compared with Group II (27.90 ± 1.11 kg/m²) (P = 0.001). Blood glucose levels during catheterization were significantly higher in Group I (138.34 ± 45.69 mg/dl) than in Group II (105.65 ± 15.06 mg/dl). HbA1c and serum triglyceride levels were significantly higher in Group I than in Group II. High-sensitive CRP (Hs-CRP) and fibrinogen levels were significantly higher in Group I than in Group II (Table 1). No statistically significant difference was observed between the two groups regarding age, sex, hypertension prevalence, peripheral arterial disease, or chronic kidney disease (Table 2).

Table 1 Clinical characteristics and laboratory results of the studied groups

Table 2 Demographics and clinical characteristics of the studied groups

After one year of follow-up, the two groups had no significant difference in the incidence of MACEs, arrhythmias such as atrial fibrillation, other arrhythmias or STEMI, NSTEMI, or unstable angina. However, a statistically significant difference was detected between the two groups when considering the sum of the composite endpoints of STEMI, NSTEMI, and unstable angina under the term ACS, with 17 cases (6.8%) in Group I and 7 cases (2.9%) in Group II (P = 0.046) (Table 3; Fig. 2).

Table 3 Clinical outcomes after one year of follow up of both groups

Fig. 2

Clinical outcomes of both groups after one year of follow up

Multivariate regression analysis was performed to identify the independent predictors of CSFP. The results showed that BMI (odds ratio (OR) = 0.694, 95% confidence interval (CI) = 0.295–0.842, P = 0.010), blood glucose during catheterization (OR = 0.647, 95% CI = 0.298–0.874, P = 0.008), serum triglycerides (TG) (OR = 0.574, 95% CI = 0.289–0.746, P = 0.010), and the NLR (OR = 0.618, 95% CI = 0.479–0.892, P = 0.001) were the independent predictors of CSFP (Table 4).

Table 4 Multivariate regression analysis shows the independent predictors of coronary slow flow

Discussion

Many theories have been proposed for the causes of CSF, including atherosclerosis, microvascular malfunction, an imbalance of vasodilator and vasoconstrictor factors, and abnormal platelet function. The present study aimed to examine this phenomenon in diabetic patients, particularly those with chronic stable angina. Many factors are involved in predicting these phenomena. The present study showed that body mass index (BMI) was an independent predictor of CSFP, consistent with the findings of Mukhopadhyay et al., who demonstrated an independent and strong correlation between BMI and CSFP. After exploring risk factors and angiographic characteristics in North Indian patients, BMI and CSFP were independently and strongly correlated [10]. Additionally, Al Suwaidi et al. reported that obese patients with normal or mildly diseased coronary arteries are at risk for CSFP due to an independent relationship between coronary endothelial malfunction and obesity [11].

The current study showed that smoking is significantly higher in the slow flow group, aligning with Ghaffari et al., who studied clinical and laboratory predictors of CSF in coronary angiography and found that smoking is a strong predictor of slow flow [12]. In contrast, Zhao et al. studied several predictors of CSFP and were unable to identify significant differences between CSFP patients and controls regarding smoking status and BMI [13].

The present study showed that CSFP is significantly associated with elevated triglyceride levels. However, previous studies examining this association have yielded different results. Some studies identified increased triglyceride levels as an independent predictor of CSFP, while others found no relationship between triglyceride levels and CSFP [14].

Consistent with our findings, Dai YX et al. reported that patients with increased triglyceride levels, fasting blood glucose levels, and low high-density lipoprotein (HDL) levels were more prone to CSFP [15]. Additionally, Pavithran et al. concluded that elevated blood glucose levels two hours after meals are linked to a higher risk of cardiovascular mortality [16]. We discovered that the blood glucose level at the time of coronary angiography strongly predicted CSFP, which aligns with the findings of Xia et al., who suggested that CSFP was caused by poor glucose metabolism. Xia et al. conducted an oral glucose tolerance test as part of their study [17]. Dharma et al. concluded that angiographic slow and no-flow patterns independently predict acute hyperglycemia (180 mg/dl) [18]. The development of free radicals, which lead to proinflammatory conditions and microvascular endothelial malfunctions, may be one of the mechanisms by which acute hyperglycemia exerts its effects, contributing to CSFP [19].

The present study showed that Hs-CRP is significantly increased in the slow flow group, consistent with the findings of Madak et al., who concluded that the TIMI frame count is positively related to elevated levels of Hs-CRP and NT-pro BNP in patients with CSF compared to the normal control group [20]. A meta-analysis performed by Zhang et al. verified the efficacy of various inflammatory and hematological indices in predicting slow flow and concluded that high CRP and Hs-CRP levels before CAG may independently predict slow flow and no-reflow [21].

The present study showed that serum fibrinogen concentration was elevated in the CSFP group, consistent with the findings of Sanghvi et al., who concluded that higher fibrinogen levels were significantly associated with CSFP [22]. Additionally, Cem et al. discovered that the presence and severity of CSF are independently related to plasma fibrinogen concentration, an easily measurable inflammatory biomarker [23]. Yurdam et al. studied the relationship between TIMI flow and the MAPH score in patients undergoing primary percutaneous coronary intervention for STEMI. They found that a high MAPH score, which consists of mean platelet volume, age, total protein, and hematocrit, may indicate coronary no-reflow [24].

Finally, our results demonstrated a considerable increase in the NLR in the CSFP group compared to the control group with normal coronary arteries. These findings agree with the conclusion that the NLR is an independent predictor of CSFP. Yılmaz et al. found that, in contrast to individuals with normal coronary morphology, the NLR was greater in patients with CAD, CSF, and coronary artery ectasia, suggesting that the NLR could predict CSFP [25]. A high NLR may independently predict atherosclerosis development in patients with CCS and acute coronary syndrome (ACS). The white blood cell (WBC) count and subtype ratio are important inflammatory indices for predicting cardiovascular outcomes and could be used as inflammatory markers in cardiovascular illnesses [26].

In the present study, MACE was higher in the CSFP group after one year of follow-up than in the control group. Although there was no significant difference between the two groups, there was a statistically significant difference when considering the composite endpoints of STEMI, NSTEMI, and unstable angina under the term ACS. These findings are consistent with those of Amirzadegan et al., who did not observe any significant difference in midterm outcomes between patients with CSF and those in the normal group [27]. Additionally, Yu et al. reported that individuals with CSF have poor clinical outcomes and are at significant risk for cardiovascular events [28].

Conclusions

Various multifactorial abnormalities, such as endothelial dysfunction, subclinical atherosclerosis, inflammatory status, and functional and structural dysfunction in the coronary microcirculation, can cause the coronary slow flow phenomenon. All of these abnormalities can contribute to temporary or permanent myocardial hypoperfusion. Our study demonstrated that serum triglycerides, body mass index, neutrophil/lymphocyte ratio, and high blood glucose levels during catheterization are independent predictors of CSFP in diabetic patients. While MACEs were higher in diabetic patients with CSF, the difference was not statistically significant, except for the composite endpoints of acute coronary syndromes.

Limitations of the study

The study had some limitations, including a relatively small number of patients and its single-center design, which necessitates multicenter studies to validate the results. Additionally, the follow-up period was relatively short (one year), and longer-term follow-up may be needed for a more comprehensive comparison of outcomes between the two groups. Furthermore, the study did not measure or include other variables and predictors of blood viscosity, such as plasma proteins (excluding fibrinogen), which may impact the incidence of slow flow phenomena.