Correlation between glycosylated hemoglobin and syntax score

ACHAIKI IATRIKI | 2020; 39(3): 167–173

Research Article

Stavros Mantzoukis¹, Marina Gerasimou²

¹Department of Cardiology, General Hospital of Ioannina “G. Hatzikosta”, Ioannina, Greece
²Department of Microbiology, General Hospital of Ioannina “G. Hatzikosta”, Ioannina, Greece

Received: 1 June 2020; Accepted: 3 July 2020

Corresponding author: Stavros Mantzoukis, MD, MSc. Department of Cardiology, General Hospital of Ioannina “G. Hatzikosta”, 1 Makrigianni Str, Ioannina 45500, Greece, E-mail: stavrosmantzoukis@gmail.com

Key words: Glycosylated hemoglobin, coronary heart disease, diabetes mellitus, coronary angiography, SYNTAX score

Abstract

Background: Glycosylated hemoglobin is used in both diabetes diagnosis and glycemic control assessment. The aim of this study was to demonstrate the possible correlation between glycosylated hemoglobin levels and the severity of coronary heart disease as expressed by the SYNTAX score.

Methods: All patients who were admitted to the Cardiology Clinic of the General Hospital of Ioannina from 16/11/2018 to 14/1/2019 due to either stable angina or acute coronary syndromes and were subjected to coronary angiography which demonstrated coronary artery disease were enrolled. A total of 93 patients were included in the study. In all participants, glycosylated hemoglobin was measured and SYNTAX score was calculated after the coronary angiography.

Results: Higher SYNTAX score was observed in patients with elevated levels of glycosylated hemoglobin. Glycosylated hemoglobin level did not emerge as an independent prognostic factor for the severity degree of coronary artery disease when the SYNTAX score was used as a severity index. However, the history of diabetes mellitus was found to be an independent prognostic factor for angiographic severe coronary disease.

Conclusion: The history of diabetes mellitus and its long-term effects on coronary arteries appear to be the most important independent risk factor for severe coronary heart disease. Glycosylated hemoglobin levels could be an important prognostic marker for the severity of coronary heart disease even in subjects with glycosylated hemoglobin within normal limits. However, this should be confirmed by larger clinical studies.

Introduction

Diabetes mellitus (DM) is one of the major risk factors for cardiovascular disease. Western lifestyle is associated with increased incidence of type 2 DM and increased cardiovascular morbidity and mortality. Glycosylated hemoglobin (HbA1c) is used both in the diagnosis of DM and in the assessment of glycemic control for a period of 2-3 months prior to sampling. Higher levels of HbA1c in diabetic patients appear to be associated with an elevated risk for cardiovascular events [1]. The purpose of the present study was to demonstrate the possible association between HbA1c levels and the angiographic severity of coronary heart disease both in patients with a history of DM and patients without a history of DM. The severity of coronary heart disease was quantitatively expressed by the value of SYNTAX SCORE.

Methods

All patients who were admitted to the Cardiology Clinic of Ioannina General Hospital from 16/11/2018 to 14/01/2019 either due to acute coronary syndrome or stable coronary artery disease and underwent coronary angiography were eligible to enter the study. More specifically, participants had been admitted due to unstable angina, myocardial infraction with ST elevation (STEMI), myocardial infraction without ST elevation (NSTEMI) or stable angina. Written informed consent was obtained from each patient included in the study and the study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki. In all of these patients, serum HbA1c (normal values: 4.3-6.1%) levels were measured one day after the coronary angiography. SYNTAX score was calculated based on coronary angiography findings in all patients. The calculation of the SYNTAX score takes into account the localization and the individual characteristics of the lesions. Stenosis is defined as a reduction in the diameter of the lumen by more than 50% compared to the diameter of the proximal healthy segment for vessels > 1.5 mm in diameter. The degree of stenosis is not included in the calculation formula except in cases of total occlusion. It is considered whether the right or left coronary artery is the dominant vessel and in which part of the coronary artery the lesion is located. For chronic obstructions, the score calculation algorithm takes into account whether obstruction dates >3 months, whether there is blind occlusion or bridging, the first segment beyond the total occlusion that is visualized by antegrade or retrograde contrast and whether there are smaller arterial branches before occlusion, and what size are they. Other features of lesions considered to be unfavorable and rated higher are: ostial lesions, bifurcation or trifurcation lesions, severe tortuosity, large length of the lesion (> 20mm), severe calcification, the presence of thrombosis as well as diffuse coronary artery disease [2]. At the same time, the existence of other risk factors that increase cardiovascular risk and may be a confounding factor in the study was recorded. More specifically, we recorded whether there was a history of arterial hypertension or a diagnosis of hypertension during hospitalization, a history of dyslipidemia or a diagnosis of dyslipidemia during hospitalization, a history of smoking in the past five years, a history of DM or a diagnosis of DM during hospitalization and a family history of coronary artery disease. In total there were 98 recordings from patients admitted to the cardiology clinic during the aforementioned period due to acute coronary syndrome or stable angina and in whom coronary angiography revealed coronary artery disease. However, the statistical analysis only included 93 individuals as 5 records were excluded for the following reasons:

Three patients because the first blood sample was not analyzed due to technical issues, and no second sample was taken as the patients had already been discharged.

Two patients enrolled twice during the study as they were twice admitted due to acute coronary syndrome during the study period. At these admissions, these patients had a fixed glycosylated hemoglobin value and a similar angiographic image. Therefore, only their first admission was recorded.

It is noteworthy that one patient died before taking a sample for HbA1c measurement. This patient’s data were used in the sample description but were not used in the correlation analysis (Figure 1).

Figure 1. Flow chart with registered and ‘dropout’ patients of the study.

None of the participants had anemia, hemoglobinopathy or a history of recent blood transfusions (events that may affect HbA1c measurement). In addition, all patients with a history of coronary heart disease who underwent coronary angiography at that time had more angiographically severe coronary heart disease.

Inclusion and exclusion criteria of the study are described in Table 1.

Descriptive measurements were described using frequencies and percentages for categorical variables while mean values and standard deviations were used for continuous variables. For the correlations between the categorical data, the x2 test was used, and in cases of non-fulfillment of the conditions, the Fisher’s exact test. The Mann Whitney test was used to detect differences between the two groups in continuous parameters, while the Pearson test was used to detect correlations between continuous variables [3-5]. Based on the results of the univariate analyses, a logistic regression model was created with the SYNTAX score as the dependent variable. Statistical analysis was performed with the SPSS v22 software and statistical significance was set at 0.05 in all cases.

Results

Descriptive data

A total of 93 patients were included in the study with a mean age of 68.83 years (40 to 88 years) for whom a set of demographic characteristics was recorded. Specifically, 79 patients were men and 14 women, 30 were smokers, 17 had a family history of coronary heart disease, 71 had hypertension, 53 had dyslipidemia, 34 had DM and 31 had known coronary artery disease. Glycosylated hemoglobin levels ranged from 4.3 to 11.1 with a mean of 6.358. By categorizing these values, it appears that 39 (42.4%) exceeded the normal threshold. The cause for admission to the cardiology clinic was 33% NSTEMI, 18% STEMI, 13% unstable angina and 29% stable angina. The mean value of the SYNTAX score was 14.81 (0-45). The SYNTAX score can also be assigned to categories depending on the values recorded. Table 2 shows that most patients had low rates (77.4%) while only 7 had a “very high” SYNTAX score.

First degree correlations

The Pearson correlation table shows that there was a statistically significant positive correlation of SYNTAX score with glycosylated hemoglobin levels, meaning that higher SYNTAX score values were expected for higher glycosylated values (p = 0.002) (Table 3). Differences were also found for the SYNTAX score depending on whether glycosylated hemoglobin level was normal or abnormal. Patients with abnormal glycosylated level had higher SYNTAX score than patients with normal glycosylated (p = 0.002) (Table 4). Differences were also found for the SYNTAX score depending on the presence of DM history. SYNTAX score values were significantly higher for patients with history of DM compared with patients without history of DM (p = 0.009) (Table 5).

Regression analysis

Logistic regression analysis was performed to determine the independent predictors. The approach of categorizing the SYNTAX score at the threshold value of 22 was adopted as this value is a critical point for the choice of coronary artery bypass grafting or not. The analysis showed that the model including DM history and left main disease can improve the prediction of whether the patient will have a SYNTAX score greater than or below 22 by approximately 7.6% while correctly predicting 1/3 of the patients who had a moderate to very high SYNTAX score (i.e. above 22). Model fit was good with p = 0.480. The analysis showed that patients with left main coronary artery disease were about 17 times more likely to have SYNTAX score greater than 22, compared to patients without left main coronary artery disease (p = 0.001). The 95% confidence interval for this estimation was approximately 3.4 to 87 times. This large discrepancy was due to the fact that only 10 patients had left main coronary artery disease. At the same time, patients with DM were about 5.5 times more likely to have SYNTAX score greater than 22, compared to patients without DM (p = 0.023). The 95% confidence interval for this estimation was approximately 1.2 to 24 times. This large discrepancy could be attributed to the relatively small number of patients with DM that did not allow to reduce uncertainty. (Table 6, Figure 2 and 3)

Figure 2. Correlation between left main coronary artery disease and SYNTAX score

Figure 3. Correlation between history of diabetes mellitus and SYNTAX score

Discussion

The study involved 93 people, the majority of whom were men and presenting multiple cardiovascular risk factors at the same time. HbA1c level was selected as an indicator of patients’ chronic glycemic status, as it takes into account both post-operative hyperglycemia episodes that are positively related to diabetes complications and mainly cardiovascular complications [6]. Regarding the distribution of clinical manifestations of coronary heart disease, the results of this study coincide with data from international literature. In particular, coronary heart disease occurs either in the form of stable angina or in the form of acute coronary syndromes. Acute coronary syndromes according to the guidelines of the European Cardiology Society appear in decreasing frequency as NSTEMI, STEMI and less as unstable angina (as is the present study) [7].

The present study showed a correlation between glycosylated hemoglobin levels and SYNTAX score values. The SYNTAX score is an angiographic score to describe the severity or complexity of a coronary artery disease. SYNTAX stands for “SYNergy between PCI with TAXUS and Cardiac Surgery”. The SYNTAX Score I is calculated using a computer program that asks sequential and interactive questions. The algorithm consists of 12 main questions, which in turn can be divided into 2 groups: the first 3 determine the dominance, the total number of vascular segments and the number of segments involved per lesion. The last 9 questions refer to the adverse lesion characteristics (e.g. calcification, degree of occlusion and length of the lesion) and are repeated for each lesion. The SYNTAX-Score I takes neither the patient characteristics nor the treatment strategy into account, but only the coronary anatomy. The SYNTAX Score II takes account age, gender, left ventricular ejection fraction, creatinine clearance, chronic obstructive pulmonary disease and peripheral artery disease. In our study, elevated SYNTAX score values are observed in subjects with elevated glycosylated hemoglobin. The above results are consistent with data from international literature showing that glycosylated hemoglobin levels constitute a risk factor for cardiovascular disease and are associated with more severe coronary heart disease as defined on the basis of the SYNTAX score [8]. The above applies not only to people with abnormal glycosylated hemoglobin but also to individuals with normal glycosylated hemoglobin levels. Even in these cases, individuals with a higher glycosylated hemoglobin value show an increased SYNTAX score. It therefore appears that the role of early intervention in glycemic control (even in patients with normal glycosylated hemoglobin levels) should be explored to reduce cardiovascular risk. It appears that in subjects with chronic hyperglycemia and glycosylated hemoglobin values at higher normal levels, there is a greater likelihood of angiographic severe coronary artery disease possibly through the same mechanisms acting in individuals with diabetes such as endothelial dysfunction and oxidative stress [8].

The role of HbA1c as an independent risk factor was not corroborated when the SYNTAX score was used as a coronary artery disease severity index. It appears that other cardiovascular risk factors interact with DM in the development of coronary heart disease. This is not consistent with studies suggesting that the value of HbA1c is an independent factor that determines angiographic severity of coronary disease even in non-diabetics regardless of the type of clinical manifestation for which they underwent coronary examination [6,8-13]. It is worth noting that there are limited studies that show that HbA1c is not an independent cardiovascular risk factor in non-diabetic patients [14] or that patients with severe angiographic lesions usually have higher HbA1c values but without being an independent risk factor [15].

In the present study, the history of DM and left main coronary artery disease appeared to be independent risk factors for the severity of coronary heart disease as expressed by the SYNTAX score. This finding could explain the absence of a statistically significant association between HbA1c and angiographic severe coronary artery disease as many patients with known diabetes mellitus (which is an independent risk factor as mentioned above) counteract the pathophysiological effects (oxidative stress – inflammation – atherosclerotic lesions) of diabetes which usually existed many years before glycemic control was achieved via dietary supplementation and medication. It should also be remembered that HbA1c levels reflect the patient’s glycemic status during the last trimester and not over a longer period of time.

This study has two main limitations. The first is that a single measurement of HbA1c was used and therefore no reliable conclusions can be drawn on the effect of HbA1c on coronary vessels over time. This is perhaps one of the reasons that HbA1c was not statistically proven to be an independent prognostic factor for severe coronary heart disease in the present study. Another limitation of the study is the failure of coronary angiography to provide accurate information on the composition of atherosclerotic plaques and thus the possibility of a minor angiographic lesion leading to a clinically significant event in the future. To avoid this limitation, intravascular ultrasound could be used in future studies in addition to coronary angiography, in order to provide important information on the arterial wall and atherosclerotic plaque formation.

In conclusion, subjects with higher HbA1c values had more severe coronary heart disease as expressed by the SYNTAX score. Glycosylated hemoglobin levels reflecting patients’ chronic glycemic status appeared to be a useful tool in the future to distinguish high-risk patients who may benefit from earlier intervention to reduce the risk of cardiovascular events. However, it remains unclear at this time whether therapeutic agents should be administered to individuals with normal HbA1c levels. Larger studies should be carried out on this topic and certainly an extensive interdisciplinary discussion should be conducted regarding the role glycosylated hemoglobin as a prognostic marker for coronary heart disease and as an indicator of early intervention even in non-diabetic patients. Studies are also needed to investigate the possible association of glycosylated hemoglobin levels, especially in non-diabetic patients, with clinical outcome, morbidity and mortality. Existing data in diabetic patients have already demonstrated the association of HbA1c with clinical outcome and morbidity [16-17]. It should also be remembered that people with coronary artery disease that cause stenosis >50% may never develop acute coronary syndrome while people with <50% lesions can develop acute coronary syndrome that can even lead to death. After all, coronary angiography only depicts the lumen of the vessel and does not provide information on the arterial wall and the status of atherosclerotic plaques (stable or unstable plaque). Therefore, in addition to the association of glycosylated hemoglobin levels with angiographic severity, it is also necessary to investigate its association with the clinical outcome.

Conflict of interest disclosure

None to declare.

Declaration of funding sources

None to declare.

Author contributions

All authors had equal contribution regarding conception and design; analysis and interpretation of the data; drafting of the article; critical revision of the article for important intellectual content and final approval of the article.

References

1. Farhan S, Jarai R, Tentzeris I, Kautzky-Willer A, Samaha E, Smetana P, et al. Comparison of HbA1c and oral glucose tolerance test for diagnosis of diabetes in patients with coronary artery disease. Clin Res Cardiol. 2012;101(8):625-30.
2. Head J, Farooq V, Serruys W, Kappetein P. The SYNTAX score and its clinical implications. Heart. 2014;100(2):169-77.
3. Agresti A. Categorical data analysis. Wiley-Interscience; 2013.
4. Draper R, Smith H. Applied regression analysis. Vol. 326. John Wiley & Sons; 1998.
5. Van Belle G, Fisher L. Biostatistics: a methodology for the health sciences. Vol. 519. John Wiley & Sons; 2004.
6. Ikeda N, Iijima R, Hara H, Moroi M, Nakamura M, Sugi K. Glycated hemoglobin is associated with the complexity of coronary artery disease, even in non-diabetic adults. J Atheroscler Thromb. 2012;19(12):1066-72.
7. Roffi M, Patrono C, Collet P, Mueller C, Valgimigli M, Andreotti F, et al. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Eur Heart J. 2016;37(3):267-315.
8. Garg N, Moorthy N, Kapoor A, Tewari S, Kumar S, Sinha A, et al. Hemoglobin A1c in Nondiabetic Patients: An Independent Predictor of Coronary Artery Disease and Its Severity. Mayo Clin Proc. 2014;89(7):908-16.
9. Arbel Y, Zlotnik M, Halkin A, Havakuk O, Berliner S, Herz I, et al. Admission glucose, fasting glucose, HbA1c levels and the SYNTAX score in non-diabetic patients undergoing coronary angiography. Clin Res Cardiol. 2014;103(3):223-7.
10. Ayhan S, Tosun M, Ozturk S, Alcelik A, Ozlu MF, Erdem A, et al. Glycated haemoglobin is correlated with the severity of coronary artery disease independently of traditional risk factors in young patients. Endokrynol Pol. 2012;63(5):367-71.
11. Cai A, Li G, Chen J, Li X, Wei X, Li L, et al. Glycated hemoglobin level is significantly associated with the severity of coronary artery disease in non-diabetic adults. Lipids Health Dis. 2014;13:181.
12. Farrag A, Ammar W, Hady A, Samhoon E. Haemoglobin A1c as a marker predicting extent and severity of coronary artery disease in non-diabetic patients. Acta Cardiol. 2016;71(5):581-585.
13. Dutta B, Neginhal M, Iqbal F. Glycated Hemoglobin (HbA1c) Correlation with Severity of Coronary Artery Disease in Non-diabetic Patients – A Hospital based Study from North-Eastern India. J Clin Diagn Res. 2016;10(9):OC20-OC23.
14. Salim A, Tai S, Tan Y, Welsh AH, Liew R, Naidoo N, et al. C-reactive protein and serum creatinine, but not haemoglobin A1c, are independent predictors of coronary heart disease risk in non-diabetic Chinese. Eur J Prev Cardiol. 2016;23(12):1339-49.
15. Verdoia M, Schaffer A, Cassetti E, Barbieri L, Di Ruocco MV, Perrone-Filardi P, et al. Atherosclerosis Study Group (NAS). Glycosylated Hemoglobin and Coronary Artery Disease in Patients Without Diabetes Mellitus. Am J Prev Med. 2014;47(1):9-16.
16. Stolar M. Glycemic Control and Complications in Type 2 Diabetes Mellitus. Am J Med. 2010;123(3 Suppl):S3-11.
17. Zhao W, Katzmarzyk T, Horswell R, Wang Y, Johnson J, Hu G. HbA1c and Coronary Heart Disease Risk Among Diabetic Patients. Diabetes Care. 2014;37(2):428-35.