Relationship between hemoglobin A1c and serum troponin in patients with diabetes and cardiovascular events

Abstract

Objectives

Diabetes mellitus is a group of metabolic disorders associated with high risk for cardiovascular disease. Although troponins are primarily clinically used for the diagnosis of acute coronary syndrome, they are also used in risk assessment in patients with acute coronary syndrome as well as in a number of other conditions. The aim of this review was to investigate the relationship between hemoglobin A1c and serum troponin in patients with diabetes and cardiovascular events.

Methods

Hemoglobin A1_c has been chosen as the best clinical indicator of glucose control and risk of micro and macrovascular complications. We investigated cardiac troponins as a group of markers of muscle injury which includes troponin T, troponin I and troponin C. Troponin T and I are specific for myocardial injury, compared to C which is specific for skeletal muscle.

Results

In this review, we showed that there was a causal relation between hemoglobin A1_c levels and serum troponin concentrations. Hemoglobin A1_c has shown to be a positive predictive factor of incidence, mortality and morbidity of conditions such as acute coronary syndrome, arrhythmias, stroke, pulmonary embolism and other conditions that causes troponin elevation by its release in circulation.

Conclusions

Chronic hyperglycemia decreases glomerular filtration and consequently decreases troponin elimination and also by affecting the heart microcirculation it leads to microvascular damage and consequently to ischemia which contribute to troponin concentration elevation. Furthermore, correlation between hemoglobin A1_c and troponin concentration manifests in their prognostic value for mortality.

Introduction

Diabetes mellitus (DM) is a group of metabolic disorders characterized by hyperglycemia that can be caused by insufficient insulin secretion, insulin action – insulin resistance (IR) or by the combination of those two factors [1]. Most commonly accepted classification of diabetes is classification on type 1, type 2, gestational diabetes and other specific types of diabetes [2]. According to International Diabetes Federation, it is estimated that in 2015 over 400 million people in the world were affected by diabetes. With the current trend of rise in the incidence and prevalence, this number will be 642 million by the year 2040 [3]. Hemoglobin A1c (HbA1_c) has shown to be the best indicator of glucose control and risk of micro and macrovascular complications in clinical trials [4–6]. HbA1_c is also used for monitoring the efficacy of antihyperglycemic therapy [7] and recently it is also used as a diagnostic test for DM [8] as well as a screening test [9]. HbA1_c shows glycemic control during the previous 2–3 months, but caution is needed in conditions that change the lifetime of erythrocytes [10, 11].

Chronic complications of diabetes include macrovascular (coronary artery disease, stroke, peripheral artery disease), microvascular (diabetic retinopathy, neuropathy and nephropathy), and diabetic foot as a result of insufficient circulation and/or sensory neuropathy [12–15]. Microvascular complications are more common than macrovascular. Diabetic nephropathy (DN) is one of the most common and most significant complication of DM [16]. DN is based on the presence of albuminuria and/or reduced estimated glomerular filtration rate (eGFR) in the absence of signs or symptoms of other primary causes of kidney damage. It is the most frequent cause of chronic kidney disease (CKD) and most common cause of kidney transplantation and it is associated with the high risk of cardiovascular (CV) mortality [17, 18].

Diabetes mellitus type 2 (DM2) is the most common type of diabetes and it accounts for 90–95% of all diabetes cases in the world [2]. Pathophysiology of DM2 is characterized by insulin resistance (IR), beta cell dysfunction and chronic inflammation; all these factors lead to hyperglycemia which leads to chronic complications [19–21]. DM2 is a multifactorial disease with genetic and environmental factors affecting clinical presentation [19]. Most DM2 patients suffer from obesity or have a high amount of abdominal fat tissue even with the normal body mass index (BMI) [2]. Adipocyte hypertrophy also causes local fat tissue inflammation resulting in a lower concentration of protective proteins like adiponectin [22, 23]. Despite screening programs, 25% of DM2 patients already have developed microvascular complications at the moment of diagnosis which means that disease already exists for at least 5 years [24, 25]. DM2 elevates the CV risk by 2–4 times and although DM2 is associated with a number of CV risk factors such as hyperglycemia, dyslipidemia and hypertension, we should stick to those and other traditional CV risk factors for CV risk assessments. Although there are several calculators for CV risk assessment [26], an early biomarker for prediction of CV disease is still missing.

Cardiac troponins: troponin T (cTnT), troponin I (cTnI) and troponin C (TnC) are proteins that by binding with calcium allow interaction between actin and myosin in the muscle. cTnT and cTnI are unique for the myocardium [27] while TnC is identical to troponin C in the skeletal muscle [28, 29]. Majority of cardiac troponins are located in the cardiac contractile apparatus while cTnT and cTnI are located in cytosolic vesicles in around 6.8% and 3.5%, respectively [30]. Although cTnT is also located in the skeletal muscle in small amounts, Ricchiuti et al. (1998.) conclude that these amounts cannot cause a false positive result [31, 32]. On the other hand, no cTnI is found in the skeletal muscle [33]. Troponins have a higher clinical value in myocardial damage diagnosis than classic enzymes such as myoglobin, creatine kinase (CK) [34, 35], cardiac specific creatine kinase (CK-MB) [36] and lactate dehydrogenase (LDH) [37]. Troponins are elevated in the range of 2–3 h after a myocardial injury, reaching the highest concentration for approximately 24 h and their concentration remains elevated for about 8 days after a myocardial injury [38]. The clinical significance of troponins is in the diagnosis of Non-ST-elevation myocardial infarction (nSTEMI) for confirmation of the diagnosis when there is no other sign besides the clinical manifestation of myocardial infarction (MI) [27, 39]. According to the new guidelines of European Cardiology Society (ESC), High-Sensitivity Cardiac Troponin (hscTn) assays should be preferred to other assays due to their higher diagnostic accuracy, sensitivity and negative predictive value [40, 41]. Troponins have been shown to be a useful indicator in the risk assessment of patients with the acute coronary syndrome (ACS), chronic kidney failure, pulmonary embolism and many other conditions [42]. Although cTnT and cTnI have a similar diagnostic value [43], cTnT has a better prognostic value for long-term mortality after MI [44, 45]. Troponins after MI are released into the bloodstream mainly as free cTnT and as a TnC-cTnI complex while other combinations are rare [46, 47]. Michielsen et al. have found that a whole cTnT molecule is excreted from the bloodstream a few hours after MI while the immunoreactive fragment remains in the bloodstream longer. [48, 49]. In conditions where there is a less increase in cTnT concentrations such as heart failure (HF) or strenuous exercise, it is assumed that renal clearance trough glomerular filtration has a much more significant role in troponin elimination which explains the inverse correlation between glomerular filtration and cTnT concentrations [50].

In DM2 patients without cardiovascular disease (CAD), cTnT concentrations are found to be higher than in the rest of the otherwise healthy individuals without CAD by using cTnT assay. Furthermore, the elevation in the troponin concentration shows a higher correlation with postprandial blood glucose than with fasting blood glucose concentrations [51]. Also, in patients without known diabetes and/or CVD, higher cTnT concentrations measured by the high sensitive cTnT assay correlate with a higher incidence of diabetes in the future [52]. In this review we will focus on the effect of hyperglycemia on conditions that cause myocardial damage and the elevation of troponin in blood. We will also explain how glycemic control affects the mechanisms of troponin degradation and elimination.

Troponin and hemoglobin A1c in acute coronary syndrome

Troponin concentration in MI is elevated [53] and in that case troponins have diagnostic and prognostic value [54, 55]. DM is an independent predictor of mortality in patients with MI [56] and in patients after coronary revascularization [57]. Among patients with ACS, one third of them suffer from DM and those have worse outcomes [58, 59], and MI can be an initial presentation of DM2 because complications of the disease can be an initial presentation of DM2 [60]. Segre et al. [61] studied the cTnI concentration in DM patients with and without underlying CAD. In their cross-sectional study they compared the cTnI concentrations in 50 DM patients with multivessel CAD and in 45 DM with the angiographically normal coronary arteries. They obtained a statistically significant cTnI elevation in the group with underlying CAD compared to the group without underlying CAD. The authors conclude that the increase in cTn concentrations correlates with CAD in DM patients although there are multiple CV risk factors associated with diabetes. Savonitto et al. [62] studied all-cause mortality in DM patients after ACS as a primary outcome as well as cardiac mortality as a secondary outcome after two years of monitoring. In their multicentric study, 1 % rise in HbA1_c had a positive predictive value for the primary and secondary outcome (HR = 1.11; 95% CI = 1.03–1.19 for all-cause mortality). Noguchi et al. showed by multivariant analysis that patients with the lower HbA1_c levels had a lower risk for major adverse cardiac and cerebrovascular event (MACCE) after ACS [63] and consequently the lower probability of troponin levels elevation [64]. Dubey et al. studied the association between HbA1_c and mortality and morbidity of 110 patients with ACS in their prospective study. They demonstrated that DM patients had a higher morbidity after ACS and a higher risk of complications like left ventricular dysfunction (LVD) and heart failure (HF) [65]. From those studies it can be inferred that there is a positive correlation between HbA1_c levels and ACS as well as a positive predictive value of HbA1_c for mortality and morbidity after ACS. Since cTns have both diagnostic and prognostic value in ACS patients, there is a correlation between HbA1_c, as an indicator of glycemic control and cTn concentrations in those patients. However, for better understanding of the complex relation between HbA1_c and cTn in ACS further clinical experiments are required.

Troponin and hemoglobin A1c in heart failure

Troponins are elevated in acute and chronic HF and have a significant prognostic value [66]. Presence of DM2 is recognized as a significant risk factor for HF development independently of CAD and hypertension, and prevalence of DM2 in HF patients is around 25% compared to 7% in overall population [67, 68]. Gerstein et al. investigated whether the HbA1_c level is a risk factor for mortality in HF patients (cardiovascular and overall mortality) and a risk factor for hospitalization. They monitored 2412 patients (907 with DM) within Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) study. Inclusion criteria were HF patients with New York Heart Association (NYHA) Classification stage II-IV, serum creatinine below 265 μmol/L, negative medical history of using ATII blockers, aorta or mitral stenosis, MI and surgical procedures on heart. They divided patients in 8 groups based on HbA1_c levels. They found a statistically significant increase in mortality and hospitalization for every 1% elevation in HbA1_c levels after adjusting for age and sex. Mortality correlates with HbA1_c levels in patients without previously diagnosed DM. Overall number of hospitalizations also correlated with HbA1_c while there were no statistically significant correlation between number of hospitalizations and HbA1_c levels in patients with previously diagnosed DM [69]. Zhao et al. studied in a prospective cohort study the correlation between HbA1_c levels and the risk of developing HF in Caucasian and Afro-Americans with DM but without the history of chronic heart disease (CHD) and HF, and the diagnosis of HF was a primary outcome. The population studied (2229 Caucasians and 2860 Afro-Americans) was monitored for a median of 6.5 years. For every 1% increase in HbA1_c levels the risk for developing HF in the Caucasians and in Afro-Americans elevated for 9% and 6%, respectively [70]. Aguilar et al. studied the effect of HbA1_c levels on hospitalization and mortality in patients with HF in a study conducted in ambulatory clinics at Veteran Affairs medical centers. There were 4048 patients of which 94% were male. Patients were distributed in 5 quartiles(Q) based on HbA1_c levels. Those in the Q3 group (HbA1_c = 7.1–7.8%) had the lowest mortality rate, 27% lower than those in the lowest Q1 group (HbA1_c < 6.4%). Those in the Q5 (HbA1_c > 9.0%) had highest mortality but their hazard ratio (HR) relative to Q1 was not statistically significant [71]. HbA1c was also a risk factor for heart failure in DM in the Atherosclerosis Risk in Communities (ARIC) study [72], while Tomova et al. concluded that the elevated HbA1_c level is a positive predictive factor of survival and decreases the need for urgent heart transplantation. In that study patients were divided in 4 groups based on glycemic control. The highest quartile patients (HbA1_c > 8.6%) had the lowest mortality rate and lowest urgent heart transplantation compared with the lowest quartile patients [73]. Authors explain that these unexpected results can be attributed to reverse epidemiology where in patients with symptomatic HF the usual risk factors such as BMI, blood pressure and high cholesterol correlate with lower mortality. Furthermore, it is assumed that in patients with low expected survival time negative effects of malnutrition/cachexia and catabolic stress outweigh the positive long-term effect of lower BMI, blood pressure and cholesterol [74, 75]. Grembowski et al. studied the reverse epidemiology of HbA1_c in HF patients in a 9-year retrospective cohort study. Patients with HbA1_c < 6.3% had a higher mortality rates than those with HbA1_c between 6.9% and 7.7% while patients with HbA1_c equal or higher than 8.7% had the lowest mortality rate. The authors conclude that there is a negative correlation between glycemic control and survival in these patients [76]. Results of this study suggest that HbA1_c level is a positive predictor of survival which is partly in contrast with other studies that found different correlations between HbA1_c levels and mortality in HF patients [77–79].

Troponin and hemoglobin A1c in diabetic cardiomyopathy

Diabetic cardiomyopathy (DCM) is defined as a myocardial dysfunction in DM patients in absence of hypertension and structural heart disease such as valvular heart disease and CAD and is characterized by lipid accumulation in cardiomyocytes, fetal genes reactivation, ventricular thickening and diastolic dysfunction that progresses towards systolic dysfunction [80, 81]. Since DM is a separate risk factor for HF development independently of age, BMI, blood pressure and CAD, it is assumed that risk for HF in DCM is exclusively under the influence of hyperglycemia and IR [82]. Diagnostic and prognostic value of cTn in patients with DCM is still unclear. Fetal hypertrophic cardiomyopathy is a well-known complication in pregnancies of DM1 mothers, it affects 40% of fetuses, causing symptoms in 5% of them while long-term complications have not been registered [83, 84]. Russel et al. measured proBrain Natriuremic Peptide (proBNP) and TnT at the time of the delivery of 45 fetuses of DM1 mothers and 39 mothers without DM1 and with normal pregnancies in a prospective study. Both biomarkers were higher in the DM1 group. The proBNP levels correlated with the thickness of the interventricular septum while the TnT levels correlated with the umbilical artery pulsatile index, which indicates a uteroplacental circulation dysfunction resulting in an elevated afterload in fetal circulation and causing myocardial ischemia. TnT levels also correlated with an unfavorable perinatal outcome (admission to the neonatal intensive care unit because of everything except hypoglycemia with or without need for assisted ventilation; pH of umbilical artery blood <=7.2, Apgar after 1 min < =3 and after 5 min < =7 and gestational age < 37 weeks) [85]. Rasmussen et al. found the correlation between HbA1_c levels of mothers suffering from DM1 and uteroplacental circulation dysfunction [86]. However, it is difficult to differentiate in older patients between symptomatology of DCM and HF due to other causes because DM is a risk for DCM [87] and for the causes of HF such as CAD and hypertension [12]. Furthermore, prevalence of DCM is unknown because there has not yet been a study large and precise enough to estimate the prevalence of DCM [87]. The effect of DCM on troponin concentrations in older patients and the effect of HbA1_c levels on the development of DCM and severity of clinical presentation will (all) need to be explored.

Troponin and hemoglobin A1c in stroke

Elevation of troponin concentration is recorded in all types of stroke including ischemic stroke (IMU), intracerebral hemorrhage (ICH) and subarachnoid hemorrhage (SAH) [88, 89]. In patients with stroke cardiac contractile dysfunction is common as well as electrocardiography (ECG) changes such as ST segment depression and T wave inversion [89]. It is assumed that cardiac injury and renal failure are the main reasons for troponin elevation in stroke [90]. It should be noted that the incidence of stroke is increased after MI [91] and that after stroke, other CV diseases are some of the major causes of mortality, especially CAD [92]. Troponin concentrations have a positive predictive value for mortality [93, 94], as well as for the degree of disability [95, 96] and for the degree of severity of stroke [97] independently of underlying cardiac and renal diseases. HbA1_c level is a positive predictive factor for stroke occurrence and this correlation is positive for every 1% of elevation of HbA1c [98]. Hjalmarsson et al. conclude that the HbA1_c level before stroke is a positive predictor of the degree and severity of stroke, worse functional outcome and overall mortality [99]. Furthermore, HbA1_c levels upon admission to hospital have a positive predictive value for the degree of severity of stroke, worse functional outcome [100] and mortality [101]. In a study done by Jing et al., newly diagnosed DM in patients with stroke has been shown to be a positive predictor of stroke reoccurrence, worse functional outcome and overall mortality, while HbA1_c levels elevation did not show the correlation with the outcomes mentioned [102]. To elucidate the causal relation between HbA1c and troponin in stroke the causes of troponin elevation in stroke as well as how HbA1_c elevation affects those causes, should be clarified.

Troponin and hemoglobin A1c in arrhythmias

Elevated troponin concentrations can be found in patients with prolonged episodes of supraventricular tachyarrhythmias (SVT) without underlying CAD [64]. It is assumed that troponin elevation is the result of reversible and/or irreversible ischemic injury of cardiomyocytes due to higher oxygen requirement caused by a higher heart rate and lower oxygen delivery due to a shorter diastole. If the duration of ischemia is longer than 15 min, troponine vesicles are being released from the cell to the extracellular matrix, consequently elevating troponin concentrations [42, 64]. Troponins are shown to be a good predictor for reoccurrence of atrial fibrillation (AF) [103] and some other forms of SVT [104]. cTnI concentrations are associated with the incidence and prevalence of ventricular arrhythmia and tachycardia [105]. Agarwall and Singh studied the effect of DM2 and glycemic control represented by HbA1_c levels on various arrhythmias in 100 patients that suffer from DM2 and arrhythmias. The authors concluded that DM2 and its comorbidities, most notably cardiac autonomic neuropathy (CAN), and poor glycemic control were the positive predictors of incidence of arrhythmias [106]. The fact that high HbA1_c levels are a separate risk factor for the development of CAN [107] potentiate significance of HbA1_c levels as a risk factor for the development of arrhythmias. Arrhythmias, especially AF, can precipitate conditions such as stroke and MI in patients without underlying CAD [108, 109] which further elevate the troponin concentrations.

Troponin and hemoglobin A1c in pulmonary embolism

Pulmonary embolism (PE) is the most dangerous form of venous thromboembolism (VTE) and despite the treatment, 10% of patients die one year after the PE episode. Conditions that predispose to VTE are plasma hypercoagulability, blood flow disturbance and endothelial dysfunction [110]. DM2 is shown to be a risk factor for the development of PE and pulmonary hypertension (PHT) independently of CAD, smoking, hypertension and HF [111]. In acute PE the troponin concentration can be elevated [64]. Although exact mechanism is still unclear, it is assumed that PHT develops due to pulmonary artery obstruction which leads to acute right ventricle dilatation with hypokinesia, which can cause myocardial ischemia with an elevation of troponin [112]. Some studies showed a positive predictive value of elevated troponin concentration for outcomes such as dilatation and dysfunction of the right ventricle, the amount of ventilation-perfusion defects as well as for overall mortality [113, 114]. Although some studies did not find any statistically significant correlation between HbA1_c levels and PE/VTE incidence [115], Bell et al. found a statistically significant correlation between HbA1_c and VTE incidence in patients with previously diagnosed DM. In patients without previously diagnosed DM, HR for VTE did not correlate with the HbA1_c levels [116]. Hyperglycemia causes glycation of proteins that take part in coagulation and fibrinolysis perpetuating the hypercoagulable plasma state. Increase in oxidative stress caused by hyperglycemia results in an increased transcription of coagulation factors. Endothelial glycocalyx damage of the blood vessels which is also caused by hyperglycemia causes the release of coagulation factors [117]. It seems that correlation between HbA1_c levels and troponin concentrations in PE is associated with the outcomes such as right ventricular dilatation and dysfunction, the amount of ventilation-perfusion defects and mortality.

Troponin and hemoglobin A1c in systemic inflammation

It is assumed that systemic inflammation can cause elevated serum troponin concentrations [118]. Studies that included patients with sepsis and without underlying MI where inflammation is constant showed the elevated troponin concentrations [119, 120]. Troponins have shown to be a prognostic factor for mortality in sepsis patients [121]. It is assumed that histones – nuclear proteins with the role of DNA condensation, during extended tissue damage in sepsis are released in the circulation and are inducing cardiomyocytes damage. Alhamdi et al. found a statistically significant increase in histone levels in the patients with sepsis which correlated with the cTnT concentrations and left ventricular dysfunction. Furthermore, in the same study they exhibit the results that in the septic mice antihistone antibodies have a protective effect on heart by reducing cTnI concentrations and preserving left ventricular contractility [122]. In addition to the mechanisms mentioned above, there are a few more suggested mechanisms of cardiac damage in sepsis. In one of them it is thought that endotoxins trough the Toll-like Receptors (TLR)-4 activate the macrophages and neutrophils, which activation is thought to cause local myocardial damage. Furthermore, cytokines such as tumor necrosis factor alpha (TNF-α) and interleukin-1 (IL-1) are thought to cause myocardial dysfunction [123]. DM2 patients are more susceptible to infections compared to the rest of the population. Septic DM2 patients will have worse prognosis, higher morbidity and mortality as well as a higher incidence of complications than the rest of the population. The cause for this is assumed to be DM which decreases immune cells function (neutrophils, lymphocyte T), consequently decreasing bacterial clearance [124]. Furthermore, elevated troponin concentrations in patients with sepsis and septic shock are a strong risk factor for short-term and long-term mortality [125]. Hospital mortality and long-term mortality in the patients with sepsis is around 10% and 60–80%, respectively. The most common cause of death in sepsis is refractory shock [126]. HbA1_c levels are a predictive factor for hospital mortality as well as for the length of hospitalization in septic patients [127]. Although troponin concentrations and HbA1_c levels are the prognostic factors for mortality in septic patients, it is hard to prove the causal relation between those two parameters because of the complexity of the whole mechanism that causes myocardial damage.

In DM2 patients’ metabolic abnormalities such as glucotoxicity, lipotoxicity, oxidative stress and IR cause specific inflammation that plays a role in pathogenesis of the disease [128–132]. Increased glucose intake and free fatty acids (FFA) cause stress in beta cells as well as in other insulin sensitive tissues (fat tissue, liver and skeletal muscle) which consequently start to release a number of cytokines and chemokines (IL-1B, IL-6, TNF). The immune cells that contribute further to inflammation are recruited consequently to the abovementioned mechanism. Furthermore, fat tissue cytokine and chemokine contribute to inflammation in other tissues as well as to IR [132, 133]. Nikiforov et al. found a positive correlation in the DM2 patients between HbA1_c and TNF-α levels which contribute to monocyte-phagocytic system activity which can indicate a correlation between HbA1_c levels inflammation intensity and monocyte-phagocytic system activity [134, 135]. Macrophages phagocytic activity does not change significantly in vitro in the high glucose environment [136, 137]. Elevated monocyte levels are shown to be a positive predictive factor of CV mortality and coronary artery plaque formation [138]. Proinflammatory cytokines and chemokines increase the monocyte-phagocytic system activity which theoretically should increase troponin clearance, but also increase the progression of atherosclerotic changes which affect the intensity of cardiac damage, consequently elevating troponin concentrations. Inflammation in DM2 is associated with lower adiponectin levels [139]. High adiponectin levels correlate positively and negatively, respectively, with high density lipoprotein (HDL) and low density lipoprotein (LDL) levels, as well as with lipid and inflammatory markers levels and with the incidence of CHD and CVD [140]. Cardioprotective effects of adiponectin are attributed to its cardiomyocytic anti-apoptotic properties, decreasing of cardiac ischemic injury and increasing revascularization [139]. Besides cardioprotective, adiponectin has an anti-inflammatory effect by increasing the level of anti-inflammatory cytokines expression such as IL-10 and IL-1RA [140]. Adiponectin also participates in atherogenesis suppression by inhibiting monocyte adhesion to plaque, decreasing the phagocytic activity, decreasing lipid accumulation on the vascular wall, and increasing nitric oxide bioavailability [141]. HbA1_c levels in the DM patients are negatively associated with adiponectin levels [142]. In the MI patients, the negative correlations between the adiponectin levels and cTnI as well as the infarct size have been observed [143].

Troponin and hemoglobin A1c in diabetic nephropathy

In ADVANCE study elevated HbA1_c levels were recognized as a risk factor for the development of diabetic nephropathy (DN) [144]. Prabhu et al. studied the correlation between HbA1_c and the estimated glomerular filtration rate (eGFR) in a cross-sectional case-control study on 100 CVD patients, 70 patients with DM and CVD and 100 healthy controls. In 100 DM patients as well as 70 patients with DM and CVD the correlation between the poor glycemic control (HbA1_c > 6.5%) and low eGFR (<60 ml/min/1.73m²) was found. Authors conclude that HbA1_c can be used as a marker for early diagnosis of reversible kidney damage because the pathologic changes are present before the laboratory ones such as creatinine and urea [16, 145]. Similar results have been obtained by Goedris et al. who found a correlation between the higher HbA1_c levels and more significant and faster eGFR decrease in DM2 patients in a retrospective 10-year cohort study [146]. Yokoyama et al. discovered that in the patients with DM2, preserved renal function and HbA1_c > 6%, eGFR decreased rapidly [147]. Kuo et al. studied the correlation between HbA1_c in patients with CKD stages 3–5 and several outcomes (initiation of hemodialysis, peritoneal dialysis and kidney transplantation), CVI and overall mortality. HbA1_c > 9% was shown to be the predictor of all outcomes in the stage 3 and 4 CKD patients while in patients with CKD there were no statistically significant correlation between the variables [148]. Oh et al. found that HbA1_c < 6.5% correlated with a lower incidence of end-stage renal disease (ESRD) development in DM patients with CKD stages 1–3 in their prospective cohort study [149]. Shurraw et al. studied the correlation between HbA1_c levels and overall mortality, progression of kidney disease, development of ESRD, CVI and overall hospitalization in the patients with eGFR<60 ml/min/1.73m². HbA1_c correlated with all outcomes except overall mortality which obtained a “U” shaped curve which means that patients with HbA1_c < 6.5% and > 8% had a higher mortality rate [150]. Phenomenon of excessive mortality in low HbA1_c levels can be explained by metabolic stress due to hypoglycemia which is more common in patients with a decrease in renal function, and it is caused by reduced renal gluconeogenesis [151, 152] as well as decreased renal insulin clearance [153]. In dialysis patients HbA1_c and mortality show a “U” shaped curve with the lowest mortality when HbA1_c is between 7% and 7.9% [154]. Michos et al. did a literature review and meta-analysis of 98 observational cohort studies and analyzed the prognostic value of cTn in CKD patients without suspected ACS. In dialysis patients overall and CV mortality correlated with cTnT and cTnI concentrations. Elevated cTnT and cTnI were positive predictors of overall mortality in patients who were not on dialysis as well as positive predictors of major adverse cardiovascular event incidence. The authors conclude that most of the studies did not take underlying CAD as a confounding factor, but ones that did obtained similar results as mentioned before [155]. The correlation between HbA1_c and troponin concentration manifests in their prognostic value for mortality. This correlation is not noticed in patients with CKD stage 5, dialysis patients as well as in patients with normal eGFR (>60 ml/min/1.73m²). In this correlation we should take into consideration that CKD is a risk factor for CV disease [156] and can cause troponin elevation by the mechanism of cardiac injury [37]. This claim is supported by the fact that there is a correlation between decrease in eGFR with increase in overall and CV mortality [157] as well as CV outcomes [158]. Since HbA1_c is a prognostic factor of CKD progression, poor glycemic control can cause troponin elevation by decreasing its elimination. To define the relation between HbA1_c and troponin concentrations in DN patients more clearly, in addition to serum troponin and HbA1_c levels, urine troponin concentrations or dialysate concentration in dialysis patients should be obtained. It should not present a problem because cTnI concentrations are found to be higher than in the blood of healthy individuals [159], and the presence of cTnI and cTnT is detected in dialysate of patients on hemodialysis [160]. Using this approach we can more precisely elucidate the role of kidney in troponin clearance and effect of HbA1_c levels on it.

Conclusion

HbA1_c has been shown to be a positive predictive factor for incidence and outcomes such as mortality and morbidity of conditions that cause troponin elevations including ACS, arrhythmias, stroke, PE and other (Table 1). However, unlike the conditions mentioned, prognostic value of HbA1_c in HF patients is still unclear. Since patients with HF are a heterogeneous group of patients with variable and frequent comorbidities which affect overall mortality, morbidity and life quality, it is hard to evaluate unambiguously the effect of HbA1_c on overall mortality and morbidity and consequently difficult to evaluate its relationship with troponin concentrations as the prognostic markers of HF. To elucidate the connection between these two parameters we should define in which part of subclinical and clinical part of HF course HbA1_c becomes the positive predictive factor of survival and comparatively measure the troponin levels as well as the dynamics of their relation. To define the target values of HbA1_c levels in DM patients with HF, research should divide patients in New York Heart Association (NYHA) groups (I-IV) and outcomes should be overall mortality, CV mortality and morbidity and progression of disease (higher NYHA stage). Furthermore, elevated HbA1_c levels correlate with decrease in GFR which lowers renal troponin clearance and consequently elevates its blood concentrations. Finally, although HbA1_c levels correlate with thrombin and monocytic-phagocyte system activity which participate in degradation and clearance of troponin, their effects are ambiguous because thrombin activity perpetuates thrombi formation while monocytic-phagocyte activity perpetuates atherosclerosis and, in both cases, increase incidence and severity of heart injury increasing troponin concentrations. To isolate the effect of glycemic control on cardiac injury from the effects of troponin degradation and elimination, we can use other markers of cardiac injury such as copeptin and heart fatty acid binding protein (HFABP) because those markers have different elimination mechanisms. Further studies should investigate the effect of glycemic control on above mentioned markers.

Table 1.

Relationship between hemoglobin A1c and serum troponin in patients with diabetes and cardiovascular events

Cardiovascular Event	Diabetic Population	Relationship with HbA1c	References
Acute coronary syndrome	-one third of patients with acute coronary syndrome suffer from diabetes - worse outcomes in diabetes	-positive correlation between troponin and HbA1c	[56–59, 62–65]
Heart failure	-prevalence 25% -increased risk of mortality and hospitalization	-prognostic value of HbA1c in heart failure is still unclear	[67–72, 77–79]
Cardiomyopathy	-diabetic cardiomyopathy in absence of hypertension and structural heart disease -fetal cardiomyopathy in pregnancies of mothers with type 1 diabetes	-positive correlation between troponin and HbA1c in fetal cardiomyopathy -unclear relationship in older patients	[80, 81, 83–87]
Stroke	- two-fold increased risk -greater severity of stroke -worse functional outcome - higher mortality	-positive correlation between troponin and HbA1c -cardiac injury and renal failure are main reasons for troponin elevation	[90, 98–102]
Arrhythmias	-comorbidities, particularly cardiac autonomic neuropathy, are positive predictors of incidence of arrhythmias	-positive correlation between troponin and HbA1c -troponin elevation is result of reversible and/or irreversible ischemic injury	[42, 64, 106, 107]
Pulmonary embolism	-increased risk for pulmonary embolism -hyperglycemia causes glycation of proteins that take part in coagulation and fibrinolysis perpetuating the hypercoagulable plasma state	-positive correlation between troponin and HbA1c levels is associated with outcomes such as right ventricular dilatation and dysfunction, amount of ventilation-perfusion defects and mortality	[111–114, 116, 117]

Considering all these facts, we can conclude that there is a causal relation between HbA1c and serum troponin in patients with diabetes and cardiovascular events. One of the most important mechanisms is the effect of hyperglycemia on glomerular filtration reduction and the consequent decrease in troponin elimination, with its additional effect on the heart microcirculation, leading to microvascular damage and consequently to ischemia which also contribute to troponin concentration elevation. Furthermore, the correlation between HbA1_c and troponin concentration manifests in their prognostic value for mortality.

Compliance with ethical standards

Conflict of interest

The authors declare that there is no conflict of interest regarding the publication of this article.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.