Serum calcification propensity is associated with HbA1c in type 2 diabetes mellitus

Objective

Despite intensive glycemic control and adequate management of cardiovascular risk factors, type 2 diabetes mellitus (T2DM) is accompanied by microvascular disease, including retinopathy, nephropathy and neuropathy, and macrovascular disease. Individuals with T2DM are prone to developing vascular calcifications, which considered playing a causal role in the etiology of diabetic complications.1

Previously, vascular calcifications were considered a result from passive precipitation of calcium and phosphate. Nowadays, the process of calcification is considered a consequence of a disequilibrium of a between calcification stimulating and inhibiting factors.2 Evidence exists that in persons with diabetes this equilibrium is unbalanced, leading to ectopic calcification in the media of the vessel wall, atherosclerotic plaque progression and subsequent cardiovascular events.1 3–6 The process of calcification is thought to be (at least partially) mediated by calciprotein particles (CPPs)7–9 that naturally circulate in the blood. Primary CPPs contain amorphous calcium phosphate, whereas secondary CPPs contain crystalline calcium phosphate.10–13 Secondary CPPs have the capability of inducing calcification of, for example, vascular smooth muscle cells,7 so the rate of primary-to-secondary CPP transitioning is viewed to be a measure of the serum anticalcification buffer capacity.

This increased formation and maturation and defective clearance of CPP may be an important novel cardiovascular risk factor (so-called mineral-stress hypothesis).14 Indeed, amorphous CPP1 exerted minor cellular responses in macrophage cell lines, while CPP2 appeared to induce oxidative stress and inflammation in macrophages,15 and oxidative stress, inflammation, and calcification in primary human aortic smooth muscle cell cultures.16 17 The T₅₀ serum calcification propensity test has been developed to allow for quantification of the serum anticalcification buffer capacity.18 This novel T₅₀ test measures in vitro how rapidly CPP2 are formed in a patient blood sample. In other words, the result of the T₅₀ test reflects the velocity of calcium phosphate crystallization in blood with lower T₅₀ values indicating increased calcification propensity. Results of the T₅₀ test have been determined to be an independent mortality predictor in both chronic kidney disease (CKD)19 and in renal transplantation patient populations.20 21

Because any serum test that can be used reliably to assess vascular calcification would be considered an asset in assessing cardiovascular risk in patients, we aimed to assess the association of the T₅₀ test with parameters of T2DM management in a large cohort of stable patients.

Research design and methods

Results

Patient baseline characteristics are shown in table 1, both overall and, for descriptive purposes, stratified by T₅₀ tertile with the middle tertile ranging from 235 to 293 min. In brief, the population had a median age of 65.8 years (62–75), 54.7% was male, mean diabetes duration was 6.5 (3.0–10.2) years, baseline HbA1c was 49 (44–54) mmol/mol and median (estimated) renal function in this study was 74 (61–86) mL/min/1.73 m². A total of 234 patients (25.1%) had a history of a macrovascular event. A total of 205 patients (30.8%) had a documented microvascular complication. Treatment of diabetes consisted strictly of dietary interventions for 19.5% of patients, whereas 67.5% took oral blood glucose lowering drugs and 13.0% used insulin therapy.

The distribution of serum T₅₀ is depicted in figure 1, with a mean value of 261±66 min. While age was not significantly different across tertiles (p=0.664), women had significantly lower T₅₀ values (p<0.001). Similarly, a lower T₅₀ value was associated with the status of current smoker, while a higher T₅₀ value was associated with using alcohol. Furthermore, the tertile with the lowest T₅₀ values had the highest Hb1Ac (p<0.001) and more often a history of retinopathy (p=0.016).

Request permission

Determinants for serum calcification propensity were identified using linear regression (table 2). Gender was found to be the strongest determinant of T₅₀, with females having a mean T₅₀ of 249±65 min, compared with 271±64 min for males (figure 2A, standardized beta −0.171, adjusted R² 0.028, p<0.001). Smoking was found to lower T₅₀ (figure 2C), as did a longer disease duration for diabetes. Interestingly, HbA1c was negatively associated with T₅₀ as well (figure 2E, standardized (st) beta −0.141, adjusted R² 0.019, p<0.001), whereas low-density lipoprotein (LDL) cholesterol exhibited a significant, positive association (figure 2F). The use of insulin therapy was associated with a lower T₅₀ (247±69 min for insulin users vs 263±64 min for non-insulin dependent patients; st. beta −0.083, adjusted R² 0.006, p=0.011; figure 2B). We could not detect an association between the T₅₀ and either documented macrovascular or microvascular disease as composite variables. Assessing all conditions separately, only the presence of diabetic retinopathy was found to be significantly associated with T₅₀.

Request permission

Backward regression was used with all variables to investigate which variables would be significantly associated with T₅₀ after correction for covariates. The final model is detailed in table 2. Sex, smoking status, systolic blood pressure, HbA1c, LDL cholesterol, total cholesterol, cholesterol/high-density lipoprotein ratio, and triglycerides all remained significant with the final model explaining 9.4% of the variance in T₅₀ (p<0.001).

Conclusions

In the present, study we explored the association between calcification propensity (measured using the T₅₀ test) and indices of T2DM management, in particular HbA1c, in a large population of outpatients. Interestingly, we found that several factors reflecting an unfavorably disease status, namely a higher HbA1c, insulin dependence, and a longer history of diabetes, were all associated with lower T₅₀ values. In multivariable analysis, only HbA1c remained significantly associated with T₅₀.

This may indicate that long-term glycemic control is a more important determinant for T₅₀ than the use of insulin and disease duration. Furthermore, HbA1c remained significantly associated with T₅₀ after adjusting for gender, smoking, LDL cholesterol, and systolic blood pressure, identifying HbA1c as an independent determinant. Importantly, poor glycemic control is thought to contribute to the development of vascular calcification,3 which is a strong predictor of cardiovascular complications and mortality.2 Therefore, the finding that functional serum calcification buffer capacity correlates to markers for glycemic control could be an important step in linking the pathophysiology of diabetes to the pathophysiology of vascular calcification. Given this relationship between better glycemic control and a lower T₅₀ value, the T₅₀ could prove to be relevant in the follow-up and cardiovascular risk management of patients with T2DM. It is an open question whether, for instance, glycation of factors involved in vascular calcification leads directly to an increased calcification propensity or whether other mechanisms are involved.

Female sex and lower LDL cholesterol levels were associated with lower T₅₀ values, indicative of an increased serum calcification propensity. Similarly, the risks of higher LDL cholesterol levels and low calcification propensity may add up to ultimately precipitate a macrovascular event. Serum T₅₀ has previously been investigated in renal disease cohorts, including predialysis CKD patients,19 24 hemodialysis patients,25 26 and renal transplantation patients.20 21 Smith et al19 found that women had a lower T₅₀ in CKD stages 3 and 4 like in our study, whereas Pasch et al25 found that women actually had a higher T₅₀ than men (with a difference of about 10 min), in a hemodialysis population. It is possible that the differences between these studies can be explained by the outsize cardiovascular risk, possibly partially CPP mediated, that is conferred by end-stage renal disease, which outweighs the effects of traditional cardiovascular risk factors, which may be more relevant in our non-CKD population and in a prehemodialysis population. On the other hand, Keyzer et al20 also did not detect an association between T₅₀ and gender in a renal transplantation cohort, most of whom had undergone dialysis for over 5 years but who had higher T₅₀ values after transplantation than our cohort of patients with diabetes. More studies will be required to elucidate whether associations between gender and calcification propensity, and between serum lipid levels and calcification propensity are disease specific.

Interestingly, serum calcification propensity was not associated with estimated glomerular filtration rate in the present T2DM population but was associated with the albumin/creatinine ratio. This indicates a window between the occurrence of incipient renal damage and the impairment of renal function in which serum T₅₀ will start to decrease. Given the pathophysiology of CPPs as a mechanism to buffer calcium and phosphate overload and the relevance of the kidney in regulating phosphate excretion through fibroblast growth factor 23 and Klotho, it is possible that the early disturbances in these factors that occur already during subtle renal injury can rapidly result in an increased serum calcification propensity.

We did not detect an association between serum T₅₀ and a history of macrovascular disease or the presence of microvascular disease. The only exception was the presence of diabetic retinopathy, but it had such a low prevalence in our cohort that independent validation in larger or more tailored cohorts will be required to provide reliable answers. It should be noted that macrovascular disease was not assessed prospectively and that a T₅₀ measurement at baseline may be a better predictor for future events, rather than reflect past events.

In summary, in this cohort of patients with T2DM, we found that serum calcification propensity was negatively and independently associated with HbA1c, which suggests that better glycemic control may correlate with a less pronounced development of vascular calcification, which ultimately might lead to better cardiovascular outcomes. We were, however, unable to establish an association with the past occurrence of macrovascular disease or with the presence of microvascular disease (with the exception of retinopathy in our univariable linear regression analysis). Prospective studies will be required to elucidate the role of serum calcification propensity and the T₅₀ measurement in the vascular burden in diabetes and cardiovascular outcomes.