Mitral annular calcification and the serum osteocalcin level in patients with chronic kidney disease

Abstract

Objective: To determine the relationships between inflammatory mediators, mitral annular calcification (MAC), and osteocalcin in patients with chronic kidney disease (CKD). Materials and methods: Echocardiographic data for 60 patients diagnosed as CKD were retrospectively evaluated. The patients were divided into 2 groups; patients with MAC (MAC+ group) and patients without MAC (MAC− group). The relationships between biochemical markers—including osteocalcin—and MAC were evaluated. Results: The study included 19 female and 41 male patients. In all, 29 patients were MAC+ and 31 were MAC−. High-sensitive C-reactive protein (hsCRP) and osteocalcin levels were significantly higher in the MAC+ group (p < 0.05). The eGFR was lower, serum calcitonin (we could not obtain calcitonin data for 15 patients), Ca, PO4, CaxPO4, the erythrocyte sedimentation rate, red cell distribution width, the neutrophil/Lymphocyte rate, and PTH were higher in the MAC+ group; however, the differences between the groups were not significant (p > 0.05). The mitral E/A ratio, mitral peak Ea velocity, tricuspid E/A ratio, hsCRP, and the osteocalcin level were strongly correlated with MAC. Multivariate logistic regression analysis showed that only the osteocalcin level and mitral E/A ratio were independent variables, each with an independent effect on MAC. Conclusion: CKD patients in the MAC+ group had higher osteocalcin levels than those in the MAC− group, and left ventricular diastolic dysfunction was more common in the MAC+ group.

• Chronic kidney disease echocardiography
• mitral annular calcification
• osteocalcin

Introduction

The life expectancy of patients with chronic kidney disease (CKD), particularly those with end-stage renal disease (ESRD), is decreased. CKD alone is considered an independent risk factor for the development of heart disease, especially coronary heart disease.1 A low estimated glomerular filtration rate (eGFR) is a powerful graded, independent predictor of cardiovascular morbidity and mortality.1 Vascular calcification plays a pivotal role in cardiovascular mortality associated with CKD.2,3 Cardiac valve calcification is a frequent complication in CKD patients and is considered a risk factor for cardiovascular mortality; however, little is known about the pathophysiological mechanisms of cardiac valve calcification and the factors associated with it. Compared to other patient groups, valve calcification may be 5–10-fold common in patients with ESRD.4 Additionally, cardiac valve calcification is associated with other vascular pathologies, including atherosclerosis and vascular calcification. Furthermore, mitral valve calcification has been associated with atrial fibrillation, stroke, and an increase in cardiovascular morbidity and mortality, both in the general population and in patients with kidney disease.5,6

Osteocalcin – also referred to as bone-Gla protein – is a small, non-collagen peptide secreted by osteoblasts.7 In addition to its role in bone metabolism, osteocalcin has been described as a circulating hormone involved in the regulation of glucose and fat metabolism.8 Recent experimental data indicate that osteocalcin – at a systemic level – differentially organizes pancreatic β-cell proliferation and adipocyte gene expression, and increases insulin secretion and adiponectin synthesis.9 Although a negative correlation between the eGFR and the plasma osteocalcin level has been reported in patients with CKD,10 there are a limited number of studies on the relationship between cardiac valve calcification and bone metabolic markers in CKD patients. The relationship between markers of bone turnover, and cardiac valve calcification and inflammatory markers remains unclear. Accordingly, the present study aimed to determine the relationships between inflammatory mediators, the osteocalcin level, and mitral annular calcification (MAC) in CKD patients.

Materials and methods

Study population

Echocardiographic data for 60 patients diagnosed as CKD were retrospectively evaluated. CKD was defined as kidney damage or a glomerular filtration rate (GFR) <60 mL·min⁻¹·per 1.73 m⁻² for ≥3 months, irrespective of the cause, according to Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines. Exclusion criteria included advanced age (>72 years), malignancy, active glomerulonephritis, immunosuppressive therapy, acute renal failure, recent renal transplant (<3 months), cardiovascular surgery (valve replacement and/or coronary artery by-pass graft surgery), and poor echocardiographic visualization. To facilitate analysis of the relationship between bone metabolism and cardiac valve calcification the CKD patients were divided into the following 2 groups: CKD patients with MAC (MAC+ group) and CKD patients without MAC (MAC− group).

The study was carried out in accordance with the Declaration of Helsinki and the Good Clinical Practice/International Conference on Harmonization, and the Ethics Committee approved the study protocol.

Echocardiography

Transthoracic echocardiography was performed using a Philips IE 33 6.0 ultrasound system (Philips Medical Systems, Andover, MA) with a 2.5-MHz transducer. The presence of dense echoes behind the posterior mitral leaflet that traveled parallel to the left ventricular posterior wall was defined as MAC. Standard 2-dimensional (2D) measurements, pulsed-wave Doppler evaluation of tricuspid and mitral flow at the level of the tip of the opened leaflets, and pulsed-wave tissue Doppler imaging with a 5-mm sample volume placed at the lateral margin of the mitral and tricuspid annuli in apical 4-chamber view at the end of expiration were obtained at rest, according to standard criteria. All data were digitally stored and analyzed offline. The mean of three consecutive heart beats was used for analysis. All analyses were performed by two investigators that were blinded to each other’s findings and patient clinical data.

Biochemical analysis

Biochemical marker data were obtained from the patients’ medical records via the hospital’s computerized database and the following variables were retrospectively analyzed: serum total cholesterol, LDL-cholesterol, uric acid, and creatinine, whole blood count, albumin, glucose, intact parathyroid hormone (iPTH), total calcium (Ca) and phosphate (Pi), magnesium, vitamin D3, high-sensitive C-reactive protein (hsCRP), calcitonin, and osteocalcin. The eGFR was measured using the 6 variable modifications of diet in renal disease (MDRD) equation. The relationships between these markers and mitral valve calcification were evaluated.

Statistical analysis

The data were tested for normality of distribution using the Kolmogorov-Smirnov test. Continuous variables are presented as mean ± SD and categorical variables as percentage. The Chi-square test was used to compare categorical data. The independent-samples t test and Mann–Whitney U test were used to compare continuous variables with normal distribution and non-normal distribution, respectively. Spearman’s and Pearson’s correlation coefficients were used for univariate correlation analysis. Following univariate correlation analysis, the effects of various variables on MAC were evaluated using backward stepwise multivariate logistic regression analysis. Statistical analysis was performed using SPSS v.20.0 for Windows (SPSS, Inc., Chicago, IL). The level of statistical significance was set at p < 0.05.

Results

In total, there were 60 patients (41 male 68.33%) with a mean age of 40.71 ± 15.80 years. The MAC+ group included 29 patients with MAC and the MAC− group included 31 patients without MAC. Baseline demographic and clinical characteristics were similar in both groups (Table 1). Transthoracic echocardiography was performed in all the patients at admission. Echocardiographic findings are shown in Table 2. Doppler imaging showed that only the mitral E/A ratio (p = 0.039) and mitral peak Ea velocity (p = 0.028) were significantly lower in the MAC+ group. Although diastolic left ventricular velocity was observed to be dependent on the presence of MAC, diastolic right ventricular velocity was not. In addition, there were differences in other diastolic left and right ventricular velocities between the 2 groups, but the differences were not significant, probably due to the small number of patients included in the study (Table 2).

Table 1. Demographic data of the patients.

	MAC (−) (n = 31)	MAC (+) (n = 29)	p
Age, (years)	42.87 ± 16.41	41.93 ± 15.83	0.739
BMI, (kg/m^2)	26.68 ± 2.63	26.36 ± 2.56	0.645
Male, n (%)	21 (67.7)	20 (69)	0.570
Smoke, n (%)	8 (25.)	11 (37.9)	0.232
Cardiovascular disease, n (%)	3 (9.7)	1 (3.4)	0.332
Diabetes mellitus, n (%)	8 (25.8)	5 (17.2)	0.313
Hypertension, n (%)	12 (38.7)	16 (55.2)	0.154
Medication usage
ACEI, ARB n (%)	23 (74.2)	16 (55.2)	0.101
Beta-blocker, n (%)	7 (22.6)	11 (37.9)	0.155

Note: MAC; Mitral annular calcification.

Table 2. Echocardiographic findings.

	MAC (−) (n = 31)	MAC (+) (n = 29)	p
IVSd, (mm)	10.35 ± 2.33	10.37 ± 1.54	0.962
LVIDd, (mm)	47.93 ± 4.21	49.10 ± 4.69	0.481
LVPWd, (mm)	10.48 ± 1.65	10.51 ± 1.50	0.493
LVEF, (%)	65.29 ± 4.56	64.86 ± 5.49	0.743
LA, (mm)	34.75 ± 5.18	35.02 ± 4.36	0.833
Mitral peak E velocity, (cm/sec)	73.24 ± 18.72	67.52 ± 15.25	0.201
Mitral peak A velocity, (cm/sec)	68.39 ± 18.00	72.72 ± 18.83	0.098
Mitral E/A ratio	1.15 ± 0.43	0.93 ± 0.28	0.039*
Mitral peak Ea velocity (cm/sec)	12.11 ± 3.53	10.07 ± 3.46	0.028*
Mitral E/Ea ratio	6.34 ± 1.71	7.27 ± 2.47	0.201
Tricuspid peak E velocity, (cm/sec)	57.09 ± 11.68	53.38 ± 10.87	0.209
Tricuspid peak A velocity, (cm/sec)	45.40 ± 10.43	52.32 ± 18.46	0.183
Tricuspid E/A ratio	1.30 ± 0.33	1.11 ± 0.35	0.068
Tricuspid peak Ea velocity (cm/sec)	11.81 ± 3.30	13.23 ± 4.11	0.145
Tricuspid E/Ea ratio	5.15 ± 1.60	4.46 ± 1.73	0.114
sTA (cm/sec)	15.05 ± 1.93	15.29 ± 2.62	0.685

Notes: MAC, Mitral annular calcification; IVSd, Interventricular septal thickness at diastole; LVIDd, Left ventricle end-diastolic internal diameter; LVPWd, Left ventricle posterior wall thickness at diastole; LVEF, Left ventricular ejection fraction; LA, Left atrium; sTA, peak systolic tricuspid annular velocity (*p < 0.05).

In the MAC+ group the serum osteocalcin level was significantly lower (26.80 ± 20.92 ng mL⁻¹ vs. 9.47 ± 6.64 ng mL⁻¹, p = 0.001) and the hsCRP level was significantly higher (19.32 ± 25.15 mg L⁻¹ vs. 6.71 ± 9.25 mg L⁻¹, p = 0.022) than in the MAC− group. The mean eGFR was 43.45 ± 21.14 mL·min⁻¹·1.73 m⁻² for the entire study population; although the eGFR in the MAC+ group was lower than in the MAC− group, the difference was not significant. Serum calcitonin Ca, PO4, CaxPO4, erythrocyte sedimentation rate, red cell distribution width (RDW), the neutrophil/Lymphocyte rate (NLR), PTH was higher in patients with MAC. However none of them were statistically significant (p > 0.05). The other hematological and biochemical parameters were similar in both groups (Table 3).

Table 3. Echocardiographic findings.

	MAC (−) (n = 31)	MAC (+) (n = 29)	p
Glucose	108.19 ± 50.00	94.31 ± 15.56	0.818
eGFR, (ml/dk)	50.33 ± 25.24	40.39 ± 19.83	0.097
Uric acid	6.81 ± 1.33	7.11 ± 1.53	0.418
GGT	35.70 ± 26.24	33.70 ± 13.70	0.651
LDL-C	167.83 ± 181.86	128.72 ± 37.82	0.589
Hemoglobin	13.06 ± 1.90	12.77 ± 2.57	0.621
Platelet	259.09 ± 70.68	237.65 ± 79.78	0.274
RDW	15.63 ± 1.74	15.81 ± 1.97	0.750
MCV	85.80 ± 6.63	88.09 ± 6.36	0.179
Erythrocyte sedimentation rate, (mm)	25.90 ± 25.67	29.28 ± 30.41	0.574
NLR	2.36 ± 1.28	2.57 ± 1.17	0.280
hsCRP	6.71 ± 9.25	19.32 ± 25.15	0.022*
Proteinuria, (mg)	2385.94 ± 3595.66	1377.57 ± 1335.79	0.882
Calcium	9.48 ± 0.91	9.72 ± 0.84	0.297
CaXP	38.05 ± 11.76	41.60 ± 10.27	0.093
Magnesium	2.10 ± 0.54	2.04 ± 0.33	0.762
Phosphate (PO4)	4.22 ± 1.31	4.45 ± 1.34	0.429
Parathyroid hormone	129.93 ± 151.24	134.57 ± 90.31	0.094
Vitamin D3	29.90 ± 31.15	30.31 ± 19.11	0.237
Osteocalsin	9.47 ± 6.64	26.80 ± 20.92	0.001*
Calcitonin	6.05 ± 7.51	8.90 ± 15.24	0.510

Notes: *p < 0.05; MAC, Mitral annular calcification; eGFR, estimated glommerular filtration rate; GGT, gamma-glutamyl transferase; LDL-C, low density lipoprotein cholesterol; RDW, Red cell distribution width; MCV, mean corpuscular volume; NLR, Neutrophil/lymphocyte rate; hs CRP, high-sensitivity C-reactive protein.

Univariate correlation analysis showed that the mitral E/A ratio, mitral peak Ea velocity, and tricuspid E/A ratio, and hsCRP and osteocalcin levels were significantly correlated with MAC (p < 0.05 for all) (Table 4). Multivariate logistic regression analysis using the backward stepwise method showed that only the serum osteocalcin level (OR = 6.205, 95% CI 1.034–1.320, p = 0.013) and mitral E/A ratio (OR = 4.594, 95% CI 0.005–0.789, p = 0.032) were independent variables with a significant effect on MAC (Table 5).

Table 4. Parameters correlated with mitral annular calcification.

Parameters	r	p
Mitral E/A ratio	−0.282	0.029*
Mitral peak Ea velocity	−0.285	0.029*
Tricuspid E/A ratio	−0.268	0.038*
hsCRP	0.322	0.013*
Osteocalcin	0.499	<0.001*

Notes: Mitral E/A ratio, mitral ratio of peak early to late diastolic filling; Tricuspid E/A ratio, tricuspid ratio of peak early to late diastolic filling; hsCRP, high sensitivity C-reactive protein (*p < 0.05).

Table 5. The effect of numerous variables on mitral annular calcification, based on multivariate logistic regression analysis.

Covariates	Adjusted OR	95% CI	p
Mitral E/A ratio	4.594	0.005–0.789	0.032*
hsCRP	3.434	0.997–1.117	0.064
Osteocalcin	6.205	1.034–1.320	0.013*
Calcium	3.233	0.861–32.666	0.072

Notes: Mitral E/A ratio, mitral ratio of peak early to late diastolic filling; hsCRP, high sensitivity C-reactive protein (*p < 0.05).

Discussion

MAC is a well-known pathologic finding that occurs mainly in elder subjects11,12 and is closely associated with such cardiovascular disorders as coronary artery disease, atherosclerosis, heart failure, and stroke.13 The clinical risk factors for cardiovascular disease are the same as those for MAC.14 Sayarlioglu et al. reported that patients with MAC were older than those without MAC and were more likely to have diabetes mellitus. In the present study MAC values of the patients were significantly higher but systolic blood pressure values were in normal range. In addition, the left ventricle mass index and pulmonary artery pressure were higher in patients with MAC, which might be indicative of volume overload.15 Nonetheless, vascular calcification and atherosclerosis are observed primarily in patients with CKD, and are associated with increased cardiovascular morbidity and mortality. The most common mechanisms responsible for calcification include inflammation and oxidative stress. Varol et al.16 reported that NLR and RDW were significantly higher in patients with MAC than in controls, and that NLR was correlated with MAC. In accordance with these earlier findings, hsCRP in the present study was higher in the MAC+ group, indicating that there was an association between calcification and inflammation in the MAC+ group; however, there wasn’t any significant differences in the other inflammatory parameters, such as RDW, serum uric acid, the erythrocyte sedimentation rate, and NLR, between the CKD patients with and without MAC.

Abnormal bone and mineral metabolism might play a pivotal role in calcification of the mitral annulus. Within the normal range, higher serum phosphate concentrations are associated with adverse cardiovascular events and mortality in patients with CKD. In a study by Adeney et al.,17 which included 439 patients with CKD, they examined the association between the serum phosphate concentration, and the presence and extent of vascular and valvular calcification at 4 anatomic sites: coronary arteries, the descending thoracic aorta, the mitral valve, and the aortic valve. They also sought to determine if serum PTH and activated vitamin D levels would partially explain the potential association between the serum phosphate concentration and calcification. The researchers concluded that although within the normal range, higher serum phosphate levels were associated with a higher prevalence of vascular and valvular calcification in patients with moderate CKD. Similarly, in the present study higher levels of Ca, PO4, CaxPO4, and PTH were observed in the CKD patients with MAC. In addition to other well-known components of mineral metabolism, an elevated calcitonin level is associated with more severe chronic artery disease; however, this relationship is not independent of traditional and non-traditional cardiovascular risk factors.18 In the present study, higher calcitonin levels were associated with MAC; however, the association was not significant, perhaps due to the small number of patients.

The role of osteocalcin in the era of cardiovascular disease is fairly new and remains a contentious issue. Janda et al.19 reported that calcification of the radial artery was significantly associated with a higher prevalence of impaired fasting glucose, diabetes, and older age, as well as higher osteoprotegerin and ADMA concentrations; however, a relationship between osteocalcin and calcification was not observed. Krzanowski et al.20 examined aortic pulse-wave velocity (AoPWV) in patients receiving peritoneal dialysis and concluded that among the known confounders of AoPWV, such as age, hypertension, and mean arterial pressure, the only significantly correlated variable was osteocalcin (r = −0.37, p = 0.005), and the adjusted OR for higher AoPWV was 0.96 (range: 0.92–0.99) per unit increase in the osteocalcin value. Yamashita et al.21 evaluated the impact of serum osteocalcin on the emergence of new cardiovascular events in hemodialyzed patients and reported that the number of cumulative cardiovascular events in the low-serum osteocalcin group was significantly higher than in the high-serum osteocalcin group. The multivariate analysis model demonstrated that a low osteocalcin level is a significant predictor of a higher incidence of cardiovascular disease.21 Kanazawa et al.22 have shown that osteocalcin is negatively correlated with fasting plasma glucose and hemoglobin A_1c, both in men and postmenopausal women (p < 0.05), and with intima media thickness in men (p < 0.05). After additional adjustments for systolic blood pressure, low-density lipoprotein-cholesterol, high-density lipoprotein-cholesterol, hemoglobin-A_1c, and the Brinkman index, there remained a significant negative correlation between osteocalcin, and brachial-ankle pulse-wave velocity and intima-media thickness in men.22

The role of osteocalcin in valvular and/or non-valvular calcification is not fully known. Avila-Diaz et al.23 reported that the serum osteocalcin level was lower in peritoneal dialysis patients without mitral and/or aortic calcification, as compared to those with baseline aortic and/or mitral calcification. Interestingly, Avila-Diaz et al.23 divided the patients into two categories (slow and rapid valvular calcification in any valve) to identify the factors that might be associated with the rate of progression of valve calcification. They observed that patients with rapid development of valvular calcification were older, diabetic, and had lower albumin and GFR levels than those with slow development of valvular calcification. Although patients with rapid development of valvular calcification had a lower level of serum osteocalcin, the difference was not significant.23

The association between a low eGFR and a high-serum osteocalcin level was reported in many studies, especially in hemodialysis patients.24–26 Moreover, vascular and valvular calcification are common in hemodialysis patients,2,4–6 which leads one to consider if the concomitant association of an elevated serum osteocalcin level and the ubiquity of calcification in CKD patients are a coincidence, or if there other factors involved.24 The present study has some limitations; primarily, its retrospective design and small patient group, as well as the lack of a comparison group. Nevertheless, we think the findings are of value. Among the present study’s patients with CKD, those with MAC had a higher osteocalcin level than those without MAC. The present findings, together with those of earlier studies, indicate that osteocalcin-mediated mechanisms alter the integrity of the mitral valve. As such, the serum osteocalcin level might be a useful marker for predicting the emergence of new cardiovascular events in CKD patients.

Declaration of interest

Financial disclosure statements have been provided, and the authors report that there are no conflicts of interest – financial or otherwise – related to the materials presented herein.