Журнал «Боль. Суставы. Позвоночник» 2 (06) 2012

Cardiovascular diseases (CVD) is an important problem of the national health service as well as worldwide. Consequently, many countries of the world appear to have vitamin D deficiency and insufficiency to be common [11], and data of the recent meta­analyses suggest vitamin D to be a new risk factor of CVD.

More than 20 cross­sectional studies [17] have exa­mined the association between plasma 25­hydroxyvitamin D (25(OH)D) and either blood pressure or prevalent hypertension. The greatest of these studies — Third National Health and Nutrition Examination Survey (NHANES­III) showed that systolic blood pressure and pulse pressure were inversely and significantly correlated with 25(OH)D levels among 12,644 participants [15]. These results were confirmed by subgroup analy­ses, in which the age­associated increases in systo­lic blood pressure were significantly lower in individuals with vitamin D sufficiency [6, 10]. The great majority of these studies demonstrate that lower circulating 25(OH)D levels are associated with higher blood pressures or a higher prevalence of hypertension. Notably, blood pressure evaluation in these studies was almost always based on a single measurement (usually taken in the physician’s office). But, in other investigations, a significant association between 25(OH)D and blood pressure has not been reported [7, 9, 14, 15]. In a 5­year follow­up of 1,739 women and men in the Framingham Offspring Study, those with low serum 25(OH)D (< 37.5 nmol/L) were 62 % more likely to develop CVD than other participants [18]. Among > 18,000 US male health professionals followed for 10 years, low 25(OH)D (≤ 37.5 nmol/L), as compared with high 25(OH)D (≥ 75), predicted a doubling in coronary heart disease incidence [2]. Laboratory studies also suggest that vitamin D confers vascular protection [8, 1] through the vascular smooth muscle cells, endothelial cells, cardiomyocytes, which have 1a­hydroxylase which converts 25(OH)D to 1.25­dihydroxyvitamin D (1.25(OH)2D), the natural ligand of the vitamin D receptor [12, 19].

The aim of the present study was to assess the prevalence of deficiency/insufficiency the level of 25(OH)D in blood plasma and establish its relationships with the indices of blood pressure and endothelial function of vessels in those suffering from arterial hypertension (AH) and ischemic heart disease (IHD).

Materials and methods

From September to March 2010 and in May 2011 we performed a clinical and diagnostic examination of 135 residents of the Western Region of Belarus with AH (39.3 %)and IHD (60.7 %), 34 males and 101 females aged 59.65 ± 8.16 years and 28 practically healthy individuals (group I). In May 2011 we performed sampling of patients in a cross­sectional study in cooperation with a group of investigators from the Department of Clinical Physiology and Musculoskeletal System Pathology of the State Institution «Institute of Gerontology named after D.F. Chebotaryov of National Academy of Medical Sciences of Ukraine». Of the total sample we examined 57 individuals in 2010 (group II) and 78 in 2011 (group III). The inclusion criteria were the following: previously confirmed IHD (G. Rose questionnaire, ECG changes, history of myocardial infarction, bicycle ergometry test, findings of coronary angiography) and/or AH. Patients with symptomatic AH, those with diabetes mellitus, acute infectious diseases, liver and kidney dysfunction, higher than stage I congestive heart failure, diseases requiring administration of hormonal agents were not included.

Some anthropometric data were obtained: height, body mass, body mass index (BMI) was calculated by the generally recognized formula (body mass in kilograms/height in meters square) as well as office (single measurement in the physician’s office) systolic (SBP) and diastolic blood pressure (DBP) and heart rate (HR).

Blood plasma calcium (Ca) and phosphorus (P) le­vels were estimated by the unified colorimetric method with the help of the spectrophotometer «Salar».

Lipid exchange indices were studied by electrophoretic method with the help of the six­channel spectrophotometer «Salar» (Belarus). We assessed total cholesterol (TCh), high density lipoproteins (HDL), low density lipoproteins (LDL) and triglycerides (TG).

On the base of the Central  Research Laboratory of the Grodno State Medical University we determined the content of parathyroid hormone (PTH) in blood serum using the DRG reagent (USA) with the help of the immunoenzymatic analyzer «Sunrise» (Tecan, Austria). In 2010 25(OH)D was assessed by means of the DRG reagent with the help of the above mentioned apparatus. In 2011 25(OH)D in blood serum was evaluated by immunoenzymatic method using the immunoenzymatic analyzer «Eleksys 2010» (Roche Diagnostics, Germany) with the help of Cobas systems.

With the help of the hardware­software complex «IMPECARD­М» (Belarus) we evaluated the forearm vascular endothelial function using the reactive hyperemia test and pulse ware velocity (PVV, m/s). By means of impedance rheography we measured baseline rate of blood flow in both forearms at rest (after 10–15 min in prone position) and within 5 min after cuff decompression at pressure exceeding the baseline by 50 mmHg. The reactive hyperemia was considered as endothelium­dependent vasodilation (EDV) — an index of relative change from the baseline in maximum rate of blood flow (Ddz/dt) within the first minute after the end of occlusion. EDV was considered to be preserved if ­Ddz/dt exceeded 12 %. Ddz/dt range from 12 to –2 % one minute after cuff removal evidenced stage I endothelial dysfunction (ED), that of from –2 to –15 % corresponded to stage II endothelial dysfunction, Ddz/dt < –15 % evidenced stage III ED. Arterial wall elasticity was considered to be normal in PVV < 12 m/s.

 Statistical analysis of the research data was done using software package STATISTICA 7.0. Data presentation corresponded to the character of their distribution: in normal distribution (by Shapiro — Wilk test) the data were expressed as mean values ± standard deviation (M ±± SD), in non­normal distribution — as median (Me) and interquartile range [LQ­UQ]. In normal distribution for testing the hypothesis about the equality of two group means we used Student t­test (t). In non­normal distribution comparison of two independent groups of the studied variable was done with the help of Mann — Whitney U test. To assess the correlations between the variables we used the Pearson correlation analysis test (r) and nonparametric Spearman’s correlation analysis (R). To assess the influence of several factors as well as their combination on the studied value multivariate analysis of variance ANOVA was used. Diagnostic significance was assessed by means of ROC­analysis. The null hypothesis was rejected at p 0.05 for each of the tests employed.

Results

General characteristic of the examined groups is shown in Table 1. The group of practically healthy individuals differed from both groups of patients by most of the mentioned indices while the groups of patients differed only by gender (in group II there were 35 % males versus 18 % (p = 0.026) in group III) and by SBP as there were more persons with AH in group III– 46 % versus 30 % in group II (p > 0.05). Variance analysis (ANOVA) revealed statistically significant differences (p < 0.05) between the groups of males and females (40 and 123, respectively) by body mass, height, HR according to Kruskal — Wallis criteria and Median test.

As demonstrated in Table 2, the main groups which were formed using the same inclusion and exclusion criteria but in different years had statistically significant differences only in 25(OH)D plasma levels. EDV remained normal in 29 % of the examined patients. Signs of stage I ED were revealed in 22 % cases, those of stage II ED — in 26 % cases, those of stage III ED — in 23 % cases. Arterial wall elasticity remained normal in all exa­mined patients with AH, in the IHD group PVV > 12 m/s was diagnosed in 11 individuals (13 %).

Incongruity of the results is due to the fact that 25(OH)D level in 2010 and 2011 was determined in different (certified) laboratories by non­identical methods. Nevertheless mathematical methods allow unite heterogeneous groups into one sample by normalization (standardization) of the baseline data. As a result of normalization the index loses its dependence on the factors by which it was standardized, in the present case — on the year and determining methods [13]. Normally distributed random variables Х with the help of the linear transformation z = (x – М)/s (where М and s² — mean value and dispersion of the studied attribute, respectively) have the same distribution called standardized distribution. For normal standardized distribution Z_mean = 0, and s= 1.

Having standardized by this method the data of the primary groups we obtained homogeneous samples. Homogeneous samples can be assembled into one and in such way we can obtain more complete information about it and consequently make more reliable conclusions [3].

The cut­points are currently the most commonly used classification of vitamin D status defined as vitamin D deficiency with 25(OH)D < 50 nmol/l, insufficiency — 51–75 nmol/l, and sufficiency > 75 nmol/l [4, 5]. The results of the analysis of vitamin D availability obtained after combining of the standardized indices of 25(OH)D serum level in the main groups and the control group (n = 156) evidence that more than 95 % of the examined individuals have vitamin D insufficiency (Fi­gure 1). There were no significant differences in frequency of 25(OH)D deficiency/insufficiency occurrence between the groups and depending on the diagnosis.

We determined significant (p < 0.05) weak inverse correlation relationships between serum level of vitamin 25(OH)D standardized and body mass (r = –0.21), 25(OH)D standardized and BMI (r = –0.19), 25(OH)D standardized and HR (r = –0.19). The inverse correlation relationship between serum level of vitamin 25(OH)D standardized and DBP (R = –0.17; p < 0.05). There were no significant correlation relationships between 25(OH)D level and other studied indices.

Assessment of the importance of 25(OH)D serum level for prognosis (diagnosis) of SBP elevation was performed by means of the ROC analysis (Receiver Operator Characteristic analysis). The space below the ROC curve built according to the 25(OH)D level and SBP values (Figure 2) was equal to 0.611 (95% confidential interval 0.512–0.711; p < 0.05). The optimal cut point which allowed us in 90 % cases to make a true positive conclusion (sensitivity) that the patients’ SBP value exceeded 140 mmHg corresponded to the 25(OH)D serum level of 17 nmol/l (and lower). When 25(OH)D plasma level was 72 nmol/l (and higher) the rate of truly negative cases (1­specificity) also amounted 90 %, that is we could reliably exclude the increase of SBP in our patients.

Using multivariate analysis of variance ANOVA (Figure 3) we determined that BMI and DBP values jointly make significant contribution to the variance of the standardized 25(OH)D serum level (F = 4.94; p < 0.03): if DBP level is elevated (higher than 90 mmHg) the 25(OH)D serum le­vel does not depend on the BMI, but if DBP level does not exceed the upper limit of normal (lower than 90 mmHg) with normal BMI the 25(OH)D serum level is considerably higher than in those with BMI > 29.0 kg/m².

Patients suffering from AH (53 individuals) differed significantly from those with IHD (82 individuals) by such parameters as age (53.77 ± 6.91 and 63.45 ± 6.50 years; p = 0.00001), serum LDL level (2.77 ± 0.79 and 3.63 ± ± 1.20 mmol/l; p = 0.0093), and TG level (1.06 ± 0.42 and 1.67 ± 0.73 mmol/l; p = 0.005) respectively. There were no frequency differences in 25(OH)D plasma deficiency/insufficiency either between the whole sample and the groups or between the groups themselves. We established a direct correlation relationship between standardized 25(OH)D plasma level and EDV indices — Ddz/dt (r = 0.33; p = 0.014) in the IHD group (n = 55) which proves the influence of vitamin D on the endothelium of vascular wall and may evidence the role of 25(OH)D insufficiency in pathogenesis of IHD.

Thus, on the basis of the obtained data the following conclusions can be made:

The prevalence of low vitamin D concentration in residents of the Western Region of Belarus with cardiovascular pathology is very high. The level of 25(OH)D in blood plasma lower than 17 nmol/l is associated with elevated SBP. Inverse correlations were established between the 25(OH)D plasma le­vel and body mass, BMI, DBP. In the IHD group there was a direct correlation relationship between the 25(OH)D plasma level and the index of endothelium dependent vasodilatation.