Relationship Among 25-Hydroxyvitamin D Concentrations, Insulin Action, and Cardiovascular Disease Risk in Patients With Essential Hypertension

Abstract

BACKGROUND

Although low plasma 25-hydroxyvitamin D (25(OH)D) concentrations have been shown to predict risk of hypertension and associated cardiovascular disease (CVD), vitamin D repletion has not consistently lowered blood pressure or decreased CVD. One possibility for this discrepancy is the presence of considerable metabolic heterogeneity in patients with hypertension. To evaluate this possibility, we quantified relationships among insulin resistance, 25(OH)D concentration, and CVD risk factor profile in patients with essential hypertension.

METHODS

Measurements were made of 25(OH)D concentrations, multiple CVD risk factors, and insulin resistance by the steady-state plasma glucose concentration during the insulin suppression test in 140 otherwise healthy patients with essential hypertension.

RESULTS

As a group, the patients were overweight/obese and insulin resistant and had low 25(OH)D concentrations. The more insulin resistant the patients were, the worse the CVD risk profile was. In addition, the most insulin-resistant quartile had significantly lower 25(OH)D concentrations than the most insulin-sensitive quartile (20.3±1.4 vs. 25.8±1.4ng/ml; P = 0.005). In the entire group, 25(OH)D concentration significantly correlated with magnitude of insulin resistance (steady-state plasma glucose concentration; r = −0.20; P = 0.02).

CONCLUSIONS

There was considerable metabolic heterogeneity and substantial difference in magnitude of conventional CVD risk factors in patients with similar degrees of blood pressure elevation. The most insulin-resistant quartile of subjects had the lowest 25(OH)D concentration and the most adverse CVD risk profile, and they may be the subset of patients with essential hypertension most likely to benefit from vitamin D repletion.

There is substantial epidemiological evidence that low plasma concentrations of 25-hydroxyvitamin D (25(OH)D) are associated with increased risk of developing hypertension,^1,2 although not all studies agree.³ Furthermore, analysis of data from the 2001–2004 National Health and Nutrition Examination Survey with linked mortality data through 2006 indicated that concentrations of 25(OH)D were inversely associated with all-cause and cardiovascular disease (CVD) mortality in the 2,609 participants with hypertension.⁴ There is also a coherent rationale for vitamin D deficiency to be associated with hypertension because it has been well documented that low vitamin D levels upregulate the renin-angiotensin-aldosterone system, increase inflammation, and cause endothelial dysfunction.^5–7 However, thus far the results of interventional studies suggest that treatment with vitamin D neither substantially lowers blood pressure (BP) nor uniformly prevents CVD.^8–10 The extensive literature is summarized in recent reviews.^11,12 One possible explanation for this somewhat paradoxical situation is that considerable heterogeneity in CVD risk exists in patients with hypertension.^13–15 More specifically, approximately 50% of patients with hypertension, treated or untreated, are insulin resistant and have the CVD risk factors usually associated with this abnormality.¹³ To the best of our knowledge, there is no available information regarding the relationship between 25(OH)D concentrations, specific measures of insulin action, and CVD risk factors in patients with hypertension. This study examines the relationships among these variables with the goal of providing new and clinically relevant information concerning the impact of insulin resistance and 25(OH)D concentrations on CVD risk in patients with hypertension.

METHODS

The study sample consisted of 140 patients with essential hypertension who were recruited from the San Francisco Bay Area through newspaper advertisements for our studies of insulin resistance. The patients were aged >30 years, had a body mass index (BMI) between 18.5 and 39.9kg/m², had a fasting glucose concentration <126mg/dl, and were not taking glucose-lowering medications or insulin. Hypertension was defined by systolic BP ≥140mm Hg and/or diastolic BP ≥90mm Hg or being treated with BP-lowering medication(s). The Stanford University’s Human Subjects Committee approved the study protocols, and all subjects gave written informed consent.

The profile of CVD risk factors included measurements of BMI, BP, fasting plasma glucose (FPG), fasting plasma insulin (FPI), and lipid and lipoprotein concentrations. Height and weight were measured while patients were wearing light clothing and no shoes, and BMI was calculated by dividing weight (in kilograms) by height (in meters squared). BP and heart rate were measured by using Dinamap automatic BP recorder (GE HealthCare, Tampa, FL) with an appropriately sized cuff after patients sat quietly for 5 minutes. Three measurements were taken at 1-minute intervals and were averaged. Blood samples for biochemical measurements were collected after an overnight fast. FPG concentration was determined on a Beckman glucose analyzer, and FPI was measured by using a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method.¹⁶ Lipid and lipoprotein concentrations were measured in the Clinical Laboratory at Stanford University Medical Center.

Concentrations of 25(OH)D₂ and 25(OH)D₃ were measured using an LC-MS/MS method that used a ThermoElectron Cohesive TLX-4 LC system coupled to ThermoElectron TSQ Ultra mass spectrometer (Thermo Fisher Scientific, Pittsburgh, PA). Analysis was completed using 2 multiple-reaction monitoring transitions per analyte, calculating the area under each individual peak, ratioing those numbers to the added stable isotope–labelled vitamin D analogues, and comparing the resultant ratios to a calibration curve. Individual results for both 25(OH)D₂ and 25(OH)D₃ were calculated and then summed to provide a total 25(OH)D result.¹⁷ For 25(OH)D₂ and 25(OH)D₃, the intra-assay coefficients of variation were 6%–10% and 4%–9%, respectively; the inter-assay coefficients of variation were 9%–12% and 10%, respectively; and the analytical sensitivity was 4ng/ml. The assay was 100% specific for the 25(OH)D₂ and 25(OH)D₃, and there was no cross-reaction with vitamin D₂ or vitamin D₃; 1α,25(OH)₂D₂; 1α,25(OH)₂D₃, calcitriol; 25,26(OH)₂D₃; 1α(OH)D₂, doxercalciferol; or 1α(OH)D₃, alfacalcidiol. Vitamin D measurements were performed at Quest Diagnostics in December 2012 on heparin plasma samples that were collected and frozen between November 2000 and April 2008 at the same time as the insulin suppression test described below.

The ability of insulin to dispose of a continuous intravenous glucose infusion was quantified by a modified version¹⁸ of the insulin suppression test as introduced and validated by our research group.^19,20 After an overnight fast, an intravenous catheter was placed in 1 arm for a 180-minute infusion of octreotide acetate (0.27 µg/m²/minute), insulin (32 mU/m²/minute), and glucose (267mg/m²/minute), and another catheter was placed in the contralateral arm to obtain blood for measurement of plasma glucose and insulin concentrations before and 150, 160, 170, and 180 minutes after starting the infusion. The mean of the 4 values obtained during the last 30 minutes of the infusion provided the steady-state plasma glucose (SSPG) and steady-state plasma insulin concentrations for each individual. Because octreotide suppresses endogenous insulin secretion, steady-state plasma insulin concentrations are similar, both qualitatively and quantitatively, in all individuals. Consequently, the height of the SSPG concentration provides a direct measure of how effective insulin is in mediating disposal of the infused glucose, a value that is highly correlated with the results of the euglycemic, hyperinsulinemic clamp.^19,21

Summary statistics are presented as mean ± SEM, median (interquartile range), geometric mean (95% confidence interval (CI)), or number (percentage) of patients. On the basis of the Endocrine Society’s guidelines for evaluation of vitamin status, patients were classified as vitamin D deficient, insufficient, or sufficient if their 25(OH)D concentrations were ≤20, 21–29, or ≥30ng/ml, respectively.²² The Institute of Medicine guidelines define individuals with 25(OH)D concentration <20ng/ml as vitamin D deficient and ≥20ng/ml as vitamin D sufficient.²³ FPI, triglyceride, total cholesterol, and low-density lipoprotein (LDL) cholesterol concentration values were log-transformed to improve normality for parametric tests. Bivariable Pearson correlation coefficients were calculated to determine the strengths of association of CVD risk factors with 25(OH)D and SSPG concentrations. Partial correlations were calculated to adjust for smoking status (smoker vs. nonsmoker) and antihypertensive drug therapy (treated vs. untreated). Means were compared by using 1-way analysis of variance and proportions were compared by χ² test. In addition, 1-way analysis of covariance models were used to adjust for differences in race (non-Hispanic white vs. nonwhite), multivitamin use, and season of blood sample collection (summer/fall (1 June–30 Novemeber) vs. winter/spring (1 December–31 May)). When an omnibus analysis of variance or analysis of covariance yielded a significant result (P ≤ 0.05), pairwise comparisons were conducted by using the least-significant difference tests. Statistical analyses were performed using SPSS-IBM software, version 20.0 (IBM, Armonk, NY).

RESULTS

The clinical and metabolic characteristics of patients with essential hypertension are presented in Table 1. Half of the patients were women, the majority (71.4%) was non-Hispanic white, and 19 (13.6%) were cigarette smokers. In the entire group, 53 (37.9%) individuals reported using multivitamins, and only 1 patient was taking oral vitamin D supplementation. One hundred eleven (79.3%) patients were receiving ≥1 antihypertensive medications. Of these individuals, 47.7% were being treated with 1 drug, 39.6% were being treated with 2 drugs, and 12.6% were being treated with 3 drugs. Approximately half of the patients (51.4%) were receiving a diuretic, 41.4% were receiving an angiotensin-converting enzyme inhibitor, 28.8% were receiving a beta-blocker, 21.6% were receiving a calcium channel blocker, and 18.9% were receiving an angiotensin II receptor blocker. Of the 111 patients receiving antihypertensive drug therapy, 78 (70.3%) had controlled BP (<140/90mm Hg). More than half of the patients (58%) were obese, 34% were overweight, and 8% were normal weight. The SSPG concentrations of the whole group varied approximately 7-fold from insulin sensitive to insulin resistant (43–312mg/dl).

Table 1.

Clinical and metabolic characteristics of patients with essential hypertension (n = 140)

Variable	Summary statistic
Age, y	55±1
Women, no. (%)	70 (50.0)
Non-Hispanic white, no. (%)	100 (71.4)
Cigarette smoker, no. (%)	19 (13.6)
Multivitamin use, no. (%)	53 (37.9)
Treated for hypertension, no. (%)	111 (79.3)
BP controlled, no. (%)	78 (70.3)
BMI, kg/m2	30.8±0.4
Systolic BP, mm Hg	134±1
Diastolic BP, mm Hg	78±1
Heart rate, beats per minute	68±1
FPG, mg/dl	99±1
FPI, µU/mla	11.4 (7.9–15.9)
SSPG, mg/dl	170±6
Triglycerides, mg/dl	124 (85–184)
HDL cholesterol, mg/dl	45±1
Total cholesterol, mg/dl	189 (159–217)
LDL cholesterol, mg/dl	115 (91–138)
25(OH)D, ng/ml	23.6±0.7

Data are mean ± SEM or median (interquartile range) unless otherwise noted.

Abbreviations: 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; BP, blood pressure; FPG, fasting plasma glucose; FPI, fasting plasma insulin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SSPG, steady-state plasma glucose.

^aFPI summary statistics were computed by using data from 133 patients; 7 individuals had missing FPI values.

Patients of non-Hispanic white descent had a significantly higher mean 25(OH)D concentration than those who were nonwhite (25.5±0.8 vs. 18.8±1.2ng/ml; P <0.001). Furthermore, patients whose blood samples were collected during the summer/fall season had a significantly higher mean 25(OH)D concentration than patients whose samples were collected during the winter/spring season (25.7±1.0 vs. 21.9±0.9ng/ml; P = 0.006).

Table 2 presents the correlations of CVD risk factors with 25(OH)D and SSPG concentrations. There were no significant associations between 25(OH)D concentration and age, BMI, systolic BP, diastolic BP, and heart rate or FPG, triglyceride, high-density lipoprotein (HDL) cholesterol, total cholesterol, and LDL cholesterol concentrations. However, significant inverse correlations were observed between 25(OH)D and FPI and SSPG concentrations. Regarding, the association of CVD risk factors with the measure of insulin action, SSPG concentration significantly and positively correlated with BMI and heart rate, as well as FPG, FPI, and triglyceride concentrations, and negatively correlated with HDL cholesterol and LDL cholesterol concentrations. Finally, the inverse correlations of CVD risk factors with 25(OH)D and SSPG concentrations were essentially unchanged after adjustment for smoking status or antihypertensive drug therapy.

Table 2.

Associations of cardiovascular disease risk factors with 25-hydroxyvitamin D and steady-state plasma glucose concentrations in patients with essential hypertension (n = 140)

Variable	25(OH)D		SSPG
	r	P value (2-tailed)	r	P value (2-tailed)
Age	0.15	0.08	0.10	0.23
BMI	−0.13	0.13	0.45	<0.001
Systolic BP	−0.02	0.82	−0.10	0.26
Diastolic BP	−0.06	0.46	−0.08	0.32
Heart rate	−0.01	0.88	0.22	0.008
FPG	0.11	0.22	0.29	<0.001
FPIa	−0.18	0.03	0.71	<0.001
SSPG	−0.20	0.02	—	—
Triglycerides	0.01	0.94	0.33	<0.001
HDL cholesterol	0.15	0.08	−0.31	<0.001
Total cholesterol	−0.06	0.52	−0.14	0.11
LDL cholesterol	−0.09	0.33	−0.24	0.004

Data are Pearson correlation coefficients (r) and accompanying P values. Fasting plasma insulin (FPI), triglyceride, total cholesterol, and low-density lipoprotein (LDL) cholesterol concentration values were log-transformed for statistical tests.

Abbreviations: 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; FPG, fasting plasma glucose; HDL, high-density lipoprotein; SSPG, steady-state plasma glucose.

^aCorrelations of FPI with 25(OH)D and SSPG were calculated by using data from 133 patients; 7 individuals had missing FPI values.

Comparison of CVD risk factors in patients with essential hypertension in quartiles of SSPG concentration is shown in Table 3. The quartiles were similar in terms of distribution of sex, race, smoking status, multivitamin use, antihypertensive medication intake, BP control, season of blood sample collection, mean age, systolic BP, and diastolic BP. However, mean BMI and heart rate and FPG, FPI, and triglyceride concentrations were significantly higher and HDL cholesterol and LDL cholesterol concentrations were significantly lower in patients in the most insulin-resistant quartile than in those in the most insulin-sensitive quartile. The mean 25(OH)D concentration was significantly lower in individuals who were the most insulin resistant than in those who were the most insulin sensitive. Furthermore, this difference remained significant after adjustment for race, multivitamin use, and season (adjusted mean 25(OH)D concentrations: 20.5±1.2 and 25.0±1.2ng/ml in SSPG quartiles IV and I, respectively; P = 0.01).

Table 3.

Cardiovascular disease risk factors in patients with essential hypertension divided into quartiles of steady-state plasma glucose concentration

Variable	SSPG concentration quartiles				P value (2-tailed)
	I Insulin sensitive (n = 35)	II (n = 35)	III (n = 35)	IV Insulin resistant (n = 35)	Overall difference	Quartile IV vs. Ia
SSPG, mg/dl	<107	107–173	174–229	>229
Age, y	55±1	54±1	55±1	56±1	0.79c	—
Women, no. (%)	17 (48.6)	21 (60.0)	16 (45.7)	16 (45.7)	0.58d	—
Non-Hispanic white, no. (%)	27 (77.1)	21 (60.0)	26 (76.5)	26 (76.5)	0.38d	—
Cigarette smoker, no. (%)	5 (14.3)	6 (17.1)	3 (8.6)	5 (14.3)	0.76d	—
Multivitamin use, no. (%)	15 (42.9)	13 (37.1)	14 (40.0)	11 (31.4)	0.79d	—
Treated for hypertension, no. (%)	27 (77.1)	28 (80.0)	26 (76.5)	30 (85.7)	0.68d	—
BP controlled, no. (%)	17 (62.3)	17 (62.3)	19 (73.1)	25 (83.3)	0.22d	—
Samples collected during summer/fall, no. (%)	19 (54.3)	12 (88.5)	17 (48.5)	14 (40.0)	0.34d	—
BMI, kg/m2	28.0±0.7	30.1±0.7	31.7±0.6	32.9±0.6	<0.001c	<0.001
Systolic BP, mm Hg	135±3	136±3	134±3	131±2	0.54c	—
Diastolic BP, mm Hg	80±2	78±2	78±2	77±1	0.81c	—
Heart rate, beats/min	63±1	68±2	69±2	70±1	0.02c	0.003
FPG, mg/dl	96±2	97±1	100±2	102±1	0.05c	0.02
FPI, µU/mlb	6.4 (5.6–7.4)	10.4 (9.0–11.9)	13.2 (11.4–15.2)	19.2 (16.8–22.1)	<0.001c	<0.001
Triglycerides, mg/dl	106 (88 – 128)	104 (86–125)	133 (110–160)	172 (143–207)	0.001c	<0.001
HDL cholesterol, mg/dl	52±2	47±2	43±2	39±2	<0.001c	<0.001
Total cholesterol, mg/dl	192 (179–207)	195 (181–209)	180 (168–193)	180 (167–193)	0.24c	—
LDL cholesterol, mg/dl	117 (106–130)	123 (118–136)	104 (94–115)	100 (90–111)	0.02c	0.03
25(OH)D, ng/ml	25. 8±1.4	23. 4±1.4	24.8±1.4	20.3±1.4	0.03c	0.005

Data are mean + SEM or geometric mean (95% confidence interval) unless otherwise noted. Fasting plasma insulin (FPI), triglyceride, total cholesterol, and low-density lipoprotein (LDL) cholesterol concentration values were log-transformed for statistical tests.

Abbreviations: 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; FPG, fasting plasma glucose; HDL, high-density lipoprotein; SSPG, steady-state plasma glucose.

^aLeast-significant difference pairwise comparison test.

^bData from 34, 33, 32, and 34 patients with complete FPI observations in the SSPG quartiles I, II, III, and IV, respectively.

^cOne-way analysis of variance.

^dχ² test.

Table 4 presents the CVD factors in patients with essential hypertension divided on the basis of clinical categories of vitamin D status. The groups were similar in terms of sex distribution, smoking status, antihypertensive drug therapy, and BP control. However, in the vitamin D–deficient group, fewer patients were of non-Hispanic white descent, a lower proportion of individuals was taking multivitamins, and a smaller percentage of samples was collected during the summer/fall season. The mean SSPG concentration was significantly higher (greater insulin resistance) in patients who were vitamin D insufficient or vitamin D deficient than in those who were vitamin D sufficient. In addition, these differences remained significant after adjustment for race, multivitamin use, and season (P = 0.04). Specifically, the adjusted mean SSPG concentrations were 176±10 and 142±13mg/dl in the vitamin D–insufficient and vitamin D–sufficient groups, respectively (P = 0.03) and 183±11 and 142±13mg/ml in the vitamin D–deficient and vitamin D–sufficient groups, respectively (P = 0.02). No other significant differences were observed in the CVD risk factors in the clinical categories of vitamin D status.

Table 4.

Cardiovascular disease risk factors in patients with essential hypertension divided into groups based on vitamin D status

Variable	Sufficient (n = 34)	Insufficient (n = 55)	Deficient (n = 51)	P value (2-tailed)
25(OH)D, ng/ml	≥30	21–29	≤20
Age, y	57±6	54±8	55±7	0.32b
Women, no. (%)	18 (52.9)	27 (49.1)	25 (49.0)	0.93c
Non-Hispanic white, no. (%)	31 (91.2)	40 (72.7)	29 (56.9)	0.003c
Cigarette smoker, no. (%)	5 (14.7)	4 (7.3)	10 (19.6)	0.18c
Multivitamin use, no. (%)	18 (52.9)	25 (45.5)	10 (19.6)	0.003c
Treated for hypertension, no. (%)	28 (82.4)	47 (85.5)	36 (70.6)	0.15c
BP controlled, no. (%)	20 (71.4)	33 (70.2)	25 (69.4)	0.99c
Samples collected during summer/fall, no. (%)	17 (50.0)	31 (56.4)	14 (27.5)	0.008c
BMI, kg/m2	30.6±0.7	30.2±0.6	31.6±0.6	0.19b
Systolic BP, mm Hg	135±3	133±2	135±2	0.77b
Diastolic BP, mm Hg	78±2	78±1	79±1	0.86b
Heart rate, beats/min	68±2	65±1	70±1	0.08b
FPG, mg/dl	100±2	99±1	98±1	0.84b
FPI, µU/mla	10.1 (8.3–12.2)	11.0 (9.4–12.8)	12.9 (11.0–15.1)	0.13b
SSPG, mg/dl	141±12	176±9*	182±10**	0.04b
Total cholesterol, mg/dl	184 (171–197)	189 (178–200)	186 (176–198)	0.84b
Triglycerides, mg/dl	118 (97–144)	144 (123–168)	114 (97–134)	0.10b
HDL cholesterol, mg/dl	47±2	45±2	43±2	0.40b
LDL cholesterol, mg/dl	108 (97–120)	110 (101–119)	115 (105–125)	0.65b

Data are mean ± SEM or geometric mean (95% confidence interval) unless otherwise noted. Fasting plasma insulin (FPI), triglyceride, total cholesterol, and low-density lipoprotein (LDL) cholesterol concentration values were log-transformed for statistical tests.

Abbreviations: 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; FPG, fasting plasma glucose; HDL, high-density lipoprotein; SSPG, steady-state plasma glucose.

^aData from 33, 52, and 48 patients with complete FPI observations in the sufficient, insufficient, and deficient groups, respectively.

^bOne-way analysis of variance.

^cχ² test.

*P = 0.04 (insufficient vs. sufficient) by least-significant difference pairwise comparison test.

**P = 0.01 (deficient vs. sufficient) by least-significant difference pairwise comparison test.

DISCUSSION

The results of our study emphasize the clinical heterogeneity of CVD risk in subjects with essential hypertension and what we believe to be new information concerning the relative roles of insulin resistance and vitamin D status in this context. To begin with, it is clear from the results in Table 2 that insulin resistance, as estimated by SSPG concentration, was significantly correlated with essentially every CVD risk factor measured. Furthermore, as seen in Table 3, the more insulin resistant the experimental quartile was, the worse the CVD risk profile was. In fact the only risk factors that did not get worse with increasing degree of insulin resistance were age, BP, total cholesterol, and LDL cholesterol. These findings are consistent with previous publications demonstrating that approximately half of the patients with high BP, treated or untreated, are insulin resistant, with a significantly more adverse CVD risk profile.^13,14 More important, there is also evidence that it is the subset of patients with essential hypertension who are insulin resistant that also develop more coronary heart disease.^15,24

The association between 25(OH)D concentration and CVD risk in patients with hypertension is a more complicated one. Thus, FPI and SSPG concentrations were the only CVD risk factors significantly associated with 25(OH)D concentration (Table 2). SSPG concentrations were higher in patients with hypertension who had 25(OH)D concentrations <30ng/ml, which includes both the vitamin D–insufficient and vitamin D–deficient groups. Comparing the mean SSPG values between the insufficient and the sufficient subjects and the deficient and the sufficient subjects, the differences were significant in both comparisons (P = 0.04 and P = 0.01, respectively) (Table 4).

This finding is consistent with the recent analyses by Heaney et al.,²⁵ which indicated that insulin resistance was associated with 25(OH)D concentrations below the same cutpoint of 30ng/ml. Using the Institute of Medicine suggested cutpoint of 20ng/ml, the insufficient group (20–29ng/ml) would be considered to have sufficient 25(OH)D, yet they would still be in the insulin-resistant category. Finally, the only subset of patients with significantly lower 25(OH)D concentrations were those in SSPG quartile IV, the most insulin-resistant group with the worst CVD risk profile (Table 3).

Our study was cross-sectional in design, and the association between low vitamin D concentrations and greater insulin resistance does not imply causation. On the other hand, there are a number of cellular/molecular mechanisms that are consistent with the notion that low 25(OH)D concentrations could contribute to insulin resistance and a decrease in insulin-mediated glucose disposal. For example, it has been shown that 1,25(OH)₂D increases the expression of insulin receptors by interacting with the vitamin D response element in the promoter region of the human insulin receptor gene.²⁶ Furthermore, 1,25(OH)₂D has been shown to activate peroxisome proliferator-activated receptor delta, a transcription factor involved in regulation of fatty acid metabolism in skeletal muscle and adipose tissue.²⁷ Vitamin D may also affect insulin sensitivity through modulation of extracellular calcium and calcium flux across cell membranes. Cytosolic calcium concentrations maintained within a narrow range are required for optimum insulin signal transduction and glucose uptake in insulin responsive tissues.²⁸ Finally and importantly, low vitamin D levels upregulate the renin-angiotensin-aldosterone system, which can be reversed by treatment with vitamin D or its analogues,⁵ and elevations in angiotensin II have been shown to be associated with insulin resistance and hypertension.^29,30

There are obvious issues that could confound the results of this study. First, the total patient sample is relatively small and was divided into quartiles for data analysis. Furthermore, the analysis was based on the entire sample, without taking into consideration that approximately 80% of subjects were taking BP-lowering drugs. However, in a prior study,¹³ we demonstrated that degree of insulin resistance and CVD risk profile were essentially identical when we compared 70 treated and 56 untreated patients with hypertension. Furthermore, although blood for measurement of 25(OH)D was obtained at different times of the year, we have adjusted our findings in terms of that variable, as well as for possible ethnic differences. Finally, we believe no other study has used a specific method to quantify insulin-mediated glucose disposal in evaluation of the relative impacts of differences in insulin action and 25(OH)D concentrations in otherwise healthy patients with hypertension, a major strength of our observations.

In conclusion, although not the goal of this study, the findings provide a formulation that may help explain why vitamin D repletion has not seemed to be clinically useful in decreasing BP and/or decreasing CVD risk in patients with hypertension.^8–10 Insulin resistance is an independent predictor of essential hypertension, approximately 50% of patients with hypertension are insulin resistant, and incident CVD is increased in this subset of individuals.^13–15,31 Furthermore, both SSPG and 25(OH)D concentrations vary substantially in patients with hypertension, and the highest 25(OH)D concentrations are seen in the most insulin-sensitive individuals—a subset at the least risk of CVD.^31,32 Consequently, it could be argued that phenotypic specificity will play a major role in determining the benefits of administrating vitamin D to patients with hypertension, and the greater the proportion of insulin sensitive subjects in the experimental population the less likely that vitamin D repletion will have a discernible clinical benefit. Put most simply, clinical benefit from administration of vitamin D would be much more likely to occur in subjects in quartile IV (the most insulin resistant), and achieving such an outcome will be unlikely the greater the proportion of subjects comparable with those in quartile I (the most insulin sensitive) that are included in the experimental population. We believe that a vitamin D intervention trial is warranted in subjects whose degree of insulin resistance is known to clarify the benefits of vitamin D repletion in the hypertensive population at risk for CVD.

DISCLOSURE

F.M.H. and M.P.C. are employees of Quest Diagnostics and own stock in the company. No other conflicts of interest exist.