Obese women having bariatric surgery have lower serum 25-OHD and higher serum PTH than healthy control women, but low serum 25-OHD is not associated with vitamin D insufficiency or with comorbid conditions.
Abstract
Background:
Low vitamin D status and hyperparathyroidism occur in obesity and may be involved in pathogenesis of obesity-associated comorbid conditions.
Aims:
Our aims were to determine in obesity whether there was vitamin D insufficiency, assessed by serum 25-hydroxyvitamin D (s25D) and serum PTH (sPTH) and whether it related to comorbid conditions.
Methods:
We conducted a case-control study of 48 women having bariatric surgery and 50 healthy women frequency matched for race, age, year, and season of study. Height, weight, s25D, sPTH, serum 1,25-dihydroxyvitamin D (s1,25D), serum bone alkaline phosphatase, serum cross-linked N-telopeptides of type 1 collagen, and serum calcium, phosphate, creatinine, glucose, and insulin were measured, and comorbid conditions were documented from patient files.
Results:
Weight (140 vs. 76 kg, P < 0.001), sPTH (44.4 vs. 25.6 pg/ml, P < 0.001), s1,25D (39 vs. 24 pg/ml, P < 0.001), serum bone alkaline phosphatase (19 vs. 12 ng/ml, P < 0.001), serum cross-linked N-telopeptides of type 1 collagen (9.6 vs. 7.9 nm bone collagen equivalents, P = 0.007), serum phosphate (3.45 vs. 3.24 mg/dl, P = 0.043), and serum creatinine (1.05 vs. 0.87 mg/dl, P < 0.001) were higher, and s25D (16 vs. 23 ng/ml, P <.001) was lower in bariatric-surgery women than control women. s25D was lower in bariatric-surgery women than controls in summer (17 vs. 26 ng/ml, P < 0.0001) but not winter (15 vs. 18 ng/ml, P > 0.2). Multiple regression analysis demonstrated that weight predicted s25D (P < 0.001) and sPTH (P = 0.001), but s25D did not predict sPTH or the presence of comorbid conditions except for osteoarthritis.
Conclusion:
Women having bariatric surgery had lower s25D and higher sPTH. The major determinant of s25D and sPTH was weight. Hyperparathyroidism in obesity did not indicate vitamin D insufficiency. Low s25D was not associated with comorbid conditions, apart from osteoarthritis.
Obesity is an increasingly prevalent condition in industrialized societies. In 2007, all but one state in the United States had a prevalence of obesity greater than 20%, and 30 states had a prevalence of at least 25% (1). In a 2008 report on obesity from Europe, 35.9% of the population over age 18 were overweight [body mass index (BMI) = 25–29.9 kg/m2], and 17.2% were obese (BMI > 30 kg/m2) (2). Obesity is associated with an increased prevalence of a number of diseases including type 2 diabetes, hypertension, hypercholesterolemia, osteoarthritis (3), gastroesophageal reflux disease (GERD) (4), sleep apnea (5), and chronic renal disease (6).
Vitamin D deficiency occurs at serum 25-hydroxyvitamin D (s25D) levels persistently lower than 10 ng/ml and results in marked calcium malabsorption, negative calcium balance, secondary hyperparathyroidism, increased bone turnover, a large increase in serum 1,25-dihydroxyvitamin D (s1,25D) in response to an increase in 25D substrate, and osteomalacia in adults and rickets in children (7). On the other hand, vitamin D insufficiency occurs when s25D is chronically less than about 30 ng/ml but above deficiency levels. Although there is widespread belief that vitamin D insufficiency manifests biochemically as mild secondary hyperparathyroidism, the effects of vitamin D insufficiency on mineral homeostasis remain controversial (8).
In obesity, s25D levels are in the insufficiency range (9–12), and there is evidence of vitamin D insufficiency as judged by an inverse relationship between sPTH and s25D (12–15). It has been speculated that this vitamin D insufficiency may play a role in the pathogenesis of a number of the comorbid conditions associated with obesity. Vitamin D insufficiency has been implicated in the pathogenesis of insulin resistance (16), type 2 diabetes (17), cardiovascular disease (17), hypertension (18), and osteoarthritis (19).
The aims of this study were to examine whether vitamin D insufficiency was present in obesity as assessed by the relationship between sPTH and s25D, evaluate predictors of s25D and sPTH, and assess whether vitamin D insufficiency was related to the presence of comorbid disease associated with obesity.
Subjects and Methods
Subjects
Subjects were 48 obese White women presenting to the St. Vincent Bariatric Surgery Center in Carmel, IN, for Roux-en-Y gastric bypass surgery for obesity between February 2005 and December 2006. These subjects were drawn from an ongoing study evaluating anthropometric, hormonal, and biochemical markers after bariatric surgery that began in January 2005. The Roux-en-Y gastric bypass surgery, which included a 1-oz gastric pouch, 50- to 100-cm Roux limb, and a 100-cm biliopancreatic limb, was performed by three surgeons (R.J., B.M.C., and C.E.G., ∼700–800 patients/yr). All subjects provided informed consent for participation in the study. The protocol was approved by the Institutional Review Board at Indiana University-Purdue University (Indianapolis, IN) and at St. Vincent Hospital (Indianapolis, IN). Only White women were included to avoid the influence of race and sex on the variables measured. None of the women were taking a vitamin D supplement. Fasting blood samples were collected within 2 h before surgery.
The presence of comorbid obesity-related diseases was collated from information in the patient case files collected from patient interviews before surgery. Comorbid conditions were those whose prevalence has been shown to increase with obesity and included diabetes, hypertension, hypercholesterolemia, osteoarthritis, sleep apnea, and GERD. The majority of the bariatric-surgery patients were recorded as being premenopausal with regular periods. However, menstrual history was not recorded in every subject, and FSH was not measured to assess for postmenopausal status.
Control subjects were 50 women randomly selected from a database of healthy women who had consented and donated fasting blood to the Indiana Clinical Research Center for observational research between January 2005 and October 2006. Control subjects were all premenopausal and healthy and were not on medications or vitamin D supplement. No survey instrument was used to collect specific information on symptoms of arthritis, sleep apnea, and GERD, but none of them had symptoms, and their medical history was negative. They were frequency matched with patients having bariatric surgery for race, age, year in which blood was collected, and season. Season divided the subjects into two groups: those whose blood was drawn from April to September (summer) and those whose blood was drawn from October to March (winter).
Methods
Height was measured using a Harpenden stadiometer and weight measured using a Scale-Tronix scale. Chronological age was recorded at the time of the visit, and BMI was calculated as a determinant of obesity. s25D and s1,25D were measured by RIA [coefficient of variation (CV) of 8.1 and 9.1% respectively; DiaSorin, Stillwater, MN]. sPTH was measured by a two-site immunoassay (CV 9.7% at 17.5 pg/ml; Nichols Institute Diagnostics, San Juan Capistrano, CA). Cross-linked N-telopeptides of type 1 collagen (NTX) was measured by ELISA (CV 7.5%; Osteomark; Ostex International, Inc., Seattle, WA). Serum bone alkaline phosphatase (sBAP) was measured by ELISA (CV 4.1%; Quidel, San Diego, CA). Serum insulin was measured by RIA (interassay CV 3.9%, intraassay CV 3.2%; Millipore Corp., Billerica, MA). Serum creatinine (sCr), serum calcium (sCa), serum phosphate (sPi), and glucose were measured using Roche COBAS MIRA Clinical Analyzer (Roche Diagnostic, Indianapolis, IN). Insulin resistance was determined by calculating the homeostasis model assessment index from measured glucose and insulin levels (20).
Data analysis
The Statistical Product and Service Solutions program (SPSS Inc., Chicago, IL; version 15.0, 2006) was used to perform statistical analysis. Descriptive statistics were computed. Mean values for all variables were compared between the bariatric-surgery and control groups (overall and stratified by season) using two-sample t tests. Two-sample tests were also used to compare seasons within each group. Calculation of Pearson correlation coefficients and linear regression analyses were performed to assess relationships between key variables. Multiple regression analyses were performed to measure to what extent 1) age, weight, and sCr predicted s25D; 2) weight, s25D, sCa, and sCr predicted sPTH; 3) age, weight, s25D, sPTH, sCa, and sCr predicted s1,25D; and 4) weight, s25D, sPTH, s1,25D, sCa, and sCr predicted sBAP. Multiple regression models were based on physiologically expected outcomes. Both R2 and squared semipartial correlations (SR2) were calculated. SR2 quantifies the unique contribution each explanatory variable makes to the total R2. R2 is the proportion of variation in the outcome of multiple regression model that is explained by all the explanatory variables combined. Within the bariatric-surgery group, two-sample t tests were used to compare mean s25D levels based on the presence or absence of comorbidities.
Results
Anthropometrics and biochemistry (Table 1)
Table 1.
Anthropometric and biochemical variables in bariatric-surgery and control groups
| Mean ± sd (range) |
P value | ||
|---|---|---|---|
| Bariatric surgery, n = 48 | Control, n = 50 | ||
| Age (yr) | 42.7 ± 9.7 (25–59) | 39.7 ± 7.4 (26–55) | 0.081 |
| Height (cm) | 163 ± 6 (150–174) | 165 ± 6 (153–177) | 0.136 |
| Weight (kg) | 140.2 ± 37.2 (91.8–251.8) | 76.3 ± 15.8 (45.0–115.3) | <0.0001 |
| BMI (kg/m2) | 52.3 ± 12.5 (35.0–90.2) | 28.1 ± 5.9 (16.7–43.0) | <0.0001 |
| s25D (ng/ml) | 16 ± 8 (3–42) | 23 ± 10 (8–54) | <0.0001 |
| sPTH (pg/ml)a | 44.4 ± 34 (11.5–215.0) | 25.6 ± 9 (10.2–54.1) | <0.0001 |
| s1,25D (pg/ml) | 39 ± 19 (13–107) | 24 ± 17 (3–73) | <0.0001 |
| sBAP (ng/ml) | 19 ± 5 (11–34) | 12 ± 4 (5–20) | <0.0001 |
| sNTX (nm bone collagen equivalents) | 9.6 ± 3.2 (3.3–22.1) | 7.9 ± 2.8 (4.5–16.4) | 0.007 |
| sCa (mg/dl)a | 8.92 ± 0.44 (8.0–10.0) | 8.78 ± 0.34 (8.1–9.4) | 0.079 |
| sPi (mg/dl) | 3.45 ± 0.52 (2.2–4.7) | 3.24 ± 0.52 (2.3–4.5) | 0.043 |
| sCr (mg/dl) | 1.05 ± 0.24 (0.7–1.9) | 0.87 ± 0.11 (0.7–1.1) | <0.0001 |
When the outlier PTH = 215 was removed, significance of difference between control and bariatric-surgery women changed only for sCa, with control sCa lower than bariatric-surgery group; P = 0.042.
The bariatric-surgery group was heavier than the control group but did not differ in height or age. All the bariatric-surgery subjects had a BMI above 30 kg/m2, whereas 14 control subjects were obese with a BMI above 30 kg/m2. In the bariatric-surgery group, the s25D was lower, and sPTH, s1,25D, sBAP, sNTX, sPi, and sCr were higher than in the control group. Mean sCa did not differ between groups. When data were analyzed without an outlying sPTH value in the bariatric-surgery group (215 pg/ml), sPTH remained higher (40.8 vs. 25.6 ng/ml, P < 0.001), sCa became significantly higher (8.94 vs. 8.78 mg/dl, P = 0.042), and significance was not affected for other variables.
Season (Tables 2 and 3)
Table 2.
Comparison of bariatric-surgery group vs. control group during summer season (April to September)
| Mean ± sd (range) |
P value | ||
|---|---|---|---|
| Bariatric surgery, n = 33 | Control, n = 35 | ||
| Age (yr) | 44 ± 9 (25–59) | 39.4 ± 7 (26–49) | 0.036 |
| Height (cm) | 163 ± 6 (150–173) | 164 ± 5 (153–174) | 0.214 |
| Weight (kg) | 138.0 ± 35.7 (91.8–230.9) | 73.8 ± 16.2 (45.0–114.1) | <0.0001 |
| BMI (kg/m2) | 51.9 ± 12.8 (35.0–90.2) | 27.3 ± 6.0 (16.7–41.2) | <0.0001 |
| s25D (ng/ml) | 17 ± 9 (4.3–41.5) | 26 ± 10 (8.0–58.4)a | <0.0001 |
| sPTH (pg/ml) | 43.6 ± 36.9 (11.5–215.0) | 26.6 ± 9.7 (11.1–54.1) | 0.011 |
| s1,25D (pg/ml) | 37 ± 17 (16–89) | 24 ± 17 (5–73) | 0.003 |
| sBAP (ng/ml) | 19 ± 5 (10.6–27.6) | 11 ± 3 (5.4–17.1) | <0.0001 |
| sNTX (nm bone collagen equivalents) | 10.1 ± 3.4 (3.3–22.1) | 8.2 ± 2.8 (4.5–15.6) | 0.011 |
| sCa (mg/dl) | 8.89 ± 0.41 (8.0–9.6) | 8.79 ± 0.37 (8.1–9.4) | 0.31 |
| sPi (mg/dl) | 3.52 ± 0.51 (2.4–4.7) | 3.24 ± 0.48 (2.3–4.1) | 0.026 |
| sCr (mg/dl) | 1.07 ± 0.26 (0.7–1.9) | 0.85 ± 0.11 (0.7–1.1) | <0.0001 |
Significantly different by season, P = 0.01.
Table 3.
Comparison of bariatric-surgery group vs. control group during winter season (October to March)
| Mean ± sd (range) |
P value | ||
|---|---|---|---|
| Bariatric surgery, n = 15 | Control, n = 15 | ||
| Age (yr) | 41 ± 11 (25–58) | 40 ± 8 (28–55) | 0.91 |
| Height (cm) | 165 ± 6 (155–174) | 166 ± 7 (156–177) | 0.395 |
| Weight (kg) | 145.0 ± 41.0 (103.1–251.8) | 82.3 ± 13.3 (61.1–115.3) | <0.0001 |
| BMI (kg/m2) | 53.2 ± 12.3 (37.8–83.2) | 29.8 ± 5.2 (22.3–43.0) | <0.0001 |
| s25D (ng/ml) | 15 ± 8 (3.0–34.4) | 18 ± 8 (8.2–39.8)a | 0.267 |
| sPTH (pg/ml) | 46.2 ± 26.3 (19.8–102.0) | 23.3 ± 7.2 (10.2–41.5) | 0.003 |
| s1,25D (pg/ml) | 46 ± 23 (13–107) | 23 ± 16 (5.0–59) | 0.003 |
| sBAP (ng/ml) | 21 ± 5 (15–34) | 13 ± 5 (6–20) | <0.0001 |
| sNTX (nm bone collagen equivalents) | 8.6 ± 2.6 (3.9–13.8) | 7.5 ± 3.0 (4.5–16.4) | 0.285 |
| sCa (mg/dl) | 8.99 ± 0.51 (8.1–10.0) | 8.75 ± 0.28 (8.2–9.1) | 0.121 |
| sPi (mg/dl) | 3.31 ± 0.54 (2.2–4.3) | 3.22 ± 0.62 (2.4–4.5) | 0.667 |
| sCr (mg/dl) | 1.01 ± 0.20 (0.7–1.5) | 0.91 ± 0.11 (0.7–1.1) | 0.098 |
Significantly different by season, P = 0.01.
When visits were categorized by season, there were 33 bariatric-surgery subjects and 35 control subjects studied in summer (April to September) and 15 bariatric-surgery and 15 control subjects studied in winter (October to March).
Studied in summer (Table 2), the bariatric-surgery group had higher age, weight, BMI, sPTH, s1,25D, sBAP, sNTX, sPi, and sCr, and lower s25D, and no difference in height and sCa.
Studied in winter (Table 3), the bariatric-surgery group had higher weight, BMI, sPTH, s1,25D, and sBAP than the control group. Age, height, s25D, sNTX, sCa, sPi, and sCr were not different between groups.
Comparison between summer and winter mean values in and within bariatric-surgery and control subjects showed that s25D was greater during summer than in winter for controls (26 vs. 18 ng/ml, P = 0.01), and was not different for bariatric-surgery subjects (15 vs. 17 ng/ml, P = 0.46). No other variables were significantly different between seasons for bariatric-surgery subjects or controls. When data were analyzed without the sPTH outlier, significance was not affected.
Relationships among age, weight, and biochemical variables
Pearson correlations (Table 4)
Table 4.
Pearson correlation coefficients in the bariatric-surgery and control subjects combined (n = 98)
| sPTH | Age | Weight | BMI | s1,25D | sBAP | sNTX | sCr | sCa | |
|---|---|---|---|---|---|---|---|---|---|
| s25D | |||||||||
| r | −0.30 | −0.27 | −0.48 | −0.50 | 0.15 | −0.31 | −0.39 | −0.21 | 0.04 |
| P value | 0.003 | 0.01 | <0.001 | <0.001 | 0.16 | 0.002 | 0.70 | 0.042 | 0.68 |
| sPTH | |||||||||
| r | −0.14 | 0.43 | 0.42 | 0.033 | 0.086 | 0.09 | 0.25 | −0.15 | |
| P value | 0.19 | <0.001 | <0.001 | 0.75 | 0.40 | 0.36 | 0.012 | 0.14 | |
| Age | |||||||||
| r | 0.16 | 0.20 | −0.03 | 0.27 | 0.08 | −0.06 | 0.18 | ||
| P value | 0.12 | 0.049 | 0.78 | 0.01 | 0.42 | 0.54 | 0.08 | ||
| Weight | |||||||||
| r | 0.98 | 0.13 | 0.51 | 0.13 | 0.35 | 0.08 | |||
| P value | <0.001 | 0.19 | <0.001 | 0.22 | <0.001 | 0.43 | |||
| BMI | |||||||||
| r | 0.13 | 0.53 | 0.11 | 0.37 | 0.12 | ||||
| P value | 0.20 | <0.001 | 0.29 | <0.001 | 0.22 | ||||
| s1,25D | |||||||||
| r | 0.29 | 0.06 | −0.03 | 0.008 | |||||
| P value | 0.004 | 0.59 | 0.76 | 0.94 | |||||
| sBAP | |||||||||
| r | 0.26 | 0.21 | 0.14 | ||||||
| P value | 0.009 | 0.036 | 0.16 | ||||||
| sNTX | |||||||||
| r | 0.055 | 0.09 | |||||||
| P value | 0.59 | 0.36 | |||||||
| sCr | |||||||||
| r | −0.07 | ||||||||
| P value | 0.51 |
In the bariatric-surgery and control subjects combined (n = 98), there was a negative correlation between s25D and sPTH, age, weight, sBAP, and sCr, whereas sPTH negatively correlated with s25D and positively with weight and sCr. In the control group, there was no significant relationship between s25 and sPTH (r = −0.14; P = 0.33), s25D and weight (r = −0.258; P = 0.07), and sPTH and weight (r = 0.27; P = 0.06), but the direction of the relationships were in the same direction as those of the combined group.
Multiple regression analysis (Table 5)
Table 5.
Predictors by multiple regression analysis in the bariatric-surgery and control subjects combined (n = 98); SR2 = semi-partial correlation
| Estimate | se | SR2 | P value | |
|---|---|---|---|---|
| Predictors of s25Da | ||||
| Age | −0.23 | 0.101 | 0.04 | 0.024 |
| Weight | −0.1 | 0.022 | 0.17 | <0.001 |
| sCr | −3.35 | 4.43 | 0.01 | 0.45 |
| Predictors of sPTHb | ||||
| Weight | 0.23 | 0.67 | 0.09 | 0.001 |
| s25D | −0.27 | 0.28 | 0.01 | 0.34 |
| sCa | −11.1 | 6.02 | 0.03 | 0.07 |
| sCr | 11.3 | 12.1 | 0.01 | 0.35 |
| Predictors of s1,25Dc | ||||
| Age | −0.25 | 0.25 | 0 | 0.092 |
| Weight | 0.14 | 0.06 | 0.05 | 0.03 |
| s25D | 0.055 | 0.24 | 0.05 | 0.03 |
| sPTH | 0.001 | 0.09 | 0 | 0.99 |
| sCa | −1.37 | 5.22 | 0 | 0.79 |
| sCr | −7.55 | 10.21 | 0.01 | 0.46 |
| Predictors of sBAPd | ||||
| Weight | 0.058 | 0.015 | 0.11 | <0.001 |
| s25D | −0.10 | 0.059 | 0.02 | 0.092 |
| sPTH | −0.037 | 0.021 | 0.02 | 0.08 |
| s1,25D | 0.076 | 0.026 | 0.06 | 0.004 |
| sCa | 1.20 | 1.25 | 0.01 | 0.34 |
| sCr | 2.14 | 2.48 | 0.01 | 0.39 |
R2 = 0.27.
R2 = 0.24.
R2 = 0.08.
R2 = 0.36.
In the bariatric-surgery and control subjects combined (n = 98), predictors of s25D were age (SR2 = 0.04; P = 0.024) and weight (SR2 = 0.17; P < 0.001) (Table 5). The only predictor of sPTH was weight (SR2 = 0.09; P = 0.001), with s25D, sCa, and sCr having no predictive value (Table 5). Both weight (SR2 = 0.05; P = 0.03) and s25D (SR2 = 0.05; P = 0.03) were predictors of s1,25D (Table 5).
Predictors of sBAP were weight (SR2 = 0.11; P < 0.001) and s1,25D (SR2 = 0.06; P = 0.004). sPTH had no predictive value and was negatively related (Table 5).
Comparison of bariatric-surgery subjects with and without comorbid conditions (Table 6)
Table 6.
s25D in relation to prevalence of comorbid conditions in bariatric-surgery subjects
| Comorbidity | n (%) | S25D (ng/ml) ± sd | P value |
|---|---|---|---|
| Diabetes | 0.55 | ||
| Present | 13/48 (27) | 14.9 ± 9.0 | |
| Absent | 35/48 (73) | 16.5 ± 8.3 | |
| Insulin resistancea | 0.07 | ||
| Present | 25/45 (56) | 14.0 ± 7.7 | |
| Absent | 20/45 (44) | 18.4 ± 8.8 | |
| Hypertension | 0.87 | ||
| Present | 27/48 (56) | 15.9 ± 8.2 | |
| Absent | 21/48 (44) | 16.3 ± 8.9 | |
| Hypercholesterolemia | 0.37 | ||
| Present | 15/48 (31) | 17.8 ± 9.5 | |
| Absent | 33/48 (69) | 15.3 ± 7.9 | |
| Osteoarthritis | 0.03 | ||
| Present | 19/48 (40) | 12.9 ± 4.2 | |
| Absent | 29/48 (60) | 18.2 ± 19.7 | |
| Sleep apnea | 0.44 | ||
| Present | 19/48 (40) | 14.9 ± 5.6 | |
| Absent | 29/48 (60) | 16.9 ± 9.6 | |
| GERD | 0.52 | ||
| Present | 26/48 (54) | 16.8 ± 6.5 | |
| Absent | 22/48 (46) | 15.3 ± 10.4 |
Insulin resistance measured by homeostasis model assessment higher than 5.
Of the 48 bariatric-surgery subjects, 27% had diabetes, 56% had insulin resistance, 56% had hypertension, 31% had hypercholesterolemia, 40% had osteoarthritis, 40% had sleep apnea, and 54% had GERD. Only with osteoarthritis was the s25D lower in those with the comorbid disease than those without (12.9 vs. 18.2 ng/ml, P = 0.03). Patients with osteoarthritis were also older than those without (47 vs. 39 yr, P = 0.002).
Discussion
In this study, s25D was lower in White women who were candidates for bariatric surgery than healthy control women. This finding agrees with a number of studies reporting that s25D is low in obesity and inversely related to body fat (10–12, 14). The simple correlation between sPTH and s25D was also negative in this study, supporting the commonly held concept that as s25D decreases below about 30 ng/ml, vitamin D insufficiency develops, a manifestation of which is secondary hyperparathyroidism (8). However, such a causal relationship was not supported by multiple linear regression analysis, which demonstrated that weight itself accounted for both the decreased s25D and for the increased sPTH. Low s25D did not contribute to the elevated sPTH. The independence of sPTH from s25D levels in obesity has been noted by others (21). Furthermore, 1 yr after bariatric surgery, sPTH remained increased in 20% of patients despite normal s25D levels (22). Importantly, an 8-wk intervention study of vitamin D supplementation in obese subjects who were considered to be vitamin D insufficient showed no effect on intact sPTH and only a questionable decrease in sPTH (1–84) despite a substantial increase in s25D (23). Thus, our data support the hypothesis that the inverse relationship between sPTH and s25D in obesity is not causative, but that both biochemical abnormalities are a direct effect of obesity itself. The presence of an inverse relationship between sPTH and s25D cannot be taken to indicate the presence of vitamin D deficiency in obesity.
The cause of low s25D in obesity is unclear. A number of suggestions have been put forward, including sequestration of vitamin D by fat, making it less bioavailable for conversion to s1,25D (24, 25) and decreased sunlight exposure (23). The latter appeared to be a factor in our study because the expected seasonal variation in s25D did not occur in the bariatric-surgery women but did in the controls. Possible explanations for the lack of seasonal variation in s25D are that obese women expose a minimum amount of skin for cosmetic reasons and exercise outdoors less often because of general immobility from increased body mass and osteoarthritis. In this study, sun exposure was not estimated, but the bariatric-surgery patients with osteoarthritis had the lowest s25D levels. In addition to weight, a proportion of the decrease in s25D in this study was accounted for by age, which is a well documented association (26).
Our finding of increased sPTH in obesity agrees with other studies (27). Multiple regression analysis demonstrated that weight was the main determinant of increased sPTH. There was no contribution to sPTH by s25D, sCa, or renal function. The pathogenesis of hyperparathyroidism in obesity remains unexplained. Hyperparathyroidism in morbid obesity regresses with weight loss (28), and we have shown that it develops and regresses with weight changes in healthy women (29), suggesting a causal relationship. A plausible mechanism may be a direct effect of adipokines on PTH secretion. As might be expected, a positive relationship between sPTH and serum leptin has been shown in obesity (30). Furthermore, in the leptin-deficient (ob/ob) mouse, injections of leptin greatly increase sPTH (31), suggesting that leptin is a PTH secretagogue. The positive relationships between serum sPTH and leptin may underlie the well documented direct relationship between bone mass and body fat mass (32) because PTH regulates sclerostin secretion (33), which is a key regulator of bone mass (34).
In the current study, both bone formation and resorption markers were increased in the bariatric-surgery patients. Of the variables measured, weight accounted most for the increase in sBAP, perhaps also suggesting a direct effect of adipokines on bone formation. The leptin-deficient (ob/ob) mouse has low bone mineral density and shortened femora (31), and leptin increases osteoblast differentiation and proliferation in vitro (35, 36) and reduces bone fragility (36). Thus, there is evidence for a direct action of leptin on bone turnover. s1,25D was higher in obese than normal-weight women and predicted bone formation independent of weight. However, published data are mixed regarding s1,25D levels in obesity; some studies showed increased s1,25D levels (9), whereas others found the opposite (37).
The bariatric-surgery subjects in this study had increased sCr. Obesity is associated with renal impairment through a number of mechanisms including hyperfiltration, hypertension, diabetes, and glomerular fibrosis (6). Chronic renal failure leads to secondary hyperparathyroidism, but in this study, the sCr did not contribute to the increased sPTH or decreased s25D.
The only obesity comorbid condition that was found to be associated with s25D in this study was the occurrence of osteoarthritis. The most likely cause underlying the association is reduced outdoor activity resulting in decreased sunlight exposure (19).
There are several strengths to this study. The sample of women having bariatric surgery was compared with a sample of healthy women, some of whom were obese, matched for age and season of study, and was a relatively large sample. Because only White women were studied, however, the findings may not hold for men or Black women. Because we did not measure adipokines, some of the possible underlying mechanisms underlying the effects of obesity could not be examined. Also, the analysis was cross-sectional, making it difficult to infer causality. A limitation in the analysis of vitamin D status in comorbid conditions of obesity was that diagnoses were taken from patient charts and were not confirmed by physical exam or biochemical assessment.
In summary, we found that sPTH is elevated in obesity and that although s25D levels were low, the increase in sPTH was due to an increase in weight and not due to s25D levels considered to be in the insufficiency range. Additionally, bone turnover was increased in obesity, but again as a direct effect of weight and not to hyperparathyroidism. Except for osteoarthritis, there was no relationship between comorbid diseases associated with obesity and s25D. In osteoarthritis, the s25D was likely reduced because of low exposure to sunlight and not causative on the osteoarthritic process. The low s25D and high sPTH levels in obesity illustrate the problem of defining vitamin D insufficiency by the simple inverse relationship between sPTH and s25D (8). Although it is well documented that secondary hyperparathyroidism is a feature of vitamin D deficiency, the level of s25D insufficiency at which hyperparathyroidism manifests, the mechanisms involved, and the role of confounding conditions such as obesity render this definition of vitamin D insufficiency at best problematic and at worst clinically misleading. Obesity is becoming an increasing feature of the normal population, and the s25D concentration that sustains vitamin D sufficiency in obese subjects needs to be established.
Acknowledgments
This work was supported by UL1RR025761.
Disclosure Summary: E.G., R.M., C.E.G., R.J., B.M.C., D.D., A.D.F., S.M.P., R.V.C., and M.P. have nothing to declare.
Footnotes
- BMI
- Body mass index
- CV
- coefficient of variation
- GERD
- gastroesophageal reflux disease
- NTX
- cross-linked N-telopeptides of type 1 collagen
- sBAP
- serum bone alkaline phosphatase
- sCA
- serum calcium
- sCr
- serum creatinine
- s1,25D
- serum 1,25-dihydroxyvitamin D
- s25D
- serum 25-hydroxyvitamin D
- sPi
- serum phosphate
- SR2
- squared semipartial correlations.
References
- 1. Centers for Disease Control and Prevention 2010. U.S. obesity trends: trends by state 1985–2009. http://www.cdc.gov/obesity/data/trends.html
- 2. Berghöfer A, Pischon T, Reinhold T, Apovian CM, Sharma AM, Willich SN. 2008. Obesity prevalence from a European perspective: a systematic review. BMC Public Health 8:200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. 1999. The disease burden associated with overweight and obesity. JAMA 282:1523–1529 [DOI] [PubMed] [Google Scholar]
- 4. Festi D, Scaioli E, Baldi F, Vestito A, Pasqui F, Di Biase AR, Colecchia A. 2009. Body weight, lifestyle, dietary habits and gastroesophageal reflux disease. World J Gastroenterol 15:1690–1701 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lopez PP, Stefan B, Schulman CI, Byers PM. 2008. Prevalence of sleep apnea in morbidly obese patients who presented for weight loss surgery evaluation: more evidence for routine screening for obstructive sleep apnea before weight loss surgery. Am Surg 74:834–838 [PubMed] [Google Scholar]
- 6. Chalmers L, Kaskel FJ, Bamgbola O. 2006. The role of obesity and its bioclinical correlates in the progression of chronic kidney disease. Adv Chronic Kidney Dis 13:352–364 [DOI] [PubMed] [Google Scholar]
- 7. Peacock M. 1993. Osteomalacia and rickets. In: Nordin BEC, Need AG, Morris HA. eds. Metabolic bone and stone disease. London: Churchill Livingstone; 83–118 [Google Scholar]
- 8. Chapuy MC, Preziosi P, Maamer M, Arnaud S, Galan P, Hercberg S, Meunier PJ. 1997. Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int 7:439–443 [DOI] [PubMed] [Google Scholar]
- 9. Bell NH, Epstein S, Greene A, Shary J, Oexmann MJ, Shaw S. 1985. Evidence for alteration of the vitamin D-endocrine system in obese subjects. J Clin Invest 76:370–373 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Buffington C, Walker B, Cowan GS, Jr, Scruggs D. 1993. Vitamin D Deficiency in the Morbidly Obese. Obes Surg 3:421–424 [DOI] [PubMed] [Google Scholar]
- 11. Looker AC. 2005. Body fat and vitamin D status in Black versus White women. J Clin Endocrinol Metab 90:635–640 [DOI] [PubMed] [Google Scholar]
- 12. Snijder MB, van Dam RM, Visser M, Deeg DJ, Dekker JM, Bouter LM, Seidell JC, Lips P. 2005. Adiposity in relation to vitamin D status and parathyroid hormone levels: a population-based study in older men and women. J Clin Endocrinol Metab 90:4119–4123 [DOI] [PubMed] [Google Scholar]
- 13. Carlin AM, Rao DS, Meslemani AM, Genaw JA, Parikh NJ, Levy S, Bhan A, Talpos GB. 2006. Prevalence of vitamin D depletion among morbidly obese patients seeking gastric bypass surgery. Surg Obes Relat Dis 2:98–103 [DOI] [PubMed] [Google Scholar]
- 14. Goldner WS, Stoner JA, Thompson J, Taylor K, Larson L, Erickson J, McBride C. 2008. Prevalence of vitamin D insufficiency and deficiency in morbidly obese patients: a comparison with non-obese controls. Obes Surg 18:145–150 [DOI] [PubMed] [Google Scholar]
- 15. Hamoui N, Anthone G, Crookes PF. 2004. Calcium metabolism in the morbidly obese. Obes Surg 14:9–12 [DOI] [PubMed] [Google Scholar]
- 16. Liu E, Meigs JB, Pittas AG, McKeown NM, Economos CD, Booth SL, Jacques PF. 2009. Plasma 25-hydroxyvitamin D is associated with markers of the insulin resistant phenotype in nondiabetic adults. J Nutr 139:329–334 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Parker J, Hashmi O, Dutton D, Mavrodaris A, Stranges S, Kandala NB, Clarke A, Franco OH. 2010. Levels of vitamin D and cardiometabolic disorders: systematic review and meta-analysis. Maturitas 65:225–236 [DOI] [PubMed] [Google Scholar]
- 18. Forman JP, Curhan GC, Taylor EN. 2008. Plasma 25-hydroxyvitamin D levels and risk of incident hypertension among young women. Hypertension 52:828–832 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Ding C, Cicuttini F, Parameswaran V, Burgess J, Quinn S, Jones G. 2009. Serum levels of vitamin D, sunlight exposure, and knee cartilage loss in older adults: the Tasmanian older adult cohort study. Arthritis Rheum 60:1381–1389 [DOI] [PubMed] [Google Scholar]
- 20. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. 1985. Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28:412–419 [DOI] [PubMed] [Google Scholar]
- 21. Pitroda AP, Harris SS, Dawson-Hughes B. 2009. The association of adiposity with parathyroid hormone in healthy older adults. Endocrine 36:218–223 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Fish E, Beverstein G, Olson D, Reinhardt S, Garren M, Gould J. 2010. Vitamin D status of morbidly obese bariatric surgery patients. J Surg Res 164:198–202 [DOI] [PubMed] [Google Scholar]
- 23. Stein EM, Strain G, Sinha N, Ortiz D, Pomp A, Dakin G, McMahon DJ, Bockman R, Silverberg SJ. 2009. Vitamin D insufficiency prior to bariatric surgery: risk factors and a pilot treatment study. Clin Endocrinol (Oxf) 71:176–183 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF. 2000. Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 72:690–693 [DOI] [PubMed] [Google Scholar]
- 25. Blum M, Dolnikowski G, Seyoum E, Harris SS, Booth SL, Peterson J, Saltzman E, Dawson-Hughes B. 2008. Vitamin D3 in fat tissue. Endocrine 33:90–94 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Baker MR, Peacock M, Nordin BE. 1980. The decline in vitamin D status with age. Age Ageing 9:249–252 [DOI] [PubMed] [Google Scholar]
- 27. Kamycheva E, Sundsfjord J, Jorde R. 2004. Serum parathyroid hormone level is associated with body mass index. The 5th Tromso study. Eur J Endocrinol 151:167–172 [DOI] [PubMed] [Google Scholar]
- 28. Andersen T, McNair P, Hyldstrup L, Fogh-Andersen N, Nielsen TT, Astrup A, Transbøl I. 1988. Secondary hyperparathyroidism of morbid obesity regresses during weight reduction. Metabolism 37:425–428 [DOI] [PubMed] [Google Scholar]
- 29. Gunther CW, Legowski PA, Lyle RM, Weaver CM, McCabe LD, McCabe GP, Peacock M, Teegarden D. 2006. Parathyroid hormone is associated with decreased fat mass in young healthy women. Int J Obes (Lond) 30:94–99 [DOI] [PubMed] [Google Scholar]
- 30. Maetani M, Maskarinec G, Franke AA, Cooney RV. 2009. Association of leptin, 25-hydroxyvitamin D, and parathyroid hormone in women. Nutr Cancer 61:225–231 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Matsunuma A, Kawane T, Maeda T, Hamada S, Horiuchi N. 2004. Leptin corrects increased gene expression of renal 25-hydroxyvitamin D3–1 α-hydroxylase and -24-hydroxylase in leptin-deficient, ob/ob mice. Endocrinology 145:1367–1375 [DOI] [PubMed] [Google Scholar]
- 32. Reid IR, Ames R, Evans MC, Sharpe S, Gamble G, France JT, Lim TM, Cundy TF. 1992. Determinants of total body and regional bone mineral density in normal postmenopausal women: a key role for fat mass. J Clin Endocrinol Metab 75:45–51 [DOI] [PubMed] [Google Scholar]
- 33. Bellido T, Ali AA, Gubrij I, Plotkin LI, Fu Q, O'Brien CA, Manolagas SC, Jilka RL. 2005. Chronic elevation of parathyroid hormone in mice reduces expression of sclerostin by osteocytes: a novel mechanism for hormonal control of osteoblastogenesis. Endocrinology 146:4577–4583 [DOI] [PubMed] [Google Scholar]
- 34. Winkler DG, Sutherland MK, Geoghegan JC, Yu C, Hayes T, Skonier JE, Shpektor D, Jonas M, Kovacevich BR, Staehling-Hampton K, Appleby M, Brunkow ME, Latham JA. 2003. Osteocyte control of bone formation via sclerostin, a novel BMP antagonist. EMBO J 22:6267–6276 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Gordeladze JO, Drevon CA, Syversen U, Reseland JE. 2002. Leptin stimulates human osteoblastic cell proliferation, de novo collagen synthesis, and mineralization: Impact on differentiation markers, apoptosis, and osteoclastic signaling. J Cell Biochem 85:825–836 [DOI] [PubMed] [Google Scholar]
- 36. Cornish J, Callon KE, Bava U, Lin C, Naot D, Hill BL, Grey AB, Broom N, Myers DE, Nicholson GC, Reid IR. 2002. Leptin directly regulates bone cell function in vitro and reduces bone fragility in vivo. J Endocrinol 175:405–415 [DOI] [PubMed] [Google Scholar]
- 37. Konradsen S, Ag H, Lindberg F, Hexeberg S, Jorde R. 2008. Serum 1,25-dihydroxy vitamin D is inversely associated with body mass index. Eur J Nutr 47:87–91 [DOI] [PubMed] [Google Scholar]