True Fractional Calcium Absorption is Decreased After Roux-En-Y Gastric Bypass Surgery

Abstract

Objective

Roux-en-Y gastric bypass (RYGB) is considered to be the gold standard alternative treatment for severe obesity. Weight loss after RYGB results primarily from decreased food intake. Inadequate calcium (Ca) intake and metabolic bone disease can occur after gastric bypass. To our knowledge, whether malabsorption of Ca contributes to an altered Ca metabolism in the RYGB patient has not been addressed previously.

Research Methods and Procedures

We recruited 25 extremely obese women in order to study true fractional Ca absorption (TFCA) before and 6 months after RYGB surgery, using a dual stable isotope method (⁴²Ca and ⁴³Ca) and test load of Ca (200 mg). Hormones regulating Ca absorption and markers of bone turnover were also measured.

Results

In 21 women (BMI 52.7 ± 8.3 kg/m², age 43.9 ± 10.4 years) who successfully completed the study, TFCA decreased from 0.36 ± 0.08 to 0.24 ± 0.09 (p < 0.001) after RYGB. Bone turnover markers increased significantly (p < 0.01). TFCA correlated with estradiol levels (r = 0.512, p < 0.02) and tended to correlate with 1,25 (OH)₂D (r = 0.427, p < 0.06) at final measurement. Stepwise linear regression indicated that estradiol explained 62% of the variance for TFCA at 6 months post-surgery (p < 0.01).

Discussion

TFCA decreases (0.12 ± 0.08) after RYGB surgery but remains within normal range. Although only some patients were estimated to have low Ca absorption after surgery, all of the patients showed a dramatic increase in markers of bone resorption. The alteration in Ca metabolism after RYGB-induced weight loss appears to be regulated primarily by estradiol levels and might ultimately affect bone mass.

Introduction

Since the development of bariatric surgery, several surgical methods for the treatment of severe obesity have been developed over the past decades. Roux-en-Y gastric bypass (RYGB)¹ surgery is considered to be the gold standard alternative treatment for severe obesity because it should result in less severe malabsorption and complications than traditional malabsorptive procedures (i.e., jejuno-ileal bypass) (1–3). The malabsorptive procedures have been recognized as a risk factor for developing bone disease (4–9) as a result of altered calcium (Ca) metabolism and compromised Ca absorption (10–16). Only a few studies have investigated Ca absorption prospectively in jejuno-ileal bypass patients and have shown that Ca absorption decreases by ~50% after surgery (10,13,16). To our knowledge, the change in Ca absorption after RYGB surgery has not been addressed previously. In addition, because inadequate Ca intake is common after gastric bypass (17–19), this may also contribute to an altered Ca metabolism and bone loss. Understanding the extent to which Ca absorption and intakes are decreased after gastric bypass surgery was a goal of this study.

Hormones regulating Ca metabolism are often disturbed in severe obesity or after bariatric surgery. For example, serum parathyroid hormone (PTH) is increased in severely obese subjects, compared with non-obese subjects (20), and has been shown to decline with dramatic weight loss (21). Yet, persistently elevated PTH levels are commonly described after bariatric surgery (22–28). Decreased 25-hydroxy-vitamin D (25OHD) levels are also common findings among patients after bariatric surgery (20,22,25,29–32). In addition, serum estrogen levels are typically elevated in severe obesity (33) and decline with weight loss (26,34,35). This hormonal profile may be regulating the disturbed Ca metabolism and the decline in true fractional Ca absorption (TFCA) (36,37) after weight loss surgery. Bone-regulating hormones in bariatric patients and their relationship to Ca absorption and bone loss (9,25,26,30,38) have not been studied previously in bariatric patients, despite increasing numbers of this type of obesity treatment (1,39). In this study, we hypothesized that the decrease in TFCA after RYGB-induced massive weight loss can be partially explained by bone-regulating hormones.

Research Methods and Procedures

Subjects and Protocol

Twenty-five obese and extremely obese (BMI 39 to 74 kg/m²) women were recruited to participate in a 6-month prospective clinical trial. Recruitment was completed in local bariatric support group meetings over a 3-year period (2001 to 2004) and from individual patient interaction with the dietitian at the surgeon’s office. From 10 support group meetings, ~70 patients showed initial interest in participating in the study, of which 18 decided to enroll into the study. Individual meetings with the dietitian resulted in another 23 potential subjects who were invited to participate in the study, of which 7 enrolled in the study. Women filled out questionnaires to assess their medical and nutrition history. Women were excluded if they were taking any estrogen, progesterone, or osteoporosis medication, or if they previously had undergone any type of surgery for weight loss. The study was approved by Rutgers University Institutional Review Board, and all subjects signed an informed consent form. Weight loss was achieved through open or laparoscopic RYGB surgery with an RY-limb length of 75 or 150 cm. As part of standard practice, post-surgical patients were advised to take two children’s chewable multivitamin/mineral tablets daily, containing 400 IU vitamin D/tablet. Furthermore, they were encouraged to consume 1 g/d Ca supplement in addition to their dietary Ca intake.

Laboratory Methods

An electronic scale (ScaleTronix ST 5002; ScaleTronix, White Plains, NY) and a stadiometer (Detecto, Webb City, MO) were used to measure weight and height to the nearest 0.25 kg and 1.0 cm, respectively. Fasting urine and serum were collected before surgery (month 0), and again at 1, 3, and 6 months after surgery, and samples were measured for markers of bone turnover. Serum hormones were measured at baseline and at 6 months. TFCA was determined before and 6 months after surgery using dual stable Ca isotope methods, inductively coupled plasma mass spectrometry, and calculations, as described previously (37,40). Briefly, on the day of the Ca absorption test, women were admitted at 7 AM after an overnight fast. After blood collection (10 mL), subjects were asked to void and then were served a standard breakfast. This meal contained a total Ca load of 200 mg, with ⁴³Ca that had been mixed in half cup of skim milk (153 mg Ca), which had been equilibrated overnight (~12 hours). The milk was consumed in its entirety under supervision, and the cup was rinsed with deionized water three times. All of the rinse water was also consumed by patients. After surgery, although milk was always consumed in its entirety, some patients were unable to finish all of their other low Ca breakfast foods. Immediately after breakfast, an intravenous injection of ⁴²Ca was administered over ~3 minutes. Syringes containing the isotopes (that were mixed with the milk or infused intravenously) were weighed before and after administration on a precision balance scale. The inductively coupled plasma mass spectrometry instrument precision and accuracy for this method is <±1% and the day-to-day coefficient of variation (CV) for 6 women measured twice was 1.2%. Serum 1,25 dihydroxy-vitamin D (1,25(OH)₂D) and 25OHD were measured by ¹²⁵I radio-immunoassay (RIA; DiaSorin, Stillwater, MN) (CV: <15.3% and <6.7%, respectively). Estrone (E₁), estradiol (E₂), and cortisol were also measured by ¹²⁵I RIA (DSL, Webster, TX) (CV: <9.4%, <8.9%, <8.3%, respectively). Intact PTH was determined by immuno-radiometric assay (DSL, Webster, TX) (CV: <5.2%). Serum osteocalcin, a marker of bone formation, was measured by RIA (BTI, Stoughton, MA) (CV: <9.0%). Markers of bone resorption were also measured. Serum n-telopeptide of type I collagen (sNTx) was measured by enzyme-linked immunosorbent assay (Osteomark, OSTEX International Inc., Seattle, WA) (CV: 4.6%). Pyridinoline (PYD, CV: < 8%) and deoxypyridinoline (CV: < 10%) were measured in 24-hour urine samples by reverse phase high performance liquid chromatography and fluorescence detection, as described previously (37). Urinary Ca and creatinine (No. 587, Sigma, St Louis, MO) (CV: <3.2%; and No. 555, CV: <11.0%) were also measured in 24-hour urine samples. Urinary creatinine was used to estimate skeletal muscle mass using the following formula: muscle mass = creat g/d*29.1 + 7.38 (41).

Dietary intakes were monitored by analyzing food records (3-day averages, at 0, 1, 3, and 6 months). Nutrients were analyzed by Nutritionist Pro software (version 2.1, First DataBank, Inc., San Bruno, CA).

Statistics

The Kolmogorov-Smirnov goodness-of-fit-test was applied to test for normal distribution of values at baseline and final measurement. Non-normally distributed variables were transformed for subsequent regression analyses. The changes (%) from baseline to final measurement were analyzed using one-way ANOVA. Pearson’s correlation coefficients were used to evaluate the associations among the different variables measured at baseline and final measurement, as well as their absolute and percentage changes. Stepwise multiple regression was performed to predict the dependent variable TFCA at baseline and at month 6, and independent variables included PTH, 1,25(OH)₂D, 25(OH)D, E₂, and Ca intake. The total estimated amount of Ca absorbed (mg/d) was calculated as the product of TFCA and Ca intake (mg/d). Predictors of the total estimated amount of Ca absorbed were assessed in a two-stage least squares analysis, with amounts of Ca absorbed as the dependent variable, and PTH, 1,25(OH)₂D, 25(OH)D, and E₂ as independent variables. To evaluate the effect of menopausal status on TFCA responses, we conducted separate analyses of covariance. p Values <0.05 were considered significant. Data are presented as mean ± standard deviation unless otherwise indicated. All analyses were conducted using the SAS statistical package (version 8.2; SAS Institute, Inc., Cary, NC).

Results

Of the 25 women recruited, 21 women were included in the final analysis. One woman was unable to complete the study due to complications after surgery. Three women had to be excluded from analysis because exclusion criteria were not revealed at the time of screening. On laboratory analysis of estradiol levels, it was apparent that two postmenopausal women were on hormone-replacement therapy, and one woman had a history of vertical banded gastro-plasty 11 years before RYGB. Of the 21 women included in the analysis, the Roux-en-Y limb length was 150 cm and 75 cm, in 17 and 4 women, respectively. Five women had the procedure done through open surgery, and 16 women underwent laparoscopic surgery. Baseline characteristics are shown in Table 1. The mean age of the 21 subjects was 43.9 ± 10.4 years (range 29 to 62 yrs), and BMI range was 38.9 to 73.5 kg/m². Based on creatinine levels at baseline, estimated skeletal muscle mass was 44.9 ± 11.5 kg before surgery.

Table 1.

Weight, BMI, calcium measurements, and creatinine before and after RYGB surgery (n = 21)^*

	Baseline	Final	Change (δ)	Change (%)	p value†
Weight (kg)	139.8 ± 23.0	101.3 ± 19.9	−38.5 ± 8.0	−27.7 ± 5.4	<0.0001
BMI (kg/m2)	52.7 ± 8.3	38.2 ± 7.3	−14.5 ± 2.9	−27.7 ± 5.4	<0.0001
TFCA (estimated from breakfast)	0.36 ± 0.08	0.24 ± 0.09	−0.12 ± 0.08	−34.0 ± 19.2	<0.0001
Ca absorbed (mg/d)‡	416.7 ± 271.3	227.2 ± 197.0	−189.5 ± 264.2	−38.7 ± 43.8	0.0006
Urine calcium (mg/d)	173.7 ± 87.8	89.5 ± 39.9	−84.2 ± 76.6	−39.3 ± 38.5	<0.0001
Urine creatinine (mg/d)	1542 ± 394	1076 ± 263	−463.2 ± 269.9	−28.2 ± 14.7	<0.0001

RYGB, Roux-en-Y gastric bypass; Ca, calcium; TFCA, true fractional Ca absorption.

^*Data (mean ± standard deviation).

^†ANOVA [for changes (%) from baseline to final measurements].

^‡Estimated Ca absorbed (TFCA × total Ca intake in Table 2).

Weight Loss and Nutrient Intake

Women lost 38.5 ± 8.0 kg (range: 16.9 to 49.5 kg) 6 months after surgery. Twenty-four hour urinary creatinine analysis showed a decrease of 463 ± 270 mg/d (p < 0.0001), suggesting an estimated loss of skeletal muscle mass of 35% of total body weight loss. As expected, caloric intake decreased after surgery (p < 0.0001), due to decreases (p < 0.001) in carbohydrate, protein, and fat intake (Table 2). Of the micronutrients assessed, only sodium and phosphorus intakes decreased (p < 0.0001) after surgery. Although dietary Ca intake decreased significantly (p < 0.0001) after surgery, total Ca intake remained the same due to an increase in Ca supplements. Ca supplementation was consumed by 48% of the subjects before surgery and by 66% after surgery. There tended (p < 0.08) to be greater vitamin D intake after surgery (Table 2) that was associated with greater Ca intake (r = 0.85, p < 0.001).

Table 2.

Nutrient intake before and after RYGB surgery (n = 21)^*

	Baseline	Final	p value
Kcal/d	2260 ± 738	794 ± 240	<0.0001
Carbohydrate (g/d)	247.1 ± 103.6	92.3 ± 40.2	<0.0001
Protein (g/d)	95.8 ± 37.2	42.3 ± 12.4	<0.0001
Fat (g/d)	96.1 ± 38.9	29.0 ± 11.1	<0.0001
Total Ca (mg/d)†	1104 ± 516	935 ± 679	0.4451
Vitamin D (μg/d)†	7.3 ± 5.9	11.0 ± 9.6	0.1126
Vitamin K (μg/d)†	51.9 ± 61.3	37.2 ± 32.9	0.1084
Phosphorus (mg/d)†	1241 ± 583	547 ± 175	<0.0001
Magnesium (mg/d)†	214 ± 111.2	218 ± 214	0.2665
Sodium (mg/d)	3785 ± 1418	1380 ± 628	<0.0001

RYGB, Roux-en-Y gastric bypass; Ca, calcium.

^*Data (mean ± standard deviation); ANOVA (for % changes from baseline to final measurements).

^†Nutrient amounts for Ca, vitamin D, vitamin K, phosphorus, and magnesium coming from supplement: baseline and final intake, respectively, for Ca: 209 ± 327 mg/d and 542 ± 699 mg/d; Vitamin D: 5.4 ± 5.9 μg/d and 9.7 ± 9.8 μg/d; Vitamin K: 4.9 ± 9.6 μg/d and 10.0 ± 14.0 μg/d; phosphorus: 38.2 ± 64.9 mg/d and 59.2 ± 115.5 mg/d; magnesium: 32.2 ± 56.9 mg/d and 107.8 ± 219.5 mg/d. Vitamin D: 1 μg = 40 IU.

Ca Absorption and Excretion

TFCA and other calcium variables before and 6 months after surgery (final values) are shown in Table 1. Individual changes in TFCA and estimated Ca absorbed from baseline to 6 months post-surgery are shown in Figure 1. TFCA, total Ca absorbed, and 24-hour urinary calcium excretion all decreased (p = 0.0001) (Table 1). The absolute decrease in TFCA was 0.12 ± 0.08. When women were divided by menopausal status, postmenopausal women (n = 9) had lower (p < 0.05) TFCA (0.32 ± 0.06) at baseline compared with premenopausal women (n = 12) (0.39 ± 0.08). Postmenopausal women also showed lower urinary Ca excretion (121.4 ± 61.0 mg/d) compared with premenopausal women (212.9 ± 86.1 mg/d) at baseline (p = 0.0003). There was no significant difference in the response to massive weight loss in any of the variables measured between pre- and postmenopausal women, whether or not values were corrected for baseline differences.

Figure 1.

(A) Individual changes (solid lines) in TFCA and (B) total estimated calcium absorbed (mg/d) from baseline to final measurement 6 months after RYGB surgery in 21 extremely obese women. Dashed line and diamonds represent group means at baseline and 6 months after RYGB surgery.

Hormones and Bone Turnover

Ca-regulating hormones and bone turnover markers before and 6 months after surgery (final values) are shown in Table 3. Serum levels of 25OHD were low (<25 ng/mL) in 11 subjects (52%) before and after surgery, but only 7 patients showed continuously low levels both before and after surgery (18.9 ± 5.8 ng/mL). Secondary hyperparathyroidism (>65 pg/mL) was prevalent in 15 women (71%) at baseline, as well as 6 months after surgery, of which 13 of these patients had elevated PTH levels both before and after surgery (94.1 ± 22.5 pg/mL). Ca absorption was not influenced by serum PTH. Estrogen levels did not differ between pre- (66.9 ± 51.9 pg/mL) and postmenopausal (36.6 ± 30.5 pg/mL) women at baseline, nor did they differ after weight loss (41.1 ± 31.0 and 28.4 ± 37.0 pg/mL, pre- and postmenopausal women, respectively). None of the hormonal markers measured changed with weight loss.

Table 3.

Calcium-regulating hormones and bone turnover markers before and after RYGB surgery (n = 21)^*

	Baseline	Final	Change (Δ)	Change (%)	p value†
1,25(OH)2D (pg/mL)	42.1 ± 16.4	45.6 ± 20.5	3.4 ± 20.9	17.8 ± 57.4	0.1700
25OHD (ng/mL)	25.4 ± 9.5	28.6 ± 14.3	3.2 ± 12.1	23.6 ± 68.9	0.1320
PTH (pg/mL)	81.3 ± 31.4	77.2 ± 28.3	−4.1 ± 30.6	3.4 ± 45.0	0.7354
Cortisol (μg/dL)	9.2 ± 3.0	10.5 ± 6.7	1.3 ± 7.1	31.5 ± 106.8	0.1916
Estradiol (pg/mL)	53.9 ± 45.7	35.7 ± 33.4	−18.2 ± 52.5	−13.1 ± 73.2	0.4222
Estrone (pg/mL)	69.6 ± 57.3	48.1 ± 48.6	−21.5 ± 73.2	−16.5 ± 58.8	0.2139
Osteocalcin (ng/mL)	10.2 ± 2.8	14.3 ± 2.9	4.1 ± 2.5	45.2 ± 29.0	<0.0001
sNTx (nM BCE)	16.6 ± 5.5	25.6 ± 6.8	9.0 ± 5.3	62.1 ± 44.2	<0.0001
PYD (nM/d)	436.8 ± 210.1	849.1 ± 244.5	409.9 ± 210.2	163.9 ± 234.8	<0.0001
DPD (nM/d)	50.2 ± 37.1	118.8 ± 71.1	67.9 ± 59.7	203.8 ± 318.2	0.0099

RYGB, Roux-en-Y gastric bypass; 1,25(OH)₂D, serum 1,25 dihydroxy-vitamin D; 25OHD, serum 25-hydroxy-vitamin D; PTH, parathyroid hormone; sNTx, serum N-telopeptide of type I collagen; PYD, pyridinoline; DPD, deoxypyridinoline.

^*Data (mean ± standard deviation).

^†ANOVA [for changes (%) from baseline to final measurements].

The bone resorption marker sNTx, was increased (p < 0.0001) as early as 1 month post-surgery, while bone formation marker, serum osteocalcin, showed a gradual rise and did not increase significantly until 3 months after surgery (p < 0.0003) (Figure 2). Urinary bone resorption markers, PYD and deoxypyridinoline, increased by ~2-fold (p < 0.01) 6 months after surgery.

Figure 2.

Changes (%) in bone resorption (sNTx) and bone formation (osteocalcin) markers from baseline to final measurement 6 months after RYGB surgery (p < 0.0001) in 21 extremely obese women. Diamonds and solid line represent sNTx; open squares and dashed line represent osteocalcin.

Correlations and multiple regression analysis

At baseline, TFCA was not correlated to any Ca-metabolism regulating hormones. After six months of weight loss, TFCA correlated with estradiol levels (r = 0.512, p < 0.02) and tended to correlate with 1,25(OH)₂D (r = 0.427, p < 0.06) levels. Not surprisingly, a higher TFCA correlated with increased urinary Ca excretion after surgery (r = 0.681, p < 0.001). A greater weight loss tended to be associated with a greater increase in bone resorption (sNTx: r = −0.423, p < 0.06). We found that women who increased dietary Ca intake after surgery (range of % change: −76 to +89) showed an increase in the bone formation marker osteocalcin (r = 0.518, p < 0.02), and tended to decrease serum PTH levels (r = −0.403, p < 0.07), but that Ca intake had no influence on resorption markers. In addition, total vitamin D intake after surgery (range of 0 to 30.4 μg/d) tended to correlate with serum 25OHD levels (r = 0.420, p < 0.06) and inversely correlate with PYD (r = −0.406, p < 0.08).

Stepwise multiple regression analysis showed that before surgery, Ca intake alone explained the variance in TFCA (52%, p < 0.02), while at six months after surgery only serum estradiol levels explained the variance in TFCA (62%, p < 0.01) (Figure 3) and was independent of weight loss (r = 0.132, p = 0.568). When pre- and postmenopausal women were analyzed separately, estradiol levels continued to predict TFCA in the postmenopausal (r = 0.732, p < 0.05) but not premenopausal women (r = 0.373, p = 0.283). The estimated amount of Ca absorbed after surgery was correlated to vitamin D intake (r = 0.803, p < 0.0001), and serum 25(OH)D explained 44% of the variance (r = 0.437, p < 0.05).

Figure 3.

Association between TFCA and estradiol levels (r = 0.512, p < 0.02) 6 months after RYGB surgery in 21 extremely obese women.

Discussion

The goal of this study was to examine the extent of Ca malabsorption associated with RYGB surgery and determine mechanisms regulating TFCA absorption and bone turnover before and after RYGB-induced massive weight loss. The results show that Ca absorption efficiency is relatively high at baseline, compared with previous reports (37,42), and although there is a 0.12 reduction of Ca absorption after RYGB surgery, TFCA values remain within normal range (at 0.24) for most women (42). Although the decrease in Ca absorption results in inadequate amounts of Ca absorbed during massive weight loss, we found no further alteration of the Ca-PTH axis when compared with before surgery. Furthermore, bone resorption is dramatically increased after surgery, relative to the rise in bone formation.

In previous studies, the percentage (rather than absolute) decrease in the fractional Ca absorption is reported. In the current dataset, there was a 34% decrease, which is less than that observed by Sellin et al. (10) in jejuno-ileal bypass patients (−50%) that was also measured 6 months after surgery. Other studies in jejuno-ileal bypass patients that measured Ca absorption at different time-points reported decreases in Ca absorption, compared with baseline of 43%, at 3 months (16), and 34% to 52% (13,16) 1 year after surgery. Based on these previous reports, we suggest that there is a less severe decrease of TFCA in RYGB patients compared with jejuno-ileal bypass, and that this may be due to more functional small intestine remaining with RYGB. However, to our knowledge, no study has compared Ca absorption in different methods of gastric bypass surgery to traditional jejuno-ileal bypass.

We expected a decrease in TFCA with moderate weight loss due to decreased caloric intake (37), hence, some of the decrease in TFCA could be due to a reduction in caloric intake, rather than solely attributed to intestinal bypass. Cifuentes et al. (37) found that moderate caloric restriction and a weight loss of 0.7 kg/wk in postmenopausal women resulted in a decrease in Ca absorption of 0.02 (baseline TFCA of 0.23). In the current study, after RYGB surgery, the more severe caloric restriction resulted in a 1.5 kg loss/wk and a decrease in absorption of 0.12 (baseline TFCA of 0.36). Hence, the rate of weight loss in the RYGB patients is about double compared with moderate weight loss, yet the decrease in Ca absorption is ~6 times greater. This decrease in Ca absorption could be due to a number of factors, including a change in gastric physiology due to the surgery alone, the effect of a more rapid weight loss, or a different pattern of consuming Ca and other nutrient intake (i.e., smaller more frequent meals). In addition, the decrease in Ca absorption spans a wide range (0.03 to 0.34), and the reasons for this remain unclear. Furthermore, we hypothesize that there is some adaptation to increase the new lower Ca absorption rate over time, since two patients measured for a third time at 18 months after RYGB surgery showed an increase in absorption of 0.05 ± 0.01 (absolute percentage increase of 5.4 ± 0.8%) from 6 to 18 months after surgery (preliminary data).

In the extremely obese women before surgery, we found no specific hormonal regulators of Ca absorption. After surgery, estradiol was the only hormonal predictor of Ca absorption. Estrogen has been shown to be a regulator of active Ca uptake in the duodenum through modulation of intestinal vitamin D receptors (43,44) and by vitamin D-independent effects (45), and a more recent finding has now shown a role of estrogen in Ca uptake in ileal and colonic cells (46). Ileal and colonic Ca uptake may play a more important role in gastric bypass patients, whose active Ca absorption sites in the upper intestine have been bypassed. The absence of an observed relationship between Ca absorption and estradiol levels in premenopausal women is likely due to highly fluctuating levels throughout the menstrual cycle. These results are similar to our findings, showing an association between Ca absorption and serum estradiol levels during caloric restriction in estrogen-deplete conditions but no association with higher estrogen levels (37,47).

25OHD levels were low in 52% of the subjects before surgery and did not change significantly after surgery, which is consistent with other studies (20,25,28,30). Although not a direct predictor of Ca absorption in our study, we show that subjects with lower 1,25(OH)₂D levels tended to have lower Ca absorption after surgery. 1,25(OH)₂D increases active cellular Ca absorption in the duodenum, proximal jejunum, and in the colon (48,49), a site that might become more important for Ca uptake in gastric bypass patients. For example, Grinstead et al. have shown that 1,25(OH)₂D can enhance colonic absorption in short bowel syndrome (49). There is also evidence that 1,25(OH)₂D may increase the paracellular diffusion of Ca through tight junctions (50–52). This would be important for the ileum, a site of passive absorption, where most of the Ca is absorbed due to the long sojourn time in this intestinal segment (53,54). Furthermore, there is also newer evidence of vitamin D-dependent active absorption in the ileum (55). Overall, despite bypassing most of the active absorptive sites of the duodenum and the majority of the jejunum, there is evidence that vitamin D can enhance Ca absorption along the remainder of the intestine by stimulating active transport as well as passive diffusion. In addition, a higher vitamin D intake after surgery, which was associated with greater serum 25OHD levels, was also associated with greater amounts of total estimated Ca absorbed. Those patients with higher vitamin D intake may also have greater nutrient intake, in general, including dietary Ca. Nevertheless, to maximize the low serum 25OHD levels in the bariatric patients and attenuate PTH (56) and bone resorption after surgery, higher vitamin D intake might be necessary.

Bone turnover increased after surgery, which is consistent with findings by Coates et al. (25). Typically, there is an increase in bone resorption with moderate weight loss (35,57). We found a more immediate and dramatic rise in biomarkers of bone resorption (60 to 200%) relative to bone formation (45%) with massive weight loss due to gastric bypass surgery. The observed “uncoupling” of bone formation and resorption markers in this study is similar to findings in bed rest studies, where a significant increase in bone resorption without concomitant increase in bone formation has been reported (58,59). Reduced mobilization and/or increased catabolism associated with surgery (60) in RYGB patients could partially explain the uncoupling of bone turnover immediately following surgery. However, the sustained increase in bone biomarkers suggests other regulators of bone turnover and mobilization and release of Ca from bone (i.e., decreased mechanical loading due to massive weight loss). Patients who increased their Ca intake after surgery showed greater increases in the bone formation biomarker compared with those consuming less Ca. It might be possible that a greater rate of bone formation in an increased state of bone turnover is beneficial in offsetting bone loss (25,38).

Normally, TFCA is decreased at high Ca intakes (61), yet this relationship was not observed either before or after gastric bypass surgery. TFCA is relatively high (0.36) in the severely obese patient before surgery, and, therefore, the decrease in TFCA (0.12) due to RYGB remains within normal range (0.24). This implies that some RYGB patients could absorb sufficient amounts of Ca after surgery, as long as dietary Ca intake is high enough. It is estimated that the total amount of Ca absorbed decreased by 190 mg/d or perhaps more, since there may be inaccurate reporting of nutrients, especially in the obese subject at baseline who often underestimate intakes (62,63). Assuming that 200 and 240 mg Ca absorbed/d is necessary to achieve at least zero Ca balance in pre- and postmenopausal women, respectively (37,64), a regression equation based on Ca absorbed and Ca intake shows that, theoretically, a Ca intake of ~800 mg/d for premenopausal and ~1500 mg/d for postmenopausal women is sufficient to achieve Ca balance after RYGB surgery. However, some women with lower absorption still would not achieve Ca balance at those intakes. To our knowledge, there is no specific recommended Ca intake after surgery, although most clinics are recommending 1.0 to 1.5 g Ca/d. Despite the fact that this level of Ca intake should achieve Ca balance in most women, the high levels of serum PTH both before and after surgery, and a marked and sustained increase in bone resorption, suggest that another mechanism is regulating bone loss, and this area requires further investigation.

Several studies have shown that RY limb length influences weight loss, with shorter limb length resulting in less weight loss (65–68), which led us to retrospectively examine whether RY limb length influences the degree of malabsorption. Although there were no differences in TFCA response or the amount of total Ca absorbed/d between patients with 75 cm compared with the majority with 150 cm RY limb length, our data are limited by the few patients with the shorter RY limb length.

In conclusion, to our knowledge, this is the first study to examine TFCA absorption after RYGB surgery. We found that serum estradiol levels influenced Ca absorption. In addition, despite the dramatic decrease in absorption due to surgery, absolute levels of absorption are not remarkably low due to high rates of Ca absorption before surgery. At least 50% of the postmenopausal women would have a negative Ca balance even with 1.2 grams Ca/d. The increased rate of bone turnover in all subjects appears to be regulated by some other mechanism that was not addressed in this paper. Although increased Ca intake can positively influence bone turnover after RYGB surgery, it is unlikely that further increases in Ca supplementation beyond current recommendations will attenuate the elevated levels of bone resorption. Despite bypassing most of the active sites for Ca absorption in these patients, Ca absorption is influenced by traditional Ca metabolism regulating hormones, estradiol, and vitamin D after RYGB surgery.