Relationships Between Vitamin D Status and PTH over 5 Years After Roux-en-Y Gastric Bypass: a Longitudinal Cohort Study

Purpose Secondary hyperparathyroidism (SHPT) after obesity surgery may affect bone health. Optimal vitamin D levels have not been established to prevent SHPT postoperatively. We investigated whether SHPT differed across threshold levels of serum 25-hydroxyvitamin D (S-25(OH)D) from 6 months up to 5 years after Roux-en-Y gastric bypass (RYGB). Materials and Methods We included 554 patients at follow-up 5 years postoperatively. Blood samples were analysed for S-25(OH)D, ionized calcium (iCa) and parathyroid hormone (PTH) during follow-up. Results PTH and prevalence of SHPT increased from 6 months to 5 years postoperatively, while S-25(OH)D and iCa decreased (all P < 0.001). PTH and SHPT development are related with S-25(OH)D, and PTH differed between all subgroups of S-25(OH)D. SHPT occurred less frequently across all subgroups of S-25(OH)D ≥ 50 nmol/l during follow-up: odds ratio (OR) 0.44 (95% CI 0.36–0.54) in patients with S-25(OH)D ≥ 50 nmol/l, OR 0.38 (0.30–0.49) with S-25(OH)D ≥ 75 nmol/l and OR 0.19 (0.12–0.31) with S-25(OH) D ≥ 100 nmol/l, all compared with S-25(OH)D < 50 nmol/l. At 5 years, 208/554 patients (38%) had SHPT; SHPT was found in 94/188 patients (50%) with S-25(OH)D < 50 nmol/l, in 69/222 (31%) with S-25(OH)D 50–74 nmol/l, in 40/117 (34%) with S-25(OH)D 75–99 nmol/l and in 5/27 (19%) with S-25(OH)D ≥ 100 nmol/l. An interaction existed between S-25(OH)D and iCa. Bone alkaline phosphatase remained increased with SHPT. Conclusions A significant relationship existed between S-25(OH)D and development of PTH and SHPT. The prevalence of SHPT was lower with threshold levels 25(OH)D ≥ 50 nmol/l and ≥ 75 nmol/l over the 5 years, and lowest with S-25(OH)D ≥ 100 nmol/l. Electronic supplementary material The online version of this article (10.1007/s11695-020-04582-5) contains supplementary material, which is available to authorized users.

We aimed to study the development of PTH and SHPT over 5 years after Roux-en-Y gastric bypass (RYGB), and relationships between these and different threshold levels of S-25(OH)D ≥ 50 nmol/l. We assessed whether the prevalence of SHPT would be lower among patients with higher S-25(OH)D up to 5 years postoperatively.

Patients and Study Design
This longitudinal observational cohort study was analysed prospectively, and the report was written to comply with the STROBE checklist [27].
Morbid obesity was defined as BMI ≥ 40 kg/m 2 , or BMI ≥ 35 kg/m 2 with obesity-related comorbidities [28]. Obesity surgery was offered at Oslo University Hospital, Aker, after failed weight loss by other means. Laparoscopic RYGB was the preferred procedure in the period, with construction of a gastric pouch 25-30 ml, a 150-cm antegastric, antecolic alimentary limb and a 50-cm biliopancreatic limb [29,30].
We aimed for minimum 500 patients in a population with high follow-up. Candidates were RYGB patients operated 2004-2009. They were evaluated preoperatively and postoperatively with weight, height and blood samples. Follow-up visits were after 6 weeks, 6 months, 1 year, 2 years, 3-4 years and 5 years. At 5 years, all were contacted by letter and eventually by telephone. Body weight was measured electronically (platform weight, Seca 635 0-300 kg, class III), height with wall-fixed steel measure and blood samples were drawn after overnight fast.
Patients with signed consent and valid PTH and S-25(OH)D at 5 years were candidates, while patients with primary hyperparathyroidism and elevated creatinine were excluded.

Supplementation
Recommended daily supplements included one multivitamin (cholecalciferol 200 IU) and two combination tablets, each containing calcium carbonate 500 mg and cholecalciferol 400 IU. Compliance was defined by use of calcium ≥ 500 mg and vitamin D ≥ 600 IU minimum 5 days a week, noncompliance as less or no use. Supplements were adjusted to keep blood values within normal reference range, from 2012 to maintain S-25(OH)D ≥ 50 mmol/l, or S-25(OH)D ≥ 75 nmol/l in cases with SHPT [7,8]. Our supplementation regimen also included oral iron (100 mg daily) and intramuscular vitamin B12 injections (1 mg per 3 months).

Statistical Analyses
Statistical analyses were performed with IBM SPSS for Windows, version 25. Continuous and categorical variables were tested with t test and chi-square test as appropriate. Regression analyses were performed with linear mixed model, diagonal covariance matrices for PTH and B-ALP using individual repeated measurements, time-dependent covariates and random intercept. Variables were included in multivariate analyses of PTH with stepwise backward elimination of nonsignificant variables. Gender, age and BMI were included as covariates. We tested multiplicative interactions for S-25(OH)D, iCa and time on PTH, and PTH and time on B-ALP. B-ALP was adjusted for gender. Generalized estimating equations (GEE), unstructured covariance matrices were used for SHPT. Missing data were not imputed. We analysed two periods: from baseline to 6 months postoperatively and from 6 months to 5 years. The second period was the main focus, as S-25(OH)D and PTH were routinely assessed. Continuous variables are presented with means and standard deviations (± SD), categorical variables in percentages, and odds ratios (OR) and relative ratios (RR) with 95% confidence intervals (95% CI).

Results
Of 823 operated, 584 (71%) attended 5-year follow-up. Included were 554 patients (67%), after exclusion of 4 with no signed consent, 10 with a suspicious primary hyperparathyroidism, 3 with elevated creatinine, and 13 with missing data of S-25(OH)D and PTH at 5 years. Follow-up period was 5.3 ± 0.4 years. Three patients had moved from our region and could not be contacted, and 7 had died during the 5 years. Table 1 summarizes preoperative characteristics.

Vitamin D, PTH and SHPT
S-25(OH)D was inversely related with development of PTH and occurrence of SHPT from 6 months to 5 years (P < 0.001), and PTH differed between all subgroups of S-25(OH)D (P < 0.001) (Fig. 1). Mean PTH response by change in vitamin D levels was.

Ionized Calcium and BMI
Serum iCa decreased in all subgroups of vitamin D from 6 months to 5 years postoperatively, and percentage with iCa in the lower range (iCa ≤1.23) increased from 26% to 61% (P < 0.001) (Fig. 4). PTH increased in all 3 subgroups of iCa over 5 years, and was higher in the lower range. iCa was inversely related with PTH development (P < 0.001). Correspondingly, SHPT was consistently lower with iCa above lower range during follow-up.
An interaction existed between S-25(OH)D and iCa. At 5 years, SHPT was most prevalent with combined vitamin D deficiency and iCa in the lower range, and there was no SHPT with S-25(OH)D ≥ 100 nmol/l and iCa above the lower range (Fig. 3).

Multivariate Analyses
In multivariate analyses of PTH from 6 months to 5 years (Table 3), the relationship with S-25(OH)D and iCa remained robust, also with BMI. Interactions for S-25(OH)D and iCa with time over the 5 years were not significant in multivariate analyses. SHPT occurred more often in men, with lower age and with higher BMI over the period (Supplementary Table). The trend of lower OR with higher S-25(OH)D remained across strata of gender, age and BMI.    and PTH lower in users over the period (P < 0.001), but not significantly at 5 years. PTH related positively with B-ALP and ALP over the 5 years (P < 0.001), with interaction for PTH and time. B-ALP declined by 1.3 U/l per year, but remained increased with SHPT (P < 0.001 for group difference). Total ALP declined slower with SHPT (P = 0.004).

Discussion
PTH and the prevalence of SHPT decreased the first 6 months after RYGB, and thereafter increased up to 5 years. S-25(OH)D was related with PTH development. Patients with vitamin D deficiency had the largest increase in PTH from 6 months, while patients with high S-25(OH)D ≥ 100 nmol/l had lowest PTH and prevalence of SHPT.
The observed increase in PTH from 6 months is in accordance with other long-term evaluations [9][10][11][12][13]. S-25(OH)D was strongly related with PTH over time, however, with limited differences in SHPT by traditional target thresholds,

Calcium and SHPT
Calcium absorption seems reduced after RYGB [31,32]. However, few have reported a relationship between calcium and PTH [18,20,23]. Extracellular calcium is a determinant of PTH secretion, and even calcium within the lower normal range may increase PTH [20,33,34]. With the feedback mechanisms involved, PTH may increase above reference range, while serum calcium still remains within normal reference range.
Our observations suggest a role of iCa on PTH. The proportion of patients with iCa in the lower range increased over time, and iCa was related with PTH development. This relationship was independent of S-25(OH)D. An interaction existed between S-25(OH)D and iCa on PTH. Still, iCa declined in all subgroups of S-25(OH)D but more slowly with higher levels.

Supplementation, SHPT and Bone Effects
Several studies have failed to document benefits of vitamin D and calcium supplementation, and the regimens have been questioned [22,24]. This study supports a modest effect with supplements of calcium ≥ 500 mg and vitamin D ≥ 600 IU on PTH. These doses are however lower than recommended by many [2,4,6].
We also found higher B-ALP in patients with SHPT up to 5 years postoperatively, suggesting higher bone turnover. SHPT might therefore help explain increased bone turnover, which is observed up to 5 years after RYGB [18,19,22]. SHPT may also lead to reduced BMD, which is observed after weight stabilizes 1-2 years postoperatively [15,19]. We recently found SHPT related with lower BMD 10 years after RYGB [21].

Implications
In clinical practice, SHPT may be considered as a marker of vitamin D and calcium insufficiency, and it is of concern after RYGB. Further research should address whether increasing S-25(OH)D levels can suppress SHPT and improve bone health. Higher doses seem necessary to achieve sufficient vitamin D levels and suppress SHPT after RYGB [2,4,24,35].
Attention to calcium status seems relevant to identify risk for SHPT. SHPT was more frequent with iCa in the lower range. Higher S-25(OH)D can increase calcium levels and lower PTH, and some individuals may need higher vitamin D levels than others [20,26,36]. The interaction between S-25(OH)D and iCa may be relevant in clinical practice.
SHPT was not prevalent with S-25(OH)D ≥ 100 nmol/l and iCa in the upper two tertiles of reference range at 5 years, which we previously reported 2 years postoperatively [20].
Optimal S-25(OH)D levels are not established after RYGB and obesity surgery in general. Achieving S-25(OH)D ≥ 100 nmol/l may be needed to suppress SHPT more effectively in some individuals, with an aim to improve long-term bone health.

Strengths and Limitations
The main strength of this study was a large sample size with high 5-year follow-up rate and repeated measurements, providing statistical strength. Bias of primary hyperparathyroidism was minimized. The single centre study design with standard surgical and follow-up procedures strengthen internal validity, but the findings need testing in other populations. As relationships up to 5 years were the primary focus, only patients with 5-year data on S-25(OH)D and PTH were included. Prevalence of SHPT in nonattenders may be higher than observed, assuming less compliance [37]. Data on preoperative S-25(OH)D and PTH, and S-25(OH)D ≥ 100 nmol/l and B-ALP, during follow-up were limited. Supplemental use was self-reported. iCa determinations at the Hormone Laboratory were adjusted during the period; however, the drop in iCa was also found in some parallel analyses performed at the Central Laboratory, with unchanged methodology.

Conclusions
S-25(OH)D levels related inversely with PTH development and occurrence of SHPT up to 5 years after RYGB. The prevalence of SHPT was lower with S-25(OH)D thresholds ≥ 50 nmol/l and ≥ 75 nmol/l. Some patients may need S-25(OH)D ≥ 100 nmol/l to suppress SHPT more effectively.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.