AIM: To investigate the effect of vitamin D (VD) concentrations and VD supplementation on health related quality of life in inflammatory bowel disease (IBD) patients.

METHODS: A cohort of 220 IBD patients including 141 Crohn’s disease (CD) and 79 ulcerative colitis (UC) patients was followed-up at a tertiary IBD center. A subgroup of the cohort (n = 26) took VD supplements for > 3 mo. Health related quality of life was assessed using the short IBD questionnaire (sIBDQ). VD serum concentration and sIBDQ score were assessed between August and October 2012 (summer/autumn period) and between February and April 2013 (winter/spring period). The mean VD serum concentration and its correlation with disease activity of CD were determined for each season separately. In a subgroup of patients, the effects of VD supplementation on winter VD serum concentration, change in VD serum concentration from summer to winter, and winter sIBDQ score were analyzed.

RESULTS: During the summer/autumn and the winter/spring period, 28% and 42% of IBD patients were VD-deficient (< 20 ng/mL), respectively. In the winter/spring period, there was a significant correlation between sIBDQ score and VD serum concentration in UC patients (r = 0.35, P = 0.02), with a trend towards significance in CD patients (r = 0.17, P = 0.06). In the winter/spring period, VD-insufficient patients (< 30 ng/mL) had a significantly lower mean sIBDQ score than VD-sufficient patients; this was true of both UC (48.3 ± 2.3 vs 56.7 ± 3.4, P = 0.04) and CD (55.7 ± 1.25 vs 60.8 ± 2.14, P = 0.04) patients. In all analyzed scenarios (UC/CD, the summer/autumn period and the winter/spring period), health related quality of life was the highest in patients with VD serum concentrations of 50-59 ng/mL. Supplementation with a median of 800 IU/d VD day did not influence VD serum concentration or the sIBDQ score.

CONCLUSION: VD serum concentration correlated with health related quality of life in UC and CD patients during the winter/spring period.

The well known north-south gradient of inflammatory bowel disease (IBD) prevalence, epidemiological data, data from animal models, and genetic association studies of vitamin D (VD) receptor polymorphisms suggest that VD plays an important role in the pathogenesis of IBD[1-3]. There is increasing evidence of the influence of VD on disease activity and its possible therapeutic use.

The role of VD in calcium metabolism and healthy bone development has been recognized for more than a century. However, the discovery that the VD receptor is present in most tissues and cells of the body provided new insight into the non-calcemic functions of VD, such as cell proliferation, immunomodulation and cell differentiation[4,5]. The VD receptor is abundantly expressed in immune cells. VD stimulates the production of T-regulatory lymphocytes and expression of interleukin 4 and transforming growth factor β, inhibits the differentiation of CD4⁺ T lymphocytes into Th1 cells, and suppresses the effector functions of T lymphocytes[4]. VD deficiency is suggested to play a role in the pathogenesis of various immune-mediated diseases, including IBD, multiple sclerosis, rheumatoid arthritis, type I diabetes, systemic lupus erythematosus and psoriasis[6]. VD receptor agonists have favorable effects in animal models of these diseases[7].

Modern lifestyles appear to result in widespread VD deficiency, especially at the end of winter[8,9]. In a large population study from Germany, 81% of men and 89% of women had a VD intake from food of lower than 200 IU/d and 57% of men and 58% of women were VD-deficient (< 20 ng/mL)[9].

VD deficiency is common among IBD patients, including those who have been recently diagnosed with such diseases[10-12]. The positive effects of VD are well documented in many animal models of IBD[7]. Limited clinical data suggest that there is an association between low VD concentration and increased disease activity in ulcerative colitis (UC) and Crohn’s disease (CD)[13,14].

There are limited clinical data on the effect of VD replacement therapy on disease activity or health related quality of life. Miheller et a[15] observed that the CD activity index (CDAI) score and the concentration of C-reactive protein were significantly reduced in a small cohort of 35 CD patients in remission who received 1000 IU VD supplementation per day. In a small placebo-controlled randomized trial of CD patients, oral supplementation with 1200 IU VD3/day reduced the risk of relapse from 29% to 13% (P = 0.06)[16]. However, in another small prospective study, VD replacement therapy did not influence disease activity in IBD patients[17].

Given the controversy concerning the effects of VD in IBD patients, the primary aim of this study was to evaluate the effect of VD serum concentration on health related quality of life in CD and UC patients. In addition, we also assessed the impact of supplementation with currently recommended doses of VD on the serum concentration of VD and health related quality of life.

This was a retrospective study of a cohort of IBD patients that were followed-up at a tertiary IBD center between August 1, 2012 and April 30, 2013.

During the summer/autumn period and the winter/spring period, 28% and 42% of IBD patients in this study were VD-deficient (< 20 ng/mL), respectively. A further 31% were VD-insufficient (20-29.9 ng/mL) in both seasons. The prevalence of VD deficiency and insufficiency did not differ between CD and UC patients. This is in agreement with several other Western studies that reported that the prevalence of VD deficiency (< 20 ng/mL) was 18%-39% and 50%-57% at the end of the summer/autumn period and the winter/spring period, respectively, and that about a further 35% of people were VD-insufficient (20-29.9 ng/mL)[12,21-23]. In the current study, the mean VD serum concentration was 4.2 ng/mL higher in the summer/autumn period than in the winter/spring period, which is similar to the seasonal difference reported by Bours (3 ng/mL)[23]. This difference in VD serum concentration between seasons is logical because production of VD3 depends heavily on exposure to sunshine. In the West, only a small amount of VD is obtained from the diet.

sIBDQ score, an indicator of disease specific health related quality of life, was significantly higher in the summer/autumn period than in the winter/spring period in UC patients (54.8 ± 1.6 vs 51.3 ± 2.1, P = 0.02) but not in CD patients. There is considerable controversy concerning seasonal variation in disease activity in CD and UC patients[24]. Some studies reported that the rate of flare-ups is low in the summer period, particularly in UC patients[25,26].

In the current study, a low VD serum concentration (< 30 ng/mL) was associated with a lower sIBDQ score in the winter/spring period in both CD and UC patients. In UC patients, the winter/spring period sIBDQ score correlated significantly with the winter/spring period VD serum concentration (r = 0.35) and the winter/spring period sIBDQ score correlated even more strongly with the mean of winter/spring period and the summer/autumn period VD serum concentrations (r = 0.46). In CD patients, there was a trend towards significant correlation between the winter/spring period sIBDQ score and the winter/spring period VD serum concentration (r = 0.17, P = 0.06).

Several recent studies made similar observations. In a recent trial of 504 IBD patients (403 CD and 101 UC), vitamin D deficiency was associated with lower sIBDQ in CD [regression coefficient -2.21, 95%CI: -4.1-(-0.33)] but not in UC (regression coefficient 0.41, 95%CI: -2.91-3.73), as well as with increased disease activity in CD (regression coefficient 1.07, 95%CI: 0.43-1.71)[11]. In a small study of CD patients, serum 25(OH)D3 concentration correlated with disease activity, as assessed by the Harvey-Bradshaw index (r = -0.484, P < 0.004)[14]. In another open label study of CD patients, after 12 wk of VD supplementation to a target serum VD concentration of more than 40 ng/mL, the mean IBDQ scores rose from baseline 156 ± 24 to 178 ± 22 points (P = 0.0006)[27]. Blanck recently reported that UC patients who were VD-deficient (< 30 ng/mL) were more likely to have high disease activity (68%, n = 19) than UC patients who were not (33%, n = 14) (P = 0.04)[13]. An inverse correlation between 25(OH)D concentration and calprotectin concentration has been reported in CD (Pearson's r = -0.35, P = 0.04), UC (r = -0.39, P = 0.04) and in all IBD together (r = -0.37, P = 0.003)[28]. IBD patients who were VD-insufficient (< 30 ng/mL) prior to anti-tumor necrosis factor-α treatment stopped this therapy earlier than patients who were not[29]. This effect was significant in patients who stopped treatment owing to the loss of response (HR = 3.49; 95%CI: 1.34-9.09) and was stronger for CD than for UC (HR = 2.38; 95%CI: 0.95-5.99). In a Dutch study of 316 IBD patients, low VD concentration, which was defined as the lowest quartile (< 16.8 ng/mL), was associated with increased disease activity[23].

In the current study, in all analyzed scenarios (UC/CD, summer/autumn period/winter/spring period), health related quality of life was the highest in patients with VD serum concentrations of 50-59 ng/mL, indicating a possible target level for therapeutic VD supplementation. In a study by Hawai, the highest 25(OH)D3 concentration recorded following natural exposure to ultraviolet B was 60 ng/mL and the authors recommended this as the upper limit when prescribing VD supplements[30].

With regards to a mechanistic explanation for our observations, VD might influence the disease activity and consequently the disease specific health related quality of life. The positive immunomodulatory effects of VD are well documented in many animal models of IBD[7]. VD also plays an important role in mucosal barrier homeostasis and preserving the integrity of epithelial junctions[31]. VD is also suggested to function in the physiological packaging of mucins in goblet cells[32].

In the current study, supplementation with a median of 800 IU/day VD (range: 400-1000 IU) did not influence the summer/autumn period VD serum concentrations. However, the change in VD serum concentration from the summer/autumn period to the winter/spring period was reduced from -6.2 ± 1.5 ng/mL in non-supplemented patients to -1.0 ± 2.8 ng/mL in VD-supplemented patients (P = 0.10, n.s.). Supplementation with such doses of VD for 4-6 mo did not affect disease related quality of life. In a small study on 55 IBD patients, the level of VD supplementation (mean, 371 IU/d) and 25[OH] serum concentration were only correlated during non-sunny months[33]. A small study from Ireland on a cohort of 58 CD patients identified VD supplementation (median, 340 IU/d) as an independent factor associated with adequate VD status in both the summer/autumn period and the winter/spring period[12]. In a recent study of 83 pediatric CD patients, supplementation with 400 IU or 2000 IU VD daily for 3 mo increased VD serum concentrations by a mean 2.8 ng/mL and 16 ng/mL, respectively, vs the baseline[34]. In an open label study on 19 CD patients, twenty-four weeks supplementation with up to 5000 IU/d vitamin D3 effectively raised serum 25(OH)D3 and reduced CDAI scores[27].

Although experimental animal studies and in vitro studies have provided evidence that VD supplementation may affect disease activity in CD and UC patients, there are limited clinical data. In a small double-blinded placebo-controlled randomized trial, oral supplementation of CD patients with 1200 IU/d VD3 non-significantly reduced the risk of relapse from 29% to 13% (P = 0.06)[16]. In another small study of 35 CD patients in remission, treatment with 1,25(OH)2D3 and 25(OH)D3 significantly decreased clinical activity, as measured by the CD activity index, as well as mean C-reactive protein concentration[15].

The current study has several limitations. First, all analyses of health related quality of life were univariate, despite this variable being modulated by multiple factors. The study design and the numbers of patients in the UC and CD cohorts prevented multivariate analyses and confounding bias was possible. The correlation between VD serum concentration and sIBDQ score was not affected by disease location, disease behavior, previous IBD surgery or smoking. However, we did not examine the effects of disease medication. Second, the association between VD serum concentration and health related quality of life can be interpreted in several ways. A lower VD serum concentration may result in higher disease related quality of life. Alternatively, higher disease activity reflected in lower sIBDQ might result in a lower VD serum concentration level owing to reduced intestinal absorption and/or changes to diet or lifestyle. Third, the effects of VD supplementation are dose-dependent. The patients took 400, 800 or 1000 IU/d VD. To evaluate the effects of VD supplementation more precisely, patients need to be prescribed a standardized dose of VD in a prospective study.

To sum up, VD supplementation with adequate doses and saturation of 25(OH)D3 reserves might be a novel therapeutic approach for IBD that is simple, effective, safe and inexpensive. Based on the findings of this study, we believe that patients with UC, and perhaps also those with CD, who are VD-insufficient might benefit from VD supplementation with adequate doses to increase their VD serum concentration to around 50 ng/mL. It is likely that currently recommended doses of VD supplementation are too low to achieve this level and that higher doses are needed. Further studies are needed to confirm the association of VD with disease activity and to explore the optimal treatment protocol.