Ankylosing spondylitis disease activity and serum vitamin D levels: A systematic review and meta-analysis : Medicine

Journal Logo

Research Article: Systematic Review and Meta-Analysis

Ankylosing spondylitis disease activity and serum vitamin D levels: A systematic review and meta-analysis

Chen, Menglu MMa; Li, Wen MDa; Li, Lailai MDa; Chai, Yihui MMa; Yang, Yuqi MDa; Pu, Xiang MDa,*

Author Information
doi: 10.1097/MD.0000000000031764

Abstract

1. Introduction

Ankylosing spondylitis (AS) affects the spine joints, sacroiliac joints, and hip joints and is a chronic inflammatory rheumatic illness that falls under the category of spondyloarthritis. Osteophytes, thoracolumbar kyphosis, osteophytes, and inflammatory back pain are the major clinical symptoms of AS.[1,2] A major impact on the patients’ quality of life is caused by the advanced stage of AS, which can also present as trouble sitting up straight, lying flat and looking straight ahead, and limitation of everyday activities.[3] Arthritic complex formation and cortical bone erosion are the 2 main pathological manifestations of AS.[4] The Bass Ankylosing Spondylitis Disease Activity Index (BASDAI), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) are 3 indicators that are often used clinically to evaluate and track disease activity in AS.[5,6] As a result, postponing bone loss and lowering disease activity are the top priorities for enhancing AS patients’ quality of life.

In addition to aiding in calcium homeostasis and bone metabolism, vitamin D (VD), as a fat-soluble vitamin, lowers the body’s levels of pro-inflammatory cytokines, which helps to lessen inflammatory reactions.[7] It has been demonstrated that AS is connected with vitamin D-related gene polymorphisms. A study by Zhang et al revealed that long noncoding RNA (lncRNA) H19, which has been demonstrated to be considerably overexpressed in individuals with AS, has been found to compete with the vitamin D receptor for binding and to promote the production of IL-17A and IL-23 cytokines. As opposed to this, VD can bind to the vitamin D receptor in vivo and lower levels of pro-inflammatory IL-17A and IL-23, which benefits AS sufferers.[8] Additionally, it has been demonstrated that VD can control the cellular regulation of cytokines to control the activity of the innate and adaptive immune systems.[9,10] Despite the fact that no pertinent experiments have been able to confirm that VD interferes with the targets and pathways of AS, several studies have evaluated and analyzed the relationship between VD and rheumatic disease activity and found that VD can regulate the function of Th17-related cytokines to prevent the persistence of inflammation in several rheumatic diseases.[11] Additionally, clinical research has demonstrated that slowing VD metabolism and increasing bone resorption can both worsen AS.[12,13] Therefore, we assume that better outcomes in AS are related with greater in vivo VD level.

The primary marker used to assess the body’s vitamin D levels is serum 25-hydroxyvitamin D (25(OH)D).[14] Previous clinical investigations have demonstrated that individuals with AS had lower blood levels of 25(OH)D relative to controls, and that this is related with a higher risk of all-cause death in these patients.[15–17] However, according to several other research, there is no connection between VD levels and AS.[18] In order to confirm the effect of high and low VD levels on the treatment of illness in AS patients, we conducted this meta-analysis to further evaluate the link between serum VD levels and the main monitoring indicators of AS disease activity, including BASDAI, CRP, and ESR.

2. Materials and methods

2.1. Search strategy

The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) Statement served as the reporting guidelines for this systematic review (see Table S1, Supplemental Digital Content, https://links.lww.com/MD/H920, Supplementary Content, which describes the PRISMA checklist), which was registered beforehand in the International Platform of Registered Systematic Review and Meta-analysis Protocols (INPLASY) with the ID INPLASY202290030 (see Protocol S1, Supplemental Digital Content, https://links.lww.com/MD/H921, Supplemental Content, describing the protocol registered in advance with the INPLASY).[19]

In order to include as much research on the connection between VD and AS as feasible, 2 of our writers (CML and CYH) did a literature search of PubMed, the Cochrane Library, EMBASE, and the China National Knowledge Infrastructure. A random combination of the following search terms was used to conduct the literature review: “vitamin D” or “vitamin D levels” or “25-hydroxyvitamin D” or “25-hydroxyvitamin D levels” or “25 (OH) D” or “25 (OH) D levels” and “ankylosing spondylitis” or “ankylosing spondyloarthritis” or “Bechterew’s disease,” with a time frame limited to August 2022 following a third reviewer arbitrated any disputes between the 2 writers (LW). To ensure that no potentially pertinent researches were overlooked, references to the primary article and associated reviews were thoroughly examined as well.

2.2. Literature selection

The following criteria had to be met for inclusion: studies had to be case-control, cohort, or cross-sectional studies; healthy people had to make up the control group; neither patients nor controls were taking VD supplements; data at 25 (OH) D levels in patients and controls; and data on the correlation index between serum VD levels and BASDAI, correlation index with ESR, or correlation index with CRP in patients and controls.

Exclusion criteria include the following: the type of literature was a review, systematic evaluation, meta-analysis, case report, animal study, in vitro study, and editorial article; the control group population was not healthy; patients or controls were taking VD supplements; and patients’ or controls’ 25 (OH) D levels were not available.

2.3. Risk of bias assessment

The effectiveness of the included studies was evaluated separately by 2 reviewers (CML and CYH). Cross-sectional studies were evaluated using the Agency for Healthcare Research and Quality (AHRQ), whereas cohort and case-control studies were evaluated using the Newcastle-Ottawa Scale (NOS).[19] Finally, for statistical analysis, we chose case–control studies with NOS scores 7 or cross-sectional studies with AHRQ ratings 4.

2.4. Data extraction

The following data was taken from the included studies by 2 separate authors: the first author’s name and the year the study was published, the nation, the subjects’ basic characteristics, such as the study category, sample size, mean age, mean BMI, and mean disease duration, the levels of serum 25 (OH) D and/or ESR and CRP in patients and controls, and the correlation coefficients between serum VD and BASDAI, ESR, and CRP in patients and controls. We followed the procedure outlined in other research[20,21] for converting Spearman correlation coefficients to Pearson correlation coefficients for inclusion in the paper.

2.5. Statistical analysis methods

All extracted data were evaluated using Review Manager software version 5.4 for data analysis. Serum 25(OH)D, ESR, and CRP levels, as well as correlations between serum VD levels and significant indicators of AS disease activity, were investigated between AS patients and controls using mean differences (MD) and 95% confidence intervals (CI). In cases where homogeneity was suboptimal (I2 ≤ 50%), a fixed-effects model was applied. Otherwise, a random-effects model was used. Differences were considered statistically significant when P < .05. Subgroup analyses were done based on the continents where the article studies were completed. Sensitivity analyses were performed by eliminating each study one at a time to examine the impact of each study on the overall meta-analysis.

3. Results

3.1. Literature search result

We discovered 709 records after evaluating the literature, including those from PubMed (n = 98), EMBASE (n = 470), the Cochrane Library (n = 10), and China National Knowledge Infrastructure (n = 131). After the software had removed 413 duplicate entries, we went through the titles and abstracts of the remaining studies, discarding 267 articles from noncase–control, cohort, and cross-sectional studies. Then, after carefully reading the entirety of each research to evaluate eligibility, we rejected 23 of them in accordance with the inclusion and exclusion criteria we had already established. Finally, 6 papers[13,16,22,23,24] and [25] were incorporated into our meta-analysis. Figure 1 shows the literature screening procedure.

F1
Figure 1.:
The procedure for screening literature.

3.2. Characteristics of the studies

In the end, 6 studies totaling 503 AS patients and 398 healthy volunteers were included in our meta-analysis. 3 of these investigations were conducted in Turkey, while the other 3 were conducted in Germany, Morocco, and China. One cross-sectional study and 5 case–control studies made up the total. Table 1 displays the fundamental features of the collected studies.

Table 1 - The basic characteristics of the included studies.
Study Country Study design Sample size
(AS/control)
Mean age
(AS/control)
Mean BMI (kg/cm2) (AS/control) Mean disease duration (year) Outcomes Statistical methods NOS/AHRQ
Bedriye
2009
Turkey Case–control 100/58 39.9/38.1 26.05/25.9 10.5 Serum 25(OH)D
ESR
CRP
COR of VD and BASDAI
COR of VD and ESR
COR of VD and CRP
Student’s t test
Mann–Whitney U test
Pearson correlation test
Spearman correlation test
7
Burhan
2018
Turkey Case–control 68/34 41.51/39.11 27.2/26.11 N Serum 25(OH)D
ESR
CRP
COR of VD and BASDAI
COR of VD and ESR
COR of VD and CRP
Chi-square test
independent sample t test
Mann–Whitney U test
Spearman’s rho test
9
Durmus
2012
Turkey Cross-sectional 99/42 36.8/36.1 N/N 10.48 Serum 25(OH)D
COR of VD and BASDAI
COR of VD and ESR
COR of VD and CRP
Shapiro–Wilk test
Pearson correlation
one-way analysis of variance
HOC Tukey’s test
Chi-square test
7
Hmamouchi
2013
Morocco Case–control 70/140 40/42 23.1/24.8 12.1 Serum 25(OH)D
CRP
COR of VD and BASDAI
Kolmogorov Smirnov test
Student’s t test
chi-squared test
Fisher’s exact test
Pearson correlation test
7
Lange
2001
Germany Case–control 70/45 38.4/N N/N N Serum 25(OH)D
ESR
COR of VD and BASDAI
COR of VD and ESR
COR of VD and CRP
Spearman’s correlation test
Mann–Whitney U test
7
Mu
2014
China Case–control 108/66 40/41 N/N N Serum 25(OH)D
COR of VD and ESR
t test
Pearson correlation test
8
25(OH)D = serum 25-hydroxyvitamin D, AHRQ = Agency for Healthcare Research and Quality, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, NOS = Newcastle-Ottawa Scale, VD = vitamin D.

3.3. Literature quality assessment

The inclusion studies’ methodological quality was evaluated using NOS and AHRQ. Four of the 6 studies received 7, one received 8, and one received a 9. Overall, all included studies had moderate to high methodological quality.

3.4. Meta-analysis of the differences in serum 25(OH)D levels between AS patients and controls

The results of the meta-analysis and subgroup analysis are summarized in Tables 2 and 3, respectively. Among the included studies, 6 studies provided data on the serum 25(OH)D levels of 503 AS patients and 398 controls. Figure 2 illustrates the statistical findings that 25(OH)D levels in AS patients were substantially lower than in controls (MD = −7.53 ng/mL, 95% CI: −9.78 to −5.28, P < .001). We ran a series of subgroup analyses based on the various continents where the studies were done in order to identify possible influencing factors after high statistical heterogeneity was identified (I2 = 67%, P = .009). As shown in Figure 3, notably, statistical heterogeneity was significantly lower in the Asian subgroup (I2 = 50%, P = .11), and statistical results remained statistically different in the European and African subgroups. Furthermore, the stability of the pooled results was further supported by the results of the sensitivity analysis (Table 4).

Table 2 - Results of meta-analysis of the associations of serum 25(OH)D, ESR, and CRP between AS patients and controls, and the relationships between serum VD levels and BASDAI, ESR, and CRP in AS patients.
Outcomes Eligible studies Sample size (AS/Control) MD (95% CI) P value Heterogeneity test
25(OH)D 6 503/398 -7.53 (-9.78, -5.28) <0.001 I2  = 67%, P = .009
ESR 3 226/150 11.75 (4.20, 19.31) 0.002 I2  = 94%, P < .001
CRP 3 238/232 15.36 (4.95, 25.77) 0.004 I2  = 96%, P < .001
VD and BASDAI 5 395 -0.34 (-0.50, -0.18) 0.001 I2  = 59%, P = .04
VD and ESR 5 433 -0.38 (-0.61, -0.16) < 0.001 I2  = 81%, P < .001
VD and CRP 4 325 -0.35 (-0.69, -0.01) = 0.04 I2  = 89%, P < .001
25(OH)D = serum 25-hydroxyvitamin D, AS = ankylosing spondylitis, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, VD = vitamin D.

Table 3 - Results of subgroup analysis of the association of serum 25(OH)D between AS patients and controls, and the relationship between serum VD levels and BASDAI in AS patients.
Outcomes Subgroup Eligible studies Sample size (AS/control) MD/Fisher’Z (95% CI) P value Heterogeneity test
25(OH)D Asian 4 375/200 −7.77 (−10.38, −5.16) <.001 I2  = 50%, P = .11
European 1 58/58 −9.95 (−12.39, −7.51) <.001 None
African 1 70/140 −4.40 (−7.01, −1.79) <.001 None
VD and BASDAI Asian 3 267 −0.34 (−0.50, −0.18) .001 I2  = 11%, P = .33
European 1 58 −0.67 (−0.93, −0.41) <.001 None
African 1 70 −0.33 (−0.57, −0.09) .007 None
25(OH)D = serum 25-hydroxyvitamin D, AS = ankylosing spondylitis, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, ESR = erythrocyte sedimentation rate, VD = vitamin D.

Table 4 - Sensitivity analysis of the differences in serum 25(OH)D, ERS and CRP levels between AS patients and controls.
Omitted studies Estimation 95% CI I2 (P)
Serum 25(OH)D
 Bedriye 2009 −6.84 −9.12 −4.56 62% (0.03)
 Burhan 2018 −7.52 −10.03 −5 74% (0.004)
 Durmus 2012 −7.93 −10.34 −5.52 71% (0.008)
 Hmamouchi 2013 −8.37 −10.48 −6.26 51% (0.09)
 Lange 2005 −6.91 −9.36 −4.46 62% (0.03)
 Mu 2014 −7.62 −10.65 −4.6 73% (0.005)
 Combine −7.53 −9.78 −5.28 67% (0.009)
ESR
 Bedriye 2009 8.43 −1.4 18.27 96% (<0.001)
 Burhan 2018 15.69 10.4 20.98 81% (0.02)
 Lange 2005 10.99 −4.19 26.17 97% (<0.001)
 Combine 11.75 4.2 19.31 94% (<0.001)
CRP
 Cervellati 2016 14.19 −1.86 30.24 97% (<0.001)
 Liu 2010 20.02 15.48 24.56 56% (0.13)
 Zuo 2016 11.89 0.38 23.39 96% (<0.001)
 Combine 15.36 4.95 25.77 96% (<0.001)
25(OH)D = serum 25-hydroxyvitamin D, AS = ankylosing spondylitis, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, VD = vitamin D.

F2
Figure 2.:
Forest plot of differences in serum 25(OH)D levels between AS patients and controls. AS = ankylosing spondylitis, 25(OH)D = serum 25-hydroxyvitamin D.
F3
Figure 3.:
Forest plot of subgroup analysis of differences in serum 25(OH)D levels between AS patients and controls. AS = ankylosing spondylitis, 25(OH)D = serum 25-hydroxyvitamin D.

3.5. Meta-analysis of the differences in ERS levels between AS patients and controls

Three studies out of the total of the included studies reported serum ERS levels for 226 AS patients and 150 controls. Figure 4 demonstrates that ERS levels were significantly higher in patients with AS compared to controls (MD = 11.75 mm/h, 95% CI: 4.20–19.31, P = .002), and significant statistical heterogeneity was found (I2= 94%, P < .001); however, subgroup analyses were not carried out because these studies did not cover all continents involved in the studies. Notably, the sensitivity analysis’s findings did not support the stability of the pooled results because statistical shifts occurred when one study was excluded at a time (Table 4).

F4
Figure 4.:
Forest plot of differences in ESR levels between AS patients and controls. AS = ankylosing spondylitis, ESR = erythrocyte sedimentation rate.

3.6. Meta-analysis of the differences in CRP levels between AS patients and controls

Serum CRP levels were measured in 238 individuals with AS and 232 controls across 3 investigations. Figure 5 demonstrates that CRP levels in AS patients were significantly higher than in controls (MD = 15.36 mg/l, 95% CI: 4.95–25.77, P = .004). High statistical heterogeneity was also noted (I2= 96%, P < .001), but subgroup analyses were not carried out because these studies did not cover all of the involved continents. Additionally, due to statistical variance after excluding one research at a time, sensitivity analysis could not support the stability of the pooled results (Table 4).

F5
Figure 5.:
Forest plot of differences in CRP levels between AS patients and control. AS = ankylosing spondylitis, CRP = C-reactive protein.

3.7. Meta-analysis of the relationship between serum VD levels and BASDAI in AS patients

Five of the 6 studies that were included in the analysis looked at the link between AS patients’ BASDAI and blood VD levels and produced Pearson’s or Spearman’s correlation coefficients. To run a pooled analysis to ascertain the link between VD levels and BASDAI in AS patients, these Pearson’s or Spearman’s correlation coefficients were changed to Fisher’s Z values. We did a number of subgroup analyses based on the various continents where the study was done because, as shown in Figure 6, the meta-analysis produced a Fisher’s Z value of −0.34 (95% CI, −0.50 to −0.18, P < .001), but there was significant statistical heterogeneity (I2= 59%, P = .04). As shown in Fig. 7, statistical heterogeneity was significantly lower in the Asian subgroup (I2 = 11%, P = .33), while statistical results remained significant in the European and African subgroups, which indicated that serum VD levels were negatively correlated with BASDAI in all subgroups. The stability of the pooled results was further supported by the results of the sensitivity analysis, which showed that the results remained stable even when one study was excluded at a time (Table 5).

Table 5 - Sensitivity analysis of the correlation between serum VD levels and BASDAI, ESR as well as CRP in AS patients.
Omitted studies Estimation 95% CI I 2 (P)
Serum 25(OH)D
 Bedriye 2009 −0.35 −0.56 −0.14 69% (0.02)
 Burhan 2018 −0.39 −0.54 −0.24 46% (0.13)
 Durmus 2012 −0.35 −0.56 −0.14 69% (0.02)
 Hmamouchi 2013 −0.34 −0.55 −0.14 70% (0.02)
 Mu 2014 −0.27 −0.38 −0.17 0% (0.47)
 Combine −0.34 −0.5 −0.18 59% (0.04)
ESR
 Bedriye 2009 −0.43 −0.69 −0.17 82% (<0.001)
 Burhan 2018 −0.46 −0.68 −0.24 77% (0.005)
 Durmus 2012 −0.38 −0.68 −0.08 86% (<0.001)
 Lange 2005 −0.31 −0.56 −0.07 81% (0.001)
 Mu 2014 −0.32 −0.56 −0.08 78% (0.003)
 Combine −0.38 −0.61 −0.16 81% (<0.001)
CRP
 Bedriye 2009 −0.44 −0.88 0 90% (<0.001)
 Burhan 2018 −0.44 −0.87 −0.02 91% (<0.001)
Durmus 2012 0.35 0.85 0.15 93% (<0.001)
 Lange 2005 −0.18 −0.37 0 55% (0.11)
 Combine −0.35 −0.69 −0.01 89% (<0.001)
25(OH)D = serum 25-hydroxyvitamin D, AS = ankylosing spondylitis, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, VD = vitamin D.

F6
Figure 6.:
Forest plot of the correlation between serum VD levels and BASDAI in AS patients. AS = ankylosing spondylitis, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, VD = vitamin D.
F7
Figure 7.:
Forest plot of subgroup analysis of correlation between serum VD levels and BASDAI in AS patients. AS = ankylosing spondylitis, BASDAI = Bass Ankylosing Spondylitis Disease Activity Index, ESR = erythrocyte sedimentation rate, VD = vitamin D.

3.8. Meta-analysis of the relationship between serum VD levels and ESR in AS patients

Five of the included studies looked at the connection between AS patients’ blood VD levels and ESR. Figure 8 illustrates the statistical results, which produced a Fisher’s Z value of −0.38 (95% CI: −0.61 to −0.16, P < .001), despite the fact that there was significant statistical heterogeneity (I2= 81%, P < .001) and that subgroup analyses were not conducted. Notably, the sensitivity analysis findings supported the reliability of the combined results (Table 5).

F8
Figure 8.:
Forest plot of the correlation between serum VD levels and ESR in AS patients. AS = ankylosing spondylitis, ESR = erythrocyte sedimentation rate, VD = vitamin D.

3.9. Meta-analysis of the relationship between serum VD levels and BASDAI, ESR as well as CRP in AS patients

Among the included studies, 5 studies examined the relationship between serum VD levels and CRP in patients with AS. As shown in Fig. 9, the statistical results yielded a Fisher’s Z value of −0.35 (95% CI: −0.69 to −0.01, P = .04), while high statistical heterogeneity was detected (I2= 89%, P < .001), but no subgroup analysis was performed. However, the results of the sensitivity analysis did not confirm the stability of the pooled results (Table 5).

F9
Figure 9.:
Forest plot of the correlation between serum VD levels and CRP in AS patients. AS = ankylosing spondylitis, CRP = C-reactive protein, VD = vitamin D.

4. Discussion

AS is a lifelong condition with a prevalence of 0.2% to 1% in adults, and men are 2 to 3 times more likely than women to develop the disease.[26,27] There is no cure for AS, and available treatments can only delay the process of spinal stiffness and fusion as much as possible, maximize physical function and long-term quality of life, and reduce patient symptoms.[28,29] Current clinical recommendations advocate nonsteroidal anti-inflammatory medicines as the first-line treatment. Additionally, interleukin (IL)-17 inhibitors and antitumor necrosis factor medications, which are newer treatments for AS, can also significantly reduce AS symptoms.[30–32] When AS is not adequately treated, one of its complications, osteoporosis, frequently causes pathological fractures,[33] which can easily happen in the intervertebral space or vertebral body and are also known as Andersson’s lesions, which are closely related to bone resorption and decreased bone density in advanced AS.[34] The surgical management of AS fractures is likewise fraught with risks, including as surgical site infection, nonbacterial pneumonia, implant failure, and mortality.[35] As a result, bone preservation and fracture prevention for AS patients are crucial since they are strongly tied to disease management and AS patients’ quality of life.

It is well known that VD is a significant steroid hormone in the body, with its major roles being to increase intestine synthesis for calcium and phosphate absorption as well as to regulate bone metabolism. But recent research has demonstrated that VD also plays a key role in autoimmune and rheumatic illnesses.[36,37] In order to undertake this meta-analysis, we statistically evaluated the findings from 6 research that looked at the relationship between VD and AS. The results of our meta-analysis showed that serum 25(OH)D levels were generally lower in AS patients than in controls, and the overall results were confirmed by our subgroup analysis. We hypothesized that low serum 25(OH)D levels in AS patients are closely related to the continent and ethnicity of the subjects, but since fewer studies were included in Europe and Africa, more case-control trials are required to further confirm these findings.

A meta-analysis was carried out to look at the connection between serum VD levels and BASDAI, ESR, and CRP in order to further highlight the association between serum VD levels and AS disease activity. The BASDAI is a score derived from a patient-self-administered questionnaire that is used to quantify disease activity in AS patients. A cutoff of BASDAI ≥ 4 is used to indicate a high level of disease activity.[5,38] Our subgroup analysis supported the negative correlation between serum VD levels and BASDAI that was identified by the findings of our meta-analysis. According to this, it can be evidenced that higher serum VD levels are closely linked to better disease management and quality of life in AS patients. It also supports the idea that continental and ethnic differences may significantly affect the negative correlation between serum VD levels and BASDAI. Additionally, research has shown that regular exposure to sunshine greatly increases the blood levels of VD in people.[39,40] As a result, individuals with AS may see an improvement in their quality of life if they receive an acceptable amount of sun exposure. Meanwhile, it has been demonstrated that lower VD levels are linked to higher all-cause mortality in AS patients, and it has been proposed that VD supplementation, used as a secondary prevention measure in cases of VD deficiency or insufficiency, may lessen AS patients’ disease activity and comorbidity.[15,41]

ESR and CRP are nonspecific inflammatory biomarkers that are frequently used to monitor AS disease activity and assess whether patients’ symptoms improve while receiving therapy.[42,43] As a result, using meta-analysis, we looked at the variations in ESP and CRP levels between AS patients and controls as well as the relationship between serum VD levels and ESR and CRP levels in AS patients. The findings demonstrated that serum VD levels in AS patients had a negative correlation with total ESR and CRP levels, which were also considerably higher in AS patients compared to controls.

While the stability of the pooled results was not supported by the results of all sensitivity analyses, it is noteworthy that our meta-analysis of ESR and CRP levels and their correlation with serum VD levels revealed a significant overall heterogeneity (I2≥ 89%), indicating that a number of unidentified factors contributed to the heterogeneity of these data sets’ results. These factors may include the mean BASDAI score and mean disease duration of the subjects, but we did not perform further subgroup.

Our meta-analysis has several advantages: first, our meta-analysis focused on the relationship between serum VD levels and 3 key markers of AS disease activity, including BASDAI, ESR, and CRP; second, we divided studies of serum 25(OH)D levels in AS patients and controls, as well as the relationship between serum VD levels and BASDAI, according to ethnicity; and thirdly, despite the large heterogeneity in the results of a subset of our studies, the quality of the included studies was moderately high.

There are certain nonnegligible restrictions despite the fact that our investigation produced some fresh findings. We first conducted a thorough search for clinical studies pertaining to AS and VD, but due to our stringent inclusion and exclusion criteria, we were only able to find 6 that met our standards. These studies were primarily focused on Asia, with few studies in Europe and Africa, and there is a lack of information from studies on other continents, so the results of our analysis should be interpreted with caution for populations on other continents. Second, the Ankylosing Spondylitis Disease Activity Score[44] is the primary index for tracking and assessing AS disease activity in addition to BASDAI, ESR, and CRP; however, since only one of our included studies[16] recorded the correlation coefficient between VD levels and Ankylosing Spondylitis Disease Activity Score, we did not examine it separately. Third, despite our best efforts to account for confounding influences, some unidentified factors continued to contribute to the high heterogeneity of some of our findings, which had an impact on the overall conclusions. Fourth, despite the fact that several of our results had a high degree of heterogeneity, which may greatly reduce their validity, our sensitivity analysis supported the stability of some findings.

5. Conclusion

In conclusion, our findings revealed a negative correlation between serum VD levels and the primary monitoring markers of disease activity in AS patients, including BASDAI, ESR, and CRP levels. This finding suggests that higher serum VD levels are strongly linked to both disease control and an improved quality of life for AS patients. Additionally, we confirmed that the total serum 25(OH)D levels and the inverse relationship between serum VD levels and BASDAI in AS patients may be significantly influenced by continental and ethnic variations. Therefore, patients with AS are advised to supplement VD by sun exposure or supplements in order to lessen clinical symptoms and enhance quality of life. ESR and CRP levels were examined in our meta-analysis in relation to the occurrence of AS, but their correlation with serum VD levels was based on less conclusive evidence. Therefore, more clinical high-quality studies in these areas should be conducted in the future to further validate our results.

Author contributions

Menglu Chen participated in the design of the study and was involved in data extraction and completed the manuscript writing.

Wen Li participated in the design of the study.

Lailai Li edited the manuscript.

Yihui Chai participated in the data extraction.

Yuqi Yang performed the statistical of the collected data.

Xiang Pu revised and reviewed the manuscript.

Conceptualization: Wen Li, Lailai Li.

Data curation: Menglu Chen, Yihui Chai.

Formal analysis: Menglu Chen, Yuqi Yang.

Funding acquisition: Xiang Pu.

Investigation: Yihui Chai.

Methodology: Wen Li, Lailai Li.

Project administration: Xiang Pu.

Software: Menglu Chen.

Visualization: Yuqi Yang.

Writing—original draft: Menglu Chen.

Writing—review and editing: Wen Li, Lailai Li, Xiang Pu.

    References

    [1]. Ward MM, Learch TJ, Gensler LS, et al. Regional radiographic damage and functional limitations in patients with ankylosing spondylitis: differences in early and late disease. Arthritis Care Res (Hoboken). 2013;65:257–65.
    [2]. Schett G. Bone formation in psoriatic arthritis: a report from the GRAPPA 2013 Annual Meeting. J Rheumatol. 2014;41:1218–9.
    [3]. Qian BP, Jiang J, Qiu Y, et al. The presence of a negative sacral slope in patients with ankylosing spondylitis with severe thoracolumbar kyphosis. J Bone Joint Surg Am. 2014;96:e188.
    [4]. Hanson A, Brown MA. Genetics and the causes of ankylosing spondylitis. Rheum Dis Clin North Am. 2017;43:401–14.
    [5]. Chen YH, Huang WN, Chen YM, et al. The BASDAI cut-off for disease activity corresponding to the ASDAS scores in a Taiwanese cohort of ankylosing spondylitis. Front Med (Lausanne). 2022;9:856654.
    [6]. Kwon OC, Park MC. BASDAI cut-off values corresponding to ASDAS cut-off values. Rheumatology (Oxford). 2022;61:2369–74.
    [7]. Yazmalar L, Ediz L, Alpayci M, et al. Seasonal disease activity and serum vitamin D levels in rheumatoid arthritis, ankylosing spondylitis and osteoarthritis. Afr Health Sci. 2013;13:47–55.
    [8]. Zhang X, Ji S, Cai G, et al. H19 increases IL-17A/IL-23 releases via regulating VDR by interacting with miR675-5p/miR22-5p in ankylosing spondylitis. Mol Ther Nucleic Acids. 2020;19:393–404.
    [9]. Konijeti GG, Arora P, Boylan MR, et al. Vitamin D supplementation modulates T cell-mediated immunity in humans: results from a randomized control trial. J Clin Endocrinol Metab. 2016;101:533–8.
    [10]. Martens PJ, Centelles-Lodeiro J, Ellis D, et al. High serum Vitamin D concentrations, induced via diet, trigger immune and intestinal microbiota alterations leading to type 1 diabetes protection in NOD mice. Front Immunol. 2022;13:902678.
    [11]. Heidari B, Hajian-Tilaki K, Babaei M. Vitamin D deficiency and rheumatoid arthritis: epidemiological, immunological, clinical and therapeutic aspects. Mediterranean J Rheumatol. 2019;30:94–102.
    [12]. Lange U, Jung O, Teichmann J, et al. Relationship between disease activity and serum levels of vitamin D metabolites and parathyroid hormone in ankylosing spondylitis. Osteoporos Int. 2001;12:1031–5.
    [13]. Lange U, Teichmann J, Strunk J, et al. Association of 1.25 vitamin D3 deficiency, disease activity and low bone mass in ankylosing spondylitis. Osteoporos Int. 2005;16:1999–2004.
    [14]. Gutierrez JA, Parikh N, Branch AD. Classical and emerging roles of vitamin D in hepatitis C virus infection. Semin Liver Dis. 2011;31:387–98.
    [15]. Ben-Shabat N, Watad A, Shabat A, et al. Low Vitamin D levels predict mortality in ankylosing spondylitis patients: a nationwide population-based cohort study. Nutrients. 2020;12:1400.
    [16]. Kocyigit BF, Akyol A. Vitamin D levels in patients with ankylosing spondylitis: Is it related to disease activity? Pak J Med Sci. 2018;34:1209–14.
    [17]. Kültür T, Öztaş D, Keskin D, et al. The relationship of serum vitamin D receptor levels with disease activity and clinical parameters in patients with ankylosing spondylitis. Turk J Phys Med Rehabil. 2019;65:389–93.
    [18]. Jiang J, Shao M, Wu X. Vitamin D and risk of ankylosing spondylitis: A two-sample mendelian randomization study. Hum Immunol. 2022;83:81–5.
    [19]. Page MJ, McKenzie JE, Bossuyt PM, et al. Updating guidance for reporting systematic reviews: development of the PRISMA 2020 statement. J Clin Epidemiol. 2021;134:103–12.
    [20]. Li X, He J. The association between serum/plasma leptin levels and obstructive sleep apnea syndrome: a meta-analysis and meta-regression. Front Endocrinol (Lausanne). 2021;12:696418.
    [21]. Surov A, Meyer HJ, Wienke A. Associations between apparent diffusion coefficient (ADC) and KI 67 in different tumors: a meta-analysis. Part 1: ADC(mean). Oncotarget. 2017;8:75434–44.
    [22]. Mermerci Başkan B, Pekin Doğan Y, Sivas F, et al. The relation between osteoporosis and vitamin D levels and disease activity in ankylosing spondylitis. Rheumatol Int. 2010;30:375–81.
    [23]. Durmus B, Altay Z, Baysal O, et al. Does vitamin D affect disease severity in patients with ankylosing spondylitis? Chin Med J (Engl). 2012;125:2511–5.
    [24]. Hmamouchi I, Allali F, El B, et al. The relation between disease activity, vitamin D levels and bone mineral density in men patients with ankylosing spondylitis. Rheumatol Reports. 2013;5:e3.
    [25]. Yahong M, Xiaoling L, Jinhua C. Clinical correlation between 25-hydroxyvitamin D levels and ankylosing spondylitis. Hebei Med J. 2014;06:848–9.
    [26]. Cross MJ, Smith EU, Zochling J, et al. Differences and similarities between ankylosing spondylitis and rheumatoid arthritis: epidemiology. Clin Exp Rheumatol. 2009;27(4 Suppl 55):S36–42.
    [27]. Li G, Lv CA, Tian L, et al. A retrospective study of alendronate for the treatment of ankylosing spondylitis. Medicine (Baltim). 2018;97:e10738.
    [28]. Smolen JS, Braun J, Dougados M, et al. Treating spondyloarthritis, including ankylosing spondylitis and psoriatic arthritis, to target: recommendations of an international task force. Ann Rheum Dis. 2014;73:6–16.
    [29]. Han D, Ouyang G, Pan P, et al. Upregulated lncRNA-NEF predicts recurrence and poor treatment outcomes of ankylosing spondylitis. Immun Inflamm Dis. 2022;10:e627.
    [30]. van der Heijde D, Baraliakos X, Gensler LS, et al. Efficacy and safety of filgotinib, a selective Janus kinase 1 inhibitor, in patients with active ankylosing spondylitis (TORTUGA): results from a randomised, placebo-controlled, phase 2 trial. Lancet. 2018;392:2378–87.
    [31]. Claudepierre P, Van den Bosch F, Sarzi-Puttini P, et al. Treatment with golimumab or infliximab reduces health resource utilization and increases work productivity in patients with ankylosing spondylitis in the QUO-VADIS study, a large, prospective real-life cohort. Int J Rheum Dis. 2019;22:995–1001.
    [32]. Dubash S, Bridgewood C, McGonagle D, et al. The advent of IL-17A blockade in ankylosing spondylitis: secukinumab, ixekizumab and beyond. Expert Rev Clin Immunol. 2019;15:123–34.
    [33]. Hinze AM, Louie GH. Osteoporosis management in ankylosing spondylitis. Curr Treatm Opt Rheumatol. 2016;2:271–82.
    [34]. Guo C, Li T, Zhang H, et al. Treatment of ankylosing spondylitis complicated with a thoracolumbar Andersson lesion by posterior closed osteotomy, debridement and fusion through the fracture line. BMC Musculoskelet Disord. 2022;23:815.
    [35]. Ull C, Yilmaz E, Hoffmann MF, et al. Factors associated with major complications and mortality during hospitalization in patients with ankylosing spondylitis undergoing surgical management for a spine fracture. Global Spine J. 2022;12:1380–7.
    [36]. Malik AA, Baig M, Butt NS, et al. Bibliometric analysis of Global Research productivity on Vitamin D and bone metabolism (2001-2020): learn from the past to plan future. Nutrients. 2022;14:542.
    [37]. Booth DR, Ding N, Parnell GP, et al. Cistromic and genetic evidence that the vitamin D receptor mediates susceptibility to latitude-dependent autoimmune diseases. Genes Immun. 2016;17:213–9.
    [38]. Garrett S, Jenkinson T, Kennedy LG, et al. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol. 1994;21:2286–91.
    [39]. Qureshi AZ, Zia Z, Gitay MN, et al. Attitude of future healthcare provider towards vitamin D significance in relation to sunlight exposure. Saudi Pharm J. 2015;23:523–7.
    [40]. Pereira LA, Luz FB, Carneiro C, et al. Evaluation of vitamin D plasma levels after mild exposure to the sun with photoprotection. An Bras Dermatol. 2019;94:56–61.
    [41]. Essouma M, Noubiap JJ. Are systematic screening for Vitamin D deficiency and Vitamin D supplementation currently feasible for ankylosing spondylitis patients? Int J Inflam. 2017;2017:7840150.
    [42]. Wu J, Yan L, Chai K. Systemic immune-inflammation index is associated with disease activity in patients with ankylosing spondylitis. J Clin Lab Anal. 2021;35:e23964.
    [43]. Dougados M, Gueguen A, Nakache JP, et al. Clinical relevance of C-reactive protein in axial involvement of ankylosing spondylitis. J Rheumatol. 1999;26:971–4.
    [44]. Kwan YH, Tan JJ, Phang JK, et al. Validity and reliability of the Ankylosing Spondylitis Disease Activity Score with C-reactive protein (ASDAS-CRP) and Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) in patients with axial spondyloarthritis (axSpA) in Singapore. Int J Rheum Dis. 2019;22:2206–12.
    Keywords:

    Vitamin D level; ankylosing spondylitis; disease activity; osteoporosis

    Supplemental Digital Content

    Copyright © 2022 the Author(s). Published by Wolters Kluwer Health, Inc.