Clinical and Imaging Outcomes after Vitamin D Supplementation in Patients with Multiple Sclerosis: A Systematic Review

The link between vitamin D and multiple sclerosis (MS) has been suggested in epidemiological, genetic, immunological, and clinical studies. The aim of the present systematic review of the literature was to assess the effects of vitamin D supplementation on clinical and imaging outcomes in patients with MS. The outcomes we assessed included relapse events, disability progression, and magnetic resonance imaging (MRI) lesions. The search was conducted using PubMed, ClinicalTrials.gov, and EudraCT databases, and it included records published up until 28 February 2023. The systematic review was reported according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines. Nineteen independent clinical studies (corresponding to 24 records) were included in the systematic review. The risk of bias in randomized controlled trials (RCTs) was analyzed using the Cochrane risk-of-bias tool. Fifteen trials investigated relapse events, and most of them reported no significant effect of vitamin D supplementation. Eight of 13 RCTs found that vitamin D supplementation had no effect on disability [assessed by Expanded Disability Status Scale (EDSS) scores] compared to controls. Interestingly, recent RCTs reported a significant reduction in new MRI lesions in the central nervous system of MS patients during supplementation with vitamin D3.


Introduction
Multiple sclerosis (MS) affected 2.8 million people worldwide in 2020 and is the most common disabling neurological disease affecting young adults [1]. MS is an immunemediated demyelinating disease of the central nervous system characterized by focal perivenular infiltrates of immune cells and plaque formation. Diagnostic criteria for MS are a combination of clinical, imaging, and laboratory evidence according to the McDonald criteria from the International Panel on Diagnosis of MS [2]. The typical symptoms of MS include muscle weakness, loss of coordination, tremor, fatigue, pain, and visual disturbance. MRI is the most useful imaging test for the diagnosis of MS and provides evidence for the dissemination in space and time during the follow-up of the disease. The typical focal hyperintense lesions observed in the central nervous system of MS patients are searched by the acquisition of T2-weighted fluid-attenuated inversion recovery (FLAIR) MRI sequences (Figure 1) [3]. In addition, the use of gadolinium (Gd)-based contrast agents and T1-weighted MRI sequences is interesting to detect active inflammatory lesions.
Both genetic variants and environmental factors are associated with MS risk. The main environmental risk factors include obesity, Epstein-Barr virus infection, smoking, and vitamin D deficiency. Vitamin D originates from both dietary sources (vitamin D2 and vitamin D3) and endogenous synthesis in skin after exposure to ultraviolet-B radiation (vitamin D3). Then, vitamin D is hydroxylated in the liver to form 25- [4]. Image d is rep duced from Lopaisankrit et al. without changes (CC BY/4.0 license) [5].
The association between vitamin D and MS has been suggested from epidemiolo cal, genetic, immunological, and clinical data. First, the relationship between MS pre lence and latitude, a major variable of vitamin D status, is well established. More precise the prevalence increases by 4.3 cases/100,000 per degree of latitude [6]. In addition, a cent study conducted in two Swedish cohorts reported that low sun exposure increa MS risk, and nearly 30% of the effect was due to vitamin D deficiency [7]. Second, mend lian randomization studies have revealed that genotypes associated with higher vitam D levels protect against MS [8]. Third, low blood levels of 25(OH)D, the biomarker of v amin D status, are associated with an increased risk of MS and clinically isolated sy drome (CIS) [9,10]. In addition, low 25(OH)D levels are an important risk factor for M activity and progression [11]. Last but not least, the immunomodulatory properties of v amin D are now well established. Vitamin D receptors are expressed in almost every i mune cell, and calcitriol exerts pleiotropic actions on several immune processes involv in the pathogenesis of MS [12].
Whether vitamin D plays a causative role in MS activity and progression remain matter of debate. Interventional trials of vitamin D supplementation in animals and h mans are therefore crucial to assess whether vitamin D supplementation is beneficial MS patients. An excellent review recently summarized the data obtained in animals af supplementation with vitamin D [13]. In brief, vitamin D prevents the development a decreases the severity of experimental autoimmune encephalomyelitis, a well-recogniz murine model of MS [14][15][16][17][18][19][20]. Moreover, vitamin D mitigates inflammatory infiltrates, d myelination, and neuron loss in mice with experimental autoimmune encephalomyel [14,19].
In the last decade, several clinical trials were conducted in MS patients to evalu the effects of vitamin D supplementation. In particular, robust data were provided in last few years by publications based on the randomized controlled trials (RCTs) SOLA CHOLINE, and EVIDIMS [21][22][23]. Here, we aimed to conduct a systematic review of d from interventional trials of vitamin D supplementation in MS patients focusing on th major outcomes: relapses, disability, and MRI lesions. Whether vitamin D supplemen tion prevents incident MS remains another interesting matter of debate, but one that mains outside of the scope of this review. The recently published results of the VIT RCT showed that a daily supplementation with 2000 international units (IU) of vitam The association between vitamin D and MS has been suggested from epidemiological, genetic, immunological, and clinical data. First, the relationship between MS prevalence and latitude, a major variable of vitamin D status, is well established. More precisely, the prevalence increases by 4.3 cases/100,000 per degree of latitude [6]. In addition, a recent study conducted in two Swedish cohorts reported that low sun exposure increases MS risk, and nearly 30% of the effect was due to vitamin D deficiency [7]. Second, mendelian randomization studies have revealed that genotypes associated with higher vitamin D levels protect against MS [8]. Third, low blood levels of 25(OH)D, the biomarker of vitamin D status, are associated with an increased risk of MS and clinically isolated syndrome (CIS) [9,10]. In addition, low 25(OH)D levels are an important risk factor for MS activity and progression [11]. Last but not least, the immunomodulatory properties of vitamin D are now well established. Vitamin D receptors are expressed in almost every immune cell, and calcitriol exerts pleiotropic actions on several immune processes involved in the pathogenesis of MS [12].
Whether vitamin D plays a causative role in MS activity and progression remains a matter of debate. Interventional trials of vitamin D supplementation in animals and humans are therefore crucial to assess whether vitamin D supplementation is beneficial for MS patients. An excellent review recently summarized the data obtained in animals after supplementation with vitamin D [13]. In brief, vitamin D prevents the development and decreases the severity of experimental autoimmune encephalomyelitis, a well-recognized murine model of MS [14][15][16][17][18][19][20]. Moreover, vitamin D mitigates inflammatory infiltrates, demyelination, and neuron loss in mice with experimental autoimmune encephalomyelitis [14,19].
In the last decade, several clinical trials were conducted in MS patients to evaluate the effects of vitamin D supplementation. In particular, robust data were provided in the last few years by publications based on the randomized controlled trials (RCTs) SOLAR, CHOLINE, and EVIDIMS [21][22][23]. Here, we aimed to conduct a systematic review of data from interventional trials of vitamin D supplementation in MS patients focusing on three major outcomes: relapses, disability, and MRI lesions. Whether vitamin D supplementation prevents incident MS remains another interesting matter of debate, but one that remains outside of the scope of this review. The recently published results of the VITAL RCT showed that a daily supplementation with 2000 international units (IU) of vitamin D3 for a median of 5.3 years decreased the incidence of autoimmune diseases by 22% [24]. In addition, vitamin D intake during early infancy may reduce the incidence of type 1 diabetes [25].

Materials and Methods
This systematic review follows the PRISMA 2020 guidelines [26]. Table 1 shows the inclusion and exclusion criteria of studies according to the Population, Intervention, Comparison, Outcome, Study (PICOS) format. The review protocol is registered in the PROSPERO registry (#CRD42023411095). Records were also excluded if MS patients retrospectively self-reported data on outcomes.

Study design
Articles/records published in English and in peer-reviewed journals or in recognized databases of trial registration.
Articles not published in English, methodological articles, reviews, meta-analyses, comments, letters to the editor, studies conducted in animals, case reports.

Information Sources and Search Strategy
The search procedure was conducted using PubMed, ClinicalTrials.gov, and EudraCT databases. All records up to 28 February 2023 were identified. The search strategy is detailed in Table S1.

Selection and Data Collection Process
We independently screened the titles and abstracts and discarded records and studies that were not applicable. Relevant studies were selected regarding the inclusion and exclusion criteria. We resolved disagreements through discussion and consensus. We aimed to investigate possible causes of heterogeneity by collecting data on intervention procedures (dosing, duration of supplementation, form of vitamin D, and administration route). We independently collected all data in Excel sheets for each predefined outcome. No automation tool was used in the process.

Study Risk of Bias Assessment
We independently judged the risk of bias in the included randomized trials using the Cochrane risk-of-bias tool (RoB 2). Disagreements in judgement were resolved through discussion and consensus. The risk of bias is presented in Table S2.

Study Risk of Bias Assessment
We independently judged the risk of bias in the included randomized trials using the Cochrane risk-of-bias tool (RoB 2). Disagreements in judgement were resolved through discussion and consensus. The risk of bias is presented in Table S2.

Study Selection
The PRISMA flowchart is shown in Figure 2. Twenty-four records were identified using our systematic approach, corresponding to 19 independent clinical trials. The main characteristics of the selected studies are reported in Table 2. Only one international study was identified and included 11 European countries [21]. The remaining studies were conducted in a single country in Europe [22,23,[27][28][29][30], the Middle East [31][32][33][34][35][36][37][38], North America [39][40][41], and Australia [42]. Of the 19 selected studies, there were 9 RCTs against placebo, 4 RCTs using a low dose of vitamin D as a control group [22,34,39,40], 1 RCT with routine care as a control group [36], and 5 uncontrolled trials [28,31,32,35,41]. Two of the 14 RCTs were open-label designs [36,40], and the remaining 12 trials were double-blind.      Table 3 presents the patients' characteristics and the inclusion and exclusion criteria of the studies included in the present systematic review. The numbers of recruited patients were <50 in eight studies, between 50 and 100 in seven studies, and >100 in four studies. The selected studies were conducted in populations of patients with CIS [27], CIS or RRMS [22,32], undefined presentation of MS [30,33,36,38], and, for the majority of studies, in RRMS patients. In the selected trials, participants were recruited according to the serum 25(OH)D level at baseline: <50 nM in two studies [35,36], <75 nM in two studies [22,33], <85 nM in one study [29], between 50 and 125 nM in one study [39], and >100 nM in one study [37]. There were no specific criteria for 25(OH)D levels in the other studies. Inclusion: RRMS; age 18-55 y; adequate renal and hepatic function; early-stage MS on brain or spinal MRI; first clinical event in the last 5 years; EDSS score lower than 4.0; active disease with either one relapse or MRI new lesion within the last 18 months; no or low vitamin D supplementation (lower than 1000 IU/d). Exclusion: lactation or pregnancy; other disease than MS that could explain symptoms; relapse in the last 30 days before inclusion; use of corticosteroids within 30 days before inclusion; complete transverse myelitis or bilateral optic neuritis; abnormalities of vitamin D metabolism other than low dietary intake or decreased sun exposure; urinary calcium higher than 1.0 mmol/mmol of creatinine or hypercalcemia (11 mg/dL); hepatic impairment (alanine or aspartate aminotransferase higher than three times the upper limit of normal (ULN); bilirubin higher than 1.5 times ULN if associated with any elevation of alanine aminotransferase or alkaline phosphatase; or alkaline phosphatase higher than 2.5 times ULN); drugs other than corticosteroids that affect vitamin D metabolism; vitamin D supplementation higher than 400 IU/d; conditions with susceptibility to hypercalcemia (e.g., treatment with digitalis or hydrochlorothiazide, arrhythmia or heart disease, nephrolithiasis).    Table 4 shows the intervention protocols regarding doses, the dosing frequency, and the duration of the trial. The duration of vitamin D supplementation was <6 months for two trials [32,35], 6 months for seven trials [27,28,33,36,38,39,42], 11 months for one trial [41], 12 months for six trials [21,29,31,34,37,40], 18 months for one trial [22], and 24 months for two trials [23,40]. Vitamin D was administered by the intra-muscular route in one study [38] and by the oral route in all other studies. Lastly, vitamin D was given as 1,25(OH) 2 D3 in two studies [37,41], as 1α(OH)D3 in one trial [33], and in native forms (i.e., vitamin D3 or vitamin D2) in the remaining studies.

Relapses
Fifteen (17 reports) of the 19 trials selected in the systematic review reported results regarding the impact of vitamin D supplementation on relapse in MS patients ( Table 5). The 15 trials assessed 554 vitamin D-treated and 456 control patients (368 with placebo, 21 with routine care alone, and 67 with low doses of vitamin D). The ARR and/or the raw number of relapses during the intervention were reported in all of the 15 trials, and the time to first relapse was reported in 2 studies. Ten controlled trials (corresponding to 342 vitamin D-treated and 300 control patients who completed follow-up) found no effect of vitamin D supplementation on ARR or the raw number of relapses. In particular, the SOLAR trial, which was the largest RCT with 174 RRMS patients in each arm, reported no significant difference in ARR between the placebo group and the group supplemented with 14,000 UI/day of vitamin D3 for one year [21]. The proportion of patients free of relapses was also similar between the two groups. In addition, the recently published RCT EVIDIMS also found a similar relapse rate in MS patients treated for 18 months with either a nutritional dose of vitamin D3 (i.e., 400 IU/d) or 20,400 IU/day [22].
Three trials (186 vitamin D-vs. 144 placebo-treated MS patients) found beneficial effects of vitamin D on relapses. First, Achiron et al. observed that 6 months of 1α(OH)D3 reduced the number of relapses, resulting in a higher proportion of relapse-free patients [33]. This effect was reported at 4 and 6 months of treatment and deteriorated 2 months after drug discontinuation. In an uncontrolled trial by Laursen et al., there was an independent association between the increase in circulating 25(OH)D levels after supplementation with vitamin D3 and a decrease in ARR [28]. Finally, a post-hoc analysis of the CHOLINE RCT, conducted to avoid potential bias due to early dropout, showed a significant reduction in the ARR after vitamin D3 supplementation [23].
Finally, Stein et al. reported a higher proportion of RRMS patients with relapses in those treated with 7000 IU/day of vitamin D2 compared to those supplemented with only 1000 IU/day (36 vs. 0%), but the RCT included only 23 participants [48]. Wingerchuck et al. observed relapses in 27% of MS patients treated with 1,25(OH) 2 D3, but the trial design did not include a control group [41].

Disability
The EDSS score, ranging from 0 to 10 points, is a well-established method for assessing disability in MS patients and monitoring changes in the degree of disability over the course of time. We identified 15 trials reporting changes in the EDSS score after vitamin D supplementation in MS patients (Table 6). They assessed 469 vitamin D-treated patients and 353 control patients (312 with placebo, 9 with routine care alone, and 32 with low-dose vitamin D).  No significant difference in the change in the EDSS score during the trial between the vitamin D3 and placebo groups (−0.23 vs. 0.30, p > 0.05). Lower proportion of patients who completed the trial with a higher EDSS score in the vitamin D3 group than in the placebo group (8 vs. 38%, p = 0.02). Five RCTs (corresponding to 132 vitamin D-treated and 133 control patients who completed the follow-up) reported that vitamin D had a beneficial effect on the EDSS score [23,29,36,37,40]. In the CHOLINE trial, the progression of the EDSS score was slower in the MS patients supplemented with vitamin D3 compared to those who received the placebo [23]. In the Finnish Vitamin D Study RCT, the EDSS score slightly decreased in vitamin D3-treated patients but not in the placebo group [29]. Shaygannejad et al. found a lower progression of the EDSS score in MS patients taking 1,25(OH) 2 D3 compared to the placebo [37]. Burton et al. observed no difference in the change in EDSS scores between the two groups, but the proportion of MS patients with a progression of the EDSS score was lower in the vitamin D3 group [40]. In addition, one trial showed that in 32 RRMS patients, the EDSS score significantly decreased after 8 weeks of vitamin D treatment, but without randomization against a control group [35].
In eight RCTs, vitamin D supplementation had no effect on the EDSS score compared to control patients (including 334 vitamin D-treated and 298 control MS patients who completed the follow-up) [21,22,27,30,33,34,38,42]. In the SOLAR trial, the proportion of patients who were free from EDSS progression after one year was similar between the groups supplemented with either 14,000 IU daily of vitamin D3 or the placebo [21]. In the EVIDIMS trial, the change in the EDSS score after 18 months of intervention was the same as that in MS patients treated with either 20,400 or 400 IU/day of vitamin D3 [22]. In addition, Achiron et al. reported that 1,25(OH) 2 D3 did not significantly change the EDSS progression compared to the placebo [33].
Lastly, Stein et al. observed a significantly higher EDSS score in MS patients treated for 6 months with 7000 IU/day of vitamin D2 compared to those treated with 1000 IU/day [42].

MRI Lesions
We identified nine clinical trials reporting results relative to MRI lesions after vitamin D supplementation in MS patients (Table 7). They assessed 278 vitamin D-treated patients and 228 control patients (199 with placebo and 29 with low-dose vitamin D). The numbers of new or enlarging lesions were reported in all of these nine trials, and the change in the total volume of lesions was reported in five studies.  Compared with the placebo, vitamin D3 supplementation decreased the number of Gd-enhancing T1 lesions or new/enlarging T2 lesions in the SOLAR and Finnish Vitamin D Study RCTs after one year [21,29]. Camu et al. also found a significant reduction of new T1 lesions after 2 years of treatment with 100,000 IU of vitamin D3 every 2 weeks compared to the placebo [23]. In the EVIDIMS RCT, the MS patients supplemented with 20,400 IU/day of vitamin D3 developed fewer cumulative new Gd lesions compared to those treated with 400 IU/day, without reaching significance [22]. On the contrary, O'Connell et al. found no effect after 6 months of supplementation with vitamin D3 in CIS patients [27].
In the CHOLINE RCT, the total volume of MRI lesions decreased more in the vitamin D3-supplemented group than in the placebo group [23]. In SOLAR, the change in the total volume of T2 lesions was lower after one year of high-dose vitamin D3 compared to the placebo [21]. However, the EVIDIMS and Finnish Vitamin D Study RCTs found no such improvement with vitamin D3 [22,29].

Discussion
In the present systematic review, we identified 19 clinical trials reporting data on the influence of vitamin D supplementation on relapses, disability, or MRI lesions. Among the studies reviewed here, it is worth noting that the SOLAR and CHOLINE RCTs included the largest number of MS patients, and that they were controlled against placebos and conducted in a double-blind manner [21,23]. The EVIDIMS RCT is also of particular interest because it compares supplementation with a low or high dose of vitamin D3, and it included patients with either RRMS or CIS, which is relatively original [22]. We have to mention that we excluded Jelinek's study from our systematic review [49]. Although they included a very large number of MS patients (>2000), the data on vitamin D supplementation, relapses, and disability were obtained through retrospective self-reporting, which conferred a high degree of the risk of bias [49].
After reviewing the 15 selected trials reporting results on relapses, we determined that there is no solid evidence that vitamin D supplementation is effective at preventing relapses in MS patients. Overall, 10 controlled trials including 63% of the supplemented MS patients in the selected studies reported that there was a lack of significant effects on relapses following vitamin D supplementation. Among the well-designed RCTs, only the CHOLINE RCT reported a beneficial effect of vitamin D3 [23]. In this study, only clinically active RRMS patients treated with IFN-β1 and with a 25(OH)D concentration <75 nM at baseline were included. The 25(OH)D levels were around 50 nM at baseline in the vitamin D3 and placebo groups and tripled to reach 157 nM after supplementation [23]. In contrast, the SOLAR RCT, which also included clinically active RRMS patients treated with IFN-β, did not find a beneficial effect of vitamin D3 on relapses. It may be possible that vitamin D3 has beneficial effects on relapses only if MS patients are vitamin D deficient before supplementation. However, 75% of the MS participants in the SOLAR RCT also had 25(OH)D blood levels lower than 75 nM, and the 25(OH)D levels progressed similar in SOLAR to what was seen in CHOLINE [21]. In addition, two other RCTs performed in MS patients with vitamin D deficiency at baseline found no beneficial effects of vitamin D3 on relapses [29,34]. It is worth noting that CHOLINE had the longest duration of supplementation (i.e., 2 years), and therefore a time effect cannot be definitively excluded.
A previous review conducted by Hanaei et al. also concluded that supplementing MS patients with vitamin D has no significant effect on the relapse rate [50]. However, they did not include the results of the large and well-designed CHOLINE, SOLAR, and EVIDIMS RCTs. We found that the conclusions were not altered when we added the analysis of these new RCTs. A meta-analysis from the Cochrane Library published in 2018 found that vitamin D3 had no effect on the ARR, obviously without including the results from the SOLAR, CHOLINE, and EVIDIMS RCTs, which were published in 2019 and 2020 [51]. However, we cannot fully exclude to date that 1α(OH)D3 has a specific beneficial effect. While only one RCT investigated the effect of 1α(OH)D3 on relapses, it reported a higher proportion of MS patients free of relapses with 1α(OH)D3 compared with the placebo [33].
Regarding the disability in MS patients, we found the results too heterogeneous to confirm that vitamin D has a beneficial effect on the EDSS score. Indeed, five RCTs reported beneficial effects, whereas eight RCTs did not. The CHOLINE RCT found a moderately but significantly lower progression of the EDSS score in the vitamin D3-treated group compared to the placebo group. However, as already mentioned above, because CHOLINE was the longest-running trial, a time effect cannot be ruled out. Interestingly, no difference in the proportion of patients free of EDSS progression was observed after the one-year supplementation in SOLAR [21]. Consistently, the change in the EDSS score was not significantly different between the MS patients treated with a high or a low dose of vitamin D3 in EVIDIMS. In addition, the controlled trials conducted with 1α(OH)D3 or 1,25(OH) 2 D3 also yielded evidence of a lack of benefit [33,37]. Hanaei et al. concluded that vitamin D had no effect on the EDSS score in a previous systematic review, but without considering the results of the CHOLINE, SOLAR, and EVIDIMS RCTs [50]. The most recent metaanalysis available in the Cochrane Library also showed that one year of supplementation with vitamin D3 did not improve the EDSS score more than the placebo [51]. As already mentioned, this meta-analysis is from 2018 and therefore does not include the results from the SOLAR, CHOLINE, and EVIDIMS trials published in 2019 and 2020.
Regarding the MRI lesions, the present systematic review demonstrates that supplementation with vitamin D3 has beneficial effects in MS patients on the development of new lesions in the central nervous system. The placebo-controlled, double-blind, randomized SOLAR, Finnish Vitamin D Study, and CHOLINE trials reported a reduction in the appearance of new lesions of IFN-treated RRMS patients after supplementation with vitamin D3 for 12 or 24 months [21,23,29]. In addition, a similar trend was observed in the EVIDIMS RCT in patients with RRMS or CIS treated with 20,400 IU daily of vitamin D3 compared to those supplemented with 400 IU/day [22]. Three RCTs in which supplementation lasted 6 months found no effects associated with vitamin D, suggesting that the beneficial effect on new MRI lesions may occur only with more than 6 months of supplementation [27,38,42]. However, it must be mentioned that, beyond the shorter duration, these three RCTs are also unique because they were conducted only in CIS patients [27], or with vitamin D2 [42], or with vitamin D administered by the intramuscular route [38]. Contrary to our present conclusions, the most recent meta-analysis in the Cochrane Library concluded that vitamin D3 had no effect on new MRI lesions after one year of supplementation [51]. However, their conclusions were based on only two RCTs with 12-month time horizons. Since then, the results from the SOLAR, CHOLINE, and EVIDIMS RCTs were published, yielding significant new data that show a reduction in new MRI lesions in patients receiving vitamin D3 supplementation. On the other hand, we considered that the level of evidence suggesting a beneficial effect of vitamin D on the total volume of lesions remains too weak to properly draw conclusions. Indeed, the apparent discrepancies between the results from the CHOLINE and SOLAR RCTs versus the EVIDIMS and Finnish Vitamin D Study RCTs remain to be further investigated.

Conclusions
The present systematic review aimed to assess the influence of vitamin D supplementation on relapses, disability, and MRI lesions in MS patients. We concluded that vitamin D3 supplementation has a beneficial effect on new MRI lesions. In contrast, we found that there is no robust evidence to date that vitamin D supplementation is effective for preventing relapses or the progression of disability.
Here, we chose to focus on three major outcomes in MS, but other clinical outcomes are worth investigating relative to the quality of life of MS patients. Interestingly, Xie et al. recently concluded in a systematic review of RCTs that vitamin D has beneficial effects on MS-related depression [52]. In addition, the systematic review by Głąbska et al. concluded that most studies suggested that vitamin D supplementation in MS patients has a beneficial influence on the quality of life [53].
Supplementary Materials: The following supporting information can be downloaded at https://www. mdpi.com/article/10.3390/nu15081945/s1, Table S1: The applied electronic search strategy for the systematic review of databases; Table S2: Assessment of the risk of bias of the randomized trials selected in the systematic review according to the RoB 2 tool.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.