Relationship of vitamin D status with advanced liver fibrosis and response to hepatitis C virus therapy: A meta-analysis

Authors

  • Mónica García-Álvarez,

    1. Unidad de Infección Viral e Inmunidad, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Daniel Pineda-Tenor,

    Corresponding author
    1. Unidad de Infección Viral e Inmunidad, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
    • Address reprint requests to: Salvador Resino, Ph.D., or Daniel Pineda-Tenor, Ph.D., Centro Nacional de Microbiología, Instituto de Salud Carlos III (Campus Majadahonda), Carretera Majadahonda-Pozuelo, Km 2.2, 28220 Majadahonda, Madrid, Spain. E-mail: sresino@isciii.es or dpineda@isciii.es; fax: +34 918 223 269.

    Search for more papers by this author
    • These authors contributed equally to this work.

  • María A. Jiménez-Sousa,

    1. Unidad de Infección Viral e Inmunidad, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
    Search for more papers by this author
  • Amanda Fernández-Rodríguez,

    1. Unidad de Infección Viral e Inmunidad, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
    Search for more papers by this author
  • María Guzmán-Fulgencio,

    1. Unidad de Infección Viral e Inmunidad, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
    Search for more papers by this author
  • Salvador Resino

    Corresponding author
    1. Unidad de Infección Viral e Inmunidad, Centro Nacional de Microbiología, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain
    • Address reprint requests to: Salvador Resino, Ph.D., or Daniel Pineda-Tenor, Ph.D., Centro Nacional de Microbiología, Instituto de Salud Carlos III (Campus Majadahonda), Carretera Majadahonda-Pozuelo, Km 2.2, 28220 Majadahonda, Madrid, Spain. E-mail: sresino@isciii.es or dpineda@isciii.es; fax: +34 918 223 269.

    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

  • Supported by grants from Fondo de Investigación de Sanidad en España (FIS) [Spanish Health Founds for Research], grant number PI11/00245; and Red Española de Investigación en SIDA (RIS) [AIDS Research Network] grant number RD12/0017/0024; D.P.T., M.G.F., M.A.J.S., and M.G.A. are supported by “Instituto de Salud Carlos III,” grant numbers CM12/00043, RD12/0017/0024, CD13/0001 and CD12/00442, respectively.

Abstract

There is growing evidence that vitamin D is related to chronic hepatitis C (CHC) pathogenicity. We analyzed the relationship of vitamin D status with advanced liver fibrosis (ALF) in CHC treatment-naïve patients and sustained virologic response (SVR) in CHC patients on pegylated interferon alpha plus ribavirin (pegIFNα/ribavirin) therapy. We performed a meta-analysis of all eligible studies published to date (April, 2014) in PubMed, SCOPUS, LILACS, and the Cochrane Library, assessing plasma/serum vitamin D levels related to ALF and/or SVR. Pooled odds ratios (ORs) were estimated by either fixed or random effects models. Fourteen studies were selected from the literature search, seven for ALF (1,083 patients) and 11 for SVR (2,672 patients). For liver fibrosis, low vitamin D status was related to a diagnosis of ALF, with the cutoffs of 10 ng/mL (OR = 2.37, 95% confidence interval [CI] = 1.20, 4.72) and 30 ng/mL (OR = 2.22, 95% CI = 1.24, 3.97) being significant, and a near-significance for 20 ng/mL (OR = 1.44, 95% CI = 0.99, 2.12). Regarding SVR, a significant heterogeneity among studies was found (P < 0.001), and we only found a significant association with SVR for a vitamin D cutoff of 20 ng/mL (OR = 0.53, 95% CI = 0.31, 0.91). When meta-analysis was performed excluding the outliers, significant pooled ORs were found for all patients (10 ng/mL [OR = 0.48, 95% CI = 0.34, 0.67] and 20 ng/mL [OR = 0.58, 95% CI = 0.45, 0.76]) and GT1/4 patients (10 ng/mL [OR = 0.53, 95% CI = 0.34, 0.81] and 20 ng/mL [OR = 0.54, 95% CI = 0.39, 0.74]). Conclusion: Low vitamin D status in CHC patients is associated with a higher likelihood of having ALF and lower odds of achieving SVR following pegIFNα/ribavirin therapy. (Hepatology 2014;60:1541–1550)

Abbreviations
ALF

advanced liver fibrosis

CHC

chronic hepatitis C

HCV

hepatitis C virus

HIV

human immunodeficiency virus

IFNα

interferon-alpha

IL

interleukin

SVR

sustained virologic response

VDR

vitamin D receptor

Nearly 200 million people worldwide are chronically infected with hepatitis C virus (HCV), which leads to the development of cirrhosis, endstage liver disease, hepatocellular carcinoma, and liver transplantation.[1] Advanced liver fibrosis (ALF) in chronic hepatitis C (CHC) is believed to be progressive and largely irreversible.[2] Such fibrosis progression is probably due to multifactorial interactions between viral and host characteristics such as age at HCV infection, gender, daily alcohol intake, intravenous drug use, obesity, metabolic syndrome, HCV genotype, and coinfection by other viral pathogens such as human immunodeficiency virus (HIV).[3]

Throughout the past decade, CHC therapy has consisted of pegylated interferon alpha plus ribavirin (pegIFNα/ribavirin). This treatment produces rates of sustained virologic response (SVR) of around 40-50% in patients infected with HCV genotype (HCV-GT) 1, 60% in HCV-GT 4, and 80% in HCV-GT 2/3. Additionally, recent analyses have revealed that the use of pegIFNα/ribavirin in HIV/HCV-coinfected patients may provide SVR rates which approximate those of HCV-monoinfected patients.[4] Current CHC therapy involves new direct-acting antivirals plus pegIFNα/ribavirin, which has further improved the SVR rate.[5] Several host factors that influence the efficacy of IFN therapies have been identified, including age, sex, liver fibrosis, HCV genotype, HCV viral load, interleukin-28B (IL28B) polymorphisms, and obesity.[6] However, an unexplained variability in HCV treatment outcomes still remains.

There is growing evidence that vitamin D status is related to chronic liver disease.[7] The vitamin D receptor (VDR) is widely expressed in the liver and its expression is negatively associated with the severity of liver histology in CHC patients.[8] Moreover, vitamin D has antiproliferative and antifibrotic effects on the liver, and may have potential therapeutic value.[9]

Vitamin D is produced naturally during exposure to ultraviolet B radiation. It is then metabolized in the liver forming 25-hydroxyvitamin D (25(OH)D), which is later metabolized to the active form, 1,25-dihydroxyvitamin D (1,25(OH)2D), in the kidneys. In the blood, 25(OH)D is the main circulating form of vitamin D, and its concentration in plasma is the most reliable indicator of vitamin D status.[10] There is consensus that levels of 25(OH)D below 25 nmol/L (10 ng/mL) are qualified as deficient, and that over 75 nmol/L (30 ng/mL) may be required for optimal health.[11] In HCV infection, vitamin D status has been associated with CHC-related outcomes such as liver fibrosis progression in treatment-naïve patients and SVR in patients on pegIFNα/ribavirin therapy.[10] This evidence suggests the potential for vitamin D supplementation as a preventive and/or early treatment strategy for CHC.

Recently, a many publications have reported on vitamin D status and CHC, but some conflicting conclusions have been reached. To our knowledge, only one meta-analysis has been published to date about the association between vitamin D and SVR,[12] but it only involved a literature search through March 2012. This study did not analyze liver fibrosis or any other factors that might affect the relationship between the study factor and outcome. For that reason, our aim was to carefully analyze the relationship between vitamin D status and ALF in HCV treatment-naïve patients and SVR in HCV-infected patients on pegIFNα/ribavirin therapy, by conducting a meta-analysis of all eligible studies published to date (April, 2014).

Materials and Methods

Search Strategy

Relevant studies were identified by a PubMed, SCOPUS, LILACS, and the Cochrane Library literature search with the following terms: (“hepatitis C” or “HCV” or “chronic hepatitis C”) and (“vitamin D” or “ergocalciferol” or “cholecalciferol”). The information contained in this report is based on articles published before April 2014.

The meta-analysis was conducted following guidelines from Sutton et al.,[13] and the data were reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.[14]

Study Selection

We developed strict criteria for categorizing the studies by two independent reviewers:

Inclusion criteria: (1) patients infected with HCV or coinfected with HCV/HIV; (2) ALF data were exclusively derived from hepatic biopsy (not from noninvasive markers) in treatment-naïve HCV patients and/or SVR data in patients undergoing pegIFNα/ribavirin therapy; and (3) data of serum or plasma 25(OH)D levels related to ALF and/or SVR.

Exclusion criteria: (1) coinfection with hepatitis B virus (HBV); (2) absent or inadequate information about 25(OH)D levels, study population, HCV status, or not enough information to calculate the odds ratio (OR) and 95% confidence intervals (CI); (3) studies with sample size less than 40 subjects; (4) patients who received vitamin D supplementation; (5) reviews, letters, editorials, or clinical cases.

In order to select the candidate studies, we carefully screened the title and abstract of each publication. When articles fulfilled the inclusion criteria, we examined the full text and extracted data from the study. When studies included several subgroups and some of them did not fulfill the inclusion criteria, we only incorporated into the meta-analysis those subgroups that did meet the inclusion criteria. When more than one article studying the same cohort was found, only the study with the most extensive cohort was reviewed, excluding the remaining overlapping studies or data.

Data Extraction

Data were extracted independently by two investigators (M.G.A. and D.P.T.) and then cross-checked. When data were unclear or required assumptions to be made, other investigators (M.A.J.S., M.G.F., A.F.R., and S.R.) were consulted so that a consensus could be reached before recording an entry in the database. When the data were not explicitly reported, we contacted some authors of individual studies to request the data.

Outcome Variables

  1. ALF according to Metavir (F≥3). Data containing a different score were transformed to the Metavir score.[15]
  2. SVR was defined as an undetectable serum HCV-RNA level up through 24 weeks after the end of HCV treatment. HCV-GT 1/4 patients were considered as difficult-to-treat and HCV-GT 2/3 patients were considered as easy-to-treat.[16]

Quality Assessment

In order to evaluate the quality of the included studies, two investigators (M.G.A. and D.P.T.) appraised them independently using an evaluation system modified from the Newcastle-Ottawa Scale.[17] The included studies were judged on several aspects, including the selection of study populations (for both deficiency and sufficiency of vitamin D), comparability between groups, ascertainment of exposure, and demonstration of the outcomes of interest. The full score was 9 stars, and a high-quality study was defined as a study with 7 or more stars.

Statistical Analysis

All analyses were performed using Stata software (v. 11.0; Stata, College Station, TX). P < 0.05 was considered significant.

Overall, meta-analysis was performed only when two or more articles studying the same outcome were available. In all analyses, pooled ORs and 95% CIs were calculated. The significance of the pooled OR was calculated by the Z-test. A fixed effect model (Mantel-Haenszel method) was used for homogeneous studies.[18] When significant heterogeneity existed, a random effect model was applied (DerSimonian and Laird method).[19] The study heterogeneity was assessed using the Cochran's Q statistic and I2 statistic, considering a Q statistic P < 0.1 or I2 > 50% as significant heterogeneity. In addition, when heterogeneity was detected meta-regression analysis was also performed with the aim of defining the potential effect of the covariates on the outcome variables. The regression coefficients obtained describe how ALF or SVR changes with each unit increase in the covariate. Significance of the linear relationship was identified by the P value.[20] The covariables analyzed were: the cutoff for insufficiency/deficiency of vitamin D, study design, HIV coinfection, country of the study, ethnicity, and methodology for vitamin D level estimation.

Publication bias was assessed by Begg's funnel plot and the Egger linear regression test,[21] which detects funnel plot asymmetry. Publication bias was assumed to exist when the Egger test reported P < 0.05. The Galbraith plot was used to detect possible outliers of the heterogeneity,[22] which could have biased the combined estimate. This graphical method allowed us to check those studies that had a strong influence on the pooled results. The sensitivity analyses were also conducted to assess the consistency of results and to investigate the influence of one study on the overall meta-analysis.[21] It was carried out by sequential omission of individual studies.

Results

Search Results

The search strategy yielded 280 entries (Fig. 1). Forty-two of those 280 were considered to have potential value, and the full texts were retrieved for detailed evaluation. After exclusion based on detailed assessment, 14 studies were eligible for inclusion. These consisted of seven that were selected for ALF meta-analysis[23-29] and 11 for SVR meta-analysis.[23-25, 28, 30-36] Four studies were included in both analyses.[23-25, 28]

Figure 1.

Flowchart showing the selection of articles included in the meta-analysis. *Four studies were eligible for meta-analysis of both liver fibrosis and SVR.

Article Characteristics

The main characteristics of the studies analyzed are summarized in Table 1 for ALF meta-analysis and Table 2 for SVR meta-analysis. The publication year of the studies ranged from 2011 to 2014, and meta-analyses involved 1,083 individuals for ALF outcome and 2,672 for SVR outcome. Data were collected from several countries with different latitudes for long periods of time. The seasonal differences on vitamin D levels could not be considered because the majority of studies did not provide this information, and when there was, the information was very heterogeneous.

Table 1. Characteristics of Studies Included in Meta-Analysis for Advanced Liver Fibrosis (ALF)
YearFirst AuthorDesignNFibrosis ScorePeriod (Start/End)Age (Years)Gender (% Male)CountryEthnicity (%C-%AA)HCV GenotypeHIVCutoff (ng/mL)Assay
  1. AA, African Americans; C, Caucasians; CI, chemiluminescent immunoassay; EIA, enzyme immunoassay; F, fibrosis stage; HCV, hepatitis C virus; HIV, human inmunodefficiency virus; ICMA, immunochemiluminometric; LC, liquid chromatography; NR, not reported; RIA, radio immuno assay; SVR, sustained virologic response.

2014Guzmán-Fulgencio (25)Cross-sectional174Metavir (51)2000/200840.874.7Spain100% C1-4Yes10EIA
2013Petta (26)Cross-sectional260Scheuer (52)NR52.849.2Italy100% C1No30LC
2013Ladero (27)Cross-sectional89Metavir (51)January/May 201254.362.03Spain100% C1, non 1No20CI
2013Mandorfer (28)Cross-sectional65Metavir (51)NR38.671Austria100% C1-4Yes10/30EIA
2012Lange (29)Cohort303Metavir (51)2000/20054563.62Switzerland100% C1-4No20RIA
2011Milazzo (23)Case-control48Ishak (53)2007/20104574.19Italy100% C1, non 1Yes30RIA
2011Bitetto (24)Cross-sectional114Ishak (53)2005/20094751.7Italy100% C1-5No20ICMA
Table 2. Characteristics of Studies Included in Meta-Analysis for the Analysis of Response to HCV Treatment
YearFirst AuthorDesignNHCV TreatmentPeriod (Start/End)Age (Years)Gender (% Male)CountryEthnicity (%C-%AA)HCV GenotypeHIVCutoff (ng/mL)Assay
  1. AA, African Americans; C, Caucasians; CI, chemiluminescent immunoassay; EIA, enzyme immunoassay; F, fibrosis stage; HCV, hepatitis C virus; HIV, human inmunodefficiency virus; ICMA, immunochemiluminometric; LC, liquid chromatography; NA, north africans; NR, not reported; RIA, radio immuno assay; SVR, sustained virologic response.

2014Guzmán-Fulgencio (25)Cross-sectional125pegIFNα/RBV2000/200840.874.7Spain100% C1-4Yes10/30EIA
2013Mohamed (31)Case-control50pegIFNα/RBV2010/ 20114068Egypt100% NA4No30ELISA
2013Mandorfer (28)Cross-sectional65pegIFNα/RBVNR38.671Austria100% C1-4Yes10/30CI
2013Bitetto (32)Case-control190pegIFNα/RBV2005/20094751.7Italy100% C1-5No10/20ICMA
2012Weintraub (33)Cross-sectional171pegIFNα/RBV2001/200947.2541.52USA62% AA 38% C1No20/30RIA
2012Terrier (34)Cross-sectional106pegIFNα/RBV2001/201058,248.93France100% C1,4, otherNo12/30RIA
2012Falleti (35)Cross-sectional206pegIFNα/RBV2005/20094851.5Italy100% C1-5No20ICMA
2011Milazzo (23)Case-control51pegIFNα/RBV2007/20104574.19Italy100% C1, non 1Yes30RIA
2011Bitetto (24)Cross-sectional211pegIFNα/RBV2005/20094751.7Italy100% C1-5No10/20ICMA
2011Terrier (30)Cross-sectional189pegIFNα/RBV4th trimesters39,577.24France100% C1-4, otherYes10/20/30RIA
2011Bitetto (36)Cohort42IFNα/RBV;pegIFNα/RBV1996/20065471.42Italy100% C1, otherNo10/20ICMA

The quality scores of the included studies are summarized in Supporting Table 1 (case-control and cross-sectional analysis) and Supporting Table 2 (cohort studies). Overall, the quality score was high (7.83 ± 0.71 and 7.5 ± 0.71, respectively).

Plasma/Serum Vitamin D Levels

Nearly 70% of all patients had suboptimal 25(OH)D levels (below 20 or 30 ng/mL, depending on the cutoff of the study), and almost 50% of the HCV-infected patients had deficient 25(OH)D levels (less than 10 or 20 ng/mL). Vitamin D status was lower in HIV/HCV-coinfected patients, as 82.7% of them had suboptimal levels, compared to 66.2% of HCV-monoinfected patients.

Meta-Analysis of Advanced Liver Fibrosis

A total of seven studies examined the association between vitamin D status and ALF. No publication bias was observed by Begg's Funnel plot and Egger's test (Supporting Fig. 1).

Figure 2 shows the pooled ORs for each different threshold used to define the deficiency or insufficiency of vitamin D. From seven studies analyzed, six of them showed a direct relationship between deficiency or insufficiency of vitamin D and ALF, but only three of them were statistically significant. Overall, we found a significant association between vitamin D status and ALF, including cutoffs of 10 ng/mL (OR = 2.37, 95% CI = 1.20, 4.72) and 30 ng/mL (OR = 2.22, 95% CI = 1.24, 3.97). The cutoff of 20 ng/mL was very close to significance (OR = 1.44, 95% CI = 0.99, 2.12), possibly due to the low number of articles included.

Figure 2.

Forest plot of the meta-analysis performed to investigate the association between vitamin D levels and ALF. ALF, advanced liver fibrosis; CI, confidence intervals; F, fibrosis stage; OR, odds ratio.

The Cochran's Q statistic and I2 statistic did not show heterogeneity among the studies (P > 0.1; Fig. 2). This finding was supported by Galbraith's plots (Supporting Fig. 2) and by meta-regression analysis (Supporting Table 3).

Meta-Analysis of SVR

A total of 11 studies examined the association between vitamin D status and SVR. No publication bias was observed by Begg's Funnel plot and Egger's test (Supporting Fig. 3).

Figure 3 shows the pooled ORs for each different threshold used to define vitamin D status. From 11 studies analyzed, nine of them showed a direct relationship between deficiency or insufficiency of vitamin D and SVR. Overall, the pooled OR only indicated a significant association between the cutoffs of 20 ng/mL and SVR for patients not stratified by HCV genotype (OR = 0.53, 95% CI = 0.31, 0.91). The Cochran's Q statistic and I2 statistic suggested a strong heterogeneity among the studies included in this analysis (Fig. 3).

Figure 3.

Forest plot of the meta-analysis performed to investigate the association between vitamin D levels and SVR. CI, confidence intervals; GT, genotype; OR, odds ratio; SVR, sustained virologic response.

A meta-regression analysis was conducted to examine the source of heterogeneity (Table 3), and only ethnicity had a potential effect on SVR (P = 0.047). Galbraith's plots identified four outliers as sources of heterogeneity (Supporting Fig. 4): Mohamed et al.,[31] Weintraub et al.,[33] Terrier et al.,[30] and Terrier et al.[34] When the meta-analysis was performed excluding these outliers, the heterogeneity disappeared (Supporting Fig. 5). Thus, significant pooled ORs were found for all patients (10 ng/mL [OR = 0.48, 95% CI = 0.34, 0.67] and 20 ng/mL [OR = 0.58, 95% CI = 0.45, 0.76]) and in patients infected with HCV-GT 1/4 (10 ng/mL [OR = 0.53, 95% CI = 0.34, 0.81] and 20 ng/mL [OR = 0.54, 95% CI = 0.39, 0.74]) (Supporting Fig. 5).

Table 3. Meta-Regression Analysis for SVR According to Vitamin D Deficiency
 CoefficientStandard ErrorP Value95% CI
  1. CI, confidence interval; GT, genotype; HCV, hepatitis C virus; HIV, human inmunodefficiency virus; SVR, sustained virologic response.

A) SVR (HCV-GT All)    
 Cutoff0.0930.3270.781−0.834, 0.646
 Study design−0.1240.2790.642−0.766, 0.497
 HIV coinfection−0.2190.3870.584−1.096, 0.656
 Region0.0470.1530.766−0.301, 0.395
 Racial descent−0.9810.4250.047−1.944, −0.018
 Method Vitamin D0.0880.3030.779−0.597, 0.773
B) SVR (HCV-GT 1/4)    
 Cutoff0.0760.3320.821−0.640, 0.793
 Study design−0.6200.7770.451−2.459, 1.219
 HIV coinfection−1.2180.6360.097−2.723, 0.287
 Region0.2590.3320.460−0.526, 1.045
 Racial descent−1.1960.6350.102−2.698, 0.305
 Method Vitamin D0.5100.5000.342−0.672, 1.693
C) SVR (HCV-GT 2/3)    
 Cutoff−0.6280.3910.169−1.634, 0.377
 Study design0.4171.5250.810−6.144, 6.979
 HIV coinfection0.9980.8610.366−2.707, 4.703
 Region−0.6610.6890.438−3.625, 2.302
 Method Vitamin D−0.6610.6890.438−3.625, 2.302

Additionally, we replicated the analysis only considering the studies of HCV-monoinfected patients (Supporting Fig. 6). The Cochran's Q statistic and I2 statistic suggested a strong heterogeneity, which was similar to that observed among the studies of HIV/HCV-coinfected patients. Nevertheless, the pooled OR indicated a significant relationship between the cutoff of 20 ng/mL and SVR for both patients not stratified by HCV genotype (OR = 0.44, 95% CI = 0.27, 0.71) and patients infected with HCV-GT 1/4 (Supporting Fig. 6).

Discussion

The two major results of our meta-analysis results were: (1) a significant direct association between plasma vitamin D deficiency/insufficiency and ALF and (2) a significant inverse association between low vitamin D status and achieving a successful virologic response to pegIFNα/RBV therapy.

HCV-infected subjects usually have vitamin D deficiencies, which are related to liver disease severity and a lower chance of responding to pegIFNα/RBV therapy.[7, 9] However, there is inconsistency in the published data, possibly due to the heterogeneity in the study designs, such as characteristics of patients (HCV infection or HIV/HCV coinfection, ethnicity), characteristics of HCV infection (genotype), and characteristics of vitamin D assessment (seasonality, cutoff values, and methodology). For this reason, our meta-analysis focused on displaying general conclusions about the trend of these associations (overall OR), while serving as a robust tool to investigate heterogeneous results.

Based on the global analysis, there was a significant association between suboptimal vitamin D levels and ALF. However, not all studies detected significant differences between groups, perhaps due to the different criteria considered in each individual study. However, the heterogeneity between studies was low and the covariates included in the meta-analysis did not have significant values (ethnic origin, study design, HCV genotype, HIV coinfection, vitamin D cutoff value, and methodology to measure vitamin D).

The role of vitamin D in CHC patients is controversial and the mechanisms by which these associations occur are not well established.[10] On the one hand, since the liver is responsible for 25(OH)D production, patients with liver injury may have low levels of 25(OH)D due to low vitamin D absorption and impaired 25(OH)D synthesis. However, liver function needs to be severely compromised for this impairment to occur.[37] On the other hand, low vitamin D levels have been associated with the more severe histologic changes in liver disease.[38] This effect could be explained by the antiproliferative and antifibrotic effects of vitamin D on the liver[39] and the role of vitamin D as an immune modulator that reduces inflammation and enhances protective immune responses.[40] Thus, it probably is a feedback mechanism. Severe liver failure impairs hydroxylation of vitamin D and causes vitamin D deficiency, which in turn increases hepatic fibrosis, leading to liver failure. However, since almost all of the included studies have a cross-sectional design, it is impossible to dissect the temporal relationship between 25(OH)D levels and liver disease severity.

Interestingly, severe liver disease is usually associated with vitamin D deficiency,[10] while ALF usually results in a lower SVR rate in CHC patients on HCV therapy.[41] Therefore, it would be expected that vitamin D levels would be associated with SVR. In our meta-analysis, we found a general trend towards a direct relationship between low vitamin D levels and decreased SVR rate. However, this is not always so, for example, not for all HCV genotypes nor for all vitamin D cutoffs. This may be due to the degree of heterogeneity detected during the analysis of SVR (P < 0.05; I2 > 50%). Additionally, ethnicity was detected as a covariate that affects the results found in the SVR meta-analysis. In fact, when the meta-analysis was performed excluding the four outlier articles (Mohamed et al.,[31] Weintraub et al.,[33] Terrier et al.,[30] and Terrier et al.[34]), only patients of Caucasian ethnicity remained. As a consequence, the heterogeneity disappeared and the statistical associations were stronger. Thus, low vitamin D levels were linked to lower odds of achieving SVR after pegIFNα/RBV therapy for all patients as well as for patients infected with HCV-GT 1/4 (difficult-to-treat). On the other hand, this meta-analysis did not detect significant associations for HCV-GT2/3 (easy-to-treat), possibly due to the smaller number of patients with these HCV genotypes and the high SVR rate. Another hypothesized cause of heterogeneity was the presence of HIV coinfection. Terrier et al.[30] detected a lack of significant association between vitamin D levels and response to pegIFNα/RBV therapy in HIV/HCV coinfected patients. This discordancy could be due to the more complex situation of HIV/HCV coinfected patients. However, although vitamin D metabolism is affected by HIV infection and combined antiretroviral therapy (cART),[37] we found similar association values for HCV monoinfected and HIV coinfected patients. In fact, HIV coinfection was not detected as a significant covariate in meta-regression analysis.

The major source of vitamin D is the exposure to natural sunlight, and therefore, the major cause of vitamin D deficiency is inadequate exposure to sunlight.[42] Thus, the influence of race, country latitude or study region, and seasonality in vitamin D status have been previously described.[42] People with a naturally dark skin tone have natural sun protection and require at least three to five times longer exposure to make the same amount of vitamin D as a person with a white skin tone. Besides, vitamin D deficiency has been proposed as a cause of the racial differences in response rates to antiviral therapy for CHC.[33, 43] Moreover, the season variability had a substantial impact on 25(OH) levels,[44] but, unfortunately, most studies did not establish the relationship between the seasonal levels of vitamin D and the outcome variables. Only four items studied this effect, finding that both HCV-infected individuals[29, 34] and HIV/HCV-coinfected patients[23, 25] suffered vitamin D insufficiency during all seasons. The other factor that might influence is the study region, but we did not find values in the meta-regression.

In this meta-analysis, a high prevalence of vitamin D insufficiency (25(OH)D <30 ng/mL) and deficiency (25(OH)D <10 ng/mL) was found in HCV-infected patients. Vitamin D deficiency is almost universal among patients with chronic liver disease,[45] and specifically in HCV-infected patients,[45-47] as well as in HIV/HCV-coinfected patients.[23, 28, 30] Besides, prevalence of vitamin D deficiency in HCV-infected[46] and HIV/HCV-coinfected patients[23] is higher than in the general population. However, the prevalence of insufficiency or deficiency of vitamin D has been found in the general population is not so far from the values found in patients infected with HCV. In fact, the National Diet and Nutrition Survey (NDNS) provides evidence of low vitamin D status, as defined by a plasma 25(OH)D concentration less than 10 ng/mL, in most age groups in the United Kingdom population, and highlights a greater risk of vitamin D deficiency in population subgroups.[11] Moreover, a major point of controversy is the establishment of optimal 25(OH)D levels. The latest recommendations suggest that a 25(OH)D level over 20 ng/mL is sufficient to meet the vitamin D requirement.[44] However, the Endocrine Society Clinical Practice Guideline (ESCPG) suggested that vitamin D requirements may be greater for sick patients than for healthy individuals and blood vitamin D levels above 30 ng/mL may have additional health benefits in reducing the risk of various disease conditions.[42]

Moreover, the ESCPG also recommended screening for vitamin D deficiency in individuals at risk for deficiency, including ones with hepatic failure, and supplementation with vitamin D for these deficient patients. In our meta-analysis, the cutoff 10 ng/mL showed higher pooled OR for liver fibrosis and SVR, although the cutoff 30 ng/mL also showed significant values for ALF and was nearly significant for SVR. Thus, our study suggests that the cutoff 30 ng/mL may be an appropriate threshold to prevent fibrosis and HCV therapy failure. In CHC patients, vitamin D supplementation has been suggested as a preventive and/or early treatment strategy.[9] Another major point is the vitamin D supplementation to improve the SVR rate in patients treated with pegIFNα/ribavirin.[9] However, there are conflicting data in the scientific literature. Thus, a small number of studies have reported the positive impact of vitamin D supplementation on SVR rates in patients treated with pegIFNα/ribavirin.[36, 48-50] However, further clinical investigations on the effect of vitamin D supplementation in treating CHC are needed to confirm this item.

Finally, in order to properly interpret our results, some considerations have to be taken into account. First, our meta-analysis was performed using the unadjusted raw data provided from each study, whereas most of the results given by the authors were adjusted by age, gender, HCV viral load, and/or other factors. For this reason, the pooled ORs may differ slightly from those cited by the original articles. Second, the seasonal differences in vitamin D levels, dietary intake, polymorphisms of vitamin D hydroxylating enzymes, and other variables involved in vitamin D metabolism, such as parathyroid hormone, have not been considered. IL28B polymorphisms have not been taken into account for evaluating SVR. Third, only data of blood 25(OH)D levels were used to evaluate vitamin D status rather than the 1,25(OH)2D assay. The serum 1,25(OH)2D assay is only recommended for monitoring certain conditions, such as acquired and inherited disorders of vitamin D and phosphate metabolism.[42, 44] Fourth, multiple methodologies for measuring 25(OH)D were used, including enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), high-performance liquid chromatography (HPLC), and liquid chromatography tandem mass spectroscopy. However, the vitamin D detection methodology was not a source of heterogeneity in this meta-analysis. Fifth, the majority of the studies had a cross-sectional design and the number of studies in some subgroup analyses was small, which might have led to weak results. Consequently, these results should be interpreted with caution.

In conclusion, this meta-analysis shows that a low vitamin D status in CHC patients is associated with a higher likelihood of having ALF and lower odds of achieving SVR, suggesting the utility of vitamin D screening in HCV-infected patients.

Acknowledgment

We thank those authors of the studies included in this review for providing additional information upon request.

Authors Contributions

S.R. designed and supervised the study. M.G.A. and D.P.T. collected all data, performed the statistical analysis, and drafted the report. M.A.J.S., M.G.F., and A.F.R. participated in the interpretation of the data and critically reviewed the report. All authors approved the final version of the article.

Ancillary

Advertisement