Vitamin D deficiency or supplementation and the risk of human herpesvirus infections or reactivation: a systematic review protocol

Introduction Human herpesviruses induce lifelong latent infections and may reactivate as the immune system deteriorates. Recent studies have suggested that vitamin D, an essential element of bone health, may have some effect of protecting against infections, but investigations of its potential to prevent herpesvirus infection or reactivation are limited. We will review the current literature examining vitamin D and the risk of herpesvirus infections or reactivation. Methods and analysis Our systematic review will address two research questions: (1) Do deficient/insufficient serum vitamin D levels increase the risk of herpesvirus infections and (2) Does vitamin D supplementation protect against herpesvirus infections? We will include only intervention studies with control groups, cohort studies and case-control studies. We will use subject headings and keywords to search for synonyms of ‘vitamin D’ and ‘herpesviruses’ (including herpes simplex virus type 1 and 2, varicella-zoster virus, cytomegalovirus, Epstein-Barr virus and human herpesviruses type 6, 7 and 8) in Medline, Embase, Global Health, Web of Science, Scopus and Cochrane Central Register of Controlled Trials, and the grey literature databases Open Grey, EThOS and BASE from inception to 31 August 2019. References to the included articles and relevant systematic reviews will also be examined. Two reviewers will independently screen the study titles and abstracts, and examine the full texts to decide the final eligibility. They will independently extract data from the studies and assess bias using the Cochrane Collaboration approach. A third researcher will solve any discrepancies. The results will be narratively synthesised; if an adequate number of studies is included and the homogeneity between studies is acceptable, a meta-analysis will be performed. We will assess the quality of evidence using the Grading of Recommendations, Assessment, Development and Evaluation framework, and display the results in a summary of findings table. Ethics and dissemination Ethical review is not required for a systematic review. We will publish the results in a peer-review journal. Any amendments to the protocol will be recorded in the supplementary section. PROSPERO registration number CRD42019130153.


Methods and analysis:
A systematic review approach will address two research questions: 1. Do deficient/insufficient serum vitamin D levels increase the risk of herpesvirus infections, and 2. Does vitamin D supplementation protect against herpesvirus infections? We will include only intervention studies, cohort studies and case-control studies. We will use subject headings and keywords to search for synonyms of "vitamin D" AND "herpesviruses" in databases, including Medline, Embase, Global Health, Web of Science, Scopus, and Cochrane Central Register of Controlled Trials, and the grey literature databases Prevention Information & Evidence eLibrary, Open Grey, EThOS, and BASE. The references of the included articles and relevant systematic reviews will also be examined. Two reviewers will independently screen the study titles and abstracts, and examine the full texts to decide the final eligibility. They will independently extract data from the studies and assess bias using the Cochrane Collaboration approach. The third researcher will solve any discrepancy in inclusion, data extraction and bias assessment. The results will be narratively synthesized; if an adequate number of studies is included and the homogeneity between studies is acceptable, a meta-analysis will be carried out. We will assess the quality of evidence using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) framework, and display the results in a summary of findings table.

Ethics and dissemination
Ethical review is not required for a systematic review. We will publish the results in a peer-review journal. Any amendments to the protocol will be recorded in the

STRENGTHS AND LIMITATION OF THIS STUDY:
 This systematic review will be undertaken following the predefined population, exposure, comparator and outcome framework  All available databases, including six major databases and four grey literature databases, will be searched to obtain all eligible studies.  Herpesviruses are a group of double-stranded DNA viruses that infect humans and some animals. After infecting their hosts, these viruses cannot be eradicated; instead, they establish latency and persist for life. As the host's immunity declines, these viruses can reactivate to induce various symptoms. There are eight human herpesviruses ( which may lead to blindness 2 , malignancy 3 , and increased financial burdens 4 .
Consequently, investigating immunomodulatory factors associated with infection or reactivation of this virus family is important.
Vitamin D is mainly endogenously synthesized by the skin after sun exposure and can be supplied through dietary intake and supplementation. It plays an important role in absorbing calcium and phosphate, which are essential for bone health 5 . Recently, some studies have indicated that vitamin D may have potential immunomodulatory effects. In response to pathogen exposure, immune cells, such as monocytes or macrophages, upregulate vitamin D receptors and enzymes which catalyse vitamin D to synthesize anti-pathogenic peptides 6 . One meta-analysis indicated that among patients receiving dialysis due to chronic kidney diseases, patients with higher or normal serum vitamin D levels had a 39% lower risk of infection than those who were deficient in vitamin D (relative risk ratio: 0.61, 95% confidence interval: 0.41-0.89) 7 . Furthermore, vitamin D confidence interval: 0.81-0.96) 8 . However, limited evidence exists concerning whether vitamin D is associated with herpesvirus infections.
In this study, we will comprehensively review studies of the effect of serum vitamin D levels or the use of oral vitamin D supplementation on infection with, or reactivation of any of the eight human herpesviruses.

OBJECTIVE
This review aims to explore the association between vitamin D and herpesviruses. The proposed systematic review will address two primary research questions:

METHODS
This study protocol will be reported according to the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) 9 .

Eligibility Criteria Study Design and Characteristics
To identify the possible causal association between vitamin D and herpesvirus infections, we will review observational studies, including cohort and case-control studies, and intervention studies, including randomized or non-randomized controlled trials. Descriptive studies, ecological studies, cross-sectional studies, case reports and case series will not be included. If a systematic review relevant to our topic is found, we will review its references.

Participants
Only human studies will be included in our review, and animal or cell studies will be excluded. Studies from all age groups or involving patients with any immune status are eligible.

Exposure
We have two exposures of interest for each research question in our review. For the first research question, the exposures are deficient or insufficient serum vitamin D levels in the participants. We will include articles in which the serum vitamin D levels are  Table 1). We will exclude studies using topical vitamin D analogues, because their effects on systemic serum vitamin D levels are unknown.

Comparators
The comparator groups for vitamin D insufficiency and deficiency are those with sufficient serum vitamin D levels. For those receiving oral vitamin D supplementation or treatment, the comparators are those without vitamin D supplementation or treatment, respectively.

Outcomes
The outcomes will be infection with or reactivation of all eight human herpesviruses listed in Table 1. Infection with and reactivation of herpesviruses are defined using clinical or laboratory criteria. The clinical criteria include patients presenting with classical symptoms, for instance, such as the painful rash of herpes zoster, or a diagnosis recorded by physicians. Laboratory criteria include using laboratory techniques to confirm the diagnosis, such as evaluation of the serum viral load by PCR.
Studies reporting only serum antibodies against herpesviruses will not be included.

Information sources
We will search the following database from inception until June 2019: Medline (Ovid), EMBASE (Ovid), Web of Science, Scopus, Cochrane Library, and Global Health (Ovid).
Grey literature databases, including Open Grey, Prevention Information & Evidence eLibrary, BASE, and EThOS, will also be searched. Finally, the clinical trial register at ClinicalTrials.gov will be searched to ensure the completeness of the literature.

Search strategy
We will search for synonyms of "human herpesviruses" and "vitamin D" using both controlled vocabularies and keywords in each database. A search strategy in Medline (Ovid) is listed in Supplementary Table 2. The subject headings will be modified for different databases. The results will be combined using the Boolean logic operator "AND." For some database with limited search functions, such as single line search, keywords will be split into small sections to fulfil the requirement (Supplementary Table   3). In addition to searching electronic databases, we will manually search for the references lists of the included articles and relevant systematic reviews.

Data management
One researcher will import the search results from different databases into the citation management software EndNote X9 (Clarivate Analytics, version 9.1/2019). Duplicated results will be identified and deleted.

Selection process
Two researchers will independently screen the titles and abstracts of all identified studies. We will obtain full texts of all studies fulfilling the review criteria, and the two researchers will screen all articles to establish their eligibility for inclusion. Any discrepancy in the reviewing process will be adjudicated by the third researcher.

Data collection process
Data extraction for the first three studies will be performed by two independent researchers to ensure the integrity of the process; then, one researcher will extract data from the remaining studies. If any data are missing or unclear, we will contact the authors for further clarification and information.

Data items
We will summarize and extract data from the included studies using a Population, Exposure/Intervention, Comparator, Outcomes, and Study characteristics framework as follows: 1

Risk of bias in individual studies
We will assess the risk of bias from the included studies using a template form based on the Cochrane approach for trials and observational studies, and all relevant domains of bias will be assessed for different study types. For randomized controlled trials, we will evaluate bias due to the following sources: 1. random sequence generation; 2. allocation concealment; 3. blinding of participants, personnel, or assessment; 4.
incomplete outcome data; and 5. selective reporting. For observational studies, we will consider bias due to the following sources: 1.confounding factors; 2. selection of participants or controls; 3. differential and non-differential misclassification of the exposure or outcome; 4. reverse causation; and 5. missing data (Supplementary Table  F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   4). A piloted form will be tested by applying it to extract data from the first three studies.
To ensure the quality and consistency of the risk of bias assessment, the first three studies will be evaluated by two independent researchers. Then, one researcher will complete the evaluation of the remaining included studies. Any discrepancies will be examined by the third researcher.

Data synthesis and meta-bias (es)
We will comprehensively search different databases including grey literature databases to minimize bias in our search. We will use a narrative synthesis to summarize the data and results from the eligible studies included in our review. Since each herpesvirus subtypes have differing pathogenic pathways, we will display the results separately by virus. If an adequate numbers of studies with acceptable homogeneity are included, a meta-analysis will be performed to integrate the study results. We will decide whether to use fixed-or random-effects models by considering the I 2 value for heterogeneity; values >50% will be considered to represent substantial heterogeneity 11 . If the number of included studies is sufficient, we will carry out subgroup analyses of subjects with different ages, immune statuses, or latitude to explore the source of heterogeneity. A funnel plot will be used to present the distribution of studies and examine any possible publication bias 12 . All statistical analyses will be performed by using STATA version 15.

Confidence in cumulative evidence
We will evaluate the quality of evidence using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) framework 13 , or that are well adjusted for plausible confounders will receive higher grades, whereas those with a higher risk of bias, inconsistency, indirectness, or imprecision will be downgraded. We will conclude the quality of evidence using a 'Summary of Findings' table and assign each outcome a quality rank of "high", "moderate", "low" or "very low" for the level of confidence.

Instructions to authors
Complete this checklist by entering the page numbers from your manuscript where readers will find each of the items listed below.
Your article may not currently address all the items on the checklist. Please modify your text to include the missing information. If you are certain that an item does not apply, please write "n/a" and provide a short explanation.
Upload your completed checklist as an extra file when you submit to a journal.
In your methods section, say that you used the PRISMA-Preporting guidelines, and cite them as:  Methods and analysis: 10 Our systematic review will address two research questions: 20 References to the included articles and relevant systematic reviews will also be 21 examined. Two reviewers will independently screen the study titles and abstracts, and 22 examine the full texts to decide the final eligibility. They will independently extract data  Th 1 lymphocytes and increases Th 2 69 lymphocytes, and it also intensifies the effect of regulatory T lymphocyte responses. 10 11 70 Regarding the effects of vitamin D associated AMPs on herpesviruses, a cell study 71 indicated that Cathelicidin decreased HSV-1 viral titres isolated from keratoconjunctivitis 72 patients; 12 furthermore, another cell study also showed that vitamin D supplementation 73 reduced HSV-1 viral load and mRNA expression in HSV-1 infected cells. 13 74 Vitamin D also shows some anti-infective potential in epidemiological studies.   Table   158 3). In addition to searching electronic databases, we will manually search for the 159 references lists of the included articles and relevant systematic reviews. 165 Selection process 166 Two researchers will independently screen the titles and abstracts of all identified 167 studies. We will obtain full texts of all studies fulfilling the review criteria, and the two 168 researchers will screen all articles to establish their eligibility for inclusion. Any 169 discrepancy in the reviewing process will be adjudicated by the third researcher.
170 Data collection process 171 Data extraction for the first three studies will be performed by two independent 172 researchers to ensure the integrity of the process; then, one researcher will extract data 173 from the remaining studies. If any data are missing or unclear, we will contact the 174 authors for further clarification and information.

175
Data items 176 We will summarize and extract data from the included studies using a Population, 177 Exposure/Intervention, Comparator, Outcomes, and Study characteristics framework as 178 follows:  207 We will assess the risk of bias from the included studies using a template form based on A piloted form will be tested by applying it to extract data from the first three studies. 217 To ensure the quality and consistency of the risk of bias assessment, the first three 218 studies will be evaluated by two independent researchers. Then, one researcher will 219 complete the evaluation of the remaining included studies. Any discrepancies will be 220 examined by the third researcher.

221
Data synthesis and meta-bias (es) 222 We will comprehensively search different databases including grey literature databases 223 to minimize bias in our search. We will use a narrative synthesis to summarize the data Confidence in cumulative evidence 236 We will evaluate the quality of evidence using the Grading of Recommendations, 237 Assessment, Development and Evaluations (GRADE) framework 25 . We will evaluate the 238 design and risk of bias across all included studies. Studies with significant effects, 239 strong dose responses, or that are well adjusted for plausible confounders will receive 240 higher grades, whereas those with a higher risk of bias, inconsistency, indirectness, or 241 imprecision will be downgraded. We will conclude the quality of evidence using a 242 'Summary of Findings' table and assign each outcome a quality rank of "high", 243 "moderate", "low" or "very low" for the level of confidence.

Ethics and dissemination
245 Any changes in the study protocol will be recorded and reported. Any revision of the 246 protocol will be recorded in the supplementation of the final published review.

Instructions to authors
Complete this checklist by entering the page numbers from your manuscript where readers will find each of the items listed below.
Your article may not currently address all the items on the checklist. Please modify your text to include the missing information. If you are certain that an item does not apply, please write "n/a" and provide a short explanation.
Upload your completed checklist as an extra file when you submit to a journal.

CRD42019130153
166 Two researchers will independently screen the titles and abstracts of all identified 167 studies. We will obtain full texts of all studies fulfilling the review criteria, and the two 168 researchers will screen all articles to establish their eligibility for inclusion. Any 169 discrepancy in the reviewing process will be adjudicated by the third researcher.
170 Data collection process 171 Data extraction for the first three studies will be performed by two independent 172 researchers to ensure the integrity of the process; then, one researcher will extract data 173 from the remaining studies. If any data are missing or unclear, we will contact the 174 authors for further clarification and information.

175
Data items 176 We will summarize and extract data from the included studies using a Population, 177 Exposure/Intervention, Comparator, Outcomes, and Study characteristics framework as 178 follows:  222 We will comprehensively search different databases including grey literature databases 223 to minimize bias in our search. We will use a narrative synthesis to summarize the data Confidence in cumulative evidence 236 We will evaluate the quality of evidence using the Grading of Recommendations, 237 Assessment, Development and Evaluations (GRADE) framework 25 . We will evaluate the 238 design and risk of bias across all included studies. Studies with significant effects, 239 strong dose responses, or that are well adjusted for plausible confounders will receive 240 higher grades, whereas those with a higher risk of bias, inconsistency, indirectness, or 241 imprecision will be downgraded. We will conclude the quality of evidence using a 242 'Summary of Findings' table and assign each outcome a quality rank of "high", 243 "moderate", "low" or "very low" for the level of confidence.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59

Ethics and dissemination
245 Ethical review is not required for a systematic review. We will publish the results in a 246 peer-review journal. Any changes in the study protocol will be recorded and reported.

Instructions to authors
Complete this checklist by entering the page numbers from your manuscript where readers will find each of the items listed below.
Your article may not currently address all the items on the checklist. Please modify your text to include the missing information. If you are certain that an item does not apply, please write "n/a" and provide a short explanation.
Upload your completed checklist as an extra file when you submit to a journal.