Study design
This systematic review and meta-analysis was conducted according to the PRISMA guidelines (15). The protocol of this review was registered in the open access PROSPERO database prior to conducting the review, number CRD42021254473 (https://www.crd.york.ac.uk/prospero/).
Data sources
With the help of experienced libraries and information scientists, the PubMed (https://pubmed.ncbi.nlm.nih.gov/), Scopus (https://www.scopus.com/home.uri), and EMBASE (https://www.embase.com) databases were searched for relevant English-language articles. In addition, the reference lists of all identified studies were searched for relevant articles, and gray literature was searched for in Google Scholar (https://scholar.google.com/). The search was restricted to the period from 1st January 2000 to 8th October 2022. All the articles were exported to Mendeley Desktop 1.19.8 software (Mendeley Ltd., London, UK) for further processing, and duplicates were removed.
Search strategy
The search terms were combined using Boolean operators OR for synonyms and ‘AND’ across elements of PICOS (population, intervention, comparison, outcome and study design), as follows:
The terms for the population of interest were ‘HIV’, ‘AIDS’, ‘human immunodeficiency virus’, and ‘acquired immune deficiency syndrome’. This population was restricted to sub-Saharan Africa by country name: Africa OR Algeria OR Angola OR Benin OR Botswana OR Burkina Faso OR Burundi OR Cameroon OR Canary Islands OR Cape Verde OR Central African Republic OR Chad OR Comoros OR Democratic Republic of Congo OR Djibouti OR Egypt OR Equatorial Guinea OR Eritrea OR Ethiopia OR Gabon OR Gambia OR Ghana OR Guinea OR Guinea OR Guinea Bissau OR Ivory Coast OR Kenya OR Lesotho OR Liberia OR Libya OR Libi OR Libia OR Madagascar OR Malawi OR Mali OR Mauritania OR Mauritius OR Morocco OR Mozambique OR Mocambique OR Namibia OR Niger OR Nigeria OR Principe OR Reunion OR Rwanda OR Sao Tome OR Senegal OR Seychelles OR Sierra Leone OR Somalia OR South Africa OR St Helena OR Sudan OR Swaziland OR Tanzania OR Togo OR Tunisia OR Uganda OR Western Sahara OR Zaire
The search terms for the outcome of interest were ‘oropharyngeal’, ‘candidiasis’, ‘OPC’, ‘Oral thrush’, ‘Candida’, ‘Non Albicans Candida’, ‘NAC’, ‘non Candida albicans’, Candida’, ‘C. albicans’, ‘C. parapsilosis’, ‘C. glabrata’, ‘C. tropicalis’, ‘C. dubliniensis’, C.krusei’, ‘C. norvegensis’, ‘ C. guilliermondii’, C. albicans’, ‘C. glabrata’, ‘C. tropicalis’, ‘C. krusei’, ‘C. dubliniensis’, ‘C. parapsilosis’, ‘C. guilliermondii’, ‘C. famata’, ‘C. kefyr’, ‘C. norvegensis’, ‘C. sake’, ‘C. lusitaniae’, ‘C. pintolopesii’, ‘C. pseudotropicalis’, ‘C.globosa’, ‘C. dattila’, ‘C. inconspicua’, ‘C. hellenica’, ‘C. holmii’, ‘C. pulcherrima’, ‘ C. valida’, ‘C. africana’, ‘C. fabianii’, ‘C. cacaoi’, ‘C. zeylanoides’.
The search terms for study design were ‘cross sectional’, ‘observational’, ‘descriptive’, ‘prevalence’, ‘transverse’, ‘cohort’, and ‘case control’.
Specific terms for the intervention and comparator were not applied since this review was neither an intervention nor a comparison study. This search was restricted to the period from 1st January 2000 to 8th October 2022. In addition, the reference lists of all included studies were searched, and gray literature was searched on Google Scholar for more articles.
Review question and eligibility criteria
The review question was “What is the prevalence of oropharyngeal candidiasis and the distribution of Candida species among people living with HIV in Africa?” This question was described further (Table 1) using the PICOST framework, which guided the eligibility of the included studies. Studies were included if they were published in the English language between 1st January 2000 and 8th October 2022 and if they reported outcomes of interest, that is, the prevalence of OPC and distribution of Candida species causing OPC among PLHIV in Africa. This review included only observational studies with either cross-sectional, case‒control or cohort study designs reporting OPC and Candida species among PLHIV living in Africa. We included studies that diagnosed OPC infection based on both the presence of oral lesions and the mycological identification of Candida species isolated from the oral cavity of PLHIV. We excluded studies that reported the prevalence of OPC without information on the causative Candida species. We excluded animal model reports and observational studies whose full text could not be retrieved even after request from the corresponding authors and comprehensive library search.
Study outcomes
The primary study outcome was the prevalence of OPC in PLHIV. The secondary outcome was the distribution of Candida species (Table 1).
Study selection process
Data management
Using Mendeley Desktop referencing software version 1.19.8 (Mendeley Ltd., London, UK), we imported all identified titles, excluded duplicates, and screened and grouped these into relevant eligibility categories as described in our Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flow chart (Figure 1).
Minimizing bias in study identification and selection
Two reviewers (BM and AAK) carefully conducted the literature search. Two independent reviewers (HI and GM) examined relevant studies and screened their titles and abstracts for eligibility. After the initial screening, the full texts of the eligible studies were retrieved and examined for eligibility by RN and AM. Disagreements were resolved by discussion with two reviewers (BA and OJS) to reach consensus.
Data extraction
Data extraction was performed using a spreadsheet developed from Microsoft Excel version 16 (Microsoft Corporation, Richmond, Seattle, Washington, USA). The extracted data included the first author, year of publication, country where the study was conducted, study design, sample size, gender, mean age of the study population, Candida identification method, and prevalence of OPC and Candida species. The data were extracted in duplicate by BM and RN, and any disagreements were resolved by a third party (AM).
Qualitative assessment
Two reviewers (GM and BM) independently assessed the risk of bias of the included studies, and any discrepancies between the two reviewers were resolved by reaching a consensus through discussion. Eligible studies were assessed for risk of bias using the Joanna Briggs Institute quality assessment tool for prevalence studies (16). This tool consists of 9 parameters: (1) an appropriate sampling frame to address the target population, (2) a proper sampling method, (3) an adequate sample size, (4) a description of the study subject and setting, (5) sufficient data analysis, (6) the use of valid methods for the identified conditions, (7) valid measurements for all participants, (8) the use of appropriate statistical analysis, and (9) an adequate response rate. Each parameter was scored either 1 for failure to answer the question or 0 for ability to answer the question. The risk of bias was classified as low, moderate, or high, with total scores ranging from 0 to 2, 3 to 4 or 5 to 9, respectively (Table 4).
Synthesis
The extracted data were imported into STATA 17.0 statistical software (STATA, College Station, Texas) for analysis. Descriptive statistics and narrative synthesis were used to summarize the data and present the results from the included eligible studies, respectively. A random effects cumulative meta-analysis was performed using STATA 17.0 to map the prevalence of OPC among PLHIV for each study and estimate the summary estimate (meta-analysis) in Africa. These results are displayed in a forest plot. We visually explored heterogeneity by inspecting the forest plot and statistically quantifying it using the I2 statistic. Because we found a high level of heterogeneity, we conducted a meta-regression testing the variables of year of publication and sample size before conducting a sensitivity analysis by a leave-one-out sensitivity analysis. Additionally, publication bias was evaluated using funnel plots and Egger’s test at a significance level of < 0.05. This sensitivity analysis explored the significant differences between the distribution of Candida species and the prevalence of OPC among people living with HIV in Africa. Any value with P < 0.05 was considered statistically significant at the 95% confidence interval (CI).
Sensitivity analysis
To identify the source of heterogeneity, a leave-one-out sensitivity analysis was employed. Sensitivity analysis was performed using the random effects model.
Meta-regression
Meta-regression analysis was performed to explore the associations between prevalence of OPC and the year of publication and study sample size.