Intestinal parasites and human immuno virus in Ethiopian tuberclosis patients: A Systematic review and meta-analysis.

Background: The distribution of intestinal parasites among patients with tuberculosis in Ethiopia has not been well understood. Thus, this systematic review and meta-analysis was designed to determine the pooled national prevalence of intestinal parasites and its association with HIV among patients with tuberculosis in Ethiopia. Methods: Original articles were searched in PubMed, Google Scholar, EMBASE, and World Health Organization’s HINARI portal and supplemented by the hand searching of cross-references. Data were extracted using a standard data extraction checklist. Random-effects model was used to estimate the pooled prevalence of intestinal parasites and odds ratio of the association. The I2 statistic was utilized to quantify statistical heterogeneity across studies. Funnel plot asymmetry and Egger’s regression tests were used to check for publication bias. The analysis was done by STATA version 14 for Windows. Results: Of 725 identified studies, 12 articles were eligible for inclusion in the final analysis. The pooled national prevalence of intestinal parasites among patients with tuberculosis in Ethiopia was 36.1% (95% CI: 22.1 – 50.1, I2:98.7%). Subgroup analysis based on study design indicated that the prevalence of intestinal parasite among case-control studies was 41.69% (95% CI: 28.6 – 54.8, I2: 95.1%). The odds of intestinal parasites among patients with tuberculosis- HIV co-infection was not significantly different compared with patients with tuberculosis without HIV/AIDS (OR: 0.99, 95% CI: 0.7 - 4.7, p: 0.96). Conclusion: In Ethiopia, at least one out of three patients with tuberculosis have intestinal parasite. These findings suggest the need of more attention on increasing screening tuberculosis patients for intestinal parasites and deworming intervention.


Introduction
Parasitic and tuberculosis infections are among the most prevalent infections in humans in developing countries ( 1). The overlap of TB and parasitic disease morbidity presented with high and consistent figures all over the world ( 2-6). More than half of people with latent or active tuberculosis infections have intestinal parasitic infestation, which are common in high tuberculosis (TB) burden nations ( 7). In Africa, one-third of TB patients have intestinal parasite infection(IPIs) ( 8) that contribute for high rate of therapeutic failure of pulmonary TB ( 9).
It is evident that pulmonary tuberculosis and parasitic diseases were shown to be risk factors for each other and represented with high magnitude of comorbidity in developing countries ( 8,10). Co-infection may significantly inhibit the host's immune system, increase antibacterial therapy intolerance and be detrimental to the prognosis of the disease. In addition, infection with parasitic diseases can alter the protective immune response to mycobacterium tuberculosis. This indicates that concomitant helminthic infections weaken immune resistance to mycobacterial infections ( 11,12). Furthermore, the presence of intestinal parasites affects the effectiveness of vaccine against TB and increases the chance of developing active tuberculosis diseases after vaccination ( 13).
In addition, parasitic infection presented with more severe radiological pulmonary disease in the number of involved lung zones at the end of TB treatment course. In particular, simultaneous intestinal helminth infection in patients with newly diagnosed TB alters patient's immunity profile that could favor persistent MTB infection and a more protracted clinical course of the disease ( 14,15).
In developing countries, parasitic worm infections among TB patients are common and increases the chance for TB complication. Parasitic infection increases the chance of TB lung damage and induces the progression of a latent TB infection to active tuberculosis disease ( 16). The presence of high tuberculosis progression, delay in clinical response and sustained infectiousness leads to high transmission rate of TB to healthy individual (17)(18)(19). In sum, high prevalence of TB and IPIs comorbidity can be explained on the basis of low level of socioeconomic and immunity status ( 27). Literatures about the prevalence and risk factors of intestinal parasites among these vulnerable populations is an input to policy makers and program planners to have due attention and to take appropriate measures. However, there has been a limited understanding about the extent of parasitosis co-morbidities with tuberculosis in Ethiopia. The available few evidences are inconclusive and inconsistent. Hence, pointing the overall level of comorbidity might alarm policy makers, ministry of health to develop interventional guidelines or bi-directional frameworks for effective treatment of TB and parasite infection comorbidity as well as successful implementation of the national TB control and prevention strategy. Therefore, this systemic review and meta-analysis aimed to determine the pooled national prevalence of intestinal parasites and its association with HIV among patients with tuberculosis in Ethiopia.

Search approach and appraisal of studies
Articles were accessed through web-based electronic database searches, desk reviews of the grey literature, and cross-references of identified studies. The electronic databases searched were Searching was done using the following key terms: "intestinal diseases, parasitic", "intestinal", "diseases", "parasitic", "parasitic intestinal diseases", "intestinal", "parasites", "intestinal parasites", intestinal helminthes", "intestinal protozoa", "soil-transmitted helminthes", "tuberculosis", "patients", "Ethiopia". These key terms were combined used "AND" and "OR" Boolean operators. To allow a comprehensive search strategy, these key terms were pre-defined that included all fields within records. Medical Subject Headings (MeSH terms) was also used to help expand the search in advanced PubMed search. The study period for searching of article was conducted from November comorbidity of TB/ IPIs in the study divided by the total number of patients in a study multiplied by 100. We also estimate the association (as measured by crude odds ratios) between comorbidity and patient's HIV status.

Inclusion and exclusion criteria
All articles for the study were included if they met the following inclusion criteria: written in Englishlanguage, full-text articles on observational studies (case-control or cross sectional), conducted in Ethiopia from 2004 to 2018, published in peer-reviewed journals or that available at local shelves, and used internationally accepted diagnostic criteria to diagnose intestinal parasite. On the other hand, studies which didn't report specific outcomes for intestinal parasites quantitatively were excluded from the final analysis.
Data abstraction and quality assessment After preliminary assessment, two reviewers downloaded abstracts to assess for inclusion. The abstracts were assessed for agreement with the inclusion criteria. When it was unclear whether an abstract was relevant, it was included for retrieval. At this stage articles deemed irrelevant or out of the scope of the study were excluded and the full-text of the rest articles downloaded for a detailed review. Two reviewers then assessed the quality of potentially eligible articles using Newcastle-Ottawa Scale (NOS) criteria ( 28) that was validated previously( 29). There are three main categories for NOS criteria: Selection, Exposure, and Comparability. A study can be awarded a maximum of one star for each numbered item within the selection and exposure categories. A maximum of two stars can be given for comparability categories. Then the number of stars that was given for each numbered item was computed to determine the quality score of each study. Discrepancies in quality assessment scores were resolved with a third reviewer whenever appropriate. The average of two independent reviewer's score was used to determine the quality of each article. The current systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocol (PRISMA-P) guidelines ( 30).

Data analysis
Using a Microsoft Excel spreadsheet template that develop based on Joanna Briggs Institute format, author name, publication year, region, dominant parasites, the age-range of patients, and funder were extracted from each study ( 31). These data were then imported to Stata version 14 software for Windows to compute the pooled prevalence of intestinal parasites and to examine the association with HIV. The statistical heterogeneity of study outcomes was assessed using the I 2 statistics ( 32). Methodological heterogeneity was evaluated by comparing study design.
Additionally, clinical heterogeneity that can arises from differences in participant characteristics (e.g., sex, age, baseline disease severity, ethnicity, comorbidities) and types or timing of outcome measurements was assessed with face-to-face discussion by authors. First, we set and agree on definition of Variations in sex, age, baseline disease severity, ethnicity, comorbidities) and types or timing of outcome measurements. Next, we have discussed and had a consensus about how much this heterogeneity affects our pooled analysis. Random-effects model was used to estimate the pooled prevalence of IPs at a 95% confidence interval (CI). Because theoretically we expected that the study settings and socio-economic contexts might differ radically across these studies, subgroup analysis based on study design and group of parasites. We used visual examination of funnel plot asymmetry and Egger's regression tests to check for publication bias ( 33). Meta-regression analysis was employed to identify the source of heterogeneity using publication years, sample size and region as covariates. Sensitivity analysis was also performed to examine influence of each study on the overall effect size. We appraised the risk of bias using the 10item rating scale developed by Hoy et al. for prevalence studies ( 34). The tool assesses studies on 10 domains, including sampling, data collection, reliability and validity of study tools, case definition, and prevalence periods. Each study was assigned a score of 1 ("yes" answers to domain questions) or 0 ("no" answers to domain questions) for each domain, and these scores were summed to provide an overall study quality score. Scores of 8-10 were considered as having a "low risk of bias", 6-7 a "moderate risk", and 0-5 a "high risk". For the final risk of bias classification, disagreements between the reviewers were resolved via consensus (Additional file 1).

Identification and description of studies
Of 725 identified studies, 260 duplicate articles were excluded after reviewing the titles and abstracts. Next, 452 articles were excluded with a reason of irrelevance. Finally, the full-text of the remaining 13 articles were downloaded and assessed for quality and for the presence of all required information. One article was excluded since the outcome was not clearly stated ( 35). The remaining 11 studies were included in the final meta-analysis (Figure 1).

Meta-regression analysis
The univariate meta-regression analysis, showed that a one person increases of sample size brought a variation on prevalence of each study by 0.2 (p = 0.03) ( Table 2).  Our finding was high compared to a report in Egypt (16%)( 9) and China (21.7%) ( 47). The Egyptian study ( 9) was only assessed the burden of a single parasite, which was not a dominant parasite in our finding. In our study, Ascaris lumbricoides was the dominant parasite. Thus, high prevalence of Ascaris lumbricoides in our study might increase the overall IPIs prevalence and can be accounted for this variation. Socioeconomic difference between Ethiopia and China can be contributed for this discrepancy. Ethiopia is low-income country whose peoples live under poverty whereby risk factors, such as poor sanitation, low health awareness, low urbanization and low health care service, high On the other hand, our finding was lower than the study report in brazil (65%) ( 7). This difference can be explained by the variation in the study area. Our study estimated the prevalence from studies conducted without consideration of socio-economical class.
However, the Brazilian study (   7) was conducted among low-income community, in particular in the area where there is inadequate sanitary structure, latrines being shared with other rooms, well water was the main source of drinking, and near to half of families had low income.
Our subgroup analysis result showed that estimate from case-control studies was higher than crosssectional studies. This difference might happen related to the nature of sample selection. The unique advantage of cross-sectional surveys is that it is possible to determine the prevalence of an outcome because the study samples are selected randomly ( 51). Conversely, in case-control studies, cases and/or controls are selected on criteria related to the exposure of interest that can affect the overall prevalence ( 52). Based on the category of parasite subgroup analysis, the highest percentage was reported for helminths group with a prevalence of 43.3%. This finding was consistent with previous studies that conducted among people with TB (

STUDY STRENGTH AND LIMITATIONS
The present meta-analysis is the first in Ethiopia that can give nationally pooled evidence about the overall national burden of IPs among patients with TB.
This meta-analysis has also limitations. First, most of the studies reviewed were published in small regional journals making it difficult to gauge the extent of peer-review and quality of studies.
However, all included studies passed the quality screening criteria. Second, due to the absence of data, unadjusted odds ratios were used to estimate the effect size of HIV, which prevented us excluding confounding effect of other factors. The findings of this meta-analysis would be best interpreted keeping these analytical limitations and the limitations of the original studies in mind.

Conclusion
In Ethiopia, at least one out of three patients with tuberculosis have intestinal parasites. These findings suggest the need of more attention on increasing screening tuberculosis patients for intestinal parasites and deworming to reduce its effect on patient's overall health status.