Prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia: a systematic review and meta-analysis

Background Co-trimoxazole is used as a prophylaxis for human immunodeficiency virus (HIV) patients to prevent opportunistic infections. Its widespread use results in the emergence of co-trimoxazole resistance, which is a significant problem. This systematic review and meta-analysis determined the pooled prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia. Methods The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline was applied to report this study. The protocol has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the assigned number CRD42024532240. Article search was performed using electronic databases such as PubMed, Medline, EMBASE, Google Scholar, Hinari, Web of Science, Science Direct, and African Journals Online. Data were extracted using a Microsoft Excel spreadsheet and analyzed using STATA version 11.0 software. A random-effects model was used to estimate the pooled effect size of co-trimoxazole resistance across studies with a 95% confidence interval. The heterogeneity was checked using I2 statistic. The presence of publication bias was determined using a funnel plot and Egger’s test with a p-value <0.05 evidence of statistically significant bias. Subgroup and sensitivity analyses were performed. Results Twenty-two studies with 5,788 HIV-infected individuals were included. The pooled prevalence of co-trimoxazole resistance was 61.73% (95% CI: 53.10–70.37%), with heterogeneity (I2 = 87.7%) and statistical significance (p < 0.001). A higher co-trimoxazole resistance was observed in HIV-infected individuals with urinary tract infection; 82.10% (95% CI: 75.03–89.17%). Among the bacterial spp., higher resistance to co-trimoxazole was observed in Escherichia coli; 70.86% (95% CI: 53.44–88.27%) followed by Salmonella spp.; 67.66% (95% CI: 41.51–93.81%) and Proteus spp.; 66.23% (95% CI: 34.65–97.82%). Conclusion There is a higher prevalence of co-trimoxazole resistance in HIV-infected individuals in Ethiopia. This alarms WHO’s recommendation of co-trimoxazole prophylaxis guidelines to review and update it. Additionally, a nationwide assessment of co-trimoxazole resistance in Ethiopia as a whole is required. Systematic review registration: identifier: CRD42024532240.


Introduction
Human immunodeficiency virus (HIV), which later causes acquired immunodeficiency syndrome (AIDS), is one of the global health threats (1).According to the Joint United Nations Programme on HIV/AIDS, 2022 Global HIV and AIDS Statistics fact sheet, 84.2 million people have become infected with HIV since the HIV epidemic began, and 40.1 million have died from AIDS-related illnesses.Besides, around 650,000 people died from AIDS-related illnesses worldwide in 2021 (2).Ethiopia has made remarkable strides in controlling the HIV/AIDS epidemic over the past decade.However, the prevalence remains relatively high in urban areas where estimates indicate a 3 % rate compared to under 1 % nationally (3).
Co-trimoxazole, a combination of trimethoprim and sulfamethoxazole antibiotics, is recommended as a prophylaxis for HIV-infected individuals to prevent opportunistic infections; however, the indication may vary according to CD4 cell count, age, geographical setting, and prior medical history (4,5).The use of co-trimoxazole prophylaxis in low-income settings is recommended by the World Health Organization (WHO) for all children with HIV and for adults with advanced HIV infection or who are living in areas where bacterial infections are prevalent, irrespective of antiretroviral therapy (ART) (4,6).
Previously, research had shown that co-trimoxazole prophylaxis decreases morbidity and death in HIV-infected individuals (7).However, the current scenario might be different with over 80% of bacteria causing opportunistic infections are resistant to co-trimoxazole (8).The widespread use of co-trimoxazole prophylaxis in HIV patients is thought to be the cause of an increasing burden of antimicrobial resistance throughout the AIDS era.The emergence of co-trimoxazole resistance among HIV-infected individuals has become a significant problem (8).A study in Tanzania reported a 75% bacterial resistance to co-trimoxazole from HIV patients and 81.3% resistance in patients with prophylaxis (8).
In Ethiopia, different individual studies were conducted on the pattern of co-trimoxazole resistance to bacterial infections or colonizations in HIV-infected individuals.This systematic review and meta-analysis report is necessary to provide valuable data on the status of co-trimoxazole resistance among HIV-infected individuals in Ethiopia, which is crucial for revising the recent guideline of co-trimoxazole prophylaxis.Therefore, this systematic review and meta-analysis determined the pooled prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia.

Study design and protocol registration
This systematic review and meta-analysis were performed as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (9) (Supplementary Table S1).The protocol has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the assigned number CRD42024532240.

Database search strategy
The co-trimoxazole resistance in bacterial pathogens that infect or colonize HIV-positive people was the main focus of this metaanalysis.The appropriateness of the included studies for this metaanalysis was assessed using the COCOPOP (Condition, Context, and Population) paradigm.Ethiopia served as the context (CO), HIV-infected individuals as the population (POP), and the study's condition (CO) was the prevalence of co-trimoxazole resistance.Studies published before 2024 were included in the search, and it was last conducted until March 25, 2024.Electronic databases such as PubMed, Medline, EMBASE, Google Scholar, Hinari, Web of Science, Science Direct, and African Journals Online were searched to identify articles reporting the prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia.We used search terms alone and in combination with Boolean operators such as "OR" or "AND."An example of a PubMed search strategy used was as follows: (((((bacteria) OR (infection)) OR (colonization))) AND ((((((antimicrobial resistance) OR (antibiotic resistance)) OR (co-trimoxazole resistance)) OR (trimethoprim-sulfamethoxazole resistance)) OR (antibiogram))) AND ((((human immunodeficiency virus) OR (immunocompromised)) OR (HIV/AIDS)) OR (HIV))) AND (Ethiopia).A manual search was conducted for relevant papers in the references of the included studies and other reviews.The articles retrieved were imported into EndNote X9 bibliographic software manager (Clarivate Analytics, Philadelphia, PA, United States).

Outcome of interest
The main finding in this systematic review and meta-analysis was the national assessment of the pooled prevalence of co-trimoxazole resistance among HIV-infected individuals.Co-trimoxazole resistance based on the type of bacteria spp.was the second outcome of this study.

Study selection and eligibility criteria
Two independent reviewers (MA and GG) screened the titles and abstracts of the identified studies to determine their eligibility.Full-text articles were then assessed for eligibility, and any disagreements between the reviewers were resolved through discussion.Articles published in the English language with a cross-sectional study design that included the prevalence of co-trimoxazole resistance in HIV-infected individuals in the results, studies conducted in Ethiopia and without limit on specimen type, type of infection or colonization, and study period.
Studies not reporting co-trimoxazole resistance profile, case reports, review articles, meta-analysis studies, conducted on tuberculosis, sexually transmitted infections (viral and protozoa), or parasites, studies not conducted in Ethiopia, and study populations other than HIV-infected individuals were excluded from the study.

Quality assessment
After removing duplicated papers, all potentially eligible papers were reviewed.Full-text papers were retrieved for review and relevant information was extracted.The Joana Briggs Institute (JBI) critical appraisal checklist for prevalence study was used to assess the quality of included studies (10).Accordingly, articles with high and medium quality (fulfilling 50% of quality assessment) were included in the final analysis (Supplementary Table S2).Two independent authors (MA and GG) assessed the quality of the studies, and any disagreement was solved by discussion.

Data extraction
Data were extracted from each study using the designed tool in Microsoft Excel 2019 (Microsoft Corp., Redmond, WA, United States) by two independent authors (MA and GG).Any ambiguity and difference during extraction were resolved through discussion.The data extracted from eligible studies were author name, year of publication, area in which the study was conducted, study design, type of sampling method, number of HIV-infected individuals, sex (male), age group, type of infection or colonization, number of total isolates, bacterial group as Grampositive or Gram-negative, number of bacterial species, co-trimoxazole resistance, and co-trimoxazole prophylaxis.

Data analysis and synthesis
The relevant data were extracted from the eligible studies into a Microsoft Excel spreadsheet and then exported to the STATA version 11.0 software for meta-analysis.The pooled prevalence of co-trimoxazole resistance and 95% confidence intervals were visually displayed using a forest plot.Subgroup analysis was performed based on region, sampling technique, year of study, type of infection or colonization, and bacterial group.The heterogeneity of the included studies was evaluated using the inverse variance index (I 2 statistic).No, low, moderate, and high levels of heterogeneity were considered when the values of I 2 were 0, 25, 50, and 75%, respectively (11).In all pooled analyses, heterogeneity resulting from differences in effects across different studies was accounted for using a random-effects model.A sensitivity analysis of each study's impact on the overall prevalence was also conducted.Publication bias was statistically investigated using Egger's test (12) and visual inspection of funnel plots.A p-value <0.05 in Egger's test was considered as evidence of statistically significant publication bias.

Search results
A total of 1761records were retrieved during an electronic database search.The 155 were screened after removing records for several reasons such as studies with irrelevance of topics, not published in the English language, studies on tuberculosis, sexually transmitted infections, or intestinal parasites, studies not conducted in Ethiopia, and study populations other than HIV-infected individuals.Moreover, 61 duplicates were removed and 94 full-text articles were assessed for eligibility.Seventy-two full-text articles were excluded because of no co-trimoxazole resistance data, not fulfilling study quality criteria (Q-score), case reports, review articles, and meta-analysis studies.Finally, 22 studies were included in the systematic review and metaanalysis (Figure 1).

Studies characteristics
A total of 22 studies (13-34) comprising 5788HIV-infected individuals were included in this systematic review and metaanalysis.All of these studies used a cross-sectional study design.Regarding the sampling technique, 11 studies used systematic random sampling, 8 studies used convenient sampling, and the remaining 3 studies did not report their sampling techniques.Studies included 9 from Amhara, 5 from South Ethiopia, 3 from Addis Ababa, 2 from Sidama, and 1 from (Oromia, Harari, and Tigray).The minimum and maximum numbers of study participants were 100 in Gondar (19) and 450 in Addis Ababa (16), respectively.Of the 5788HIV-infected individuals, 3,318 (57.3%) were females.The proportion of HIV-infected individuals on co-trimoxazole prophylaxis was not consistently reported in published articles; only 40.9% of studies have reported the status of prophylaxis.Studies reported that a total of 859 HIV-infected individuals were given the prophylaxis (Table 1).

Publication bias
Studies were assessed for potential publication bias with a subjective examination of funnel plot and statistics using Egger's test and funnel plot (p-value = 0.028), indicating the presence of publication bias (Table 2; Figure 8).

Sensitivity analysis
We conducted a sensitivity analysis for the prevalence of co-trimoxazole resistance using a random effects model by step-by-step removal of each study.The results showed that the exclusion of studies does not have a significant effect on the pooled prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia (Figure 9).

Discussion
This systematic review and meta-analysis is the first to determine the pooled prevalence of co-trimoxazole resistance in bacteria causing different types of infection or colonization among HIV-infected individuals in Ethiopia.The study included According to recent studies from the African continent, Gram-negative pathogenic bacteria show significant rates of co-trimoxazole resistance, ranging from 50 to 97% (35)(36)(37)(38).In this study, the pooled prevalence of co-trimoxazole resistance among HIV-infected individuals was 61.73% in all types of bacteria.In addition, the prevalence of co-trimoxazole resistance was higher in E. coli, followed by Salmonella and Proteus spp.It is suggested that co-trimoxazole prophylaxis may cause resistance in HIV-infected individuals for a variety of reasons, including prior exposure to co-trimoxazole or other sulphonamide-based drugs, non-adherence to co-trimoxazole prophylaxis, mutation, and patient self-medication, even though it lowers the risk of antimicrobial resistance by preventing infections and minimizing hospitalizations (42,43).According to a study, there is evidence of cross-resistance between co-trimoxazole and penicillin antibiotics, and co-trimoxazole prophylaxis finally leads to the selection of co-trimoxazole resistance and MDR in pneumococci (15).
Additionally, co-trimoxazole is used for the prevention of Pneumocystis jiroveci (previously carinii) pneumonia and bacterial opportunistic infections in HIV-infected individuals.In this study, the subgroup analysis revealed a higher co-trimoxazole resistance in bacterial uropathogens, followed by bloodstream Moreover, there was a significant heterogeneity among studies (I 2 = 87.7%),which is not surprising given the variation in study settings, study period, type of bacterial infection or colonization, and type of bacterial species identified in the studies.A metaanalysis of co-trimoxazole resistance for most bacteria could not be performed because of the small number of studies reported, and we included bacteria reported in at least three studies.The presence  Forest plot showed subgroup analysis based on type of infection or colonization.Forest plot showed subgroup analysis based on bacterial group.
of high heterogeneity across studies and limited number of studies in certain regions may impact the generalizability of findings.

Limitations of the study
As a strength, this study is the first report of pooled prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia.Due to the absence of published articles, we could not include all regions in Ethiopia, which resulted lack of generalizability for the problem.In addition, viral load, CD4 count, and co-trimoxazole prophylaxis were not fully reported in our systematic review and meta-analysis.

Conclusion and recommendations
According to our study, there is a higher prevalence of co-trimoxazole resistance to bacterial species causing

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.Sensitivity analysis.

FIGURE 1 Flow
FIGURE 1Flow diagram describing the selection of studies for the systematic review and meta-analysis on the prevalnce of co-trimoxazole resistance among HIV-infected individuals in Ethiopia.

FIGURE 2 Forest
FIGURE 2Forest plot showed the pooled prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia.

FIGURE 3
FIGURE 3Forest plot showed subgroup analysis based on study region.

FIGURE 4 Forest
FIGURE 4Forest plot showed subgroup analysis based on year of study.

FIGURE 5 Forest
FIGURE 5Forest plot showed subgroup analysis based on sampling technique.

FIGURE 7
FIGURE 7 opportunistic infections or colonization in HIV-infected individuals.This provides information for the WHO recommendation of co-trimoxazole prophylaxis to consider the need for modification to the prophylaxis agent and a detailed review of current guidelines.It can include medical care and regular follow-up to monitor compliance and side effects during co-trimoxazole administration.Moreover, epidemiologic surveillance on the extent of co-trimoxazole resistance is required in Ethiopia as a whole by including regions not described in this study.

FIGURE 8 Funnel
FIGURE 8Funnel plot indicated publication bias in the studies on the co-trimoxazole resistance among HIV-infected individuals in Ethiopia.

TABLE 1
Summary of studies on the prevalence of co-trimoxazole resistance among HIV-infected individuals in Ethiopia.

TABLE 2
Publication bias using Egger's test.