The impact of diabetes mellitus on the emergence of multi-drug resistant tuberculosis and treatment failure in TB-diabetes comorbid patients: a systematic review and meta-analysis

Background The existence of Type 2 Diabetes Mellitus (DM) in tuberculosis (TB) patients is very dangerous for the health of patients. One of the major concerns is the emergence of MDR-TB in such patients. It is suspected that the development of MDR-TB further worsens the treatment outcomes of TB such as treatment failure and thus, causes disease progression. Aim To investigate the impact of DM on the Emergence of MDR-TB and Treatment Failure in TB-DM comorbid patients. Methodology The PubMed database was systematically searched until April 03, 2022 (date last searched). Thirty studies met the inclusion criteria and were included in this study after a proper selection process. Results Tuberculosis-Diabetes Mellitus patients were at higher risk to develop MDR-TB as compared to TB-non-DM patients (HR 0.81, 95% CI: 0.60–0.96, p < 0.001). Heterogeneity observed among included studies was moderate (I2 = 38%). No significant change was observed in the results after sub-group analysis by study design (HR 0.81, 95% CI: 0.61–0.96, p < 0.000). In the case of treatment failure, TB-DM patients were at higher risk to experience treatment failure rates as compared to TB-non-DM patients (HR 0.46, 95% CI: 0.27–0.67, p < 0.001). Conclusion The results showed that DM had a significant impact on the emergence of MDR-TB in TB-diabetes comorbid patients as compared to TB-non-DM patients. DM enhanced the risk of TB treatment failure rates in TB-diabetes patients as compared to TB-non-DM patients. Our study highlights the need for earlier screening of MDR-TB, thorough MDR-TB monitoring, and designing proper and effective treatment strategies to prevent disease progression.


Introduction
The existence of Type 2 Diabetes Mellitus (DM) in tuberculosis (TB) patients causes serious public health concerns (1).People across different regions of the world are becoming affected by this comorbidity (2).The occurrence rate of DM in TB patients is rising globally and causes severe health problems in TB patients (3).DM affected more than 50% of TB patients across all regions of the world (4,5).This co-morbid condition leads to many unfavorable TB treatment consequences such as treatment failure rates and increased drug resistance cases in TB patients (6,7).
One of the challenges in controlling this TB-DM comorbid condition is the development of MDR-TB, in which resistance develops for both Isoniazid and Rifampicin (8).The presence of DM enhances the chances of developing MDR-TB and DM was recognized as a predictor for its emergence (9).The prevalence of MDR-TB ranged from 10 to 30% in diabetics (10,11).The emergence of MDR-TB in TB-DM comorbid patients is very dangerous for the health of people (8).It can be the leading cause of death in people (8).The TB treatment cure rates have been reduced from 96 to 54% due to the presence of MDR in diabetic patients.The reason for reduced TB treatment cure rates and increased treatment failure rates may be due to complex treatment protocols and the adverse effects of alternative medications given to patients with MDR-TB (8).Moreover, it puts an increased financial burden on patients and becomes one of the reasons for disease progression (8,12).
The "WHO Collaborative Framework for Care and Control of Tuberculosis and Diabetes" helps to guide healthcare professionals in the effective management of both diseases at administrative and clinical levels (13).The link between DM and MDR-TB is not wellstudied in the literature.Only 13 studies were included in a previous systematic review and meta-analysis that reported the association between DM and MDR-TB and found that DM is a strong predictor to develop MDR-TB in TB patients and this TB-DM comorbid condition can potentially lead to high treatment failure rates (9).But to the best of the author's knowledge, there is still a lack of studies on the association between DM and MDR-TB and TB treatment failure results.Moreover, there is also a need for adjusting potential covariates.A more comprehensive systematic assessment of the available literature is needed for further understanding of their association in TB-DM comorbid patients.Thus, this study's goal was to investigate the impact of DM on the Emergence of MDR-TB and Treatment Failure rates in TB-DM comorbid patients versus TB-non-DM patients.

Methodology Study design
The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines were being followed by this study (14).

Literature search
In this review, the PubMed database was systematically searched until April 15, 2023 (the last date of searching).For PubMed Search terms used were as follows: "Diabetes Mellitus" [Mesh]) OR "Diabetes Mellitus, Type 2" [Mesh]) AND "Tuberculosis" [Mesh]) OR "Tuberculosis, Pulmonary" [Mesh]) AND "Treatment Outcome" [Mesh]) AND "Drug resistance" [Mesh]) OR "Multidrug resistance" [Mesh]).The studies that were fulfilling the criteria were also searched from the references of the included studies.The review was only performed on English-language published papers.

Inclusion and exclusion criteria
The inclusion criteria for this study are as follows: the research articles (case-control, cohort or cross-sectional) reporting the association of DM with both MDR-TB and treatment failure rates with adjusted covariates data, studies involving human subjects receiving anti-TB therapy, clinical outcome studies, studies investigating diabetes mellitus's impact on tuberculosis treatment failure results in tuberculosis patients, and articles published in the English language.The study excluded non-research articles, non-human studies, reports, editorials, studies not analyzing the MDR-TB and treatment failure rates in patients, studies involving pre-DM and type-1 DM patients, and studies for which full text could not be retrieved.Three reviewers independently reviewed papers that fulfilled the selection process.If there were any conflicts between the reviewers, then those were clarified through discussion or the involvement of fourth reviewer.

Study selection
The first step performed was to screen the studies by reading their title and abstracts.The entire content of the selected studies

Data extraction
From the research articles included in this meta-analysis, the relevant information was collected using a data collection form.Author information, year, country of study, study design, participants, inclusion criteria, treatment regimen, covariates, TB treatment outcomes, and regression results with 95% CI and p values were the data of the included studies extracted.The relevant data were extracted independently by two reviewers.The data of included studies extracted were reviewed by a third reviewer.Any conflicts were sorted out through discussion until a final decision was taken.

Statistical analysis
Heterogeneity among studies was assessed by using I 2 test.The pooled HR (95% CI) were calculated using the fixed effects model.Subgroup analyses were performed by study design to assess the impact of study design on results.p < 0.01 was considered statistically significant.All statistical analyses were conducted by the meta program in STATA version 12.0 (Stata Corp, College Station, Texas).

Search results
When searching the databases, 2,690 results were retrieved.A total of 1,257 articles qualified for the preliminary screening after 1,433 duplicates were removed.Based on their titles and abstracts, 1,208 studies were irrelevant to our study objective and were excluded.Out of 49 studies that were available for full-text review, 19 studies were excluded out of which seven studies were review articles, two studies were editorials, and eight studies did not use multivariable analysis for adjustment of covariates.Thirty studies were selected for inclusion in this meta-analysis after a complete article overview.Figure 1 depicts the selection process of studies included in the meta-analysis.Impact of DM on development of MDR-TB.

MDR-TB
The results of pooled data of MDR-TB for all included studies are shown in Figure 2.This study showed that TB-DM patients were at higher risk to develop MDR-TB as compared to TB-non-DM patients (HR 0.81, 95% CI: 0.60-0.96,p < 0.001).There was 19% reduced risk in TB-non-DM patients for MDR-TB emergence as compared to TB-DM patients.Heterogeneity observed among included studies was moderate (I 2 = 38%).To check the impact of the study design on the results, the subgroup analysis was conducted.Only studies that were prospective were included in the sub-group analysis.No significant change was observed in the results (HR 0.81, 95% CI: 0.61-0.96,p < 0.000) as shown in Figure 2. The heterogeneity was reduced (I 2 = 23%).The 95% CI was overlapping in both major analysis and subgroup analysis (Figure 3).

Treatment failure
Figure 4 shows the impact of diabetes on treatment failure in TB-Diabetes comorbid patients.TB-DM patients were at higher risk to experience treatment failure rates as compared to TB-non-DM patients (HR 0.46, 95% CI: 0.27-0.67,p < 0.001).TB-non-DM patients were at 46% reduced risk for experiencing treatment failure rates as compared to TB-DM patients.The heterogeneity observed among included studies was (I 2 = 61%).

Risk of bias assessment
The included studies showed NOS score range of 6-9.Out of 30 studies, the quality of 21 studies was high.The criteria for selection domain were fully met by all studies except five studies.The age factor was not adjusted in multivariable analysis in five studies in section of confounding bias.For bias in exposure measurement, the Sub-group analysis by study design.

FIGURE 4
Impact of DM on TB treatment failure rates.

Discussion
The TB-DM comorbidity has become more prevalent worldwide and it has become a major challenge for healthcare organizations to control this comorbid condition (44).As it is understood that diabetes in TB patients influences negative TB treatment outcomes including sputum positivity, high risk of relapse, and mortality (17,39).It is a matter of concern that the development of MDR-TB in TB-DM comorbid patients may further worsen the TB treatment outcomes.It puts an additional financial burden on patients that cause disease progression and ultimately leads to the extensively drug-resistant TB (XDR-TB) (45).This study showed a significant association between DM and MDR-TB.
In this study, there was a lower risk of MDR-TB emergence for TB-non-DM patients when compared with TB-DM comorbid patients (HR 0.81, 95% CI: 0.60-0.96,p < 0.001).Our results were found comparable to the previous studies and meta-analyses that found a significant association between DM and MDR-TB (9).The following factors can be the reason for the emergence of MDR-TB: (1) Due to DM, impaired immune functioning, altered microbial genomics, and uncontrolled glucose levels can enhance the susceptibility to develop MDR-TB in TB patients.(2) DM may also enhance the risk of MDR-TB emergence in TB patients due to DM include phagocytotic activity, chemotactic response, generation of oxidative species, microbe proliferation, altered drug disposition, and non-adherence to treatment (46)(47)(48)(49)(50).While results were inconsistent with few previous studies (5,27,(30)(31)(32)51).The limited number of the patients, geographical variations, and unadjusted covariates may be the reason for these discrepancies in results and should be further addressed.
Our study showed a lower risk of experiencing treatment failure rates in TB-non-DM patients as compared to TB-DM comorbid patients (HR 0.46, 95% CI: 0.27-0.67,p < 0.001).Diabetics are more prone to TB treatment failure rates.The TB treatment failure results in our study were found similar with previous studies (51-53) but discrepancies in results were found in the previous three studies (30,39,51).The small sample size and study design used in studies may be the factors causing discrepancies in results.However, most of the studies in our study assessed by the NOS scale were of high quality and there was a lack of lab data for acquiring information on risk factors.
Most of the information was based only on self-reporting or health records of patients, which can lead to assessment error.There should be complete reporting of non-response rates among cases as well as controls to assess reporting-based bias properly.There was moderate heterogeneity across all studies.We performed subgroup analysis by study design, the heterogeneity was reduced.With sub-group analysis by study design for MDR-TB, no change in results was noted when we included prospective studies (HR 0.81, 95% CI: 0.61-0.96,p < 0.000).
There are some limitations present in our study.The studies with different study designs were included that may lead to heterogeneity in results.The heterogeneity in control groups can affect the association to be determined.We assessed overall MDR-TB patients not by primary and secondary subtypes, which may cause differences in risk factors and outcomes.Moreover, there may be the possibility of publication bias, even though the study included a comprehensive set of studies.The statistical techniques used to address the variability in the study design may introduce some uncertainty in the findings.Despite few limitations, the study has several strengths.The main findings of this study given us a valuable understanding of how DM affects emergence of MDR-TB in TB-Diabetes comorbid patients.The strengths of our results lie in comprehensive analysis of a significant number of studies, focusing on a specific patient population.We focused on specific subtypes of DM and TB, providing a more refined understanding of their interaction.The screening for MDR-TB should be prioritized at an earlier stage in TB-DM comorbid patients, which could reduce the economic burden on these patients.There should also be thorough MDR-TB monitoring and necessary steps should be taken to design a proper treatment strategy to prevent the disease progression in these patients.For future concern, this study highlights the importance of conducting more thorough research with large groups of people, using consistent methods, and considering variables that might affect the results.This would help gain a better understanding of how Type 2 DM affects TB-DM comorbid patients.

Conclusion
The results showed that DM had significant impact on emergence of MDR-TB in TB-Diabetes comorbid patients as compared to TB-non-DM patients.DM enhanced the risk of experiencing TB treatment failure rates in TB-Diabetes patients as compared to TB-non-DM patients.Our study highlights the need for earlier screening of MDR-TB, thorough MDR-TB monitoring and designing proper and effective treatment strategies to prevent disease progression.

TABLE 1
Summaries of the included studies.

TABLE 2
Risk of bias among included studies.rate was not reported in seven studies by the cases as well as controls.The overall NOS mean score of included studies was seven, indicating low biasness risk and high quality of the included studies.Table2provides the risk of bias among included studies.