Nexus Between Tuberculosis and Diabetic Mellitus, A Prospective Cohort Study.

Background: This work aimed to describe the clinical presentation of TB in patient with DM, to determine the effects of DM on TB treatment outcomes, to identify the effects of TB on glycemic control, and to describe the lipid prole of TB and DM patients. Methods: This prospective cohort study design was conducted. The data were collected from September 2018 to June 2020 using patient interviews, examining the patients, chart review, and collecting blood samples. Binary logistic regression was used to identify the determinants of TB treatment outcomes in the context of DM. Kaplan Meier survival curve was used to see the effects of DM on TB clinical response. Linear regression was used to identify the determinants of the HbA1c level during TB infection. Results: A total of 1092 study participants were included giving for the response rate at 93.81 %. Good TB treatment outcome was observed in 72.5 % of the patients [95 % CI: 69 % - 76 %]. The odds of good TB treatment outcomes were at 75 % lower in the presence of DM (AOR 0.25 [95 % CI: 0.08 – 0.73]). The median time of clinical response in TB and DM patients was 45 days interquartile range (IQR) of 8 days; the median time of clinical response in DM free TB patients was 9 days [IQR 2 days]. TB increased the HbA1c level of DM patients by 1.22 % (B 1.22 [95% CI: 1.11 – 1.34]). In six months period, 60 % of TB and DM patients had got 3 episodes of acute complications. Conclusion: DM signicantly decreases the favorable treatment outcome of DOTS. TB predisposed DM patients for bad glycemic control and increased episodes of acute DM complications. functional micronutrient levels of the study International physical activity questionnaire (IPAQ) was used to measure regular physical activity High-performance liquid chromatography used to

The eligibility criteria for this study were DM or TB patients having a regular follow up at the referral hospitals of the region. The study participants unwilling to give consent for the study were excluded. The quality of the study was maintained by conducting a pretest and providing training to the data collectors and supervisors. In addition, the standard operating procedures were adhered to during the laboratory data collection. The sample size was calculated using Epi-info software with the assumption of 95% CI, power of 80%, a risk ratio of 0.8, TB and DM to DM or TB patients ratio of 2, and non-response rate of 10% gives 1164 study participants (388 DM and TB patients, 388 DM free TB patients, and 388 TB free DM patients). The simple random sampling technique was used to select the study participants using their TB or DM registration ID as a sampling frame. Data were entered into the computer using EPI-info software and transferred to the STATA version 14 for the analysis.
Descriptive statistics were used to describe the pro les of study participants. Binary logistic regression was used to identify the determinants of TB treatment outcomes in the context of DM. Kaplan Meier survival curve was used to see the effect of DM on TB clinical response. Linear regression was used to identify the determinants of the HbA1c level during TB infection. The paired t-test was used to see the effects of DOTS on lipid and micronutrient levels of the patients. Multiple imputation method was used to handle the missing data.
Ethical clearance was obtained from Bahir Dar University College of Medicine and Health sciences ethical review board (Ethical application number CMHS/IRB/124/2018). A support letter was obtained from Amhara national regional state health bureau and respective hospitals. Written informed consent was obtained from each study participant. Study participants with abnormal laboratory ndings were linked to the curative segment.

Results
A total of 1092 study participants were included in the study. The response rate was 93.81%. Twenty-six patients were excluded because of lowquality laboratory samples, 34 study participants were excluded because of their consent, and 12 study participants were excluded because of loss to follow up. The mean age of the study participants was at 32.63 years [SD ± 15.79 years] with the lowest age of the study subjects at 18 years (Table 1).   (Table 3).   DM patients with TB infection have lower baseline micronutrient and lipid pro le compared to TB or DM patients; after 6 months the micronutrient level and lipid pro le was not increasing to the su cient level for DM and TB patients ( Table 6). The baseline functional status of TB and DM patients was poor, after 6 months, the functional status of TB and DM patients did not improve su ciently (Table 7).  [23]. This is due to the high Mycobacterium burden leading to a longer time for culture conversion [24]. Additionally, DM affects the pharmacokinetics of anti-TB drugs by reducing their plasma concentration [25].
The proportion of pulmonary TB was lower in the presence of DM, only 24.1% of the TB cases were pulmonary and 55.11% of them were extrapulmonary TB which was found to be very severe. This nding was in-line with the ndings in Georgia [26]. This is due to the effects of DM on the smear production and excretion that affects the detection of TB, indicating that a signi cant number of pulmonary TB was not detected in DM patients [27].
Atypical TB was common in DM patients reaching 40% of the TB patients. This nding is consistent with the ndings in Brazil [28]. This indicates that the TB diagnostic algorithm should be revised in the case of DM as reported with the study in India [29].
In this study, 15.5% of the DM patients manifested with adverse TB reaction compared to 8.3% of the adverse drug reaction in TB patients without DM. The reason might be due to the high pills burden of TB-DM patients that nally predisposed the patients for the adverse drug reactions [30].
In this study, the median time of clinical response in TB and DM patients was 45 days, interquartile range (IQR) of 8 days; the median time of clinical response in DM free TB patients was 9 days [IQR 2 days]. This is consistent with the previous report in Ethiopia [31], suggesting that the treatment algorithm should be revised in TB-DM patients in Ethiopia.
In our study, we observed that TB increased the HbA1c level of DM patients by 1 [32]. This is due to the reason that TB induces glucose intolerance worsening the glycemic control of DM patients [33].
In our study, we observed that excess episode of acute DM complications in DM and TB patients; in the 6 months period, 60% of the TB and DM patients which had found to be 3 episodes of acute complications compared to 15% of cumulative complications in TB free DM patients. This nding is comparable with the nding in Mexico [34]. This is be due to the effects of infection on inducing the stress hormones that disturbs insulin metabolism [35,36].
After DOTS completion, Vitamin A increased by 2.65 µg/dl for DM and TB patients, as compared to 7 µg/dl increment for DM free TB patients.
This strengthens the justi cation that, higher complications rate of DM among TB patients was attributed to low vitamin A level.
The baseline serum zinc level of DM and TB patients was 88.23 µg/dl, while the baseline serum zinc level of DM free TB patients was 90.87 µg/dl. This nding agrees with ndings from India [37]. This will strengthen the necessity of zinc supplementation for DM patients [38].
In this study, vitamin D increased by 1.1 ng/dl for DM and TB patients, and its values increased by 7.46 ng/dl for DM free TB patients. This nding agrees with the nding in Korea [39]. This nding implies that the critical role of vitamin D was limited in TB-DM patients.
The study indicated that the baseline Ferritin level of DM and TB patients was 14.88 ng/ml whereas the baseline Ferritin level of DM free TB patients was 18.88 ng/ml. The nding was consistent with previously available literature regarding the association between DM and Ferritin. This is due to the high burden of anemia in our study areas [40].
In our study we observed that the HDL level of DM and TB patients was at 34.17 mg/dl, the baseline HDL level of DM free TB patients was at 43.49 mg/dl. The nding was in line with the nding in Netherland [41], suggesting high malnutrition burdens in the study area [42].
In DM and TB group, the basic activity of daily living was improved by 11.6% for the good zone, 44.5% increment for the good zone was observed in DM free TB patients. This nding is in agreement with the ndings in Netherland [43]. This urgently calls great attention of decision-makers to revise the management of the two morbidities during co-existence.

Consent for publication
Not applicable

Availability of Data and Materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declares that they have no competing interests Funding This research work was nancially supported by Bahir Dar University. The funder has no role in design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
Author contribution BEF and TEF conceived the experiment; BEF, MBK, WKA, MBK, WKA, AS, SH, TT, FB and TEF performed the experiment, plan the data collection process, analyzed and interpreted the data. BEF, MBK, WKA, MBK, WKA, AS, SH, TT, FB and TEF wrote the manuscript and approved the nal draft for publication.