Determinants of Drug Resistance in Previously-Treated Pulmonary Tuberculosis Patients Registered at a Chest Clinic in South Delhi, India

Introduction Drug-resistant tuberculosis (DR-TB) is a major concern to effective control of tuberculosis (TB) in India and the likelihood of drug resistance increases with repeated exposure to anti-TB drugs. India has emerged as one of the leading contributors of DR-TB in the world posing a major threat to TB control. In the current study, we aim to find the burden and factors associated with drug resistance in previously treated pulmonary TB patients. Methods A cross-sectional study was conducted among 230 previously treated pulmonary TB patients registered with Directly Observed Treatment, Short-course (DOTS) centers under Nehru Nagar Chest clinic in Delhi, India. The participants were selected consecutively as they registered with the chest clinic. A predesigned, pretested, semi-structured questionnaire in the Hindi language used to collect socio-demographic data and factors associated with the development of drug resistance. Physical examination of all the participants was done (height, weight, pallor). Data were analyzed using SPSS version 21. Binary logistic regression analysis was used to identify independent risk factors of drug resistance. Results Of 230 previously treated pulmonary TB patients, 80 (34.8% (95% CI:28.7-40.9%)) were drug-resistant. Age (p=0.021), ever consumption of alcohol (p= 0.001), pallor (p=0.06), BMI (p=0.028), fasting blood sugar (p=0.001), treatment failure (p=0.005) and the number of prior courses of anti-tuberculosis treatment (ATT) taken (p=0.004) were significantly associated with drug resistance. On applying binary logistic regression analysis, independently associated factors with drug resistance were ever consumption of alcohol, pallor, high fasting blood sugar level, previous treatment failure patients and the number of prior courses of ATT (p<0.05). Conclusion The findings of this study revealed that patients who had pallor, high fasting blood sugar, treatment failure and who had two or more prior courses of ATT were more likely to have DR-TB. Identifying the risk factors for drug-resistant TB is essential in facilitating the government to draw public health interventions. Further research is warranted to explore the causal associations.


Conclusion
The findings of this study revealed that patients who had pallor, high fasting blood sugar, treatment failure and who had two or more prior courses of ATT were more likely to have DR-TB. Identifying the risk factors for drug-resistant TB is essential in facilitating the government to draw public health interventions. Further research is warranted to explore the causal associations.

Introduction
Tuberculosis (TB), a major global health problem since ancient times, believed to have originated 150 million years ago. It has a major impact on a country's social and economic productivity. Alongside (Human Immunodeficiency Virus) HIV, TB is the second most common cause of death worldwide [1]. According to the Global TB Report 2018 of the World Health Organization (WHO), 10 million new cases of TB developed in 2017 worldwide and India contributed 27% of the cases [2]. In 2017, TB caused an estimated 1.3 million deaths in HIV negative individuals plus an additional 300, 000 in HIV positive people. Mortality due to drugresistant tuberculosis (DR-BT) was 410, 000 deaths approximately [3].
DR-TB has been a topic of growing interest in the last decade. Until 1994-1997 when the global project on anti-TB drug resistance surveillance was started by WHO and the International Union Against Tuberculosis and Lung Disease (IUAT-LD), the exact magnitude of TB burden was not known to the world [4]. DR-TB has emerged as a major setback to effective TB control globally. The threat being greater in countries like India, resource-limited developing countries. Many studies have reported poor treatment adherence, irregular treatment, long duration of treatment as the risk factors for the development of DR-TB.
During the last decade, there has been an increase in reported incidences of drug resistance in previously treated patients, particularly among those treated irregularly, or with incorrect regimens and doses. Some studies suggest that the most important risk factor for the development of DR-TB is the previous treatment of TB [5][6][7][8][9]. TB is a constantly changing disease as it varies with the population composition, cultural practices, and attitude of the people. Therefore, the research needs to be constantly updated to keep the scientific community and the health system of the country abreast with them, so that the government policies can keep up with the changing trend.
In Delhi, most of the studies were done to find out the pattern of drug resistance among new TB patients or Multi-Drug Resistant TB. The factors associated were not studied in detail among previously treated pulmonary TB patients. There was no focus on overall drug resistance. Identifying factors associated with the development of resistance and studying their interaction is important. Hence, in the current study, we plan to determine the burden of drug resistance and associated factors of drug-resistance in previously treated pulmonary TB patients.

Study design and setting
This was a cross-sectional study conducted between November 2016 and April 2018 at the Directly Observed Treatment, Short-course (DOTS) centers under Nehru Nagar Chest Clinic in New Delhi, India. Pretesting, modification of the questionnaire, and groundwork for data collection was done by December 2016. Data collection was done for 13 months from January 2017 to January 2018. There were a total of 25 chest clinics in Delhi under the Revised National Tuberculosis Control Program (RNTCP). The current study was conducted in the Nehru Nagar chest clinic which was selected out of these 25 chest clinics by simple random sampling. There were 3 TB Units and 42 DOTS Centers under this Chest Clinic. A total of approximately 15,000 patients registered in the preceding year i.e. 2015 at Nehru Nagar Chest Clinic. Among these, there were about 3600 patients in previously treated including 2600 pulmonary TB cases. DR-TB registering with the clinic were approximately 480 annually. Previously treated pulmonary TB patients (Recurrent TB, Treatment after Failure, Treatment after Loss to Follow-Up) both sputum smear-positive and sputum smear-negative were included in the study. Serious/debilitated patients unable to give consent and interview were excluded from the study.

Sample size and sampling technique
The sample size was calculated by taking the prevalence of drug resistance in India among retreatment cases of pulmonary TB as 15% as per the Global TB report 2015 [10]. We took the confidence interval (CI) as 95%, power as 80%, absolute precision as 5% and calculated sample size using the formula 4pq/d 2 [where p=prevalence taken; q= (1-p); d=precision]. After adding a 10% non-response rate to it, the minimum sample size came out to be 224, which was rounded off to 230. All the patients were consecutively included from the register until the sample size of 230 was achieved.
A pre-designed, pre-tested, semi-structured, interviewer-administered questionnaire in the Hindi language was used to interview the study participants to elicit the relevant information. The questionnaire included questions on socio-demographic characteristics including age, sex, place of residence. The socio-economic status of the study participants was determined by using the Modified BG Prasad's socioeconomic status scale, 2016 [11]. Information related to TB including clinical presentation in the current episode of TB, information related to drug resistance from the records, factors associated with drug resistance including the history of previous treatment for TB, smoking, alcohol consumption, co-morbidities, etc. were collected. General physical examination including anthropometric assessment and local examination of the study participants were done. Body Mass Index (BMI) cut-offs for adults ( >18 years) used were according to WHO Asian BMI guidelines (BMI < 18.5 kg/m 2 : underweight; 18.5 -22.9 kg/m 2 : normal; 23 -24.9 kg/m 2 : overweight; ≥25 kg/m 2 : obese) [12]. For study participants ≤ 18 years, WHO percentile nomograms for males and females were used [13]. Data were collected at the DOTS centers when the patients came to collect their medication.

Definitions
All the operational definitions were as per the national guideline laid out in the RNTCP 2016. "Treatment after Failure" was defined as patients who had been previously treated for TB and whose treatment failed at the end of their most recent course of treatment. "Recurrent TB" was defined as a TB patient previously declared as treated (cured/treatment completed) and was subsequently found to be a microbiologically confirmed TB case. "Treatment after Loss to Follow Up" were patients who received anti-tuberculosis treatment (ATT) for one month or more and were declared lost to follow-up in their most recent course of treatment and subsequently found to be microbiologically confirmed cases of TB [14].

Statistical analysis
Data Entry was done on Microsoft Excel spreadsheet and data analysis was done using the licensed Statistical Package for Social Sciences (SPSS) v. 21. The data were summarized and presented in the form of tables and appropriate diagrams. The qualitative data were summarized as proportions and quantitative data as mean (standard deviation) or median (Inter-quartile range (IQR)). Qualitative data were analyzed using the Chi-Square/Fisher exact test while quantitative data by t-test. The level of significance was set at p<0.05. Odds ratio (OR) and 95% interval were calculated to assess the magnitude of the association between risk factors and DR-TB.

Ethics
Ethical clearance was obtained from the Institutional Ethics Committee of Vardhman Mahavir Medical College and Safdarjung Hospital, before data collection. Written informed consent was obtained from the study participants.

Socio-demographic data
A total of 230 respondents participated in the study. The age of the study participants ranged from 10-80 years with a median age of 30 years (IQR: 21-42 years). Most of the participants belonged to the age group of 16-

Treatment history
Out of the total study participants, the majority were recurrent TB cases (128; 55.7%). More than three-fourths of the study participants had taken ATT once before the current episode (186; 80.9%), 29 (12.6%) study participants had taken two prior courses of ATT and 14 (6.1%) participants had taken 3 or more courses of ATT. More than half (138; 60.0%) of the study participants were taking regular treatment during the previous episode of TB while 92 (40.0%) took irregular treatment. The majority of the study participants had completed their last course of ATT (146, 63.5%) but more than 1/3rd (84, 36.5%) had not. The treatment history of TB patients is shown in Table 3.

Burden of drug resistance
Drug resistance was found in 80 i.e. 34.8% (CI: 28.7-40.9%) of the study participants. The drug susceptibility of the patients is depicted in Figure 1.

Associated factors with drug-resistance
On applying appropriate tests of significance, drug resistance was found to be associated with age (p=0.021), ever consumption of alcohol (p= 0.001), pallor (p=0.06), BMI (p=0.028), fasting blood sugar (p=0.001), type of patient (p=0.005) and the number of prior ATT courses (p=0.004). Factors associated with drug resistance are shown in Table 4.

Discussion
This study provides us with the burden of drug resistance in previously treated pulmonary TB patients. It also gives pertinent information on the various factors associated with drug resistance TB which can guide India in planning activities to tackle this growing problem namely the presence of pallor, fasting blood sugar ≥110 mg/dl, prior treatment failure, ≥ 2 prior courses of anti-tubercular treatment.
In the present study, drug resistance was found to higher in patients who had never consumed alcohol. This was in contrast to a study by Fregona et al. (2016) in Brazil where chronic alcohol consumption was significantly associated with drug resistance [21]. Marahatta et al. in Nepal found no association between alcohol and DR [22]. Bhat et al. (2015) in a cross-sectional study in Madhya Pradesh reported a significant association between alcohol consumption and pulmonary TB (OR 3.2; 95% CI 480.8-2254.8; p = 0.009). This may be an erroneous association that needs further exploration [16].
Multiple treatments with ATT in their lifetime was found to be significantly associated with drug resistance wherein those who had taken ATT two or more times in their life had 2.4 times higher odds of having drug resistance than those who had only one prior course of ATT. Baya [23]. Multiple exposures of the bacteria to the drugs can lead them to gain adaptive mechanisms and become resistant to the action of the drugs or can be due to the natural selection of already resistant strains of the bacteria.
In the present study, treatment after failure patients were 4.08 times more likely to be drugresistant as compared to recurrent TB patients and treatment after loss to follow up patients. Sharma et al. reported a higher proportion of drug resistance in defaulters among previously treated patients in Delhi [24]. Similar results to our study were reported by Baya et al. (OR = 3.82, 95% CI (1.87-7.79), p = 0.0002) [20]. This may be because in failure cases, the bacteria have a higher chance of developing resistance due to prolonged exposure to the drugs and also the bacteria are expected to be more resilient to the drugs which led to a failure of treatment in the first place.
One very interesting finding was that, drug resistance was found to be 1.8 times higher in study participants who had pallor on examination (45.2%) compared to those who did not have any pallor (38; 27.7%) and this difference was statistically significant (p<0.05). In a study by Nagu et al. (2014) in Tanzania, anemia prevalence was 86% in TB patients and these patients were 3 times more likely to have positive sputum smear at two months as compared to non-anemic patients (Relative Risk (RR) = 3.05; 95% CI 1.11-8.40, p = 0.03) and the risk for sputum positive smear results increased with severity of anemia (p-value for trend <0.0) [25]. The higher prevalence of drug resistance in anemia can be due to the weakened immune system. As has been studied before, delayed sputum smear conversion is a risk factor for both TB treatment failure and drug-resistant TB emergence [26]. In another study from Taiwan, among 34 patients who had delayed sputum conversion, 24 (70.6%) were found to have Isoniazid (INH) resistant strains of TB microbes [27]. In India, routine testing of hemoglobin is not done for all TB patients registering with the DOTS center and reserved only for serious cases. The program needs to include routine estimation of Hb and timely treatment of anemics for a better outcome as early detection and treatment are key. To understand better, the direct association between anemia and drug resistance needs to be explored further.
Participants with blood sugar above the cut-off level of 110 mg/dl had 2.3 times the odds of being drug-resistant compared to those with sugar level < 110 mg/dl.  [28]. This may be explained by the impaired immunity due to high blood sugar, rendering them susceptibility to infection with resistant strains. This requires strict blood sugar monitoring for all TB patients and early control of the same.
Limitations: This study has a few limitations. First, it was limited to only one district of Delhi.
There is a need for bigger studies with a larger sample size to shed light on this issue of drug resistance. Secondly, the cross-sectional nature of our study does not allow us to make any conclusion regarding the causal nature of any of the determinants. Despite the limitations, this study provides crucial information about drug resistance in Delhi among previously treated patients. Very few studies have investigated drug resistance factors in previously treated patients.

Conclusions
The present study found the burden of drug resistance in previously treated pulmonary TB patients to be 34.8% (95% CI: 28.7-40.9%). Various factors observed to be significantly associated with drug resistance: never consumption of alcohol, presence of pallor, underweight individuals, fasting blood sugar ≥110 mg/dl, treatment after failure patients, ≥ 2 prior courses of ATT. They pose an imminent threat to TB control in India. Aggressive efforts such as strengthening the laboratory capacity to ensure timely detection, treatment, and monitoring of hemoglobin and deranged blood sugar. Activities to ensure patients and their caretakers understand the consequences of irregular and incomplete treatment should be upscaled. Strategies to ensure treatment adherence and completion especially in children and the importance of a good diet to ensure normal BMI need to be intensified. Written informed consent was obtained from the study participants. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.