XDR-TB is a serious global health threat. The emergence of XDR TB reflects a failure to implement the measures recommended in the WHO’s Stop TB strategy. This strategy emphasizes expanding high quality DOTS programme 
Emergence of XDR-TB is reported worldwide. Reported prevalence rates of XDR-TB of total MDR cases are: 6.6% overall worldwide, 6.5% in industrialized countries, 13.6% in Russia and Eastern Europe, 1.5% in Asia, 0.6% in Africa and Middle East and 15.4% in Republic of Korea 
. XDR-TB has been reported from 84 countries by end of 2011; the average proportion of MDR-TB cases with XDR-TB is 9.0% and is significantly associated with worse outcome than MDR-TB 
The actual incidence and prevalence rate of XDR-TB in India is not available. A few scattered reports: Mondal et al
reported 7.4% of MDR strains as XDR 
; a study from Hinduja Hospital, Mumbai revealed 11% of MDR strains as XDR 
. Singh et al
reported 33.3% of MDR TB cases as XDR-TB in a population of HIV sero-positive patients from AIIMS, New Delhi 
. Sharma et al
found 2.4% of MDR as XDR-TB cases from Delhi 
. A study conducted by Paramasivan et al
reported 4.6% XDR-TB cases in Chennai 
and another recent study by Khanna et al
from Delhi reported 5.76% XDR-TB 
. None of these studies were carried out as per statistical significance considerations of the numbers of MDR TB isolates included.
Development of drug resistance (MDR/p-XDR/XDR) may result due to various reasons: inappropriate treatment regimens (i.e
. drug selection, duration of treatment and the correct dosage); patient factors (i.e.
mal-absorption, poor adherence); programme-related factors (i.e.
, incompetent health personnel, irregular drug supply). In fact, it has been said that the emergence of MDR-TB is evidence of systematic failure of the global community to tackle a curable disease 
We designed a cross sectional study and enrolled 611 clinically suspected drug resistant TB patients. Of 611 patients, samples from 483 grew MDR-TB cultures, and eighteen of the 483 MDR-TB isolates were XDR-TB (3.7%). The current study was only designed to study the existence of XDR-TB and the association of risk factors, and not to comment on the prevalence of XDR-TB in India.
Analysis of XDR-TB cases showed that XDR-TB was more frequent among patients with family history of TB (p 0.045) (). In 33.3% (6/18) of the families of XDR-TB patients, history of TB was recorded, though no information regarding the treatment regimen followed or drug resistance pattern was available. Of these 6 XDR-TB patients, 1 patient expired while 4 were defaulters. The data implies that family contact with TB patients may be one of the main reasons for the spread of the disease. Disease in the family may indicate the role of some genetic or immunological predisposition in families or just the increased transmission due to close and prolonged exposure. Family history of TB could lead to some patients actually getting primary XDR-TB, which may be responsible for florid disease. In addition, compromised care due to social factors such as possible callousness to a challenge that has caused protracted morbidity but no mortality, or loss of faith in the treatment regimens possibly explain the high defaulter rate.
A salient and novel observation in the study was co-morbidities in seven XDR-TB patients. These patients had Diabetes mellitus (n
4), Myasthenia gravis (n
1), Thalassemia intermedia (n
1) and transfusion associated jaundice (n
1) (). Patients with an associated pathology showed a higher tendency for acquiring resistance (p 0.001) (). Patients with co-morbidities often have compromised immunity. Hence we propose a strong role of immune competence in controlling the disease, with or without treatment. A poor immune response would enhance chronicity of disease, and would hence foster accumulation of further mutations and the subsequent selection of a highly resistant clone of bacteria.
It was observed that 83.3% (15/18) of XDR-TB patients were from poor socio-economic status (monthly income, Indian National rupees (INR) <8,000/-) (p 0.013) () due to which these patients may not find it affordable to get the necessary tests done, like DST (not part of programmatic services at the time of the study). Other social factors such as poor nutrition, poor standards of living, hence higher exposure to disease, which come with poor socio-economic status may also contribute. Thus we may infer from the study that socio-economic status may be a contributor for developing drug resistant TB.
Interestingly, patients who had taken 2nd
line injectable treatment earlier had higher chances of getting XDR-TB (p 0.001) ( and ). Jeon et al
have reported XDR-TB to be associated with the cumulative duration of previous treatment with second-line TB drugs among subjects in a tertiary care TB hospital in S. Korea 
. Dalton et al
have reported that prior use of second-line anti-TB drugs more than quadrupled the risk of extremely drug- resistant tuberculosis (XDR-TB) in an 8-country prospective study 
. Chan et al
have demonstrated in a cohort of 174 patients with multidrug- resistant TB that 12 patients with multidrug- resistant TB strains resistant to the fluoroquinolones and streptomycin had significantly better initial and long-term outcomes, compared with 10 patients with extensively drug- resistant TB, hence signaling caution towards misuse of 2nd
line injectables 
There was a distinct difference in the chances of finding XDR-TB in patients who hailed from a rural vis a vis urban background. MDR-TB patients in rural areas had higher chances of disease worsening to lead to XDR-TB; on the contrary, in patients who belonged to urban areas XDR-TB formed a smaller subset of MDR-TB patients. This was an incidental finding and may signal towards ignorance of rural patients towards available treatment options or possibly poor coverage of programmatic services in such areas ().
Our data suggests that an inadequate initial drug regimen may be associated with the development of XDR, as we observed, 94.4% XDR-TB patients were previously treated for tuberculosis (). These patients were already under medication for TB, though their compliance and dosage schedules, sources of medicine could not be ascertained. Disease worsening, accumulation of drug resistance, progression of MDR to XDR could all result from inadequate treatment.
High numbers of p-XDR-TB cases is a cause for great concern. OFX resistance seen in 7.4% of MDR strains and AMK/CAP resistance in 5.9% are only a single mutation away from converting to an XDR-TB strain. A recent study in South Africa raised the concern of those MDR-TB cases that were resistant to a single marker of XDR-TB (either OFX or KAN), as being at great risk of developing XDR-TB if not managed appropriately 
. Issues about poor/non-response to programmatic management of MDR-TB loom large due to such cases.
The XDR-TB patients were followed up and it was found that 4 (22.2%) expired during treatment, 1(5.5%) was treated successfully, 4 (22.2%) defaulted and 9 (50%) were not traceable. The treatment regimen being given to these patients is given in .
Our study has a few limitations. Data on CXR, BCG vaccination and intake of injectable/fluoroquinolone were not available for all patients. Patients were not willing to give proper residential information; hence migrants could not be traced for follow up. Patients did not have proper previous medical records; hence it was difficult to explain the poor response or progress to drug resistant TB. The small number of XDR-TB (3.7%) cases was also a limitation. It is likely that a larger patient cohort would have shown a more significant association with various variables, including migrants, co-morbidity, socioeconomic status, age, sex, underlying diseases such as chronic obstructive pulmonary disease or abnormal liver function. Another limitation was that resistance to KAN was not checked in all the isolates and only for XDR-TB isolates. Maus et al
have cautioned against the faulty practice of generalizing resistance to a class of drugs, e.g., cyclic peptides or aminoglycosides, based solely on the resistance to a single drug in the class 
The percentage of MDR and XDR-TB patients that are detected depends on the study design, the sampling frame and the study population. There is a need to reevaluate and recalculate actual prevalence of XDR-TB from different population samples residing in various regions of India. Currently, the rapid diagnosis and treatment of persons with TB, particularly any form of drug- resistant TB, are high priority public health interventions. Effective control of drug resistant TB requires massive scaling-up of culture, DST capability and novel rapid assays to detect drug resistance 
. DST is recommended universally for new and retreatment TB cases. Early treatment of MDR-TB with drugs reserved only for treating such patients would again contribute significantly to the control; though this would entail strict action to restrict the use of these drugs only for the program. A quick diagnosis of MDR and XDR-TB translates into greater likelihood of patient care and less spread of this potentially lethal strain thus benefiting the individual and the society. In addition, more exhaustive efforts should be made to manage drug resistant TB cases more effectively to improve treatment outcomes of all patients and hence minimize further development of TB resistant to all available drugs.