This was a retrospective cross-sectional study, conducted in the patients. Duration of the study was from June 2018 to May 2019, to evaluate drug resistance in patients attending Government Hospital of Thoracic Medicine at Chennai, Tamil Nadu, India.

Data for 8140 patients were retrieved for a period of June 2018 to May 2019, all data related to testing and diagnosis of TB using GeneXpert MTB/RIF Assay were collected at Central laboratory, Government Hospital of Thoracic Medicine, Tambaram Sanatorium and were documented. The study was conducted in the age groups between 6-75 years with symptoms and signs suggestive of TB like fever, cough with expectoration, haemoptysis, weight loss, fatigue and with radiological symptoms suggestive of TB like cavity, consolidation etc., and lymphadenitis. Since Govt. Hospital of thoracic Medicine is a Nodal DR TB Centre and Center of Excellence for HIV/AIDS, the patients were from all over Tamil Nadu and borderline states.

Inclusion criteria

1) Based on clinical criteria: a) All presumptive TB patients- People living with HIV/AIDS (PLHIV) and all new symptomatic patients; b) All presumptive MDR patients- New patients at the time of diagnosis (Including history of contact with known MDR patients)- Previously treated patients (Relapse, failure, defaulters)- In drug sensitive patients any follow-up smear positive, non-responders and others; 2) Based on Microbiological criteria: All smear positive patients before starting Anti Tuberculosis Treatment (ATT)- Smear negative patients if signs and symptoms suggestive of TB infection.

Exclusion criteria: Stool, Blood samples, patients being treated with ATT, Monitoring bacteriological cure, Monitoring response to treatment.

Two sputum samples, one spot and next day early morning sputum sample (A and B) collected from patients. The Extra pulmonary samples were collected by the physicians depending on the site involved. Smear was prepared from both sputum and extra pulmonary samples (pleural fluid, ascitic fluid, cerebrospinal fluid, tissue biopsy and pus) under strict aseptic precautions stained by Auramine fluorescence technique and observed under LED fluorescent microscopy for Acid Fast Bacilli (AFB). All smear positive and smear negative samples were tested for CBNAAT as per physician’s request. For GeneXpert the 2-3 mL samples are collected in two 50 mL falcon tube (one more sample as a back up for further testing). The samples were transported to the laboratory in the same day and stored in a refrigerator at 2-8°C. Both sputum and extrapulmonary samples were processed and tested as per the Cepheid MTB/RIF assay procedure guidelines [2]. The samples were processed under strict aseptic precautions in the Biosafety Cabinet Class II A2. To the collected sample, buffer added in 2:1 ratio, vortex done for few seconds and incubated at room temperature for 10 minutes, 2 mL of liquefied supernatant taken by sterile transfer pipette and transferred to the GeneXpert cartridge containing in built primers, probes, PCR reagents, barcode scanned and loaded in the GeneXpert machine. Ultrasonic lysis of the sample releases DNA and mixes with PCR reagents. Automated RT-PCR amplifies the DNA and RIF resistance detected by probes A to E specific to identify the mutant codons 507 to 533 of RRDR region of rpoB gene and the printed reports were available within two hours. The reports generated were informed to the concerned physicians and the State Treatment Laboratory Supervisor (STLS). The samples were further tested in Line Probe Assay (LPA) [8] at NIRT, Chennai for confirmation of RIF resistance and other associated resistance to isoniazid and other second line drug such as fluroquinolones and second line injectables.

All statistical analysis were performed using SPSS version 15.0. Continuous variables were presented as mean±standard deviation or median and Interquartile Range (IQR) values, as appropriate based on the variable normal distribution. Categorical data were presented as frequencies and percentages.

A total of 8140 samples were tested for CB-NAAT which include 7110 sputum samples and 1030 extra pulmonary samples. MTB detected in 3726 and RIF sensitive found in 3554 samples include 238 extra pulmonary samples and RIF resistance was detected in 172 which include 22 extra pulmonary samples [Table/Fig-1].

[Table/Fig-1]:

Flow chart for the sample distribution.

The overall sensitivity of CB-NAAT in smear negative pulmonary and extra pulmonary samples was 32%. Out of the total 6070 males, 2943 were RIF sensitive and 129 were RIF résistance. Among the 2070 females 611 were sensitive to RIF and 43 were resistant [Table/Fig-2]. Among the RIF sensitive patients, isoniazid monoresistance was found in 1.9%. The type of clinical samples used in the study and tested with CBNAAT is shown in [Table/Fig-3]. Among the TB/HIV co-infected patients 1% found to be resistance to RIF. The sensitivity of GeneXpert in smear negative sputum samples was 89% and extra pulmonary samples were 22% compared with AFB smear. In present study, 97% of the RIF resistance patients were previously treated and there is no gender preponderance as male and females are found equally susceptibility in these study. Among the 172 RIF resistance samples majority were due to mutant codon 531 of RRDR region of the rpoB gene [Table/Fig-4,5]. RIF monoresistance was found in 90% of patients which was confirmed by LPA. A 10% had isoniazid resistance 6% had fluoroquinolone resistance and 2% has resistance to Second line injectables and 1% had resistance to both. Among the 105 paediatric patients tested 10% of the patients were found MTB detected RIF sensitive [Table/Fig-6].

This study shows GeneXpert has more diagnostic accuracy in detecting RIF Resistance and done as a primary screening test for MTB detection in PLHIV patients who remain smear negative due to lack of caseous necrosis and cavitation, supported by study conducted at Delhi and Uttar pradesh [9,10]. GeneXpert is also more advantageous in paediatric TB whose samples like gastric aspirates remain paucibacilliary, supported by study conducted at Mangalore [11-13]. GeneXpert has a sensitivity of 32% in smear negative samples, lower sensitivity because bacillary count lower than to be detected by CB-NAAT especially in paucibacilliary extra pulmonary samples is the reason, this supported by study conducted at Italy [14] and Kolkata [15]. Performace of GeneXpert MTB/RIF from previous studies is represented in [Table/Fig-7]. In RIF sensitive 3554 patients, 10 had monoresistance to Isoniazid detected by LPA done at NIRT [5,7,9,12,14-19]. RIF resistance was detected in 172 samples, among which 97% are previously, treated patients and 3% were new patients. This is supported by the study conducted by Asian pacific society of respirology [16]. The emergence of resistant bacilli in previously treated patients was due to the wide spread use of drug, poor compliance, inappropriate and non-adherence to treatment. The aerobic resistant bacillus remains dormant inside the anaerobic environment of granulomas and cavities. Whenever this cavities or granulomas breaks down, bacilli starts multiplying in the aerobic environment and spreads via blood to other organs and via bronchus to others through aerosols. There is no gender preponderance in RIF resistance and both male and female are equally susceptible, this supported by study conducted at Zambia [17]. Probe E mutation (58%) of the rpoB gene, which denotes mutation of the 531 codon was found in majority of patients followed by probe D (21%), probe A (10%), probe B (10%) and probe C (1%), this supported by the study conducted at Malawi, Africa [7]. RIF mono resistance was detected in more than 80% of patients which was confirmed by LPA, this result is similar to the studies conducted in South Africa [18,19]. Isoniazid resistance occurs due to mutations at multiple genes including katG (catalase-peroxidase gene), inhA region (NADH-dependent-ACP-reductase In hA gene) kasA (β-keto-acyl-ACP-synthase gene), ndh (NADH-dehydrogenase) and a hpC region (alkyl-hydroxyperoxide-reductase) genes [12]. Recently available molecular methods doesn’t cover all the multiple mutations of the Isoniazid resistant genes, could be reason for the lesser detection of isoniazid co-resistance. A 6% patients has fluroquinolones resistance, 2% had resistance to second injectable like kanamycin, capreomycin and amikacin and 1% of patients had resistance to both fluroquinolones and Second line injectables as many patients sample culture results turn negative during MDR-TB treatment this is similar to the study conducted at Lesotho hospital [5]. So, early detection of RIF resistance by GeneXpert as a surrogate marker of MDR TB is necessary to reduce morbidity and mortality and to prevent transmission of MDR strains.

Technical: GeneXpert MTB/RIF Assay requires continuous electrical supply and back up either by UPS or Genset is needed as prolonged power failure affect processing of reports and may need repeat testing. Room temperature should be maintained between 24°C to 30°C. The catridges should be stored at 2-8°C as this temperature is needed to maintain the stability of PCR reagents and buffers contained in the catridge.

Diagnostic: GeneXpert MTB/RIF Assay cannot detect RIF resistance due to mutations occurring outside the RRDR and also doesn’t cover Isoniazid coresistance. Positive reports do not indicate about the viability of the bacilli and negative report does not exclude positivity when the bacillary count is too low to be detected by the machine.

The present study showed high prevalence of RIF resistance in previously treated patients due to the following reasons: a) Presence of previous history of ATT with poor adherence invariably leads to RIF Resistance; b) Drug sensitive patients might turn resistant as a result of mutation due to poor adherence and compliance or inappropriate regimen; c) Drug sensitive patients might turn resistant due to amplification of dormant resistant bacilli during treatment. So, all suspicion, new and previously treated patients samples should be subjected to GeneXpert MTB/RIF Assay before starting DOTS regimen as per universal DST to reduce morbidity and mortality and to prevent transmission of drug resistant strains. This study confirms that Genexpert remains the rapid diagnostic tool for the diagnosis of TB as well as to confirm sensitivity/resistance to RIF both in pulmonary and extra pulmonary samples.