Drug Resistance in Osteoarticular Tuberculosis: A Study From an Endemic Zone

Background: The present study was undertaken to determine the incidence of drug resistance against anti-tubercular drugs among patients from an endemic zone. Methodology: Forty consecutive clinico-radiologically diagnosed patients of osteoarticular tuberculosis (29: spine, 11: extraspinal) were enrolled. Pus from needle aspiration was taken in 31 cases, tissue following spinal decompression in seven, synovial in one, and sinus edge biopsy in one. The pus/tissue was subjected to acid-fast bacilli (AFB) staining and liquid culture, sensitivity to 13 anti-tubercular drugs (Isoniazid (INH), rifampicin (RIF), kanamycin (KAN), amikacin (AMK,) capreomycin (CAP), ethionamide (ETH), levofloxacin (LEV), moxifloxacin (MOX), linezolid (LNZ), para-amino-salicylic acid (PAS), bedaquiline (BDQ), delamanid (DLM), and clofazimine (CFO)) were checked, and histopathological/cytopathological examination and molecular tests were performed.  Results: The mean age of patients was 29.07(9-65) years; 21 were female and 19 were male. The diagnostic accuracy for tuberculosis was 20% by AFB smear, 65% by liquid culture, 82.5% by histopathology, and 90% by cartridge-based nucleic acid amplification testing (CBNAAT). All culture-positive isolates were identified as Mycobacterium tuberculosis with no non-tubercular Mycobacterium. The drug resistance detected on CBNAAT was 11.1%, line probe assay (LPA) first line was 15.4%, LPA second line was 4%, and liquid drug susceptibility testing (DST) 11.5%. We detected 15.4% INH resistance, 11.1% RIF, 7.6% LEV, 3.8% MOX and PAS. No resistance was detected against second-line injectable drugs (SLID), ETH, LNZ, BDQ, DLM, and CFO.   Conclusions: No single laboratory modality can ascertain the diagnosis in all cases; hence, samples should be sent for all tests in tandem. In the presence of insufficient samples, tissue may be subjected to CBNAAT and histopathology to arrive at tissue diagnosis. In this subset, overall drug resistance incidence was 12.5% (5/40) with one patient each of isolated INH and RIF resistance, one of multidrug-resistance (MDR), and two of pre-extensively drug-resistant (pre-XDR). Primary drug resistance came out to be 11.1% (4/36) with one patient each of isolated INH and RIF resistance, one of MDR, and one Pre-XDR.


Introduction
Drug-resistant tuberculosis (TB) continues to be a public health threat. The emergence of drugresistant TB has threatened the goal of global TB elimination. Drug resistance in TB is categorized as single drug resistant-TB, multiple-drug-resistant (MDR)-TB, and extensively-drug-resistant (XDR)-TB [1]. Primary and acquired drug resistance incidence is well documented for pulmonary TB but not for osteoarticular TB (OATB) [2].
The rapid detection of drug-resistant TB improves treatment outcomes and prevents disease transmission [3]. Among OATB cases, suspected (presumptive) drug-resistant TB can be identified as patients showing failure of clinico-radiological improvement, deterioration of existing lesions, or the appearance of a fresh lesion/abscess while on anti-tubercular therapy (ATT) for a minimum of four to five months [4]. The drug sensitivity testing (DST) methods include culture-based (phenotypic) and nucleic acid-based (genotypic) methods such as GeneXpert® MTB/RIF (Mycobacterium tuberculosis/rifampicin) assay (Cepheid, Sunnyvale, California,, United States), cartridge-based nucleic acid amplification test (CBNAAT), and line probe assay (LPA) for first and second-line drugs.
The prognosis for spinal/extraspinal TB is improved by early diagnosis and intervention. Delay in diagnosis of drug-resistant spinal TB often results in the development of spinal deformity and/or neurological complications. There is a mean delay of six to eight months before the diagnosis of TB is ascertained, even in endemic countries, due to difficulty in procuring samples from deep-seated lesions as well as difficulty in demonstrating acid-fast bacilli (AFB) on smear and culture owing to the paucibacillary nature of the disease [5][6][7]. Genotypic DST can reduce this delay in establishing drug resistance and the institution of appropriate treatment [8,9].

Materials And Methods
Forty consecutive patients of OATB (29 spinal and 11 extra-spinal cases) were enrolled between January 2021 and April 2022, with a mean age of 29 years (9-65 years) and a male:female ratio of 21:19. Patients diagnosed clinico-radiologically and on magnetic resonance imaging (MRI) observations were enrolled. Human immunodeficiency virus (HIV), hepatitis B surface antigen (HBsAg) or hepatitis C virus (HCV)positive patients were excluded.
None of the patients in this study have been included in any previous studies and ethical clearance was obtained from the Institutional Ethics Committee -Human Research, University College of Medical Sciences, New Delhi, India (approval number: IECHR/2020/PG/47/31). Tissue samples were obtained by aspiration of cold abscesses (n=31), after surgical intervention (n=7), synovial biopsy of the knee (n-1), and sinus edge biopsy around the hip (n=1). Seven out of 29 spinal lesions were operated for debridement/decompression with/without instrumented stabilization. Pus, fluid, cartilage, synovial, intervertebral disc material, granulation tissue, caseous tissue, and bony tissue from the affected bone were sent for histopathology/cytopathology at our institute, and AFB smear, mycobacterial culture, CBNAAT, LPA, and liquid culture DST to 13 anti-tubercular drugs (INH, RIF, KAN, AMK, CAP, ETH, LEV, MOX, LNZ, PAS, BDQ, DLM, and CFO ) at the National Institute of Tuberculosis and Respiratory Diseases (NITRD), New Delhi, India. Standard operating procedures were followed during specimen collection, transport, and processing.
CBNAAT, if valid, was reported as MTB detected/not detected and sensitive/resistant to RIF. LPA was only done in culture-positive cases. First-line LPA was reported as MTB detected and sensitivity/resistance to RIF and INH was noted. LPA second line was reported as MTB detected and sensitivity/resistance to second-line injectable drugs (SLIDs) and fluoroquinolones (FQ) were recorded. In case of indeterminate result, the same test was repeated in the specimen available. The diagnosis of MTB was ascertained by various methods: molecular methods (CBNAAT, LPA), AFB smear, culture, and histology. The diagnostic accuracy of molecular methods and culture for MTB and the percentage of resistance to each 13 drugs were calculated. The diagnostic accuracy percentage of one modality alone was calculated and then with other modalities one by one. Once the diagnosis for MTB was ascertained, the standard treatment for different sites of OATB was followed with appropriate ATT (2HRZE + 10HRE) for sensitive disease and an appropriate regimen for proven drug resistance under the programmatic management of drug-resistant TB (PMDT) guidelines was instituted [1].
The data obtained from the laboratory reports was entered into a Microsoft Excel spreadsheet (2019; Microsoft Corporation, Redmond, Washington, United States) and analyzed.

Results
The most common presenting symptom in our patients was pain, pain and swelling being the second most common presenting complaint. Constitutional symptoms (fever/night sweats/loss of appetite/weight loss) were present in 24/40 (60%) cases. Four patients out of 40 (10%) had a prior history of ATT intake before reporting to us for their existing condition.

Diagnostic accuracy
The diagnostic accuracy of AFB (

Discussion
The prevalence of MDR, any drug resistance, and XDR in pulmonary TB is reported to be 3.5%, 24.9%, and 0.06% (among new) and 26.7%, 58.4%, and 1.3% (among previously treated), respectively [2]. However, no such data exists on OATB. Hence, in this study, we aimed to give the incidence of prevalent drug resistance including primary drug resistance in consecutive OATB patients. Since no single modality of laboratory investigation can ascertain the diagnosis of TB in all cases, hence pus/tissue obtained should be subjected to AFB smear/histology/culture and molecular tests simultaneously. We agree and reiterate the observation from two previous studies conducted by Abhimanyu et al. [10] and Yadav and Jain [11] The observations analyzed on various tests in tandem are as follows: (a) On conducting CBNAAT (36/40, 90%) and LPA in tandem, only one extra case could be diagnosed with a combined diagnostic accuracy of 92.5% (37/40); hence, the inclusion of LPA does not increase diagnostic accuracy significantly.
In one case, RIF resistance was detected on CBNAAT and culture sensitivity; however, the LPA first line came out to be RIF sensitive. Neeraj et al. reported that such discordance between CBNAAT and LPA can be due to contamination, low bacterial loads, mixed organisms, and silent mutation [13].

DST interpretations
Genotypic Molecular kits only detect those resistances that have a known gene locus; therefore, resistances originating from mutations of other genes or gene regions as well as other unknown resistance mechanisms will not be detected [8]. These kits only screen the amino acid sequence and not the nucleic acid sequence so silent mutations will go undetected [14]. Therefore, solitary reliance on genotypic drug resistance detection would underestimate the disease burden and drug resistance rates [15].
Molecular tests cannot differentiate between members of the MTB complex. They are unable to differentiate between viable and nonviable bacteria DNA; hence, they cannot be used for monitoring the progression or success of treatment of patients with ATT [15].
Culture-based or phenotypic DST testing can detect all known and unknown resistances irrespective of mutations and resistance mechanisms. It has low sensitivity since MTB is fastidious and difficult to grow [16][17][18]. These tests require high biosafety laboratory infrastructures and have a longer turnaround time of four to six weeks for detection of MTB with an added one to two weeks duration for DST, limiting its role in early diagnosis and interventions in drug-resistant TB [16][17][18].  [20]. However, both these studies included only those patients who were surgically intervened and not consecutive cases. There is no study that reported drug resistance incidence and profile of anti-tubercular drugs in consecutive cases of OATB like ours.

Limitations
The limitation of our study was the small number of enrolled cases. This was because the study was done during the coronavirus disease 2019 (COVID-19) pandemic when there was a low inflow of non-COVID-19 patients. However, since no data exists in the literature on the drug resistance parameters of consecutive cases of OATB patients, our study is worth reporting.

Conclusions
Since no single modalities can ascertain diagnosis in all cases, hence samples should be sent for all the available tests: smear examination for AFB using ZN stain, TB culture, histopathology, CBNAAT, and LPA. However, an observation made in our study was that if an inadequate sample is obtained, tissue should be sent for CBNAAT and cytopathology/histopathology as combining them we got a diagnostic accuracy of 39/40 (97.5%). CBNAAT and liquid culture demonstrates drug resistance in most cases. Addition of LPA did not demonstrate extra RIF resistance cases. However, it did report an extra INH resistance case.
We found an overall drug resistance incidence of 12.5% (5/40) with primary drug resistance of 11.1% (4/36) in consecutive cases of OATB patients. Such high primary drug resistance in the existing population should give rise to a high degree of suspicion for drug resistance whenever we suspect a case of OATB and hence every effort must be made to procure pus/tissue from cold abscesses/pre/para-vertebral collection for DST (both molecular and phenotypic) so that ATT can be prescribed more effectively based on sensitivity pattern.

Additional Information
Disclosures