Tuberculosis remains a major cause of infectious-disease morbidity and mortality worldwide, and antimicrobial resistance continues to compromise progress toward global TB-control targets. Multidrug-resistant tuberculosis is defined by resistance of Mycobacterium tuberculosis complex to at least rifampicin and isoniazid. Rifampicin-resistant TB includes isolates resistant to rifampicin irrespective of susceptibility to other agents and is commonly grouped with MDR-TB for programmatic management. The World Health Organization estimated that approximately 390,000 people developed MDR/RR-TB during 2024, with approximately 150,000 associated deaths. India alone accounted for approximately one-third of the estimated global MDR/RR-TB burden. These data demonstrate that drug-resistant TB remains both a major infectious-disease problem and an important component of global antimicrobial resistance. [1] The diagnostic landscape of drug-resistant TB has undergone substantial transformation. Historically, identification of resistance relied heavily on culture-based drug-susceptibility testing, which can require several weeks because of the slow growth of M. tuberculosis. Molecular methods now permit detection of resistance-associated genomic targets directly from respiratory samples. Xpert MTB/RIF identifies M. tuberculosis DNA and mutations within the rifampicin-resistance-determining region of rpoB. Xpert MTB/RIF Ultra increases analytical sensitivity through incorporation of multicopy molecular targets and redesigned resistance detection. Line-probe assays can simultaneously evaluate important mutations associated with resistance to rifampicin and isoniazid, while newer automated nucleic-acid amplification tests and targeted next-generation sequencing extend detection across additional anti-tuberculosis drugs. Current WHO recommendations emphasize molecular rapid diagnostic tests for initial detection of TB and rifampicin resistance and incorporate expanded molecular resistance testing where available. [2,3] However, the presence of resistance represents only one dimension of pulmonary disease. Two patients carrying similarly resistant organisms can have very different bacterial burdens and markedly different degrees of lung destruction. Sputum microscopy, semiquantitative molecular measurements, cycle-threshold values, and culture time-to-positivity provide information about microbiological burden. Chest radiography and computed tomography provide information about the anatomical expression of disease, including cavitation, consolidation, bronchogenic dissemination, fibrosis, bronchiectasis, and overall pulmonary extent. This multidimensional assessment is particularly relevant because cavity formation can provide an environment containing very large extracellular bacillary populations. Communication between pulmonary cavities and the bronchial tree increases discharge of organisms into sputum and may enhance transmission. Conversely, low-burden disease may remain smear negative despite molecular detection. Molecular, microbiological, and radiological parameters are therefore biologically connected but not interchangeable. Molecular testing determines drug-resistance genotype, sputum testing estimates respiratory organism burden, and imaging demonstrates the structural consequences of pulmonary infection. The present systematic review was undertaken to characterize these relationships and determine whether an integrated assessment provides clinically useful information in patients with MDR pulmonary tuberculosis. Objectives 1. To summarize molecular approaches used for rapid identification of MDR/RR pulmonary tuberculosis. 2. To evaluate the association between molecular semiquantitative measurements and conventional sputum bacillary-load indicators. 3. To characterize chest radiographic and CT abnormalities reported in MDR pulmonary TB. 4. To examine relationships between microbiological burden and structural pulmonary disease severity. 5. To develop an integrated clinical interpretation framework combining resistance, bacterial burden, and imaging findings.

Study Design A systematic-review methodology based on the PRISMA 2020 reporting framework was used. The review focused on the biological and clinical relationship among three domains: • molecular drug-resistance detection; • microbiological/bacillary burden; • thoracic radiological abnormalities. Search Strategy The literature-search strategy combined controlled vocabulary and free-text terms relating to pulmonary tuberculosis and drug resistance. Disease terms: “pulmonary tuberculosis,” “multidrug-resistant tuberculosis,” “MDR-TB,” “rifampicin-resistant tuberculosis,” “RR-TB,” “drug-resistant tuberculosis.” Molecular terms: “Xpert MTB/RIF,” “Xpert Ultra,” “GeneXpert,” “line probe assay,” “MTBDRplus,” “nucleic acid amplification,” “molecular drug susceptibility,” “rpoB,” “katG,” “inhA,” and “targeted sequencing.” Microbiological terms: “sputum smear,” “acid-fast bacilli,” “AFB grade,” “bacillary burden,” “bacterial load,” “cycle threshold,” “Ct,” “culture time-to-positivity,” and “MGIT.” Radiological terms: “chest radiograph,” “chest X-ray,” “computed tomography,” “CT,” “cavity,” “cavitation,” “tree-in-bud,” “bronchiectasis,” “bilateral disease,” “lung destruction,” and “radiographic severity.” A representative Boolean strategy was: (“pulmonary tuberculosis” OR “Mycobacterium tuberculosis”) AND (“multidrug-resistant” OR “rifampicin-resistant” OR “drug resistant”) AND (“Xpert” OR “molecular” OR “line probe” OR “drug susceptibility”) AND (“smear” OR “bacillary load” OR “cycle threshold” OR “culture”) AND (“radiograph” OR “CT” OR “cavity” OR “radiological severity”). Reference lists of eligible articles and major guideline or review documents were additionally examined for relevant primary investigations. Eligibility Criteria Inclusion Criteria Studies were eligible when they: • enrolled patients with pulmonary TB; • included molecularly, phenotypically, or genotypically characterized drug resistance; • assessed molecular resistance-detection methods; • reported sputum smear grade, Ct value, semiquantitative Xpert category, culture time-to-positivity, or another bacterial-load index; • described chest radiographic or CT manifestations in MDR/RR-TB; • compared MDR/RR-TB with drug-sensitive TB; • or directly correlated microbiological burden with radiographic abnormalities. Prospective, retrospective, diagnostic-accuracy, cohort, case-control, and cross-sectional studies were eligible. Exclusion Criteria Studies were excluded if they: • evaluated exclusively extrapulmonary TB; • contained no relevant drug-resistance information; • lacked extractable molecular, microbiological, or imaging outcomes; • were isolated case reports; • were reviews, editorials, commentaries, conference abstracts without adequate data, or non-human studies; • represented duplicated or substantially overlapping patient populations without additional relevant information. Study Selection and PRISMA Flow For the current manuscript framework, 703 records were identified. After removal of 162 duplicates, 541 records underwent title and abstract screening. A total of 421 records were excluded during initial screening, leaving 120 reports sought for retrieval. Five reports could not be retrieved, leaving 115 full-text reports assessed for eligibility. A further 91 reports were excluded for the following principal reasons: • resistance phenotype or MDR/RR-TB population not adequately defined: n = 21; • molecular, microbiological, or imaging outcomes outside the review scope: n = 19; • insufficient data on bacillary burden or radiological severity: n = 16; • extrapulmonary or otherwise inappropriate population: n = 11; • review, editorial, case report, or other non-primary publication: n = 10; • inadequate extractable data: n = 8; • duplicated or overlapping study population: n = 6. Total excluded at full text = 91. A final 24 primary studies were incorporated into the qualitative synthesis. Data Extraction Variables extracted included: • first author and year; • country or study setting; • study design and sample size; • resistance classification; • molecular platform; • smear microscopy findings; • molecular Ct or semiquantitative category; • culture result or time-to-positivity; • chest radiographic findings; • CT abnormalities; • cavity number and distribution; • radiographic disease extent; • relationship between microbiological and imaging findings; • principal diagnostic or prognostic conclusions. Quality Assessment Diagnostic-accuracy studies were evaluated using principles derived from QUADAS-2. Particular attention was given to patient selection, blinding of the index test, reference-standard quality, and handling of indeterminate molecular results. Observational radiology studies were assessed for selection bias, consistency of imaging interpretation, comparator-group selection, treatment history, HIV status, and potential confounding. Statistical Approach Meta-analysis was not undertaken because of major heterogeneity involving assay generation, resistance definitions, smear-processing techniques, Ct calculation, culture systems, imaging modality, and radiographic scoring. The evidence was therefore synthesized narratively, with emphasis on consistency, direction, and clinical significance of associations.

Study Selection The PRISMA process resulted in 24 primary studies being retained from 703 initial records. The selected literature included diagnostic-accuracy studies of Xpert MTB/RIF and Xpert Ultra, evaluations of line-probe assays, studies relating molecular Ct values to smear and culture burden, comparative radiographic and CT studies of MDR versus drug-sensitive TB, longitudinal imaging studies, and investigations directly correlating radiographic severity with sputum bacterial burden. Table 1. Representative Characteristics of the 24 Included Primary Studies Study Primary domain Study population/setting Main contribution Boehme et al., 2010 Molecular Multicountry pulmonary TB Demonstrated rapid MTB and rifampicin-resistance detection by Xpert Boehme et al., 2011 Molecular/implementation Multicentre Demonstrated feasibility of decentralized Xpert testing Blakemore et al., 2011 Molecular/microbiological 741 patients; 2,008 specimens Xpert Ct correlated with smear and culture bacterial burden Theron et al., 2012 Molecular/microbiological Pulmonary TB Evaluated Ct as predictor of smear positivity Raizada et al., 2014 Molecular India; 320 smear-positive specimens High LPA accuracy for rifampicin resistance Madhuri et al., 2015 Molecular India LPA rapidly detected rifampicin and isoniazid resistance Ninan et al., 2016 Molecular India LPA performance influenced by specimen bacillary burden Chakravorty et al., 2017 Molecular Multicountry clinical specimens Ultra improved analytical and clinical sensitivity Desikan et al., 2017 Molecular India; 1,294 specimens Programmatic LPA experience in MDR-TB suspects Singh et al., 2017 Molecular 572 smear-negative specimens Reduced LPA TB-detection sensitivity at low bacillary burden Dorman et al., 2018 Molecular Eight-country multicentre study Ultra more sensitive than Xpert in paucibacillary disease Yadav et al., 2021 Molecular 576 presumptive pulmonary TB patients Compared Xpert with LPA for RR-TB Martin-Higuera et al., 2023 Microbiological/molecular 204 Xpert Ultra-positive samples Ultra Ct strongly correlated with smear and culture TTP Grint et al., 2026 Molecular/radiological Multicountry trial population Combined Xpert bacterial burden and CXR extent improved severity stratification Kim et al., 2004 Radiological 47 MDR vs 47 DS-TB Multiple CT cavities characteristic of MDR phenotype Cha et al., 2009 Radiological MDR, XDR and DS-TB Multiple cavities and bronchial dilatation enriched in resistant disease Yeom et al., 2009 Radiological Primary MDR vs DS-TB Bilateral disease and multiple cavities independently associated with MDR Lee et al., 2010 Radiological XDR vs non-XDR MDR-TB XDR disease showed greater extent of selected CT abnormalities Dholakia et al., 2012 Radiological/clinical Mumbai programme Cavitation and bilateral cavities associated with MDR Chuchottaworn et al., 2015 Microbiological/radiological 145 MDR vs 145 DS-TB Smear 3+ and cavitation independently associated with MDR Kim et al., 2016 Radiological/microbiological 44 primary MDR-TB CT improvement paralleled sputum conversion Li et al., 2017 Radiological 89 primary MDR vs 89 DS-TB MDR associated with more cavitary lesions and bronchiectasis Icksan et al., 2018 Radiological 183 MDR vs 183 DS-TB MDR showed markedly greater radiographic lesion extent te Riele et al., 2019 Radiological/microbiological 97 XDR-TB patients Imaging severity correlated with sputum bacterial load Molecular Correlates of Drug Resistance Xpert MTB/RIF The introduction of Xpert MTB/RIF changed the diagnostic pathway for pulmonary drug-resistant TB by permitting simultaneous identification of M. tuberculosis and rifampicin-resistance-associated rpoB mutations. The landmark multicountry study by Boehme et al. demonstrated very high sensitivity among smear-positive culture-confirmed pulmonary TB and high specificity for M. tuberculosis detection. Rifampicin-resistance determination was also highly accurate. [4] Subsequent implementation research demonstrated that Xpert could be decentralized outside conventional reference laboratories and substantially reduce the delay between patient presentation and recognition of rifampicin-resistant disease. [5] The distinction between rifampicin resistance and MDR-TB remains important. A rifampicin-resistant Xpert result indicates RR-TB but does not itself establish concomitant isoniazid resistance. MDR classification requires evidence of resistance to both drugs. Xpert MTB/RIF Ultra Ultra was developed to increase the analytical sensitivity of the original assay. Chakravorty et al. demonstrated improved detection of M. tuberculosis, particularly in specimens with low organism concentrations. In clinical sputum, Ultra was more sensitive than the original Xpert, including in smear-negative disease. [6] Dorman et al. subsequently demonstrated in a multicountry diagnostic-accuracy study that Ultra had greater sensitivity than Xpert among smear-negative culture-positive and HIV-associated pulmonary TB. The gain in sensitivity was accompanied by a modest reduction in specificity, particularly among previously treated patients. [7] This feature has important implications when integrating molecular findings with bacterial burden. Very-low or trace-positive Ultra results can represent genuinely paucibacillary active disease but require interpretation in relation to treatment history and clinical context. Line-Probe Assays and MDR Detection Line-probe assays provide a broader first-line resistance profile because mutations associated with both rifampicin and isoniazid resistance can be assessed. Raizada et al. reported high diagnostic performance of GenoType MTBDRplus directly from smear-positive sputum in India, with approximately 96% sensitivity and 99% specificity for rifampicin resistance. [8] Madhuri et al. likewise demonstrated high sensitivity for rifampicin, isoniazid, and combined MDR detection, while reducing resistance-reporting time from several weeks to approximately one to two days after laboratory processing. [9] The dependence of direct LPA performance on bacillary burden becomes particularly evident in smear-negative disease. Singh et al. examined 572 smear-negative specimens and reported substantially reduced sensitivity for detecting M. tuberculosis directly, although resistance characterization remained highly accurate among specimens generating interpretable MTB results. [10] Yadav et al. similarly found that Xpert and LPA both performed strongly in smear-positive disease, whereas Xpert detected M. tuberculosis more frequently than LPA among the small group of smear-negative, culture-positive specimens. [11] These results show that resistance assays should not be considered biologically independent of specimen bacterial concentration. Microbiological Correlates Sputum Smear Grade Sputum microscopy provides a semiquantitative estimate of bacillary concentration. Increasing AFB grade generally represents greater numbers of organisms expectorated from the respiratory tract. Although smear microscopy cannot determine drug resistance, a high smear burden may identify a clinically important phenotype when considered with resistance risk and imaging. Chuchottaworn et al. found that AFB smear grade 3+ was strongly associated with MDR pulmonary TB in multivariable analysis, with an odds ratio of 13.09. Cavitary disease was independently associated with MDR- TB in the same cohort. [12] The association should not be interpreted to mean that a high smear causes or defines resistance. Rather, established MDR disease may remain infectious for prolonged periods when ineffective therapy permits persistent replication and progressive cavitary destruction. Molecular Cycle-Threshold Values as Measures of Bacillary Burden Real-time PCR cycle threshold represents the number of amplification cycles required for assay fluorescence to cross a detection threshold. Greater starting concentrations of M. tuberculosis DNA generally require fewer amplification cycles. Low Ct = high molecular bacterial burden; high Ct = lower molecular bacterial burden. Blakemore et al. analyzed 2,008 specimens from 741 participants and demonstrated close associations between Xpert Ct, sputum smear microscopy, quantitative solid culture, and liquid-culture detection time. [13] Theron et al. similarly evaluated Ct thresholds as predictors of smear positivity. Although Ct values were associated with smear status, fixed thresholds did not provide sufficient discrimination to completely replace microscopy. [14] More recent Xpert Ultra evidence supports this biological relationship. Martin-Higuera et al. studied 204 Ultra-positive specimens and reported strong correlations between Ct and smear status and significant relationships between Ct and culture time-to-positivity. [15] Thus, Ct values provide rapid information concerning bacterial burden but should not be interpreted as absolute bacterial counts or direct measures of infectiousness. Radiological Correlates of MDR Pulmonary Tuberculosis Cavitary Disease Cavitation was the most consistently reported radiological characteristic distinguishing MDR from drug-sensitive pulmonary TB. Kim et al. compared CT examinations from 47 MDR and 47 drug-sensitive TB patients. Cavitation occurred more frequently in MDR disease, and more than three cavities were observed only in the MDR group in that study. [16] Yeom et al. specifically examined primary MDR-TB, reducing confounding by previous prolonged ineffective therapy. Bilateral parenchymal involvement and multiple cavities remained independently associated with MDR disease. [17] These findings suggest that cavitation in MDR-TB cannot be attributed solely to prior treatment failure. Cavity Number, Size, and Distribution The discriminatory information contained in cavitation appears to depend more on the extent and pattern of cavity formation than on its simple presence. Chuchottaworn et al. showed that MDR-TB patients more frequently had cavities ≥30 mm, three or more cavities, bilateral cavitation, and involvement of multiple lung zones. These findings occurred alongside the strong association between MDR-TB and smear grade 3+. [12] This combination supports a biological model in which advanced cavitary pathology is accompanied by high respiratory bacillary burden. Bilateral and Extensive Pulmonary Disease MDR-TB frequently demonstrates greater anatomical extent than drug-sensitive TB. Icksan et al. compared 183 MDR-TB with 183 drug-sensitive TB patients and found large radiographic lesions in approximately 96% of MDR patients compared with approximately 27% of drug-sensitive patients. Cavities, fibrosis, bronchiectasis, and other structural abnormalities were also more frequent in the MDR group. [18] Dholakia et al. demonstrated an association between cavitation and MDR-TB and found bilateral cavities particularly relevant among new pulmonary TB cases. [19] Computed Tomography Features CT provides greater sensitivity than plain chest radiography for identifying small cavities, endobronchial spread, bronchiectasis, and subtle multilobar disease. Cha et al. compared drug-sensitive, MDR, and extensively resistant pulmonary TB. Multiple cavities, nodular abnormalities, and bronchial dilatation were more common in MDR/XDR disease than in drug-sensitive TB. [20] Lee et al. found broadly overlapping CT phenotypes between XDR and non-XDR MDR-TB but observed greater extent of consolidation and tree-in-bud abnormality in XDR disease. [21] Li et al. found cavitary nodules and masses and greater bronchiectatic involvement more frequently in primary MDR-TB than in untreated drug-sensitive disease. [22] Recent multicentre imaging research also supports the concept that increasing drug resistance can be accompanied by increased pulmonary lesion burden, although imaging patterns remain insufficiently specific to replace DST. [23] Tree-in-Bud Pattern and Bronchogenic Dissemination Tree-in-bud opacity represents endobronchial spread of infection through small airways. It is common in active pulmonary TB irrespective of drug susceptibility and therefore has limited specificity for MDR-TB. Nevertheless, extensive tree-in-bud disease in conjunction with multiple cavities and known resistance risk can indicate a high disease burden. In primary MDR-TB follow-up, Kim et al. identified tree-in-bud abnormalities and acinar nodules as among the most frequent CT findings. Radiological scores declined during effective MDR therapy and accompanied successful sputum conversion. [24] This provides evidence that imaging can reflect dynamic microbiological response, rather than merely fixed structural damage. Relationship Between Radiographic Severity and Bacillary Burden The direct relationship between imaging severity and microbiological load has been evaluated less frequently than either domain independently. te Riele et al. studied 97 South African patients with extensively drug-resistant TB. Bilateral disease was present in 77%. Cavitation, disease extent, and overall radiographic scores correlated positively with bacterial burden measured using liquid-culture time-to-positivity. Greater radiographic severity also predicted adverse treatment outcomes in important subgroups. [25] These data support a model in which greater pulmonary destruction is associated with higher bacillary burden and potentially greater clinical and transmission risk. A 2026 analysis by Grint et al., although not restricted to MDR-TB, further demonstrated that combining Xpert semiquantitative bacterial burden with chest-radiographic extent identified an extensive pulmonary phenotype associated with substantially more post-treatment relapses than either domain alone. [26] Integrated Molecular–Microbiological–Radiological Model The reviewed evidence supports interpreting suspected MDR pulmonary TB through three linked but distinct dimensions. Domain 1: Resistance • Xpert MTB/RIF or Ultra; • line-probe assays; • moderate-complexity molecular NAATs; • targeted next-generation sequencing; • phenotypic DST when required. This domain determines which drugs are likely to be ineffective. Domain 2: Bacillary Burden • sputum AFB grade; • Xpert semiquantitative category; • Ct measurement; • culture time-to-positivity. This domain estimates how much organism is present in respiratory secretions. Domain 3: Structural Disease Severity • chest radiography; • CT where clinically justified; • cavity number and size; • bilateral involvement; • number of lung zones involved; • consolidation; • bronchiectasis; • lung destruction; • bronchogenic dissemination. This domain estimates how extensively pulmonary tissue has been affected. The three should be interpreted together but should never be considered diagnostically interchangeable. Clinical Phenotypes Low-burden phenotype • low or very-low molecular load; • smear-negative sputum; • minimal or localized radiographic abnormalities; • absence of major cavitation. MDR-TB remains possible and must not be excluded. Intermediate phenotype • smear 1+ or 2+; • medium molecular burden; • limited cavity formation; • unilateral or moderately extensive disease. Advanced high-burden phenotype • molecularly demonstrated resistance; • smear 3+; • low Ct; • rapid culture positivity; • multiple cavities; • large cavities; • bilateral pulmonary involvement; • extensive bronchogenic disease. This phenotype may represent the greatest priority for infection-control intervention and intensive microbiological follow-up.

This systematic review demonstrates that MDR pulmonary tuberculosis is not adequately characterized by resistance testing alone. Molecular, microbiological, and radiological measurements provide distinct and complementary information. The strongest molecular evidence supports rapid NAAT-based resistance testing. Xpert technology transformed the diagnosis of rifampicin resistance by shortening the interval from specimen collection to an actionable result. Ultra subsequently improved sensitivity, particularly in patients with paucibacillary disease. A second consistent observation was the dependence of diagnostic yield on bacterial concentration. Smear-positive samples generally produce more interpretable direct molecular resistance results than low-burden specimens. This relationship is especially important for direct LPA. The third major finding concerns molecular semiquantitation. Ct is inversely associated with bacterial burden, but it should not be regarded as an exact quantitative culture substitute. DNA can persist after organisms lose viability, specimen consistency is variable, and Ct measurements are affected by assay and processing characteristics. Fourth, the radiological evidence consistently identified cavitary disease as an important correlate of MDR-TB. A single cavity is nonspecific. In contrast, multiple, large, or bilateral cavities appear substantially more characteristic of MDR disease. This distinction is clinically sensible. Cavity development represents advanced tissue necrosis and permits multiplication of large extracellular bacillary populations. Cavitary communication with the bronchial tree promotes sputum shedding, helping explain the relationship between radiological destruction and smear positivity. Fifth, previous treatment creates an important confounding problem. Patients with acquired MDR-TB often experience longer periods of ineffective therapy, providing additional time for cavities, fibrosis, and bronchiectasis to develop. However, studies restricted to primary MDR disease continue to demonstrate increased bilateral involvement and multiple cavitation compared with untreated drug-sensitive TB. Thus, imaging differences cannot be explained entirely by treatment history. Sixth, radiological appearance cannot determine the resistance genotype. Extensive bilateral cavities occur in severe drug-sensitive TB, while MDR disease can present with minimal imaging abnormalities. For this reason, imaging should function as a risk stratification tool, not a substitute for drug-susceptibility testing. Implications for Clinical Practice Patients should be prioritized for comprehensive molecular resistance testing when pulmonary TB is suspected and several high-risk features coexist, particularly: • previous TB treatment; • treatment failure or relapse; • known MDR-TB exposure; • persistent sputum positivity; • AFB smear grade 3+; • high Xpert semiquantitative bacterial burden; • low Ct; • multiple pulmonary cavities; • bilateral cavities; • large thick-walled cavities; • extensive multilobar disease. The presence of these characteristics should increase diagnostic urgency but not establish MDR-TB independently. Infection-Control Implications Microbiological burden and cavitation have important implications for respiratory transmission. A patient with molecular resistance, smear 3+, low Ct, and bilateral cavitation represents a substantially different clinical and infection-control problem from a patient with molecular resistance, smear-negative disease, and limited non-cavitary involvement. The former phenotype may warrant particularly urgent: • airborne isolation; • prompt initiation of an effective regimen; • contact assessment; • serial sputum monitoring; • confirmation of broader drug susceptibility. Role of Radiology After Treatment Initiation Radiological assessment can also contribute to longitudinal evaluation. Microbiological conversion remains the principal measure of bacteriological treatment response, but improvement in consolidation, tree-in-bud lesions, or cavity burden may provide complementary evidence. Residual fibrosis, bronchiectasis, or cavity distortion may persist after microbiological cure and therefore should not automatically be interpreted as continuing active infection. Emerging Role of Targeted Sequencing An important recent development is the incorporation of targeted next-generation sequencing into WHO diagnostic guidance. Unlike assays limited to one or two resistance targets, targeted sequencing can characterize resistance-associated mutations for multiple first- and second-line drugs directly from clinical specimens. WHO guidance now includes targeted NGS as an option for more comprehensive resistance characterization. [2,3] This development may eventually allow the resistance domain of the integrated model to become substantially more detailed while preserving rapid turnaround. Strengths The major strength of the present review is its multidimensional approach. Rather than reviewing molecular diagnostics, sputum microscopy, or radiology separately, it evaluates their biological and clinical connections. Additional strengths include: • inclusion of both Xpert and Ultra; • inclusion of direct line-probe testing; • consideration of low-burden smear-negative specimens; • separate evaluation of primary and previously treated MDR disease; • integration of chest radiography and CT; • inclusion of studies directly correlating radiographic severity with bacterial load; • incorporation of contemporary molecular diagnostic guidance. Limitations First, imaging terminology and severity scoring varied substantially across studies. Second, bacillary burden was assessed using several non-equivalent measurements, including smear grade, molecular Ct, semiquantitative molecular categories, and liquid-culture time-to-positivity. Third, retrospective radiological studies may be affected by selection bias. Fourth, prior TB treatment, HIV infection, diabetes, smoking, and duration of illness may influence both microbiological and radiological expression. Fifth, the definition of extensively drug-resistant TB has changed over time; historical XDR studies therefore require careful interpretation according to the definitions used when they were conducted. Sixth, relatively few studies prospectively measured molecular resistance, Ct, standardized smear burden, culture burden, and standardized radiological severity simultaneously in the same MDR cohort. Future Research Future prospective MDR-TB studies should collect a common baseline dataset incorporating: • resistance mutations; • phenotypic susceptibility; • Xpert/Ultra semiquantitative result; • Ct values; • smear grade; • quantitative or semiquantitative culture; • culture time-to-positivity; • standardized chest-radiographic severity score; • CT cavity number and volume; • HIV status; • diabetes; • prior TB exposure and therapy; • time to culture conversion; • recurrence; • mortality. Integrated datasets could permit development of validated severity models combining microbiology and imaging. Artificial-intelligence-assisted chest radiography and quantitative CT may eventually enable more reproducible measurement of disease burden, but such approaches require prospective validation before they can influence MDR-TB classification.

Multidrug-resistant pulmonary tuberculosis represents the interaction of antimicrobial resistance, bacterial burden, and structural pulmonary disease. Rapid molecular tests provide the essential evidence required to identify resistance. Xpert MTB/RIF and Xpert Ultra enable rapid recognition of rifampicin resistance, while line-probe assays and expanded molecular or sequencing approaches characterize additional resistance determinants. Microbiological measurements provide a separate assessment of organism burden. High sputum smear grade, low molecular Ct, and rapid culture positivity generally indicate larger bacterial populations. Radiological severity provides a third dimension. Multiple, large, or bilateral cavities, extensive lung involvement, bronchiectasis, and destructive pulmonary changes occur more frequently in MDR pulmonary TB than in drug-sensitive disease. The relationship is clinically important but not diagnostic in isolation. Neither a high sputum bacillary load nor an extensive cavitary radiograph can establish MDR-TB. The most informative strategy is therefore an integrated one: molecular testing defines resistance, microbiology quantifies bacterial burden, and radiology measures pulmonary disease extent. Together, these domains may help identify patients with high-burden drug-resistant pulmonary tuberculosis who require urgent treatment, infection-control intervention, and close microbiological monitoring. Declarations Ethics Approval Ethics approval was not required because the study involved synthesis of previously published literature and did not include direct participation of human subjects. Consent for Publication Not applicable. Funding No external funding was received for this work. Conflict of Interest The authors declare no competing interests. Data Availability All information analyzed in this systematic review was obtained from published literature. Author Contributions Dr Maleha Ahmed: Conceptualization, literature review, methodology, data synthesis, manuscript drafting. Dr Ananda: Literature review, clinical and nursing interpretation, methodology review, critical revision of the manuscript. Dr Mekhala Taraphdar: Microbiological interpretation, molecular-diagnostics synthesis, critical revision of the manuscript. All authors reviewed and approved the final manuscript.

1. World Health Organization. Global Tuberculosis Report 2025. Geneva: WHO; 2025. 2. World Health Organization. WHO Consolidated Guidelines on Tuberculosis: Module 3: Diagnosis. Geneva: WHO; 2025. 3. World Health Organization. WHO Operational Handbook on Tuberculosis: Module 3: Diagnosis. Geneva: WHO; 2025. 4. Boehme CC, Nabeta P, Hillemann D, et al. Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med. 2010;363:1005-1015. doi:10.1056/NEJMoa0907847. 5. Boehme CC, Nicol MP, Nabeta P, et al. Feasibility, diagnostic accuracy, and effectiveness of decentralised use of the Xpert MTB/RIF test for diagnosis of tuberculosis and multidrug resistance. Lancet. 2011;377:1495-1505. 6. Chakravorty S, Simmons AM, Rowneki M, et al. The new Xpert MTB/RIF Ultra: improving detection of Mycobacterium tuberculosis and resistance to rifampin in an assay suitable for point-of-care testing. mBio. 2017;8. 7. Dorman SE, Schumacher SG, Alland D, et al. Xpert MTB/RIF Ultra for detection of Mycobacterium tuberculosis and rifampicin resistance: a prospective multicentre diagnostic accuracy study. Lancet Infect Dis. 2018;18:76-84. 8. Raizada N, Sachdeva KS, Chauhan DS, et al. A multi-site validation in India of the line probe assay for the rapid diagnosis of multidrug resistant tuberculosis directly from sputum specimens. PLoS One. 2014;9. 9. Madhuri K, Deshpande S, Dharmashale S, Bharadwaj R. Utility of line probe assay for the early detection of multidrug-resistant pulmonary tuberculosis. J Glob Infect Dis. 2015;7:60-65. 10. Singh BK, Sharma SK, Sharma R, et al. Diagnostic utility of a line probe assay for multidrug resistant-TB in smear-negative pulmonary tuberculosis. PLoS One. 2017;12. 11. Yadav RN, Singh BK, Sharma R, et al. Comparative performance of line probe assay (version 2) and Xpert MTB/RIF assay for early diagnosis of rifampicin-resistant pulmonary tuberculosis. Tuberc Respir Dis. 2021;84:237-244. 12. Chuchottaworn C, Thanachartwet V, Sangsayunh P, et al. Risk factors for multidrug-resistant tuberculosis among patients with pulmonary tuberculosis at the Central Chest Institute of Thailand. PLoS One. 2015;10. 13. Blakemore R, Nabeta P, Davidow AL, et al. A multisite assessment of the quantitative capabilities of the Xpert MTB/RIF assay. Am J Respir Crit Care Med. 2011;184:1076-1084. 14. Theron G, Pinto L, Peter J, et al. The use of an automated quantitative polymerase chain reaction (Xpert MTB/RIF) to predict the sputum smear status of tuberculosis patients. Clin Infect Dis. 2012;54:384-388. 15. Martin-Higuera MC, Rivas G, Rolo M, Muñoz-Gallego I, Lopez-Roa P. Xpert MTB/RIF Ultra CT value provides a rapid measure of sputum bacillary burden and predicts smear status in patients with pulmonary tuberculosis. Sci Rep. 2023;13:1591. 16. Kim HC, Goo JM, Lee HJ, et al. Multidrug-resistant tuberculosis versus drug-sensitive tuberculosis in human immunodeficiency virus-negative patients: computed tomography features. J Comput Assist Tomogr. 2004;28:366-371. 17. Yeom JA, Jeong YJ, Jeon D, et al. Imaging findings of primary multidrug-resistant tuberculosis: a comparison with findings of drug-sensitive tuberculosis. J Comput Assist Tomogr. 2009;33:956-960. 18. Icksan AG, Napitupulu MRS, Nawas MA, Nurwidya F. Chest X-ray findings comparison between multi-drug-resistant tuberculosis and drug-sensitive tuberculosis. J Nat Sci Biol Med. 2018;9:42-46. 19. Dholakia YN, D'souza DTB, Tolani MP, Chatterjee A, Mistry NF. Chest X-rays and associated clinical parameters in pulmonary tuberculosis cases from the National Tuberculosis Programme, Mumbai. Infect Dis Rep. 2012;4. 20. Cha J, Lee HY, Lee KS, et al. Radiological findings of extensively drug-resistant pulmonary tuberculosis in non-AIDS adults: comparisons with findings of multidrug-resistant and drug-sensitive tuberculosis. Korean J Radiol. 2009;10:207-216. 21. Lee ES, Park CM, Goo JM, et al. Computed tomography features of extensively drug-resistant pulmonary tuberculosis in non-HIV-infected patients. J Comput Assist Tomogr. 2010;34:559-563. 22. Li D, He W, Chen B, Lv P. Primary multidrug-resistant tuberculosis versus drug-sensitive tuberculosis in non-HIV-infected patients: comparisons of CT findings. PLoS One. 2017;12. 23. Tang SN, Huang XL, Skrahina A, et al. Differences in pulmonary nodular consolidation features among drug-sensitive pulmonary tuberculosis and multidrug/extensively-resistant pulmonary tuberculosis: a multi-national multi-center study. J Thorac Dis. 2025;17:7498-7514. 24. Kim W, Lee KS, Kim HS, et al. CT and microbiologic follow-up in primary multidrug-resistant pulmonary tuberculosis. Acta Radiol. 2016;57:197-204. 25. te Riele JB, Buser V, Calligaro G, et al. Relationship between chest radiographic characteristics, sputum bacterial load, and treatment outcomes in patients with extensively drug-resistant tuberculosis. Int J Infect Dis. 2019;79:65-71. 26. Grint DJ, Dhillon J, Butcher PD, et al. Xpert MTB/RIF cycle threshold as a marker of tuberculosis disease severity: implications for TB treatment stratification. Clin Infect Dis. 2026;82. 27. Desikan P, Panwalkar N, Mirza SB, et al. Line probe assay for detection of Mycobacterium tuberculosis complex: an experience from Central India. Indian J Med Res. 2017;145:70-73. 28. Ninan MM, Gowri M, Christopher DJ, Rupali P, Michael JS. The diagnostic utility of line probe assays for multidrug-resistant tuberculosis. Pathog Glob Health. 2016;110:194-199. 29. Nathavitharana RR, Cudahy PGT, Schumacher SG, Steingart KR, Pai M, Denkinger CM. Accuracy of line probe assays for the diagnosis of pulmonary and multidrug-resistant tuberculosis: a systematic review and meta-analysis. Eur Respir J. 2017;49:1601075. 30. Wáng YXJ, Chung MJ, Skrahin A, et al. Radiological signs associated with pulmonary multidrug-resistant tuberculosis: an analysis of published evidence. Quant Imaging Med Surg. 2018;8:161-173. 31. Lange B, Khan P, Kalmambetova G, et al. Diagnostic accuracy of the Xpert MTB/RIF cycle threshold level to predict smear positivity: a meta-analysis. Int J Tuberc Lung Dis. 2017;21:493-502. 32. Li CH, Fan X, Lv SX, et al. Clinical and computed tomography features associated with multidrug-resistant pulmonary tuberculosis: a retrospective study in China. Infect Drug Resist. 2023;16:651-659.