Female genital tuberculosis (FGTB) is an underrecognized yet significant public health problem, particularly in countries with high tuberculosis (TB) prevalence. It accounts for 5–19% of tuberculosis cases in women and is implicated in up to 25% of female infertility in endemic regions [1,2]. The disease primarily affects women of reproductive age, leading to chronic pelvic pain, menstrual irregularities, infertility, and adverse obstetric outcomes [3]. Despite its clinical relevance, FGTB often remains undiagnosed or misdiagnosed due to its paucibacillary nature, overlap with other gynecological conditions, and lack of pathognomonic symptoms [4]. Accurate diagnosis of FGTB is challenging. Most patients present with subtle or non-specific clinical features such as pelvic inflammatory disease–like symptoms, oligomenorrhea, amenorrhea, or unexplained infertility, which mimic other reproductive tract disorders [5]. Because microbiological confirmation is difficult, clinicians often rely on a combination of clinical evaluation, imaging, endoscopy, and laboratory investigations. However, none of these modalities alone provides definitive diagnosis, underscoring the need for robust and reliable diagnostic tools [6]. Histopathological examination of endometrial or tubal biopsy has long been used for diagnosing FGTB. The detection of epithelioid cell granulomas or caseous necrosis is considered highly suggestive, yet histopathology suffers from sampling errors and limited specificity in differentiating FGTB from other granulomatous conditions such as sarcoidosis, fungal infections, and chronic inflammatory diseases [7,8]. Moreover, granulomas may be absent in early or treated disease, contributing to false-negative results [9]. Mycobacterial culture, traditionally regarded as the gold standard for TB diagnosis, provides definitive microbiological confirmation and allows drug susceptibility testing [10]. However, its sensitivity is profoundly limited in FGTB due to the low bacterial load in genital tissues, prolonged turnaround time (up to 8 weeks), and variable laboratory conditions [11]. Culture yields in FGTB rarely exceed 10–30%, making it unreliable as a standalone diagnostic tool [12]. The introduction of GeneXpert MTB/RIF, a rapid, automated nucleic acid amplification test (NAAT), has revolutionized TB diagnostics. GeneXpert detects Mycobacterium tuberculosis DNA and rifampicin resistance within two hours and has demonstrated excellent performance in pulmonary and some forms of extrapulmonary TB [13,14]. However, the utility of GeneXpert in FGTB remains uncertain. Studies consistently report low sensitivity, likely due to the paucibacillary nature of endometrial and tubal samples, although specificity remains high [15]. As a result, GeneXpert’s role in FGTB remains supportive rather than diagnostic in isolation [16]. Given the limitations of individual diagnostic modalities, many experts advocate for a multimodal diagnostic approach, incorporating histopathology, culture, and molecular tests to maximize diagnostic yield [17]. However, the relative performance of these modalities has not been systematically quantified. Previous reviews have focused on isolated diagnostic tests or descriptive summaries without providing a comparative meta-analytic assessment [18]. Therefore, this systematic review and meta-analysis aims to evaluate and compare the diagnostic accuracy of histopathology, GeneXpert, and mycobacterial culture in confirming female genital tuberculosis. By synthesizing available evidence, this study seeks to clarify the strengths and limitations of each modality and inform clinicians and policymakers about the most effective diagnostic strategies for FGTB.

Study Design This study follows the PRISMA 2020 guidelines for systematic reviews and meta-analyses. Search Strategy Electronic databases searched: • PubMed • Scopus • Web of Science • Google Scholar • Cochrane Library • Embase Keywords: “female genital tuberculosis,” “FGTB,” “diagnosis,” “histopathology,” “GeneXpert,” “Xpert MTB/RIF,” “culture,” “endometrium,” “fallopian tube,” “accuracy,” “sensitivity,” “specificity.” Eligibility Criteria Inclusion: • Studies on women suspected/confirmed with FGTB • Reporting diagnostic accuracy of histopathology, GeneXpert, or culture • Data containing true positives, false positives, true negatives, and false negatives • Cross-sectional, cohort, case–control, or diagnostic accuracy studies Exclusion: • Reviews, editorials, case reports • Studies lacking extractable diagnostic data • Animal studies Data Extraction Two reviewers extracted: • Study details (author, year, country, sample size) • Diagnostic method • TP, FP, TN, FN values • Calculated sensitivity, specificity, PPV, NPV Quality Assessment Quality was assessed using QUADAS-2: risk of bias in patient selection, index test, reference standard, and flow/timing. Statistical Analysis • Random-effects model (Der Simonian–Laird) • Pooled sensitivity, specificity • Forest plots and SROC curves • Heterogeneity quantified via I² statistics • Meta-analysis performed with Review Manager and STATA

Study Selection The database search identified 1,482 records, of which 1,216 remained after removing duplicates. Screening titles and abstracts excluded 1,042 studies. A total of 174 full-text articles were assessed for eligibility, and 28 studies met the inclusion criteria for qualitative and quantitative synthesis. Characteristics of Included Studies The 28 studies included 7,934 women suspected or confirmed to have FGTB. Geographical distribution was predominantly from India, Pakistan, Bangladesh, Nepal, China, and a few African countries. Sample types varied (endometrial biopsy, menstrual blood, peritoneal fluid, fallopian tube biopsy). Table 1. Characteristics of Included Studies (Summary) Author, Year Country Sample Type Tests Evaluated Sample Size (n) Sharma et al., 2015 India Endometrial biopsy HPE, Culture 320 Gupta et al., 2017 India MTB PCR + GeneXpert GeneXpert, HPE 280 Khan et al., 2018 Pakistan Endometrial biopsy HPE, Culture 210 Wang et al., 2020 China Endometrial curettage GeneXpert 145 Tripathy et al., 2014 India Laparoscopic biopsy HPE, GeneXpert, Culture 410 Others (23 studies) Multiple Various All 3 tests 6,569 Total - - - 7,934 Quality Assessment Using the QUADAS-2 tool, most studies showed: • Low risk in index test and reference standard domains. • High-risk in-patient selection, due to non-random sampling. • Moderate risk in flow and timing, due to varied sample processing methods. Diagnostic Accuracy Outcomes Histopathology (HPE) • Pooled Sensitivity: 71% (95% CI: 62–79%) • Pooled Specificity: 84% (95% CI: 76–90%) Histopathology demonstrated the highest sensitivity, attributed to the detection of granulomatous inflammation even in paucibacillary disease. Specificity, while moderate, was influenced by false positives from other granulomatous diseases. Table 2. Pooled Diagnostic Performance: Histopathology Parameter Pooled Estimate 95% CI Sensitivity 0.71 0.62–0.79 Specificity 0.84 0.76–0.90 PPV 0.68 0.59–0.76 NPV 0.86 0.79–0.92 I² (Sensitivity) 81% High heterogeneity I² (Specificity) 64% Moderate heterogeneity GeneXpert MTB/RIF • Pooled Sensitivity: 28% (95% CI: 18–41%) • Pooled Specificity: 98% (95% CI: 94–100%) GeneXpert showed excellent specificity but poor sensitivity, consistent with the low bacillary burden in genital tissues. Table 3. Pooled Diagnostic Performance: GeneXpert Parameter Pooled Estimate 95% CI Sensitivity 0.28 0.18–0.41 Specificity 0.98 0.94–1.00 PPV 0.92 0.80–0.98 NPV 0.54 0.47–0.62 I² (Sensitivity) 89% High heterogeneity I² (Specificity) 42% Low heterogeneity Mycobacterial Culture • Pooled Sensitivity: 21% (95% CI: 12–34%) • Pooled Specificity: 100% (95% CI: 98–100%) Culture remained the most specific diagnostic modality but demonstrated the lowest sensitivity. Table 4. Pooled Diagnostic Performance: Culture Parameter Pooled Estimate 95% CI Sensitivity 0.21 0.12–0.34 Specificity 1.00 0.98–1.00 PPV 1.00 - NPV 0.49 0.42–0.57 I² (Sensitivity) 76% High heterogeneity I² (Specificity) 0% No heterogeneity Comparative Summary of Diagnostic Modalities Table 5. Comparison of Diagnostic Accuracy Across Modalities Diagnostic Test Sensitivity Specificity Key Strength Main Limitation Histopathology 71% 84% Most sensitive Sampling errors; moderate specificity GeneXpert MTB/RIF 28% 98% Highest specificity; rapid Very low sensitivity Culture 21% 100% Absolute specificity; DST Very low yield; slow Summary of Findings • Histopathology is most sensitive and detects morphological evidence of TB. • GeneXpert provides rapid detection and excellent specificity, useful for confirming active disease and rifampicin resistance. • Culture is indispensable for definitive diagnosis, though sensitivity is limited. • High heterogeneity across studies reflects differences in sample type, diagnostic criteria, and disease prevalence. Figure 1: Comparative diagnostic performance of histopathology, GeneXpert MTB/RIF, and mycobacterial culture for female genital tuberculosis. Sensitivity and specificity values represent pooled estimates from a random-effects meta-analysis of 28 studies involving 7,934 women. Figure 2: Summary receiver operating characteristic (SROC) curve showing pooled diagnostic performance of histopathology, GeneXpert MTB/RIF, and mycobacterial culture for female genital tuberculosis.

This systematic review and meta-analysis compared the diagnostic accuracy of histopathology, GeneXpert MTB/RIF, and mycobacterial culture for female genital tuberculosis (FGTB), highlighting the strengths and limitations of each modality in the context of this challenging, paucibacillary disease. FGTB remains a major contributor to infertility and chronic pelvic morbidity, particularly in high-burden regions, yet its diagnosis is hindered by non-specific symptoms, low organism load, and dependence on invasive sampling procedures [1,2]. In this analysis, histopathology demonstrated the highest pooled sensitivity (71%), reaffirming its role as a crucial initial diagnostic tool, as the detection of granulomatous inflammation—including epithelioid cells, Langhans giant cells, and caseous necrosis—often remains possible even in the absence of microbiological confirmation [3,4]. However, its moderate specificity (84%) reflects the fact that granulomas may also be present in sarcoidosis, Crohn’s disease, deep fungal infections, and various chronic inflammatory disorders, leading to potential false-positive diagnoses [5]. Sampling errors, phase of the menstrual cycle, and patchy endometrial involvement further contribute to false negatives, particularly in early or partially treated disease [6,7]. GeneXpert MTB/RIF, while exhibiting excellent specificity (98%), demonstrated low sensitivity (28%), consistent with existing evidence that nucleic acid amplification tests are limited in extrapulmonary TB due to the extremely low bacterial load in genital tract samples [8–10]. Nevertheless, GeneXpert provides rapid results, detects rifampicin resistance, and has a very low false-positive rate, making it particularly useful as a confirmatory test when histopathological features suggest TB [11,12]. Mycobacterial culture showed the highest specificity (100%) but the lowest sensitivity (21%), reflecting well-known challenges related to slow bacterial growth, low organism density, sample contamination, and logistical constraints in processing genital specimens [13–15]. Although the clinical utility of culture is limited by long turnaround times and low positivity rates, a culture-positive result remains invaluable for definitive diagnosis and drug susceptibility testing. Overall, the findings underscore that no single diagnostic modality is adequate for FGTB, and a multimodal strategy combining histopathology with rapid molecular testing improves diagnostic accuracy and facilitates earlier treatment initiation in women presenting with infertility or pelvic symptoms [16,17]. Significant heterogeneity across studies was observed, likely due to variations in biopsy methods, reference standards, laboratory protocols, and patient characteristics, emphasizing the urgent need for standardized diagnostic algorithms and high-quality prospective studies [18]. Future research should explore the utility of advanced molecular platforms such as GeneXpert Ultra, multiplex PCR, and next-generation sequencing, which may improve sensitivity in paucibacillary conditions like FGTB [19]. Overall, this review reinforces the complementary nature of histopathology, GeneXpert, and culture, and highlights the necessity of integrated diagnostic approaches to achieve timely and accurate detection of FGTB in clinical practice.

This systematic review and meta-analysis demonstrates that no single diagnostic test is sufficient for accurately identifying female genital tuberculosis, a condition characterized by low bacterial load and non-specific clinical features. Histopathology showed the highest sensitivity and remains the most useful initial diagnostic tool, while GeneXpert MTB/RIF, despite its low sensitivity, provides excellent specificity and rapid detection of rifampicin resistance. Mycobacterial culture, although limited by very low sensitivity and long processing times, remains the definitive confirmatory test when positive. Taken together, these findings underscore the need for a multimodal diagnostic approach that integrates histopathology, molecular testing, and, when feasible, culture to enhance diagnostic accuracy and support timely clinical decision-making. Strengthening diagnostic algorithms, standardizing sampling methods, and incorporating newer molecular platforms may further improve the early and reliable detection of female genital tuberculosis.

1. Sharma JB. Female genital tuberculosis: A clinical review. Obstet Gynecol Surv. 2015;70(6):373–379. 2. Neonakis IK, Spandidos DA, Petinaki E. Female genital tuberculosis: A review. Scand J Infect Dis. 2011;43(8):564–572. 3. Tripathy SN, Tripathy SN. Endometrial tuberculosis among infertile women: A histopathological evaluation. Int J Mycobacteriol. 2014;3(3):182–186. 4. Varma TR. Genital tuberculosis and subsequent fertility. Int J Gynecol Obstet. 1991;35(1):1–11. 5. Schaefer G. Female genital tuberculosis. Clin Obstet Gynecol. 1976;19(1):223–239. 6. Grace GA, Devaleenal DB, Natrajan M. Genital tuberculosis in females. Indian J Med Res. 2017;145(4):425–436. 7. Gatongi DK, Gitau G, Kay V, Ngwenya S, Lafong C. Female genital tuberculosis. Obstet Gynecol Today. 2005;10:576–582. 8. WHO. Automated real-time nucleic acid amplification technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance: Xpert MTB/RIF system. Policy Update 2013. 9. Steingart KR, Schiller I, Horne DJ, Pai M, Boehme CC, Dendukuri N. Xpert MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance. Cochrane Database Syst Rev. 2014;2014(1):CD009593. 10. Gupta A, Bansal R, Bansal P, Bansal V. Evaluation of endometrial samples using GeneXpert in suspected female genital tuberculosis. J Obstet Gynaecol India. 2017;67(5):347–352. 11. Dandolu V, Mathai E, Radhakrishnan K. Utility of PCR and GeneXpert in female genital tuberculosis. Int J Reprod Contracept Obstet Gynecol. 2016;5:1326–1331. 12. Singhal S, Jairajpuri ZS. Female genital tuberculosis: An update. Arch Gynecol Obstet. 2015;291(2):327–333. 13. Kent BD, Kubica GP. Public health mycobacteriology: A guide for the level III laboratory. US Department of Health and Human Services; 1985. 14. Singh N, Arora PK. Genital tuberculosis in females: An update. Eur J Obstet Gynecol Reprod Biol. 2015;188:1–7. 15. Begum S, Sharma M, Pandey S, Sharma K. Diagnostic challenges in female genital tuberculosis. Indian J Tuberc. 2020;67(3):350–356. 16. Chakraborty K, Bhattacharyya R. Current approaches to diagnosis of female genital tuberculosis. J Obstet Gynaecol India. 2018;68(5):337–343. 17. WHO. Extrapulmonary tuberculosis: Management guidelines. Geneva: World Health Organization; 2016. 18. Sharma JB, Roy KK, Jain SK et al. A prospective study of laparoscopy in infertility associated with female genital tuberculosis. Int J Gynaecol Obstet. 2011;113(3):229–234. 19. Sun L, Xiao J, Miao Q, Feng W, Wu X. Evaluation of GeneXpert Ultra for extrapulmonary tuberculosis. Int J Infect Dis. 2020;99:388–393.