|Year : 2021 | Volume
| Issue : 1 | Page : 66-70
Female genital tuberculosis in Pakistan – A retrospective review of 10-year laboratory data and analysis of 32 cases
Tazeen Fatima1, Rumina Hasan1, Faisal Riaz Malik1, Imran Ahmed1, Linda Alice Bartlett2, Michael G Gravett3, Sadia Shakoor1
1 Department of Pathology & Laboratory Medicine, Aga Khan University, Karachi, Pakistan
2 Bloomberg School of Public Health, Johns Hopkins School of Public Health, Baltimore, MD, USA
3 Department of Obstetrics and Gynecology, University of Seattle, WA, USA
|Date of Submission||12-Jan-2021|
|Date of Acceptance||21-Jan-2021|
|Date of Web Publication||28-Feb-2021|
Aga Khan University, Stadium Road, P.O. Box 3500, Karachi 74800
Source of Support: None, Conflict of Interest: None
Background: Female genital tuberculosis (FGTB) is an underobserved clinical entity owing to diagnostic challenges stemming from difficulty of obtaining diagnostic specimens and paucibacillary nature of the disease. Yet, FGTB is a cause of infertility, pelvic pain, or menstrual irregularities in high-burden countries. To assess laboratory and microbiology diagnostic utilization for FGTB in Pakistan, we have collected data from 2007 to 2016 to inform the need for improved laboratory diagnostics. The objectives of this study were to determine the proportion of FGTB as culture-confirmed extrapulmonary tuberculosis (EPTB) and to describe the characteristics of women with culture-confirmed FGTB in a nationwide laboratory network in Pakistan. Method: A retrospective database was established by accessing laboratory archives and analyzed by sex and source to determine extrapulmonary cases among women. Data were checked for quality, and after removing patient identifiers and duplicate samples, frequencies were calculated in MS Excel. Clinical characteristics of patients were derived from a linked hospital database for those patients who were diagnosed and managed at the affiliated university hospital in Karachi, Pakistan. Results: Over 10 years, 410,748 mycobacterial cultures were received from multiple geographic sites throughout Pakistan and processed at the study laboratory. The overall mean culture positivity rate was 5.9% ± 3.5%, while the mean culture positivity rate among females was 2.8% ± 0.8%. Among female culture-confirmed tuberculosis cases, the pulmonary-to-EPTB ratio of infection was 5. Over 10 years, a total of 32 FGTB cases were reported on the basis of positive cultures for Mycobacterium tuberculosis; 3 (9.4%) were rifampin resistant. Conclusions: FGTB currently constitutes a small but significant proportion of culture-confirmed EPTB. A fewer number of laboratory requisitions suggest the need to increase awareness and testing. The advent of high-sensitivity molecular testing on extrapulmonary specimens has the potential to improve diagnostic accuracy and improved detection of FGTB cases in high-burden regions.
Keywords: Diagnosis, endometrial, female genital tuberculosis, histopathology, mycobacterial culture, tubo-ovarian
|How to cite this article:|
Fatima T, Hasan R, Malik FR, Ahmed I, Bartlett LA, Gravett MG, Shakoor S. Female genital tuberculosis in Pakistan – A retrospective review of 10-year laboratory data and analysis of 32 cases. Int J Mycobacteriol 2021;10:66-70
|How to cite this URL:|
Fatima T, Hasan R, Malik FR, Ahmed I, Bartlett LA, Gravett MG, Shakoor S. Female genital tuberculosis in Pakistan – A retrospective review of 10-year laboratory data and analysis of 32 cases. Int J Mycobacteriol [serial online] 2021 [cited 2021 Apr 20];10:66-70. Available from: https://www.ijmyco.org/text.asp?2021/10/1/66/310517
| Introduction|| |
Female genital tuberculosis (FGTB) classically presents with infertility, pelvic pain, or menstrual irregularities, with or without other systemic symptoms of tuberculosis. Such symptoms and resulting debilitation as well as economic costs can have serious detrimental effects on women's quality of life, sexual and reproductive health, mental health, and productivity. Yet, the burden of FGTB is unclear – remaining an underdiagnosed form of extrapulmonary tuberculosis (EPTB), even among women in high-burden settings.
The widely held belief that FGTB is a rare site for EPTB, is due to the challenges of limited health-care utilization by women, covert symptoms, and paucibacillary nature of FGTB. The focus of national tuberculosis control programs on pulmonary tuberculosis (PTB) exacerbates this underestimation of prevalence. However, among women presenting with pelvic inflammatory disease (PID) and/or infertility, published data from high tuberculosis burden settings indicate a high prevalence of FGTB.
Data from the 2016 national survey of tuberculosis control program health-care facilities in Pakistan indicate that rates of EPTB are higher among females as compared to males, as has also been observed in studies from South Asia. However, tuberculosis control program data are not disaggregated by the site of EPTB and sex-disaggregated data on incidence cannot be used to estimate FGTB burden.
In an effort to better define the FGTB epidemiology in the high-burden country of Pakistan, we examined data from a referral laboratory to report culture-confirmed FGTB rates among females evaluated for tuberculosis. These data are important to inform the need for guideline development, facilitated referral for improved health-care delivery and quality of life in women, physicians training, and further research in diagnostics.
| Method|| |
We report on data from Pakistan, a high-burden country with an incidence of tuberculosis of 263 per 100,000 population, 41% of total cases occurring in women. This study was performed at the Aga Khan University Clinical Laboratories, with collection outlets in 289 different locations across Pakistan. Samples were collected in accordance with laboratory policies and transported within 48 h to the central mycobacteriology laboratory in Karachi, a biosafety level 3 facility. The laboratory has been accredited by the College of American Pathologists (CAP) since 2016 and is compliant with their standards. Laboratory records are archived for 2 years. However, electronic patient investigation records of essential variables are archived on a separate server at the Aga Khan University Information Technology Unit.
Study design and data sources
This was a retrospective descriptive analysis of FGTB rates among culture-confirmed EPTB cases among females, with an analysis of FGTB cases based on information obtained through linked electronic records. Data were retrieved for the years of 2007–2016.
[Supplementary Figure 1] describes data retrieval, abstraction, and secondary database creation stages and sequence. Laboratory records disaggregated by sex from 2007 to 2016 were retrieved, and duplicates were removed to generate denominators for the rates calculated for male versus female cultures requested and the proportion of positive cultures. A second database with culture results was retrieved from 2007 to 2016, and other variables in the database included sex, age, specimen type, cultured isolate identification (Mycobacterium tuberculosis or nontuberculous mycobacteria), susceptibility testing results, dates of specimen collection, and reporting. Linked laboratory identifiers allowed retrieval of histopathology requisitions and results, and linked hospital identifiers allowed examination of patient management details, where available.
We used the ISPOR and STROBE checklists to evaluate the standard of reporting for this descriptive analysis. An intersection of both checklists for this analysis is available in [Supplementary Table 1].
From 2007 to 2016, the laboratory used both liquid and solid media on specimens processed with 2% sodium hydroxide and N-acetyl-L-cysteine method and high-speed centrifugation. Extrapulmonary specimens (except cerebrospinal fluid) were decontaminated and liquid cultures were continuously monitored for evidence of growth for 6 weeks. The laboratory subscribed to proficiency testing program by CAP during the study period and obtained satisfactory results. Culture contamination rates during the 10-year study period remained within the recommended benchmark of 5%–8%.
Data abstraction and analysis
Data downloaded from laboratory servers were extracted from records using a preexisting coding manual for laboratory variables and by date (year). After initial extraction, data were abstracted into a separate Excel database by creating drop-down categories for the following variables: year, sex, specimen types, cultured isolated identification, and drug susceptibility (sensitive or resistant) for rifampin, isoniazid, and ofloxacin. Duplicate entries for specimen type and patient identification numbers were removed in batches. Data quality checks and verification were performed by comparison to source data. Any conflicts were resolved by consultation with the author group. Data were refined by type of FGTB specimen source, including endometrial tissue, endometrial curettage sample, tubo-ovarian samples, ovarian fluid/pus/tissue, and endocervical sample. For culture-confirmed FGTB samples, linked histopathology records were retrieved from the laboratory database and linked radiology and clinical records were retrieved from the hospital databases and a final relational database was created in MS Excel for the culture-confirmed FGTB cases. Frequencies and means or median values were calculated in MS Excel and Chi-square test for proportions was performed where applicable using Chi-square test calculator from www.socstatistics.com (https://www.socscistatistics.com/tests/chisquare2/default2.aspx).
The study was reviewed and exempted from individual informed consent by the Ethical Review Committee of the Aga Khan University, Pakistan.
| Results|| |
Study population and laboratory utilization
From 2007 to 2016, the study laboratory received 410,748 specimens for mycobacterial culture. The male: female ratio of all requests was 1.3. During the 10 study years, the mean culture positivity rate was 5.9% ± 3.5%, with 2.8% ± 0.8% of these positive cultures from females of all age groups. Data are presented in [Supplementary Figure 2] to show the variance across years.
Culture positivity and extrapulmonary tuberculosis rates among females
Among a total of 177,463 cultures from females, 11,544 (6.5%) were positive for M. tuberculosis complex (MTB), and 1889 (16.4%) of these positive cultures were from extrapulmonary specimen sources. Rifampin resistance (RR) was observed more frequently in PTB in females at 47.5% (n = 4583), while it remained low in EPTB cases in females at 15.8% (n = 298) (P < 0.0001; Chi-square). FGTB was responsible for 1.8% of all culture-positive EPTB specimens in females. [Figure 1] shows cultures received, positivity rates, and RR among PTB and EPTB in females of all age groups.
|Figure 1: Culture-confirmed tuberculosis among females in Pakistan based on data from a large laboratory network 2007–2016. Figures in the table represent total numbers of culture-positive specimens. MTB: Mycobacterium tuberculosis complex, RR: Rifampin-resistant tuberculosis, PTB: Pulmonary tuberculosis, EPTB: Extrapulmonary tuberculosis, FGTB: Female genital tuberculosis|
Click here to view
Culture-confirmed female genital tuberculosis
A total of 36 FGTB specimens were culture positive for Mycobacterium species. Two of these samples were excluded from this analysis due to growth of nontuberculous mycobacteria. Of the 34 MTB culture-positive specimens in 32 patients (two tubo-ovarian and one endometrial specimens were received from one patient), 31 were females 15–49 years, with a median age of 26.5 (32–25) years. Clinical indications, age, and linked specimen positivity details are presented in [Table 1].
|Table 1: Demographics, clinical indications, and microbiological and histopathological details of 32 women with culture-confirmed female genital tuberculosis diagnosed at a countrywide laboratory network, 2007-2016|
Click here to view
Among culture-confirmed FGTB, infertility was the most common indication for evaluation. Only one patient had a positive smear for acid-fast bacilli. Molecular tests (i.e., Xpert MTB/RIF or a nucleic acid amplification test) were not requested on any of the patients. Histopathological details were available in 28 patients, with 22 (78.6%) showing agreement with culture confirmation, i.e., histopathological examination showed chronic granulomatous inflammation with epithelioid cells and giant cells with or without caseous necrosis. RR was observed in 3 patients in endometrial specimens. Of the two RR patients, histopathological examination was available in two and did not show chronic granulomatous inflammation. One patient had a benign late secretory endometrium while the second had abnormal secretory endometrium, with no signs of inflammation.
| Discussion|| |
FGTB accounts for a small proportion (1.8%) of culture-positive EPTB in this countrywide laboratory network. This proportion, however, is not indicative of the incidence or prevalence of FGTB. Data from evaluation cohorts among women of reproductive age with infertility and PID in Pakistan have reported higher rates of FGTB (3%–3.5%)., Furthermore, when performed in subsets of women with infertility or associated symptoms, the yield of laparoscopic diagnosis is 44%, and of mycobacterial culture and/or nucleic acid, detection by polymerase chain reaction-based tests is 10%, and the two approaches are complementary. Laparoscopic diagnosis of tuberculosis among women with chronic pelvic pain from Pakistan has also demonstrated higher rates (20%), but the laparoscopic diagnosis was not confirmed on culture or histopathological examination of endometrial tissue. An integrated approach to diagnosis of FGTB is therefore recommended, with availability of all these services as part of national tuberculosis control programs in liaison with health services for women. Particularly, Xpert Ultra may be a useful tool in improving the yield of molecular diagnosis.
The low proportion of culture-confirmed tuberculosis observed in our data, despite the expected high prevalence of FGTB, may be a function of any of the following factors: paucibacillary nature and low culture yield of genital specimens (endometrium), lack of availability or reduced utilization of molecular methods with presumed higher sensitivity, use of private laboratory services with data that are not captured by national programs or in this study, or lack of access to health-care services required for the appropriate diagnosis and management of FGTB. Each of these factors likely contributes to the traditional understanding of FGTB as a “rare” form of tuberculosis in women even in high-burden settings. A recent evaluation of EPTB in Pakistan for the year 2016 in public health-care facilities notifying tuberculosis to national authorities, however, reveals that data are either not segregated by source or the proportion of genitourinary specimens submitted for women was very low, again pointing to low service utilization. Moreover, tuberculosis notification rates were based on smear positivity and/or Xpert MTB RIF (which is available in 5.6% of reporting facilities) and not culture or histopathological diagnosis.
The data presented demonstrate good agreement between histopathological diagnosis of FGTB and culture positivity. Two cases where RR MTB were cultured were not confirmed on histopathology and may well be representative of sample contamination during collection and transport or laboratory cross-contamination. Lower rates of RR among EPTB cases in Pakistan also support the hypothesis that culture results are a result of contamination rather than RR FGTB.
The data highlight the need for improving FGTB algorithms to facilitate improved diagnosis and management in Pakistan. Development of new and improved diagnostic tools and their application to diagnosis of FGTB should be advocated to improve case detection and confidence in diagnosis. Implementation of these diagnostic modalities would require strategies to improve access and service delivery to impact the health of women through reducing FGTB in Pakistan. Lack of published data on etiology of PID, especially granulomatous endometritis in Pakistan, also highlights the need for research to evaluate appropriate and innovative methods for specimen collection and integrated diagnosis.
HIV status of women was not known and linked laboratory and hospital databases for the 32 women with FGTB did not reveal if any of them had been evaluated for diabetes or prediabetes. Some limitations of this study which stem from the study design are important to note. The retrospective nature of the study did not allow us to address the proportion of FGTB among all women of reproductive age group who were evaluated for gynecological complaints. Specimen source-wise negative cultures could not be extracted from the laboratory's electronic database. We were also unable to access an independent database of histopathological examinations to evaluate the culture-negative proportion of granulomatous endometritis and salpingitis. Limitations of this study noted above highlight the need for program linkages and electronic databases.
Improved surveillance for FGTB paired with uptake of complementary and enhanced diagnostic methods in women is needed in Pakistan. Neither Xpert nor histopathology services are widely available to allow improved diagnosis which highlights the need for facilitated referral. The low positivity rates of culture in FGTB should lead to advocacy for research into and implementation of improved diagnostic methods and strategies.
The study was reviewed and exempted from individual informed consent by the Ethical Review Committee of the Aga Khan University, Pakistan.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Botha MH, van der Merwe FH. Female genital tuberculosis. S Afr Fam Pract 2008;50:12-6.
Goel G, Khatuja R, Radhakrishnan G, Agarwal R, Agarwal S, Kaur I. Role of newer methods of diagnosing genital tuberculosis in infertile women. Indian J Pathol Microbiol 2013;56:155-7.
] [Full text]
Ilmer M, Bergauer F, Friese K, Mylonas I. Genital tuberculosis as the cause of tuboovarian abscess in an immunosuppressed patient. Infect Dis Obstet Gynecol 2009;2009:745060.
Chowdhury NN. Overview of tuberculosis of the female genital tract. J Indian Med Assoc 1996;94:345-6,361.
Grace GA, Devaleenal DB, Natrajan M. Genital tuberculosis in females. Indian J Med Res 2017;145:425-36.
] [Full text]
Tahseen S, Khanzada FM, Baloch AQ, Abbas Q, Bhutto MM, Alizai AW, et al.
Extrapulmonary tuberculosis in Pakistan – A nation-wide multicenter retrospective study. PLoS One 2020;15:e0232134.
Mehraj J, Khan ZY, Saeed DK, Shakoor S, Hasan R. Extrapulmonary tuberculosis among females in South Asia-gap analysis. Int J Mycobacteriol 2016;5:392-9. [Full text]
Global tuberculosis report 2020. Geneva: World Health Organization; 2020. Licence: CC BY-NC-SA 3.0 IGO.
Motheral B, Brooks J, Clark MA, Crown WH, Davey P, Hutchins D, et al
. A checklist for retrospective database studies--report of the ISPOR Task Force on Retrospective Databases. Value Health 2003;6:90-7.
STROBE Strengthening the Reporting of Observational Studies in Epidemiology. University of Bern © ISPM 2009. Available from: https://www.strobe-statement.org/
. [Last accessed December 2020 20].
CLSI. Laboratory Detection and Identification of Mycobacteria. 2nd
ed. CLSI Guideline M48. Wayne PA: Clinical and Laboratory Standards Institute; 2018.
Shaheen R, Subhan F, Tahir F. Epidemiology of genital tuberculosis in infertile population. J Pak Med Assoc 2006;56:306-9.
Haider P, Jafarey SN. A histopathological study of endometrial tuberculosis in infertility. J Pak Med Assoc 1992;42:269-71.
Jindal UN, Bala Y, Sodhi S, Verma S, Jindal S. Female genital tuberculosis: Early diagnosis by laparoscopy and endometrial polymerase chain reaction. Int J Tuberc Lung Dis 2010;14:1629-34.
Baloch S, Khaskheli MN, Malik AM. Diagnostic laparoscopic findings in chronic pelvic pain. J Coll Physicians Surg Pak 2013;23:190-3.
Tahseen S, Ambreen A, Masood F, Qadir M, Hussain A, Jamil M, et al
. Primary drug resistance in extra-pulmonary tuberculosis: A hospital-based prospective study from Pakistan. Int J Tuberc Lung Dis 2019;23:900-6.
Munne KR, Tandon D, Chauhan SL, Patil AD. Female genital tuberculosis in light of newer laboratory tests: A narrative review. Indian J Tuberc 2020;67:112-20.