European Journal of Obstetrics & Gynecology and Reproductive Biology 160 (2012) 215–218

Contents lists available at SciVerse ScienceDirect

European Journal of Obstetrics & Gynecology and

Reproductive Biology
journal homepage: www.elsevier.com/locate/ejogrb

Targeted detection of 65 kDa heat shock protein gene in endometrial

biopsies for reliable diagnosis of genital tuberculosis
Sudha Prasad a,*, Megha Singhal a,b,c, Sanjay S. Negi b, Sunil Gupta c, Supriya Singh b,
Devendra S. Rawat b, Arvind Rai b
a
IVF and Reproductive Biology Centre, Department of Obstetrics and Gynaecology, Maulana Azad Medical College, New Delhi 110002, India
b
Biochemistry and Biotechnology, National Centre for Disease Control, New Delhi, India
c
Microbiology, National Centre for Disease Control, New Delhi, India

A R T I C L E I N F O A B S T R A C T

Article history: Objective: To evaluate the clinical utility of PCR compared with other available diagnostic modalities in
Received 26 May 2011 prompt diagnosis of female genital tuberculosis causing infertility.
Received in revised form 30 September 2011 Study design: Prospective case-controlled trial. Premenstrual endometrial biopsy specimens were
Accepted 10 November 2011
collected from 150 infertile women of reproductive age group suspected of having genital tuberculosis.
All patients underwent diagnostic endoscopy (laparoscopy and hysteroscopy) and the samples obtained
Keywords: were subjected to microscopy, culture by the BACTEC 460 TB System, histopathology and polymerase
Genital tuberculosis
chain reaction (PCR) for detection of 165 bp region of 65 kDa gene of Mycobacterium tuberculosis. The
Polymerase chain reaction
Laparoscopy
results were correlated with the laparoscopic findings.
Endometriosis Results: While the laparoscopy/hysteroscopy findings were indicative of tuberculosis in 12.6% of cases,
Infertility 14.6% of the specimens showed evidence of 65 kDa gene of M. tuberculosis and only 3.33%, 1.33% and
0.66% were positive by culture, smear and histopathology, respectively.
Conclusion: Since laparoscopy, hysteroscopy other endoscopic procedures are associated with operative
risks and may cause flaring of infection, and other conventional laboratory tests including histopathology
have poor sensitivity, PCR-based detection of 65 kDa gene of M. tuberculosis in endometrial biopsy
specimens could be a promising molecular diagnostic technique compared to conventional methods of
diagnosis.
! 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Among the various clinical presentations of tuberculosis,

female genital tuberculosis poses serious concern throughout
Mycobacterium tuberculosis (M. tuberculosis), the etiological the world because of various associated complications like
agent for tuberculosis, has been extensively studied for over a oligomenorrhea, amenorrhea, primary or secondary infertility,
century now. Despite the availability of anti-tubercular chemo- chronic pelvic pain, pelvic mass and significant mortality [1,3–7].
therapy tuberculosis still remains a major health problem and is The most commonly affected organs are the fallopian tubes and
the leading cause of morbidity and mortality in many developing/ endometrium, followed by the ovaries, cervix, vagina and vulva,
resource-poor countries. Despite the efforts that are being made to and it is often associated with tubal blockage and pelvic, peritubal
control tuberculosis worldwide, countless numbers of people die and perihepatic adhesions (Fitz–Hugh–Curtis syndrome) [3,4,8,9].
with every passing year. Efforts in averting the disease have further Female genital tuberculosis commonly causes caseation and
been impeded by the synergistic relationship between tuberculo- ulceration of the endometrium, resulting in destruction and
sis and HIV, as tuberculosis, being an opportunistic infection partial or total obliteration of uterine cavity leading to Asherman’s
worsens the immunological suppression in HIV patients. Early, syndrome [10,11]. In India, 5–16% is reportedly caused by
accurate diagnosis and immediate curative treatment, under tuberculosis, but the actual incidence rate may be underreported
proper supervision to ensure that drugs are taken for the due to asymptomatic presentation of the disease and paucity of
appropriate duration, is the key to disease control. investigations.
Female genital tuberculosis caused by M. tuberculosis, presents
invariably as a secondary complication acquired from an extra-
genital source like pulmonary or abdominal tuberculosis [3,4].
* Corresponding author. Tel.: +91 9968604341.
E-mail addresses: [email protected] (S. Prasad), [email protected]
Effective management involves rapid, accurate diagnosis and early
(M. Singhal). anti-tubercular treatment.

0301-2115/$ – see front matter ! 2011 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ejogrb.2011.11.015
216 S. Prasad et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 160 (2012) 215–218

The diagnosis of tuberculosis still relies on acid fast bacilli (AFB) 2.3. Microscopy and culture
microscopy and culture, despite the fact that both techniques
suffer several diagnostic lacunae, as reported earlier [1]. In samples The endometrial samples were processed as described by
derived from extrapulmonary sites that are often paucibacillary in Chakravorty et al. [7]. The processed endometrial extract was
nature, smear microscopy offers low sensitivity, requiring microscopically examined for AFB, and 0.1 ml of the extract was
104 bacilli/ml for detection. Culture, although being the gold inoculated into a BACTEC 12B vial along with antibiotics like
standard, requires a minimum of 10–100 bacilli/ml and a long PANTA (polymyxin B, amphotericin B, nalidixic acid, trimethoprim
incubation period (by rapid test the earliest that mycobacteria can and azlocillin) and benzyl penicillin.
be detected is within twelve days) causing delay in diagnosis and
treatment. Histopathological examination by haematoxylin and 2.4. DNA extraction
eosin for granulomatous tissue reactions compatible with tuber-
culosis infection is usually inconclusive, as reported earlier [2]. Bacterial DNA was extracted from 200 ml of the endometrial
Over the last decade, polymerase chain peaction (PCR) has biopsy sample using the commercially available Qiagen DNA
emerged as a rapid, sensitive and specific molecular method for extraction kit (Qiagen, GmbH, 40724 Hilden, Germany) following
detection of mycobacterial DNA by amplifying 65 kDa protein- the manufacturer’s protocol. An initial modification step was
encoding gene, 38 kDa antigen coding gene, the IS6110 and mpt64 carried out by keeping the preliminary processed material at 80 8C
gene in both pulmonary and extrapulmonary samples, as reported for 10 min for inactivation of the possible mycobacteria. The
by various authors [4–6,8]. Subclinical disease or latent tubercu- extracted DNA was stored at !20 8C until use.
losis infection might give a positive result in PCR, but this is
considered insignificant as prompt diagnosis is essential for 2.5. Polymerase chain reaction
averting permanent damage to genital organs and consequent
infertility. PCR for the b-globin gene was carried out to ascertain the
Endoscopic procedures like laparoscopy and hysteroscopy are quality of the extracted DNA and verify that the isolated DNA is
widely used for investigation of infertile women [12–14]. Their suitable for PCR amplification [15]. PCR was carried out to
role in the diagnosis of genital tuberculosis is well established. amplify the 165 bp region of the 65 kDa HSP gene of M.
However, subclinical, latent or early stage infection may be tuberculosis using previously reported primers [8]. The PCR was
overlooked during the procedure. A positive laboratory test may carried out in 25 ml volume consisting of the forward and
thus be helpful in timely diagnosis before extensive damage reverse primers at final concentrations of 0.01 and 1 mM,
occurs. respectively, 2.5 U of Taq polymerase (Perkin Elmer) in
Based on the aforesaid facts, the present prospective study was amplification buffer, 200 mM (each) of the four deoxyribonu-
undertaken to assess the utility of PCR in definitive diagnosis of cleoside triphosphate and 5 ml of the extracted DNA [8,9].
tuberculosis in Indian infertility patients in conjunction with Thermal profile included initial denaturation at 95 8C for 10 min
endoscopic procedures – laparoscopy and/hysteroscopy and followed by 40 cycles of denaturation at 95 8C for 20 s, annealing
conventional modalities (smear microscopy and culture based at 63 8C for 20 s, extension at 72 8C for 1 min, and final extension
on the radiometric BACTEC system). at 72 8C for 7 min. Each PCR was set up along with a positive
control (H37Rv) and several negative controls (sterile distilled
water) interspersed among samples to monitor cross-contami-
2. Materials and methods nation. The amplified products were electrophoresed on 1.5%
agarose gel and stained with ethidium bromide to check for the
The clinical part of the study was carried out at the IVF and presence of specified bands.
Reproductive Biology Centre in the Department of Obstetrics and
Gynecology, Maulana Azad Medical College, a teaching govern- 2.6. Histopathology
ment hospital in Delhi, while the molecular genotyping was carried
out at the National Centre for Disease Control (NCDC), Delhi. The Paraffin-embedded tissue sections were prepared and stained
institutional ethical committee approved the study. with haematoxylin–eosin and examined for granulomatous
reactions suggestive of mycobacterial disease.
2.1. Specimen collection
2.7. Laparoscopy/hysteroscopy
The study included 180 women between 18 and 40 years of age
presenting to the infertility clinic for a period of 18 months from All patients underwent diagnostic endoscopy (laparoscopy and
March ‘09. Additionally, 10 endometrial biopsy samples from hysteroscopy) under general anesthesia post-menstrually using
patients with a diagnosis other than tuberculosis were collected and the three puncture technique. The laparohysteroscopic findings
evaluated for the specificity of PCR in these samples. Women with a were subdivided into two types: suggestive which included
previous history of genital tuberculosis and male factor infertility affirmative findings (presence of tubercles, peritubal and/or
were excluded from the study. A detailed history comprising periovarian adhesions, tubo-ovarian mass, beaded tubes, cornual
menstrual, obstetrical and medical history, Mantoux status, blockage) and suspicious findings (hydrosalpinx, sacculated tubes,
erythrocyte sedimentation rate and previous history of tuberculosis signs of chronic inflammation, pelvic inflammatory disease, and
was taken and a systemic abdominal and vaginal examination was mild adhesions) while the second type were normal findings.
performed. Consent was obtained from every patient.
2.8. Statistical analysis
2.2. Endometrial biopsy
The population was divided into two groups of PCR positive and
Premenstrual endometrial biopsy was collected from all women PCR negative women. Data analysis was performed using SPSS
attending the infertility clinic. Endometrial tissue obtained was version 13 (SPSS Inc., Chicago, IL, USA). x2 test was used as test of
collected in normal saline in a sterile container and immediately independence. Value of P " 0.05 was considered to be statistically
transported to the tuberculosis laboratory for further processing. significant.
 S. Prasad et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 160 (2012) 215–218 217

3. Results i.e. hemosiderin, characteristic glands and stroma) was a

coincidental finding among 22.7% patients positive by PCR.
Of the 180 cases enrolled, 150 were included for the study: 18
women who did not undergo laparoscopy, 3 patients in whom 4. Comment
culture could not be done and 9 patients whose PCR results were
not reproducible or who did not undergo PCR testing were Female genital tuberculosis is a very common disease in
excluded from the study. According to Kuppuswami’s scale, the developing countries which manifests itself as pelvic inflammatory
majority of the patients (73.2%) were found to be of poor socio- disease in its acute form and later with menstrual irregularities and
economic status [16]. The mean age of the study population was infertility, and is almost always secondary to a primary lesion
28.6 # 4.6 years (range 18–40). Primary infertility was present in 108 elsewhere. The true incidence of the disease remains unknown as
women (72%) while 42 (28%) had secondary infertility. The duration the disease poses diagnostic difficulties mainly because the
of infertility ranged from 2 to 18 years. primary symptoms are usually non-characteristic. Infertility is a
A total of 22 (14.6%) women were diagnosed with genital well-known sequela. Early diagnosis invariably helps to speed up
tuberculosis on the basis of laboratory tests and laparoscopic/ the decision-making process and markedly reduces the time lag in
hysteroscopic findings. Among the samples, differences were starting anti-tubercular therapy. Although the reported incidence
observed between results for the detection of M. tuberculosis by of genital tuberculosis in Asian and western countries varies
AFB, histopathological evidence of tuberculosis infection, isolation between 0.69% in Australia and 17.4% in India, the actual incidence
by culture and detection of M. tuberculosis by PCR. PCR detected 22 may be higher because a large proportion of cases go unreported
(22/150; 14.6%), culture 5 (5/150; 0.033%) and AFB smear detected due to lack of sensitive and specific investigations [18,19]. Factors
only 2 (2/150; 0.0133%) cases. All cases positive by AFB smear and such as poverty, homelessness, a poorly functioning national
culture were also found to be positive by PCR. M. tuberculosis tuberculosis program and dismantling of public health infrastruc-
isolates were identified by the standard biochemical criteria [17]. ture have significantly contributed to the worsening situation.
Granulomatous tissue reaction compatible with tuberculosis was To qualify as a diagnostic test for genital tuberculosis, a
observed only in 1 (1/150; 0.0067%) case. combination of early diagnosis along with good sensitivity and
Laparoscopic/hysteroscopic findings were pathognomonic for specificity is required. In the present study, we evaluated the
genital tuberculosis in 19 (12.67%) women while 131 (87.33%) acceptability of PCR as a diagnostic test for female genital
women had normal findings. The various findings included tuberculosis and also correlated the findings of PCR with
adhesions; grade I (localized, covering one-third of the adnexa), laparoscopy/hysteroscopy and other diagnostic tests. A good
grade II (moderate, covering one-third to two-thirds of the correlation between PCR and laparoscopy/hysteroscopy results
adnexa), caseation, endosalpingitis and obliterated pouch of was found. The main advantage of PCR is that it is a rapid and
Douglas (POD). Tubal factor etiology was most prevalent, with specific molecular technique which allows detection of mycobac-
90% afflicted patients out of whom 15% had a history of prior teria from both pulmonary and extra-pulmonary specimens within
salpingectomy. The fallopian tube findings included cornual block, 4–5 h, compared to culture which has a poor detection rate and
hydrosalpinx, pyosalpinx and beaded-tube appearance. requires a minimum of 12 days to get the result [11,20,21].
Various clinical parameters like the Mantoux test, ESR, previous Histopathology was inconclusive in a maximum number of cases.
history of tuberculosis and type of infertility were compared This is in agreement with an earlier study by Kumar et al. [2]. In this
between the PCR positive and negative women. None of the tests prospective study, we found that PCR alone could detect the
showed a significant correlation between the two groups. Table 1 presence of M. tuberculosis DNA in 14.6% cases while AFB culture
shows the comparison of various parameters between the PCR and AFB smear showed a poor detection rate of 0.033% and 0.013%
positive and PCR negative patients. respectively, thus suggesting an appreciable and sensitive nature
Among the PCR positive patients, past history of tuberculosis of PCR in detection of mycobacteria.
was present in 31.8% of patients in the form of pulmonary (42.8%), In our study, PCR was positive in all the cases with affirmative or
lymph node (28.5%) and bone tuberculosis (28.5%). Normal suspicious findings on endoscopy while detecting an additional 5
menstrual cycles were seen in 70.6% of cases while menstrual cases which showed normal laparoscopic findings. This could be
dysfunction in the form of amenorrhoea, oligomenorrhea, menor- attributed to the fact that while laparoscopy/hysteroscopy detects
rhagia and dysmenorrhea was seen in 8%, 13.33%, 7.3% and 0.67% of conspicuous changes such as pelvic/intrauterine adhesions, tuber-
women, respectively. Oligomenorrhea was found to have a cles, grade I endometriosis with the tubes being the commonest
significant correlation between PCR positive and negative patients location, subtle changes might be overlooked. Thus an additional 3
(P < 0.05). Endometriosis grade II (consistent with endometriosis, women diagnosed with genital tuberculosis by PCR may be
considered as harboring a sub-clinical or latent infection, the early
diagnosis of which could have a significant role in regaining fertility.
Table 1
Comparison of the various characteristics between PCR positive and PCR negative
Improvement in pregnancy rates after tuberculosis treatment
patients. has been reported earlier [23,24]. It is thus imperative to diagnose
*
the disease at an early stage, thus preventing tubal damage and
Characteristics PCR positive PCR negative p-value
subsequent infertility. Endoscopic procedures like laparoscopy and
n % n % hysteroscopy, although considered as the gold standard for
No. of patients 22 128 diagnosing genital tuberculosis, have associated risks of increased
Menstrual history complication rates during the procedure and postoperative flare-
$ Amenorrhea 0 0 12 9.3 NS (.134) up of insidious disease though no operative risk and flaring of
$ Oligomenorrhea 7 31.8 13 10.1 0.005 infection was seen in our study [21,25]. High complication rates
$ Menorrhagia 2 9.09 9 7.03 NS (.732)
associated with endoscopic procedures support the use of PCR in
$ Normal 13 59.09 93 72.6 NS (.196)
$ Dysmenorrhea 0 0 1 0.78 NS (.677) diagnosis of tuberculosis.
ESR 14 63.6 58 45.3 NS (.112) Stringent quality control measures were adopted while
Mantoux test 12 54.5 53 41.4 NS (.250) carrying out PCR. Several negative controls were interspersed
Previous history of TB 7 31.8 44 34.3 NS (.815) between samples to avoid cross-contamination thus preventing
*
Statistical test: Chi square test. false positive results.
218 S. Prasad et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 160 (2012) 215–218

Menstrual dysfunction complaints were seen in 29.3% of [3] Dam T, Bose M. Paucibacillary tuberculosis—a retrospective study. J Indian
Med Assoc 2002;100(4):231–3.
patients. Oligomennorhea was found to have a significant [4] Vijaya Bhanu N, Singh Urvashi B, Chakraborty M, et al. Improved diagnostic
correlation between PCR positive and negative women, indicating value of PCR in the diagnosis of female genital tuberculosis leading to infertil-
that tuberculosis bacilli cause scarring of the endometrial lining. ity. J Med Microbiol 2005;54:927–31.
[5] Pai M, Ramsay A, O’Brien R. Evidence-based tuberculosis diagnosis. Plos Med
Partial or total destruction of endometrium by the disease process 2008;5(7):e156.
resulting amenorrhea has been reported in a few cases. [6] Singh Urvashi B, Bhanu NV, Suresh N, et al. Utility of polymerase chain reaction
The gross appearance of endometrium was mostly unremark- in diagnosis of tuberculosis from samples of bone marrow aspirate. Am J Trop
Med Hyg 2006;75(5):960–3.
able. In advanced cases, however, ulcerative or atrophic endome- [7] Chakravorty S, Tyagi JS. Novel multipurpose methodology for detection of
trium and an obliterated endometrial cavity due to extensive mycobacteria in pulmonary and extra-pulmonary specimens by smear
intrauterine adhesions was seen on endoscopy. Endometriosis microscopy, culture, and PCR. J Clin Microbiol 2005;43(June (6)):
2697–702.
grade II was observed in 22.7% of PCR positive patients. This can be
[8] Negi SS, Khan S, Gupta Sunil FB, Pasha ST, Khare S, Lal S. Comparison of the
attributed to the fact that endometriosis, being an auto-immune conventional diagnostic modalities, BACTEC culture and polymerase chain
disease, has been shown to create a defect in natural killer activity reaction test for diagnosis of tuberculosis. Ind J Med Microbiol 2005;23(1):
resulting in decreased cytotoxicity to autologous endometrium. 29–33.
[9] Shinnick TM. The 65 kilodalton antigen of Mycobacterium tuberculosis. J Bac-
This immune defect may account for the incidence of other teriol 1987;169:1080–8.
infections including tuberculosis occurring more frequently in [10] Sharma JB, Pushpraj M, Roy Kallol K, Jain Sunesh K. Weightage of different
these women [22]. diagnostic modalitites for diagnosis of tuberculosis. Int J Obs Gynaecol
2010;13:17–20.
The prevalence of female genital tuberculosis in infertility [11] Agarwal J, Gupta JK. Female genital tuberculosis—a retrospective clinicopatho-
clinics has been reported to range from 1 to 19% [4,26–28]. Atypical logical study of 501 cases. Indian J Pathol Microbiol 1993;36:389–97.
clinical presentation and failure of the laboratory tests currently [12] Sharma JB, Roy KK, Pushpraj M, Kumar S, Malhotra N, Mittal S. Laparoscopic
findings in female genital tuberculosis. Arch Gynecol Obstet 2008;278:
used have probably entailed significant underreporting of the 359–64.
actual prevalence of genital tuberculosis. [13] Al-Muhim AA. Laparoscopic diagnosis of peritoneal tuberculosis. Surg Endosc
Our results show that PCR-based detection of M. tuberculosis in 2004;18:1757–61.
[14] Volpi E, Calgara M, Ferrero A, Viagno L. Genital and peritoneal tuberculosis:
endometrial biopsy specimens is a sensitive technique for pre- potential role of laparoscopy in diagnosis and management. J Am Assoc
emptive vigilance of possible reactivation for genital tuberculosis, Gynecol Laparosc 2004;11:269–72.
which is a leading cause of infertility in developing countries. In the [15] Khoja S, Ojwang P, Khan S, Okinda N, Harania R, Ali S. Genetic analysis of HIV-1
subtypes in Nairobi, Kenya. PLoS One 2008;3(9):e3191.
absence of a gold standard, PCR is being suggested as a gold
[16] Kumar N, Shekhar C, Kumar P, Kundu AS. Kuppuswamy’s socioeconomic status
standard diagnostic modality for diagnosis of genital tuberculosis scale-updating for 2007. Indian J Paediatr 2007;74(12):1131–2.
in view of its high sensitivity and specificity. Moreover, since [17] Stager CE, Davis JR, Siddiqi SH. Identification of Mycobacterium tuberculosis in
performing laparoscopy and other endoscopic procedures on every sputum by direct inoculation of the BACTEC NAP vial. Diagn Microbiol Infect
Dis 1988;10(June (2)):67–73.
patient is not feasible in practice, PCR qualifies as a promising [18] Falk V, Ludviksson K, Agren G. Genital tuberculosis in women. Analysis of
molecular diagnostic technique in the near future. 187 newly diagnosed cases from 47 Swedish hospitals during the ten-year
period 1968–1977. Am J Obstet Gynecol 1980;138(1 December (7 Pt
2)):974–7.
Conflict of interest [19] Schaefer G. Female genital tuberculosis. Clin Obstet Gynecol 1976;19:
223–9.
None. [20] Misra R, Sharma SP, Jina R, Pant N, Srivastava DK. Female genital tract
tuberculosis with special reference to sterility in Eastern UP. J Obstet Gynaecol
India 1993;104–9.
Acknowledgements [21] Sharma JB, Pushparaj M, Roy Kallol K, Jain Sunesh K. Increased complication
rates associated with laparoscopic surgery among patients with genital tu-
berculosis. Int J Gynaecol Obstet 2010;109:242–4.
The first and second authors contributed equally to this work.
[22] King KM. Endometriosis—a systemic disease? Biochem Investig 2001.
The authors deeply acknowledge the patients for giving consent for [23] Manjunath N, Shankar P, Rajan L, Bhargava A, Saluja S, Shriniwas. Evaluation of
the study. Author MS acknowledges financial support from the a polymerase chain reaction for the diagnosis of tuberculosis. Tubercle
1991;72:21–7.
University Grants Commission, Delhi, India, during the course of
[24] Purvita D, Hasibul Hasan S, Sourendra Kanta G, et al. Role of latent genital
this study. We thank Dr. P.K. Dhamija for helpful discussions. tuberculosis in repeated IVF failures in the Indian clinical setting. Gynecol
Technical assistance from staff of IVF and Reproductive Biology Obstet Invest 2006;61:223–7.
Centre and Department of Biochemistry and Biotechnology is [25] Singh N, Sharma AK, Dadhwal V, Gupta N, Mittal S. Postoperative flare-up of
genital tuberculosis: a clinical reality. Int J Tuberc Lung Dis 2008;12(8):
highly acknowledged. 981–3.
[26] Kox LFF, Rhienthong D, Medo Miranda A, et al. A more reliable PCR for
References detection of Mycobacterium tuberculosis in clinical samples. J Clin Microbiol
1994;32(March (3)):672–8.
[1] Kolk AHJ, Schuitema ARJ, Kuijper S, et al. Detection of Mycobacterium tubercu- [27] Parikh FR, Naik N, Nadkarni SG, Soonawala SB, Kamat SA, Parikh RM. Genital
losis in clinical samples by using polymerase chain reaction and a nonradio- tuberculosis—a major pelvic factor causing infertility in Indian woman. Fertil
active detection system. J Clin Microbiol 1992;30(10):2567–75. Steril 1993;67:497–500.
[2] Kumar P, Shah NP, Singhal A, et al. Association of tubercular endometritis with [28] Roy A, Mukherjee S, Bhattacharya S, Adhya S, Chakraborty P. Tuberculous
infertility and other gynaecological complaints of women in India. J Clin endometritis in hills of Darjeeling: a clinicopathological and bacteriological
Microbiol 2008;46(12):4068–70. study. Indian J Pathol Microbiol 1993;6:361–9.