The International Journal of Mycobacteriology

: 2016  |  Volume : 5  |  Issue : 3  |  Page : 318--323

Laparoscopy in the diagnosis of tuberculosis in chronic pelvic pain

Shalini Rajarama1, Priyanka Gupta1, Bindiya Gupta1, Iqbal R Kaur2, Neerja Goel1,  
1 Department of Obstetrics and Gynecology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Dilshad Garden, Delhi, India
2 Department of Microbiology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Dilshad Garden, Delhi, India

Correspondence Address:
Bindiya Gupta
Department of Obstetrics and Gynecology, MCH 4th Floor, University College of Medical Sciences and Guru Teg Bahadur Hospital, Dilshad Garden, 110095 Delhi


Background: To estimate the prevalence of genital tuberculosis in women with idiopathic chronic pelvic pain on laparoscopy, correlate laparoscopic findings with microbiological and histological diagnosis of tuberculosis and assess the response to anti tubercular treatment (ATT) in these cases. Method: In a prospective cohort study, fifty women with idiopathic chronic pelvic pain were enrolled. Diagnostic laparoscopy was done in all women and fluid from pouch of Douglas and/or saline washings were sent for acid fast bacilli (AFB) smear, conventional and rapid culture and DNA polymerase chain reaction (PCR) analysis for diagnosis of genital TB. The results of these tests were analyzed and agreement with laparoscopy was assessed using Kappa statistics. Pain scores using visual analogue scale were compared before and after treatment. Results: Pelvic pathology was present in 44 (88%) women of idiopathic chronic pelvic pain, with a 34% prevalence rate of genital tuberculosis. Pelvic inflammation was associated with positive peritoneal fluid PCR (n = 4) and AFB culture (n = 3). Acid fast bacilli PCR had substantial agreement (kappa statistics = 0.716) with visual findings at laparoscopy. There was a significant reduction in pain scores after treatment. Conclusion: Genital tuberculosis contributes to one-third cases of chronic pelvic pain. Pelvic inflammation is an early feature of genital TB and peritoneal fluid PCR has the best co-relation with laparoscopic findings of genital tuberculosis.

How to cite this article:
Rajarama S, Gupta P, Gupta B, Kaur IR, Goel N. Laparoscopy in the diagnosis of tuberculosis in chronic pelvic pain.Int J Mycobacteriol 2016;5:318-323

How to cite this URL:
Rajarama S, Gupta P, Gupta B, Kaur IR, Goel N. Laparoscopy in the diagnosis of tuberculosis in chronic pelvic pain. Int J Mycobacteriol [serial online] 2016 [cited 2023 Jan 28 ];5:318-323
Available from:

Full Text


The global prevalence of chronic pelvic pain (CPP) ranges from 4% to 43.4%, with prevalence rates reported to be 15% and 24% in the United States and the United Kingdom, respectively [1]. The prevalence rate of CPP in Asia ranges from 5% to 43%, whereas in India it is 5.2% [1]. CPP accounts for 20% of outpatient visits to gynecologists, 40% of laparoscopies, and 10–15% of hysterectomies [2],[3].

CPP is a multifactorial entity, with a poorly understood etiology and complex pathophysiology. As a result, there is a lack of evidence-based diagnosis and management options for CPP. Management of CPP in women is difficult with normal gynecological examinations and imaging studies (idiopathic CPP) and such cases are best evaluated on laparoscopy. Common findings on laparoscopy in these cases include endometriosis, adhesions, pelvic inflammatory disease, and pelvic congestion [4].

Pelvic inflammatory disease is an important cause of CPP, of which genital tuberculosis (TB) may present as CPP in 42–80% cases, the latter being higher in developing countries [5],[6]. Developed countries have seen a resurgence in the incidence of TB mainly because of human immunodeficiency virus infections, but the prevalence of CPP due to genital TB is not known. Diagnosis of female genital TB is difficult due to its paucibacillary nature, and available conventional methods such as erythrocyte sedimentation rate (ESR), Mantoux test, and chest X-ray have low sensitivity and specificity. Isolation of acid-fast bacilli (AFB) on smear and culture and presence of granulomas on histopathology have the highest specificity for diagnosis of genital TB, but these lack sensitivity [7]. Rapid diagnostic methods such as BACTEC 460 Mycobacteria growth inhibitor tube (MGIT) and specific gene probes have recently gained popularity [8]. Polymerase chain reaction (PCR) for detecting mycobacterial DNA (mpt64 gene) has high sensitivity and specificity and may indicate disease positivity even if the number of bacteria present is as low as one to 10 organisms/mL [9].

Genital TB usually occurs as a secondary infection due to hematogenous spread and primarily affects the fallopian tubes in 95–100% cases [7]. It was hypothesized in this study that the bacilli will be shed more frequently into the pouch of Douglas (POD) from the fallopian tubes. Therefore, isolation of bacilli from the fluid samples of the POD or peritoneal washings will be useful, especially in the management of women with pelvic pain of unknown etiology.

The aim of the study was to evaluate the prevalence of genital TB in women with idiopathic CPP. Secondary study objective was to correlate laparoscopic findings with microbiological and histological diagnoses of TB to determine the best diagnostic test and assess the response to therapy after laparoscopy.

 Materials and methods


This prospective cohort study was conducted in the Department of Obstetrics and Gynecology and the Department of Microbiology at the University College of Medical Sciences and Guru Teg Bahadur Hospital (Delhi, India) between October 2011 and December 2012. Institutional Review Board Committee approval was obtained from the Institutional Ethical Committee of University College of Medical Sciences and informed written consent was obtained from each patient.

Study population

From a previously conducted pilot sample study of 20 cases, the incidence rate of genital TB in CPP was found to be 40% (unpublished data). To estimate the incidence rate of 40% with 15% absolute precision on either side with 95% confidence level, with continuity correction, a sample size of 44 patients was adequate. Based on this statistics, 50 consecutive women aged between 18 years and 50 years, attending the Gynecology Outpatient Department, with pelvic pain lasting for more than 6 months, and with normal gynecologic examination results (idiopathic CPP) were included in this prospective study to allow for 10% loss to follow-up.

Women with a known cause of CPP such as fibroid uterus, adenomyosis, tubo-ovarian masses, obvious nongynecological cause of CPP, currently on antitubercular therapy, or positive for human immunodeficiency virus infection were excluded.


Preoperative work up

After a detailed history taking and thorough clinical examination, all women underwent a complete blood count test with ESR, Mantoux test, chest radiography, and abdomen and pelvic sonogram. Endometrial biopsy in the premenstrual phase (Day 21 and Day 22) was taken in all women using a Number 4 Karman cannula. Genital TB was diagnosed by detection of AFB on smear using Ziehl–Neelsen stain, which was cultured by inoculation on Löwenstein–Jensen medium and by histopathological examination of the biopsy sample. All women then underwent diagnostic laparoscopy followed by operative procedures wherever indicated.


All laparoscopies were performed under general anesthesia by a team of three surgeons using autoclavable reusable instruments. At the time of laparoscopy, the pelvis and abdomen were visualized systematically and abnormal findings were noted. Cases with presence of tubercles, caseation, granulomas, beaded tubes, tubo-ovarian masses, fimbrial agglutination, and hydrosalpinx were considered to have TB. Red inflamed tubes and ovaries with the presence of straw-colored peritoneal fluid (pelvic inflammation) were considered suspicious for genital TB. In women with positive findings of TB on laparoscopy, endometrial biopsy was repeated during surgery and samples were sent for histopathological examination. Biopsy was performed on samples of suspicious lesions and then subjected to histopathological examination. Operative procedures such as adhesiolysis, ablation of endometriotic lesions, and cystectomy were performed in the same sitting. The samples of peritoneal fluid (10 mL) or peritoneal washings were collected from the Pouch of douglas (POD) through a 5-mm side port using an aspiration needle and stored in deep freezer at −20 °C for TB diagnostic tests.

Microbiological tests

The samples were centrifuged prior to microscopic analysis. On an average, two slides were prepared for each case and 100 fields were analyzed. Steps in the processing and analysis of fluid samples collected from the POD were the standard steps followed for detection of DNA by PCR. These steps included decontamination, DNA extraction, PCR-based DNA amplification of 240-bp fragment of the mpb64 gene followed by gel electrophoresis. AFB smear, conventional culture, and rapid culture (BACTEC MGIT 960 system; Becton Dickinson, Franklin Lakes, New Jersey, United States) tests were also performed in all cases. The microbiological tests were performed in a fully standardized laboratory (Department of Microbiology) of our medical college.


Based on laparoscopic findings, women were divided into two groups: Group A (findings suggestive of genital TB) and Group B (negative findings of genital TB). Women in Group A received antitubercular treatment, whereas those in Group B received treatment according to the specific condition in the form of oral contraceptive pills and gonadotropin analogs.

Statistical analysis

Baseline variables of Groups A and B were compared using Fisher exact test. Results of specific diagnostic tests for TB on fluid samples from the POD were compared with laparoscopic findings. Kappa statistics was applied to detect the degree of agreement between the microbiological tests and histopathology with laparoscopic evaluation. Pretreatment visual analog scale (VAS) score among women was compared using Kruskal–Wallis test and pre- and post-treatment VAS scores were compared using Wilcoxon signed-rank test. Data were analyzed by SPSS statistical software version 20 (SPSS Inc., Chicago, IL, USA) and p values <.05 were considered significant.


Fifty women were consecutively enrolled in this study and their test results were analyzed. None of the women was excluded from study analysis. The baseline variables of the 50 women enrolled are summarized in [Table 1].{Table 1}

Laparoscopic findings

Results of laparoscopic examination were normal in six women (12%), whereas in 44 patients (88%), the results showed some pelvic pathology. Gross findings of TB were seen in 13 patients (26%), followed by endometriosis in 10 (20%), adhesions in eight (16%), inflammation with straw-colored fluid in four (8%), adenomyosis in three (6%), pelvic congestion syndrome in three (6%), and miscellaneous findings (fibroid or fimbrial cyst) in three (6%) women. The details of laparoscopic findings are summarized in [Table 2]. The peritoneal fluid was present in 25 (50%) women.{Table 2}

Microbiological tests and histopathology

Of the 50 samples of peritoneal fluid/washings, one was AFB positive on microscopy, three were positive on conventional culture, and one tested positive on rapid culture with BACTEC. Laparoscopic samples from 14 women (28%) tested positive on PCR (11 peritoneal fluid samples and three samples from peritoneal saline washings). Patients with positive BACTEC culture were negative for both AFB smear and conventional culture but were positive on PCR. In two women, peritoneal biopsy showed granulomatous inflammation. The overall prevalence of genital TB was 34% (17/50); all these 17 cases are summarized in [Table 3].{Table 3}

Subgroup analysis

Group A included 17 cases in which genital TB was diagnosed laparoscopically, of which 13 patients had gross findings of TB and four had pelvic inflammation with positive culture and/or positive results on PCR. Group B included 33 cases negative for TB on laparoscopy. Only one woman was PCR positive in Group B.

Baseline characteristics

There was no significant difference in baseline characteristics between the two groups, except that there were significantly more women with constitutional symptoms and previous history of TB in Group A (p = .001). In Group A, mean ESR was 34±9.44 mm (range, 18–48 mm), whereas in Group B, it was 24.76±10.57 (p<.05). There was also a statistically significant difference in Mantoux test and abnormal chest X-ray findings in genital TB women versus nongenital TB women (p<.05).

Correlation studies

Correlation of all diagnostic tests with laparoscopy is summarized in [Table 4]. The concordance of results between conventional laparoscopic criteria of genital TB and PCR was found to be .584 (moderate agreement). All four cases of pelvic inflammation with straw-colored fluid were PCR positive (3 were also AFB culture positive), and upon including these cases, the association improved from moderate to substantial agreement (.716). On analyzing the correlation between laparoscopic criteria and positive results on either peritoneal fluid analysis or endometrial biopsy, Kappa measure of agreement was .604 (substantial agreement).{Table 4}

Treatment and pain scores

Antitubercular therapy was started in 17 women of Group A, out of which one patient discontinued treatment. Oral contraceptive pills were given in 16 (32%) women with endometriosis and pelvic congestion, antibiotics in six (12%), and gonadotropin-releasing hormone analogs in three (6%) for adenomyosis and Grade 4 endometriosis. Six women received drugs for induction of ovulation. Surgical treatment included adhesiolysis (18%), laparoscopic uterosacral nerve ablation (4%), nondescent vaginal hysterectomy (6%), and one patient each (2%) underwent salpingostomy, salpingectomy, and cystectomy. The pretreatment VAS score was comparable among all women in the treatment group (p = .332). There was a significant reduction in the severity of pain in women receiving any kind of treatment on comparing the pre- and post-treatment VAS scores ([Figure 1]).{Figure 1}


In this study, the overall prevalence of genital TB in idiopathic CPP was 34% (17/50), of which 13 had gross findings of genital TB and four had pelvic inflammation with positive culture and/or positive results on PCR. All women had symptomatic relief on anti-TB therapy (ATT).

Laparoscopic evaluation of women with CPP has been studied previously by various authors and is found to be an effective tool for diagnosing pelvic pathology [4,10–13]. However, TB as a distinct entity in the etiology of CPP and the role of laparoscopy in diagnosing genital TB as a cause of CPP have been studied inadequately.

Genital TB usually occurs following pulmonary (50% of cases) or extrapulmonary TB such as that of the lymph nodes (40% of cases), bones and joints (5% of cases), and urinary or gastrointestinal tract (5% of cases) [7]. Usually, genital TB may go unrecognized until it develops into a devastating pelvic pathology and this study attempted to explore TB presenting as CPP in the early stages. The prevalence of genital TB in women with CPP was 36% in this study. By contrast, one study from eastern Sudan reported a TB prevalence rate of 80% [12]. However, studies from India have reported a lower prevalence rate of genital TB in cases with CPP (1.2–5%) [1],[10],[13]. Coexisting genital TB may be present in 8–15% of women with pulmonary TB, and in this study 10/17 (59%) women had a previous history of TB and ATT. This higher prevalence in our study may be explained by our hospital catering to urban slums where a higher prevalence of TB is reported. In addition, a detailed search for genital TB was performed in this study using a combination of laparoscopy and microbiologic testing.

The demographic profile of our study patients indicated that a majority of women were in the reproductive age group with a mean age of 30.26±6.49 years and mean parity of 1.66±1.53, which is similar to the finding in other studies [5],[6]. The most common presentation of women was constant pain in the lower abdomen with dyspareunia in 56% and dysmenorrhea in 54%, which are similar to the figures reported by Sharma et al. [13]. Abnormal findings on laparoscopy were seen in 88%, whereas it was normal in 12%, and these are in accordance with previous studies [4],[14],[11].

Histological evidence of tubercular endometritis was low (6%), which is similar to other studies reported from India [7],[15]. Although histopathology is easy, quick, and inexpensive, bacilli are sparse in number and the sampled site may not represent the infected area. The lower rate may also reflect cyclic shedding of endometrium. Negative endometrial biopsy does not rule out genital TB because sampling errors are common and there may be tubercular lesions in the fallopian tubes, ovaries, and POD without associated tubercular endometritis.

In this study, in four women with pelvic inflammation, the peritoneal fluid was positive on TB-PCR, of which three were also AFB culture positive, suggesting that this may be considered an early sign of TB and also active disease. Although this finding needs to be corroborated by larger studies, it may be surmised here that in women with pelvic inflammation, fluid from the POD must be sampled for microbiological tests including TB-PCR and AFB culture. On analyzing the concordance of results between laparoscopy and specific diagnostic tests, only PCR showed a substantial agreement (kappa value = .716). Therefore, PCR should routinely be a part of algorithmic approach to CPP, especially if there is a high index of suspicion of genital TB. Moreover, most of the women diagnosed predominantly by PCR showed a significant improvement in pain scores after receiving ATT, thereby strengthening evidence.

Another important observation was that four women with laparoscopic features suggestive of TB did not have positive TB-PCR. This may be because these women were previously treated for extragenital TB. In addition, PCR failed to detect two cases that were positive by culture and histology. This may be attributed to the presence of inhibitors of PCR such as human blood in an amount large enough to compromise amplification of bacterial DNA.


Idiopathic CPP should be evaluated laparoscopically and genital TB should be considered as an important differential diagnosis in the etiology. In addition to the typical findings of TB, findings of pelvic inflammation on laparoscopy must be considered suspicious for TB in women with CPP and need to be corroborated with PCR of peritoneal fluid or washings and other microbiological tests.

 Conflicts of interest

All contributing authors declare no conflicts of interest.


1P. Latthe, M. Latthe, L. Say, et al, WHO systematic review of prevalence of chronic pelvic pain: a neglected reproductive health morbidity, BMC Public Health 6 (2006) 177.
2ACOG Committee on Practice, Bulletins—Gynecology, ACOG Practice Bulletin No. 51. Chronic pelvic pain, Obstet. Gynecol. 103 (2004) 589–605.
3T.A. Gelbaya, H.E. El-Halwagy, Focus on primary care: chronic pelvic pain in women, Obstet. Gynecol. Surv. 56 (2001) 757–764.
4M.B. Holloran-Schwartz, Surgical evaluation and treatment of the patient with chronic pelvic pain, Obstet. Gynecol. Clin. North Am. 41 (2014) 357–369.
5D.K. Gatongi, G. Gitau, V. Kay, et al, Female genital tuberculosis, Obstet. Gynecol. 7 (2005) 75–79.
6A.A. Ali, T.M. Abdallah, Clinical presentation and epidemiology of female genital TB in eastern Sudan, Int. J. Gynaecol. Obstet. 118 (2012) 236–238.
7R. Arora, J.B. Sharma, Female genital TB—A diagnostic and therapeutic challenge, Indian J. Tuberc. 61 (2014) 98–102.
8I.K. Neonakis, Z. Gitti, E. Krambovitis, et al, Molecular diagnostic tools in mycobacteriology, J. Microbiol. Methods 75 (2008) 1–11.
9J.B. Sharma, Tuberculosis and obstetric and gynecological practice, in: J. Studd, S.L. Tan, F.A. Chervenak (Eds.), Progress in Obstetrics and Gynaecology, vol. 18, Churchill Livingstone, Philadelphia, PA, 2008, pp. 395–427.
10S. Hebbar, C. Chawla, Role of laparoscopy in evaluation of chronic pelvic pain, J. Minim. Access Surg. 1 (2005) 116–120.
11P. Zúbor, N. Szunyogh, S. Galo, et al, Laparoscopy in chronic pelvic pain—A prospective clinical study, Ceska Gynekol. 70 (2005) 225–231 (in Slovak).
12R. Iftikhar, Outcome of laparoscopy in chronic pelvic pain, J. Surg. Pak. 13 (2008) 155–158.
13D. Sharma, K. Dahiya, N. Duhan, et al, Diagnostic laparoscopy in chronic pelvic pain, Arch. Gynecol. Obstet. 283 (2011) 295–297.
14M. Mára, Z. Fucíková, D. Kuzel, et al, Laparoscopy in chronic pelvic pain—A retrospective clinical study, Ceska Gynekol. 67(2002) 38–46 [in Czech].
15R.B. Thangappah, C.N. Paramasivan, S. Narayanan, Evaluating PCR, culture and histopathology in the diagnosis of female genital tuberculosis, Indian J. Med. Res. 134 (2011) 40–46.