Detection of genital tuberculosis among women with infertility using best clinical practices in India: An implementation study

Abstract

Background:

Diagnosis of genital tuberculosis (TB) as a cause of infertility still remains a diagnostic dilemma for clinicians, as no standard guidelines exist. The recently proposed best practices for genital TB diagnosis have not been evaluated yet in India.

Objectives:

To implement best practices to diagnose and treat likely genital TB as a cause of infertility.

Methods:

Between April 2016 and June 2018, consenting women seen at a tertiary hospital infertility clinic were assessed by thorough TB related clinical history, ultrasonography, tuberculin skin test (TST), and ESR. Those with suspected genital TB underwent laparohysteroscopy. Clinical and laboratory characteristics were compared between likely (microbiologically confirmed or probable TB) and unlikely (possible and no genital TB) genital TB. Fertility outcome was assessed among women initiated on anti-TB treatment (ATT).

Results:

Of 185 women seeking infertility care, likely genital TB was identified among 29 (15.7%) women, with 6 (21%) confirmed and 23 (79%) probable genital TB. Compared to unlikely genital TB cases, the likely genital TB group were found to have past history of TB (p < 0.001); positive TST (p = 0.002) and elevated ESR (p = 0.001). Among the likely genital TB group, all 6 confirmed genital TB were started on ATT and 2 (33.3%) conceived. Of 5 probable genital TB started on ATT, 3 (60%) conceived.

Conclusion:

Approximately 1/6th of women seeking infertility care met the criteria for likely genital TB. Conception among over-half of treated probable genital TB cases provides preliminary evidence that best clinical practices can be utilized, but needs further confirmatory studies.

1. Introduction

Female infertility is rising globally, with variation in etiology between high-income (HIC) and low-and middle-income countries (LMIC).¹ While anovulation is common in HICs, infectious etiology remains prevalent in LMICs where tuboperitoneal damage affects fertility.² Genitourinary tuberculosis (TB) is a common form of extra pulmonary TB (EPTB) worldwide, accounting for 9% of EPTB.³ Genital TB is a known cause of infertility in women where endometrial damage, tubal obstruction, ovarian and cervical affection can lead to infertility.^4,5 In the infertility clinics worldwide, an estimated 5% of the women presenting with infertility have genital TB, with prevalence rates ranging from less than 1% in HICs to as high as 3–26% in LMICs.^6,7,8,9 In India, reported annual EPTB burden was 20–25%of which 4% of EPTB cases were reported to be urogenital TB (urinary tract and genital TB).^10,11 Prior studies have attempted to define genital TB cases using individual criterion, i.e. microbiological, ultrasound (USG) and laparohysteroscopy. However, a comprehensive definition of probable or possible genital TB using all the criteria: clinical presentation, USG, laboratory and laparohysteroscopy findings, are not available.^4,13 Since genital TB is a paucibacillary disease, the yield of the newer diagnostics such as GeneXpert, known to have high diagnostic sensitivity and specificity, may still be lower. This poses a diagnostic challenge to the clinicians.^2,4,14 Therefore, similar to other EPTBs, most genital TB would likely be diagnosed based on clinical, radiologic and histopathologic findings.¹⁴ Importantly, there is a lack of standard guidelines for diagnosis of genital TB. Recent Revised National Tuberculosis Control Program guidelines recommend treatment for laboratory confirmed genital TB cases.¹² However, these guidelines do not specify any algorithm which can be used as best practice for diagnosing and treating clinical (probable) genital TB.

Recently, an algorithmic best practice approach was proposed to diagnose genital TB among infertile women.¹⁶ However, it has not yet been evaluated in India. Employing best clinical practices in diagnosing and treating genital TB may help optimize fertility rate.^1,4 Therefore, we aimed to evaluate this approach to identify likely genital TB (confirmed and probable), and unlikely genital TB (possible and no TB) among female infertility cases accessing care at a tertiary level teaching hospital in India, a country with the world’s largest TB burden.¹⁶

2. Methodology

2.1. Study design

A prospective cross-sectional study was conducted between April 2016 and June 2018 among patients visiting the weekly infertility clinic at Byramjee-Jeejeebhoy Government Medical College and Sassoon General Hospital, Pune, India, a tertiary referral center. Consenting female infertility patients between 18 and 40 years of age were eligible and enrolled in the study. Women with male partner infertility and congenital anatomical abnormalities were excluded.

After enrolment, clinical and socio-demographic characteristics, including an in-depth medical history including history of smoking, alcohol, diabetes and HIV were recorded in the case report forms designed for the study. Standard of care clinical history including menstrual history such as irregular menstrual cycles, flow, and duration of infertility, prior pregnancy outcomes and prior infertility treatments was recorded as per routine practices. Clinical examination like per abdominal, per speculum and per vaginal examinations and USG were performed as part of standard of care.

Based on the extensive literature review, we derived the best practices to identify confirmed and probable cases of genital TB as cause of infertility as shown in Fig. 1. Our best clinical practices included screening for TB symptoms, obtaining history of prior TB and recent TB contacts, erythrocyte sedimentation rates (ESR) and tuberculin skin test (TST). Those with positive TST (>10 mm induration) and/or ESR (>20) underwent further investigations such as ultrasound if not already done. In addition, endometrial TB culture or PCR and laparohysteroscopy were performed in suspected cases. At laparohysteroscopy, the uterine cavity, uterine surface, tubes, ovaries, ovarian fossa, utero sacral ligaments and peritoneal findings were assessed and tissue samples were taken if any abnormality was detected or genital TB was suspected, for both microbiological and/or histopathological diagnosis. Using this approach, genital TB case definition (confirmed, probable) was established cumulatively from clinical history, radiology, microbiology and/or laparohysteroscopy. Those categorized as confirmed and probable genital TB were referred for anti-TB treatment (ATT) initiation.

Fig. 1 –

Figure showing the best clinical practices to diagnose genital tuberculosis among women seeking care for infertility. Note: POD: Pouch of Douglas. TST: Tuberculin Skin Test. ESR:Erythrocyte Sedimentation Rate. USG: Ultra Sonography. H/o: History of. +/−: Positive or negative. CXR: Chest X-ray. HPE: Histopathological Examination. PCR: Polymerase Chain Reaction. MTB: Mycobacterium Tuberculosis.

2.2. Study definitions

Primary infertility was defined as those women who are unable to conceive after one year of unprotected intercourse.¹⁷ Secondary infertility was defined as inability to conceive after previous conception irrespective of the obstetric outcome. Genital TB was categorized as likely and unlikely TB. Likely TB was defined as microbiologically confirmed and probable TB. Unlikely genital TB was defined as possible or no TB (Fig. 1).

2.3. Ethical considerations

The study was conducted in accordance with internationally recognized standards for ethical research and the International Conference on Harmonization (ICH E6). The protocol was reviewed and approved by the Independent Ethics Committee (IEC) of BJGMC and Institutional Review Board (IRB) of Johns Hopkins University, respectively. Written informed consent was obtained for each participant prior to entry into the study.

2.4. Statistical analysis

Data were analyzed using Stata version 14.2 (Stata Corp). Continuous variables were compared across likely and unlikely genital TB groups using Mann-Whitney (Rank-sum) test, and categorical variables were compared using Fisher’s exact test. P values < 0.05 were considered significant.

3. Results

Of 246 infertile women accessing care during the study period, 185 were eligible and enrolled. Overall, the median age was 26 years (interquartile range (IQR), 24–30) and the median body mass index (BMI) was 22.96 (IQR, 21.5–25.1) (Table 1). Primary infertility was found in 155 (84%) women and secondary in 30 (16%) women (Table 1). Using the best clinical practices, 29 (15.7%; 95% CI: 11%–22%) met the definition of likely genital TB and 156 (84.3%; 95% CI: 78%–89%) met the definition of unlikely genital TB (Fig. 2). Of the 29 likely genital TB cases, 26 (89.7%) were primary infertility cases and 3 (10.3%) were secondary infertility. All 6 confirmed cases had primary genital TB and all 3 secondary infertility cases were probable genital TB. Of the 29 with likely TB, 6 (20.7%) had microbiologically confirmed TB and 23 (79.3%) had probable TB. Of the 156 unlikely genital TB group, 12 (7.7%) had possible TB and 144 (92.3%) had no TB. Overall, 14 (8%) reported a past history of TB.

Table 1 –

Characteristics of women seeking care for infertility by likely and unlikely genital tuberculosis in Pune, India.

Variable	Overall (N = 185)	Likely Genital TB (N = 29)	Unlikely Genital TB (N = 156)	p-value
Age, median (IQR**)	26 (24–30)	28 (25–30)	26 (24–29)	0.05*
Education, n (%)
≥Diploma/intermediate	5 (2%)	2 (7%)	3 (2%)
High-school	65 (35%)	5 (17%)	60 (39%)	0.03*
≤Middle school	114 (62%)	22 (76%)	92 (59%)	–
Occupation, n (%)
≥Semi-skilled worker	50 (19%)	7 (24%)	43 (28%)
Unskilled worker	56 (30%)	8 (28%)	48 (31%)	–
Unemployed/housewife	79 (43%)	14 (48%)	65 (42%)	0.85
MKS*** Scale, n (%)
Upper	2 (1%)	1 (3%)	1 (1%)	0.26
Upper middle	4 (2%)	0	4 (3%)
Lower middle	68 (37%)	8 (28%)	60 (38%)
Upper lower	111 (60%)	20 (69%)	91 (58%)
Duration of marriage (IQR)	5 (3–8)	7 (4–10)	5 (3–7)	0.03*
BMI+, median (IQR) Scale	22.9 (21.5–25.0)	22.2 (20.0–23.8)	23.2 (21.6–25.1)	0.02*
Underweight (<18.5)	8 (4%)	1 (3%)	7 (5%)	0.05*
Normal (18.5–24.9)	125 (68%)	25 (86%)	100 (65%)
Overweight (>24.9)	51 (28%)	3 (10%)	48 (31%)
BCG†† scar, n (%)
Yes	155 (99%)	29 (100%)	155 (99%)	>0.95
No	1 (1%)	0	1 (1%)
Infertility, n (%)
Primary	155 (86%)	24 (86%)	131 (86%)	≥0.95
Secondary	25 (14%)	4 (14%)	21 (14%)
Prior t/reatment for infertility, n (%)
No	149 (80%)	19 (66%)	130 (83%)	0.04*
Yes	36 (20%)	10 (34%)	26 (17%)
Past H/o†††TB, n (%)
No	171 (92%)	20 (69%)	151 (97%)	–
Yes	14 (8%)	9 (31%)	5 (3%)	p < 0.001*
Pelvic Pain, n (%)
No	182 (98%)	28 (97%)	154 (99%)	0.40
Yes	3 (1%)	1 (3%)	2 (1%)
Per Vaginal: Uterus size, n (%)
Bulky	2 (1%)	1 (3%)	1 (1%)	–
Normal	153 (98%)	28 (97%)	153 (98%)	0.5
Small	2 (1%)	0	2 (1%)	–
Mobility, n (%)
Normal	109 (59%)	16 (57%)	93 (60%)	–
Present	61 (39%)	10 (36%)	61 (39%)	0.17
Restricted	4 (1%)	2 (7%)	2 (1%)
Adnexal mass, n (%)
Present	4 (1%)	2 (7%)	2 (1%)	–
Absent	181 (98%)	27 (93%)	154 (99%)	0.12
ESR****, Median (IQR)
≤20	129 (70%)	12 (41%)	117 (75%)	–
>20	56 (30%)	17 (59%)	39 (25%)	0.001*
TST††††, Median (IQR)
≤10	108 (58%)	9 (31%)	99 (63%)	–
>10	77 (42%)	20 (69%)	57 (37%)	0.002*
Evidence of PCOS*****, n (%)
No	127 (70%)	24 (86%)	103 (67%)	0.05*
Yes	55 (30%)	4 (14%)	51 (33%)
Diagnostic Hysterolaparoscopy (n = 22)¥	22 (12%)	15 (68%)	7 (32%)	p < 0.001*
TB PCR Positive (n = 13)¥	13 (7%)	6 (46%)	7 (54%)	>0.95

^*Statistically Significant;

^**IQR: Interquartile Range;

^***MKS: Modified Kuppuswami Score Socioeconomic status in India. MKS class I-Upper, class II - Upper middle, III -Middle/lower middle, IV- Lower/upper lower),V (Lower);

⁺BMI:Body Mass Index;

^††BCG: Bacillus Calmette–Guérin;

^†††H/O:History of;

^¥for specific invasive procedures, the percentages are based on total number of testing done shown in the variable column.

^****ESR: Erythrocyte Sedimentation Rate;

^††††TST: Tuberculin Skin Test;

^*****PCOS: Polycystic Ovarian Syndrome.

Fig. 2 –

A flow-diagram showing women with likely and unlikely genital tuberculosis and their pregnancy outcomes. Note. PCOD: Polycystic Ovarian Disease. PID: Pelvic Inflammatory Disease. ATT: Anti-Tuberculosis Treatment.

Demographic and clinical characteristics were comparable between the likely and unlikely genital TB groups with a few exceptions (Table 1). Those with likely genital TB were more likely to be older (28 vs. 26 years, p = 0.05), less educated (p = 0.03), lower BMI (22.21 vs. 23.23 kg/m², p = 0.02), past history of TB (64% vs. 36%, p < 0.001), and past history of infertility treatments (34% vs. 17%, p = 0.04). In addition, those with likely genital TB were more likely to have >20 mm of ESR (p = 0.001), and positive TST (defined as >10 mm induration, p = 0.002). Furthermore, likely genital TB group were subjected to laparohysteroscopy more often (75% vs. 5%, p < 0.001).

The laparohysteroscopy findings among likely genital TB group is shown in Table 2. Peri-tubal and tubo-ovarian adhesions were seen in 10 (66.7%), hydrosalphinx in 4 (26.7%), tubal block in 3 (20%), and other tubal abnormalities in 3 (20%) women (Figs. 3 and 4). Tubercles on the fallopian tube was seen in one woman. Hysteroscopy revealed normal endometrium in 12 (80%) and pale oligemic endometrium in 3 (20%) women (Table 2).

Table 2 –

Laparohysteroscopy findings among women with confirmed and probable genital tuberculosis presenting with infertility.

	Total n (%)	Confirmed n (%)	Probable n (%)
Total likely GTB	29	6 (20.7)	23 (79.3)
Laparohysteroscopy	15 (51.7)	6 (40)	9 (60)
Laparoscopy findings
Adhesions	10 (66.7)	4 (40)	6 (60)
Hydrosalphinx	4 (26.7)	3 (75)	1 (25)
Tubal block	3 (20)	2 (66.7)	1 (33.3)
Tuboovarian mass	2 (13.3)	1 (50)	1 (50)
Other tubal abnormalities	3 (20)	1 (33.3)	2 (66.7)
Hysteroscopy findings
Normal	12 (80)	6 (50)	6 (50)
Abnormal (Pale endometrium)	3 (20)	0	3 (100)

Fig. 3 –

A laparoscopy picture showing tubercles in woman with likely genital tuberculosis.

Fig. 4 –

A laparoscopy picture showing tubo-ovarian mass in woman with likely genital tuberculosis.

Of the likely genital TB group, 2 (33.3%) of the 6 confirmed genital TB cases who initiated and completed ATT conceived (Fig. 2). Among 23 probable genital TB cases, 5 (21.7%) were initiated on ATT and 3 (60%) conceived. Of 18 patients who did not initiate ATT, 2 (11%) conceived and 16 (89%) did not. All those who conceived had no obstetric complications and delivered healthy babies at term.

In unlikely GTB group, the possible TB patients received routine care for infertility treatment. They were counselled regarding symptomatology of genital TB especially menstrual complaints and TB related symptoms. They were requested to follow up if they had any such complaint for further evaluation.

4. Discussion

Diagnostic delay of genital TB as a cause of infertility can lead to severe irreversible organ damage with poor fertility outcomes.^17,18 Furthermore, assisted reproductive techniques may become the only option to have a biological child, which is very costly and burdens the health system.^19,20 Our study that employed best clinical practices to diagnose genital TB identified a 16% prevalence of likely genital TB among women reporting infertility. Importantly, though the numbers were very small, 3 out of 5 women with probable genital TB conceived after starting treatment.

The prevalence of genital TB found in our study is consistent with other studies in India but much higher than from the United States (<1%).²¹ Similar to our findings, a study by Sharma reported 3–16% of genital TB among infertile patients in India, but extensive investigations were not used.⁹ In contrast, a North Indian study reported a very high rate of genital TB (48.5%) among infertile women.²² A nationwide survey by Indian Council of Medical Research reported an increasing prevalence of genital TB from 19% to 30% between 2011 and 2015^23,24 but noted region wise differences as well as a lack of standardized approach to diagnosing genital TB.^23,24

Importantly, our study showed that 1/3rd of confirmed genital TB participants initiated on ATT conceived. Interestingly, though small numbers, much higher proportion of conception (60%) was found among those with probable genital TB started on ATT. A prospective study reported 19.2% conception rate among women treated for infertility with ATT, with a much lower live birth rate (7.2%),⁸ in contrast to the 24.1% overall conception rate and 100% live births observed in our likely genital TB group. Using assisted reproductive technique after completion of ATT could further improve the conception rate in selected patients without endometrial damage.²⁵ However in our cohort, an assisted reproduction technique was unavailable in public sector tertiary level hospital. The observed improvement in fertility outcome could be related to successful TB treatment.

With India leading the world in the absolute burden of TB, our evaluation of best clinical practices to diagnose genital TB among infertile patient is timely and will provide guidance to clinicians on how to diagnose genital TB. Challenges to diagnose genital TB as an etiology of infertility exist due to several reasons. First, genital TB does not present with classic TB symptomatology. Furthermore, there is considerable overlap between presenting symptoms of genital TB and other infertility causes like pelvic inflammatory diseases, ovarian cyst, posing diagnostic dilemmas. In fact, our study found that 12% with likely genital TB had other gynaecological conditions, confusing the clinical picture. Second, the yield of microbiologic investigations has known to be suboptimal, even with Gene Xpert, as presence of blood in genitourinary biopsy specimens may interfere with GeneXpert tests. Finally, there are no standard approach to diagnose genital TB, which prompted our group to use the best practices to diagnose genital TB among infertile women.^16,24,26

Our study has a few limitations. We did not have a comparative group. Our approach utilized the best tools and practices available at a public tertiary hospital. Furthermore, we did not uniformly apply some invasive components of the diagnostic approach to all infertile women, but used the clinical judgment of treating clinicians to avoid unnecessary procedures for those unlikely to have genital TB or whenever alternate diagnoses were available. Genital TB being a paucibacillary disease, the decision about treatment initiation for probable TB cases was as per the treating physician, based on the combined evidence for TB in the form of abnormal clinical findings, supportive laboratory evidence like raised ESR, raised TST, as well as abnormalities seen during laparohysteroscopy (suggestive of genital TB). Our sample size was small to provide definitive evidence for fertility rate among likely genital TB group, but provides preliminary evidence for future studies. Despite these limitations, our study provides important clinical, easily adoptable best practices to diagnose and treat genital TB, a likely treatable cause of infertility.

In summary, our study demonstrates that genital TB should be considered as a cause of infertility in high TB burden settings. Importantly, the best clinical practices helped identify probable genital TB, suggesting that the proposed best clinical practice can be used in high TB burden settings as ours in India.¹⁵ Larger studies that are powered to confirm our observations, and standard application of diagnostic tools, are needed to confirm our findings.

Abbreviations:

TB

Tuberculosis

ATT

anti-TB treatment

HIC

high-income

LMIC

low-and middle-income countries

EPTB

extra pulmonary TB

BCG

Bacille Calmette-Guerin

BJGMC

Byramjee-Jeejeebhoy Government Medical College

IGRA

interferon gamma release assay

IQR

interquartile range

TST

Tuberculin skin test

ESR

erythrocyte sedimentation rates

USG

ultrasound

RNTCP

Revised National Tuberculosis Control Program

PCR

Polymerase Chain Reaction

ICH

International Conference on Harmonization (ICH E6)

IEC

Independent Ethics Committee

IRB

Institutional Review Board

BMI

body mass index

MKS

Modified Kuppuswami Score

H/O

History of

PCOS

Polycystic Ovarian Disease