Pregnancy outcomes after all modes of conception in patients with genital tuberculosis: a systematic review and meta-analysis
Article information
Abstract
This systematic review and meta-analysis aimed to summarize the pregnancy outcomes of women diagnosed with genital tuberculosis (GTB) who spontaneously conceived or underwent intrauterine insemination (IUI) or in vitro fertilization (IVF) after being treated with antitubercular therapy (ATT). Publications from the PubMed, Medline, Embase, Ovid, Scopus, Web of Science, and Google Scholar databases were searched from December 20, 2021 to March 5, 2022. The outcomes are presented as pooled averages with 95% confidence intervals. The inconsistency index (I2) test was used to measure the heterogeneity between studies. The certainty of the evidence was assessed using GRADEPro (https://www.gradepro.org/). Of the numerous articles identified, 33 met the inclusion criteria and were included in this systematic review. Generally, there was a significant increase in pregnancy rates among patients who underwent IVF compared with those who underwent ATT (37.9% vs. 23.8%; P=0.02). Conversely, there was no significant difference in pregnancy rates between patients who underwent IUI and those who conceived spontaneously (18.1% vs. 23.8%; P=0.65). In cases in which no abnormalities were found on hysterosalpingography or hysterolaparoscopy, pregnancy rates were comparable between spontaneous and IVF conceptions (48.4% vs. 49.2%). There were no significant differences in pregnancy or live birth rates between patients with GTB and those with other infertility factors undergoing IVF treatment (P>0.05). ATT, which is administered during the early stages of GTB is effective in achieving pregnancy outcomes comparable to IVF. However, in patients with advanced-stage disease, IVF is a superior treatment modality, resulting in increased pregnancy rates.
Introduction
Tuberculosis (TB) is a major global health concern. According to the World Health Organization (WHO), in 2020, TB affected approximately 10 million people worldwide, equivalent to 127 cases per 100,000 people. Most TB cases occurred in Southeast Asia (43%), Africa (25%), and the West Pacific (18%). Eight countries account for two-thirds of the total global burden, including India (26%), China (8.5%), Indonesia (8.4%), The Philippines (6.0%), Pakistan (5.8%), Nigeria (4.6%), Bangladesh (3.6%), and South Africa (3.3%) [1,2].
Female genital tuberculosis (FGTB) accounts for 9% of extrapulmonary TB cases and affects mainly women of reproductive age (15–45 years) [3]. The prevalence reported in infertility clinics ranges from 1% in the US, 2% in Italy, 4% in Saudi Arabia, 7% in Yemen, 17% in Nigeria, 2–20% in Pakistan, 6–21% in South Africa, and 3–26% in India. The condition is either asymptomatic or manifests as common gynecological symptoms, such as initial heavy menstrual flow followed by scant flow during menses, persistent pelvic or abdominal discomfort, and abnormal vaginal discharge. Most FGTB patients are underdiagnosed because of the lack of sensitive and efficient tests. Given the low sensitivity of diagnostic techniques and the challenges in obtaining specimens, a composite reference standard (CRS) is routinely used. WHO does not have any consensus guidelines for FGTB that offer recommendations for CRS definitions [4]. In most cases, patients are diagnosed once the damage incurred is significant and is frequently found during infertility workups [5–7]. FGTB patients often have a low success rate for pregnancy despite undergoing intensive therapy or surgery. To address this issue, assisted reproductive technology (ART), namely in vitro fertilization (IVF), is recommended [8].
Experience with tuberculosis-related infertility is often limited to countries where tuberculosis is endemic. Previous studies have reported varying pregnancy rates. Following antitubercular treatment, the spontaneous conception rate ranges from 0.6% to 68.3% [9,10]. The pregnancy rates ranged from 7.7% to 68.8% in FGTB IVF patients [11,12]. This meta-analysis investigated pregnancy outcomes in FGTB patients who underwent different methods of conception after antitubercular treatment (ATT): spontaneous conception, intrauterine insemination (IUI), and IVF. The pregnancy outcomes of FGTB patients who underwent IVF were compared with those of patients with other causes of infertility. We hope that this review will provide high-quality evidence to guide clinical care in this infertile population.
Methods
This meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis 13] reporting guidelines. The PubMed, Medline, Embase, Ovid, Scopus, Web of Science, and Google Scholar databases were used to collect publications from January 11, 2021 to March 5, 2022. The following search terms were used: pregnancy outcome, FGTB, antitubercular treatment, and IVF. Additional studies were identified by screening the reference lists. This meta-analysis analyzed pregnancy outcomes across studies, including pregnancy, miscarriage, live birth, oocyte retrieval, embryo transfer (ET), and implantation rates.
1. Inclusion and exclusion criteria
Studies were included if they reported pregnancy outcomes in patients with the following conditions: 1) proven GTB diagnosed through endoscopy, acid-fast bacilli culture, histopathology examination, or tuberculosis polymerase chain reaction; 2) age >18 years; 3) attended a fertility center and had proven infertility for at least 1 year; and 4) received antitubercular therapy (ATT) before the start of the pregnancy program. Studies were excluded if they reported patients with the following conditions: 1) indices of tuberculosis after undergoing or in the middle of the ART program; 2) not being treated with ATT before the start of the pregnancy program; 3) receiving an egg donor program; and 4) undergoing a surrogacy program. Studies that did not specify the number of participants were excluded. Reviews, systematic reviews, comments, editorial letters, case reports, and unpublished studies were excluded.
2. Data collection and analysis
Three authors (D.T., J.K.A., and K.T.) independently reviewed the title and abstract of each article. If the abstracts met the inclusion criteria, the full-text articles were thoroughly reviewed. The reference lists of the identified publications were screened for previously unidentified but relevant studies. The following information was retrieved: author, country, publication year, study type, patient inclusion and exclusion criteria, number of cohorts, number and criteria of the control group, treatments, reported outcomes, and other characteristics. The quality of the included studies was assessed using the JBI critical appraisal tools for case series and cohort studies [14].
3. Statistical analysis
Patient characteristics and pregnancy outcomes were analyzed using a pooled 95% confidence interval random effects model. The fixed effect or random effect Mantel-Haezel risk ratio (RR) was used to evaluate pregnancy outcomes between FGTB patients and those who underwent IVF for other infertility issues. Parameters with continuous data were analyzed using the mean difference.
The inconsistency index (I2) test, with a range of 0–100%, was used to measure the heterogeneity between studies. It quantifies the proportion of total variability observed across studies that can be attributed to heterogeneity. The formula for computing (I2) is as follows.
Where: I2=inconsistency index; df=degrees of freedom (number of analyzed studies-1); Q=Cochran’s heterogeneity statistic with the following calculation.
Where: k=number of studies; wi=weight assigned to each study; yi=effect size estimate (e.g., mean difference, odds ratio) of the i-th study; ̄=overall effect size estimate, the sum of the weighted means of effect sizes across studies.
This index ranged from 0% to 100%, with higher values indicating greater heterogeneity. Values greater than 50% or P-values less than 0.05 denote considerable heterogeneity. Subgroup analysis was performed when considerable heterogeneity was present [15].
For dichotomous data, the relative risk (RR) effect size was calculated as follows. 1) Small effect size=RR, 0.8 to 1.2. 2) Moderate effect size=RR, 0.5 to 0.8 or 1.2 to 2.0; and 3) large effect size=RR, <0.5 or > 2.0.
Cohen’s d or Hedges’ g analyses were performed to determine the continuous outcome effect size using the following criteria. 1) Small effect size=0.0 to <0.5. 2) Medium effect size=0.5 to <0.8. 3) Large effect size=0.8 to <1.4. And 4) very large effect size (>1.4).
Cohen’s d was calculated by multiplying the mean difference between the two groups by the pooled standard deviation of independent observations. When the sample sizes were unequal, Hedges’ g, which measures the effect size weighted by the relative size of the group, was used [16].
The certainty of the evidence was assessed using the GRADEPro web service (https://www.gradepro.org/). The assessment consisted of eight factors: study design (randomized controlled trial or observational study), inconsistency, indirectness, imprecision, publication bias, large effect size, plausible confounding factors, and dose-response gradient. Inconsistency was measured using I2, its P-value, and the direction of the studies [17]. The imprecision for continuous variables was graded as extremely serious, very serious, serious, or not serious if the study sample size was less than 30%, 40%, 50% or more than 50% of the optimum information size (minimum number of patients recruited to give a reliable value of α=0.05; β=0.20), respectively [18].
Results
The literature search yielded 955 studies, with an additional 10 studies identified through reference screening (Fig. 1). Among these, 53 full-text articles underwent review, ultimately resulting in the inclusion of 33 articles that met the predetermined inclusion criteria for the systematic review 9–12,19–47]. Supplementary Table 1 provides an overview of the characteristics of the reviewed full text articles.
Study flow diagram. TB, tuberculosis.
A meta-analysis was conducted on 16 studies, with four studies comparing pregnancy outcomes in GTB patients to other factors causing infertility through spontaneous conception 9,21,23,27] and 12 studies comparing outcomes via IVF modalities [9–12,23,32,35,36,38–40,43]. Additionally, six studies compared IVF outcomes to other modes of conception (spontaneous and IUI) among GTB patients 10,12,20,21,23,29]. Overall, 17 studies were included in the quantitative synthesis (meta-analysis).
1. Pooled analysis
1) Quality of the included studies
Fig. 2A, B present a quality assessment of the studies included in the qualitative synthesis. Most of the included studies were of good quality. However, one study failed to provide detailed information on patient inclusion criteria. Additionally, three studies did not report the method used to diagnose FGTB. Furthermore, the methods employed to identify conditions varied across the eight studies, with some patients undergoing laparoscopy, hysterosalpingography (HSG), culture, or polymerase chain reaction (PCR) exclusively. Nevertheless, the dropout rate was minimal across all the studies. The demographic characteristics of the participants were not described in most studies. Moreover, two studies did not specify the length of the study or follow-up period.
Risk of bias of the included studies. (A) Risk of bias graph of the included studies in the qualitative synthesis. a) Clear criteria for the inclusion. b) Method of measurement of the condition. c) Methods used for identification of the condition for all participants included in the case series. d) Drop-out rate. e) Demographic of the participants. f) Clear report of clinical information of the participants. g) Adequacy of follow-up time. (B) Risk of bias summary of the included studies in the qualitative synthesis. (C) Risk of bias graph of the included studies in the meta-analysis. a) Representativeness of the exposed cohort. b) Selection of the non-exposed cohort. c) Ascertainment of exposure. d) Demonstration that outcome of interest was not present at start of the study. e) Comparability of cohort on the basis of the design or analysis. f) Assessment of outcome. g) Length of follow-up time. h) Adequacy of follow-up cohort. (D) Risk of bias summary of the included studies in meta-analysis.
2) Characteristics of patients
① Spontaneous conception
Sixteen studies [9,10,19–30,46,47] collectively involved 1,750 GTB patients who underwent ATT treatments. The pooled average of the baseline characteristics of these patients is provided in Table 1, and detailed information for each study is available in Supplementary Table 2. The average age of the patients was 29.2 years (27.0–31.3). Primary infertility was the prevailing condition in most of the patients (74.3%). The duration of infertility ranged from one to 23 years, with an average of 4.3 years (3.0–5.7).
Baseline characteristics of GTB patients who received ATT treatments and who spontaneously conceived or underwent IVF-ET treatment in the included studies
Radiological assessments revealed diverse findings: normal findings in 12.1% (8.7–15.5%) of cases; tuberculous tubal etiology (definitive and probable) in 66.1% (57.5–74.7%) of cases, tubal blockage in 16.6% (13.7–19.5%) of cases, hydrosalpinx (some with coiled, kinked, beaded tube appearance) in 15.2% (11.0–19.4%) of cases, and extravasation of the fallopian tube in 4.1% (1.6–6.7%) of cases. Uterine abnormalities were detected in 14.0% (10.2–17.9%) of the patients. Endometrial involvement in the form of intrauterine synechia was reported in 2.4% (0.6–4.3%) of cases, whereas endometrial atrophy was observed in 0.8% (0.0–2.3%) of cases. Approximately 17.9% (12.6–23.2%) of patients presented with mild-to-moderate pelvic adhesions, whereas 5.4% (2.9–8.0%) were diagnosed with a frozen pelvis. Surgical treatment was not pursued by the majority of patients (98.7%).
The primary mode of positive diagnosis was laparoscopy (82.0%), followed by mycobacterium tuberculosis (M.tb) PCR (78.8%), ultrasound sonography (USG) (61.8%), HSG (57.4%), hysteroscopy (43.7%), histopathology (28.7%), and bacilli culture (15.6%). Only 8.1% of patients had a history of tuberculosis.
Patients received Isonidazid (INH) 300 mg+rifampicin (RMP) 450–600 mg+ethambutol (EMB) 800–1,200 mg+pyrazinamide (PZA) 1,200–1,500 mg drug combinations for 2 months, followed by INH 300 mg+RMP 450–600 mg for 4–10 months. The duration of ATT ranged from 6 to 12 months.
② IVF
Twenty-two studies, encompassing 1,275 GTB patients who received ATT and underwent IVF treatment, were analyzed 10–12,20,29,31–45] (Table 1). Detailed information on each of the included studies is provided in Supplementary Table 3. The average age of the patients ranged from 29.8 to 38.5 years, with an overall average of 31.6 years (30.9–32.4). Primary infertility was present in 83.1% of the patients. The average duration of infertility ranged from 3 to 11.2 years, with a mean of 5.4 years (4.6–6.1).
Radiographic examinations revealed normal findings in 7.5% of patients. Notably, three studies exclusively recruited patients with normal imaging findings [12,23,33] (Supplementary Table 3). Tuberculous tubal etiology was observed in 60.1% (49.3–70.9%) of the patients. Nine studies reported a tuberculous tubal etiology in all recruited patients 10,11,20,31,34,36,37,39,42]. Tubal blockage and hydrosalpinx (some with coiled, kinked, or beaded tube appearances) were found in 14.8% (9.5–20.0%) and 2.7% (0.3–5.0%) of the cases, respectively. Uterine abnormalities were observed in 9.2% (5.7–12.8%) of the patients. Intrauterine synechiae were found in 3.9% (0.0–6.2%) of all patients, whereas endometrial atrophy was reported only in 0.5% (0.0–1.1%) of the patients. Mild pelvic adhesion was observed in 1.2% (0.0–2.9%) of patients, with a frozen pelvis reported in only one study, with a prevalence of 13.9% [31].
Thirty percent (30.3%) of the patients underwent tubal surgery. Approximately 3.5% (0.7–6.4%), 2.0% (0.3–3.8%), and 3.5% (1.3–5.7%) of patients underwent salpingotomy, salpingectomy, and hysteroscopic adhesiolysis, respectively.
The diagnosis of GTB was confirmed based on endometrial biopsy histopathology in 89.8% of the patients, M.tb PCR in 86.4%, HSG in 71.2%, laparoscopy in 64.8%, hysteroscopy in 26.1%, and culture in 21.5%. Only 4.8% of patients had a history of tuberculosis.
Patients received ATT for a duration ranging from 3 to 24 months, with a regimen comprising INH 300–600 mg+RMP 450–600 mg+EMB INH 300–600 mg+RMP 450–600 mg+EMB 800–1,200 mg+PZA 1,200–1,500 mg for 2 months, followed by INH 450–600 mg+RMP 450–600 mg for 1 to 22 months. One study [32] administered a 3-month regimen to patients with PCR-positive latent TB.
3) Pregnancy outcomes
① Spontaneous conception
Approximately 23.8% (ranging from 15.0% to 32.6%) of GTB patients experienced spontaneous conception during or after the completion of ATT, as depicted in Table 2. Detailed information on pregnancy outcomes for each included study is provided in Supplementary Table 4. Among successful clinical pregnancies, miscarriages occurred in 23.8% (ranging from 8.1% to 39.9%), whereas ectopic pregnancies were noted in 9.2% (ranging from 0.0% to 24.8%), resulting in live birth deliveries in only 64.4% (ranging from 44.5% to 82.2%) of pregnancies. No instances of multiple pregnancies were observed.
Pregnancy outcome of GTB patients who received ATT treatments and spontaneously conceived or underwent IUI or IVF-ET treatment in the included studies
Significant heterogeneity in pregnancy outcomes was observed (I2>50%; P<0.05). We hypothesized that patient-specific conditions influence pregnancy outcomes. Therefore, we stratified pregnancy outcomes based on patient condition. Pooled analysis revealed the lowest pregnancy rate among patients with tubal or endometrial damage who were not treated with surgery (5.1%). Conversely, 29.3% of the patients with tubal or endometrial damage who underwent surgery were successfully pregnant. Moreover, 48.4% of the patients with normal laparoscopic/USG/HSG findings conceived. Miscarriage rates (36.8% vs. 15.8%) and the occurrence of ectopic pregnancy (6.0% vs. 2.3%) were higher among patients who experienced tubal or endometrial obstruction or damage than among those with normal findings. The incidence of live births was greater in patients with no or minor tubal or endometrial damage (81.9% vs. 59.4%).
② IUI
Only four studies [12,21,23,29] reported pregnancy outcomes after IUI as shown in Supplementary Table 5. The pregnancy rate per patient was 18.9% (6.4–31.5%) (Table 2). The heterogeneity of the data was significant (I2=67%; P=0.028).
③ IVF
Detailed information regarding pregnancy outcomes from each included study that reported on patients undergoing IVF is provided in Supplementary Table 6. The average dose of gonadotropin releasing hormone used for ovaries was 2,653.0 (ranging from 1,879.1 to 3,426.9) international unit (IU), and the average stimulation duration was 9.9 days (ranging from 8.7 to 11.1 days) (Table 2). The average endometrial thickness was considered normal, with an average of 10.3 mm (ranging from 8.6 to 12.0 mm). The mean baseline follicle-stimulating hormone (FSH) level was 7.4 (6.4–8.5) IU/mL. The average serum estradiol level on the human chorionic gonadotropin (hCG) day was 2,367.2 (1,923.1 to 2,811.3) pg/mL. The average number of oocytes retrieved per patient ranged from 3.4 to 13.8, with an average of 8.6. The average fertilization rate of the retrieved oocytes was 52.2% (ranging from 39.7% to 65.1%). The average number of embryos transferred per patient ranged from 1 to 5 (Supplementary Table 6). The average implantation rate per ET was 22.3% (ranging from 12.1% to 32.5%).
The pooled pregnancy rate per patient was 37.9% (ranging from 28.6% to 47.3%) (Table 2). Owing to the significant heterogeneity between the studies, the analysis was stratified based on imaging findings and surgical treatments. The highest pregnancy rate was achieved in patients with or without minor tubal or endometrial damage (49.2%; ranging from 29.0% to 69.4%), followed by patients with tubal damage treated with surgery (37.5%; ranging from 20.3% to 54.8%), patients with tubal damage untreated with surgery (35.6%; ranging from 10.3% to 59.9%), and patients with intrauterine synechia treated with adhesiolysis (19.7%; ranging from 4.2% to 35.2%).
The pregnancy rate was only 16.6% (range, 10.4–22.8%) per total embryo transferred. In patients with tubal involvement who underwent surgery, the pregnancy rate was 13.9% (range, 8.9–18.8%) per ET, whereas in those who did not undergo surgery, the pregnancy rate was 15.5% (range, 7.1–23.9%) per ET. The number of pregnancies per ET was 27.8% (ranging, 21.8–33.9%) in patients with or without minor tubal or endometrial damage.
The total miscarriage rate was 23.1% (13.1–33.0%) for successful clinical pregnancies, with 26.5% (12.7–40.4%) miscarriages in patients with tubal involvement treated with surgery, 39.3% (26.2–52.3%) in patients with tubal involvement who did not undergo surgery, and 12.2% (5.5–18.9%) in patients with minor tubal or endometrial involvement. Ectopic pregnancy occurred in 4.2% (1.2–7.1%) of the pregnancies. Approximately 67.4% (53.3–81.5%) of the pregnant women successfully delivered, with 43.9% (23.1–64.7%) live births from pregnant women with tubal damage treated with surgery, 38.0% (3.9–72.0%) live births from women with tubal damage not treated with surgery, and 42.3% (0.0–91.0%) from women with endometrial damage receiving adhesiolysis. Multiple pregnancies occurred in 10.9% (6.2–15.7%) of the pregnancies. There were 5.8% (1.4–10.3%) preterm live births.
2. Meta-analysis
1) Quality of the included studies
Fig. 2C, D show the quality assessment of the studies included in the quantitative synthesis (meta-analysis). Most of the included studies were of good quality and were characterized by robust methodologies. All studies demonstrated excellence in areas such as the representativeness of the exposed cohort, ascertainment of exposure, absence of the outcome of interest at the start of the study, and maintenance of a low dropout rate. However, one study identified a high risk of bias in the selection of the nonexposed cohort [22]. This study compared pregnancy outcomes between patients with positive and negative M.tb PCR results. Interestingly, 33.7% of the patients in the PCR-negative group presented with laparoscopic findings suggestive of GTB, whereas 15.4% of the PCR-positive patients presented with normal or alternative laparoscopic diagnoses. Additionally, >50% of the studies did not appropriately match the exposed and unexposed cohorts. Furthermore, one study lacked sufficient follow-up time, with only a 1-year follow-up period [12].
2) Comparison of spontaneous conception in GTB patients and other causes of infertility
Following standard ATT, patients with GTB demonstrated a significantly higher rate of spontaneous conception than individuals who experienced infertility due to other causes (RR, 1.67; 95% confidence intervals [CI], 1.16 to 2.38; P-value= 0.005) (Fig. 3A). Conversely, the rate of miscarriage between these two groups was comparable (RR, 1.43; 95% CI, 0.85 to 2.40; P-value=0.18) (Fig. 3B). Importantly, the quality of evidence for both outcomes was low. The assessment of the quality of evidence is shown in Supplementary Table 7.
Forest plot of pregnancy outcomes after spontaneous conception among patients diagnosed with GTB compared with patients who were infertile due to other causes. (A) Pregnancies per patients. (B) Miscarriages per pregnancies. GTB, genital tuberculosis; M-H, mantel-haenszel test; CI, confidence interval.
3) Comparison of IVF in GTB patients and other causes of infertility
① Baseline characteristics
Table 3 presents the baseline characteristics of patients underwent IVF treatment, distinguishing between those with GTB and those with other factors that contribute to infertility. Notably, patient age, basal FSH levels, duration of gonadotropin use, and gonadotropin dosage were comparable in both groups (Hedge’s g value <0.8), indicating that differences between the two groups can be considered negligible. However, patients with GTB exhibited a significantly longer duration of infertility by 0.8 years than those who were infertile due to other causes (Hedge’s g=2.268). Additionally, a greater proportion of GTB patients presented with primary infertility than those with other causes of infertility, with a calculated RR of 1.49.
Baseline characteristics of GTB patients compared to other factors of infertility who underwent ART treatment
② Pregnancy outcomes
Fig. 4 shows forest plots of pregnancy outcomes in GTB patients compared with pregnancy outcomes following IVF for other causes of infertility. Notably, there were no statistically significant differences observed across various parameters: the number of pregnancies per patient (P=0.53) (Fig. 4A), number of pregnancies per ET (P=0.40) (Fig. 4B), number of miscarriages per pregnancy (P=0.07) (Fig. 4C), number of live births per pregnancy (P=0.47) (Fig. 4D), average number of oocytes retrieved per cycle (P=0.77) (Fig. 4E), average number of embryos transferred per cycle (P=0.79) (Fig. 4F), and implantation rates per embryo transferred (P=0.97) (Fig. 4G).
Forest plot of pregnancy outcomes after IVF among patients diagnosed with GTB compared with patients who were infertile due to other causes. (A) Pregnancies per patients. (B) Pregnancies per cycles. (C) Miscarriages per pregnancies. (D) Live birth per pregnancies. (E) Average oocytes retrieved per cycles. (F) Average embryos transferred per cycles. (G) Implantation rates per embryos transferred. (H) Endometrial thickness. GTB, genital tuberculosis; M-H, mantel-haenszel test; CI, confidence interval; SD, standard deviation.
Stratification based on tubal or endometrial involvement was performed to determine the number of pregnancies. The meta-analysis revealed no significant difference in the number of pregnancies among patients with tubal damage, tubal and endometrial damage, or no damage compared with patients with other causes of infertility (P>0.05) (Fig. 4A, B). However, endometrial thickness was significantly lower in GTB patients than in those with other causes of infertility (mean difference, −0.52; 95% CI, −0.84 to −0.21; P=0.001) (Fig. 4H). Nevertheless, Hedge’s d analysis indicated a small effect size (0.269), suggesting that endometrial thickness was comparable between the groups.
Finally, it is important to consider that the quality of evidence for these findings was evaluated as very low to low, as indicated in Supplementary Table 7.
4) Comparison of IVF to other modes of conception in GTB patients
Fig. 5 shows forest plots illustrating the pregnancy rates in patients with GTB after different modes of conception. Analysis revealed that the pregnancy rate per patient was 5.75 times higher (RR, 5.75; 95% CI, 1.31 to 25.29; P=0.02) in patients receiving IVF treatment than in those treated with ATT alone (Fig. 5A). However, it is important to note that the quality of evidence supporting this finding was low (Supplementary Table 7).
Forest plot of the number of pregnancies observed among patients diagnosed with GTB following various modes of conception. (A) IVF vs. spontaneous conception after ATT. (B) IVF vs. IUI. (C) IUI vs. spontaneous conception after ATT. IVF, in vitro fertilization; M-H, mantel-haenszel test; CI, confidence interval; IUI, intrauterine insemination; GTB, genital tuberculosis; ATT, antitubercular therapy.
Furthermore, compared with that of IUI, the pregnancy rate was 2.29 times higher (RR, 2.29; 95% CI, 1.61 to 3.28) in IVF patients (Fig. 5B), with the quality of evidence rated as moderate. However, there was no significant difference in the pregnancy rate between patients receiving IUI and those treated with ATT alone (RR, 0.92; 95% CI, 0.65 to 1.31) (Fig. 5C). Notably, the quality of evidence supporting this comparison is very low.
The fallopian tube is the most common site of infection in genital tuberculosis patients
Genital tuberculosis is a neglected clinical problem and can easily be overlooked owing to nonspecific symptoms, chronic and cryptic protean clinical manifestations, and a lack of clinician awareness of the possibility of TB [6]. Pooled analysis revealed that the predilection site was the fallopian tube, accounting for approximately 60% of the reported cases (Table 1). Female genital tuberculosis manifests as nonspecific changes, such as tubal dilatation, tubal occlusion, irregular contour, diverticular outpouching (salpingitis isthmica nodosa), hydrosalpinx (with patterns including “cotton wool plug, pipestem tube, golf club tube, cobblestone tube, beaded tube, and leopard skin tube”), tubal occlusion, and adhesions in the peritubal region (which may present as a straight spill, a corkscrew appearance, and a peritubal halo). The endometrium was the second most common site (Table 1). Uterine involvement caused by tuberculosis might be viewed as particular features, such as a “collar-stud abscess”, a “T-shaped” uterus, or a “pseudounicornuate” uterus, or as nonspecific features such as synechiae, uterine contour distortion, obliteration of the uterine cavity, and venous and lymphatic intravasation 3]. Infection may cause ulcers in advanced stages and Asherman syndrome. When the peritoneum is involved, the manifestations include Fitz-Hugh-Curtis syndrome and a frozen pelvis [48]. In the included studies, the pooled analysis revealed 17.9%, 1.2%, and 5.4% mild, moderate, and frozen pelvic adhesions, respectively (Table 1). The reported prevalence was very low, and the heterogeneity was very high in our pooled analysis. This is due to differences in study designs, where some studies included all patients admitted to the obstetrics and gynecology department, some only recruited patients with normal findings or without uterine abnormalities, and others only recruited patients with proven tubal infertility.
Meta-analysis revealed that there were no significant differences in the ovarian reserve parameters, that is, basal FSH level, duration of gonadotropin, dosage of gonadotropin, estradiol on the day of hCG, number of oocytes retrieved, and mean number of embryos transferred between the patients with FGTB and those with other causes of infertility (Table 3, Fig. 4). This was expected because genital TB does not commonly cause extensive damage to the ovaries, allowing the follicles to remain intact and the ovaries to continue to ovulate. The lack of significant differences in the mean number of embryos transferred also indicated that there might not be significant differences in the quality of the oocytes collected or the embryos produced. This finding confirms the current understanding that genital TB infection is mainly localized in the endometrium and fallopian tubes and that genital TB-induced infertility is more likely associated with failure to fertilize rather than failure to ovulate.
Laparoscopy, HSG, and PCR are the most widely used methods for confirming GTB
GTB-positive diagnoses were mainly based on laparoscopic or HSG findings (57.2–82.0%). As previously stated, GTB induces particular alterations in the structure of affected organs; hence, imaging tools for typical structural diagnostics are critical. Laparoscopy can directly visualize the ovaries, fallopian tubes, and peritoneal cavity and perform biopsies. It may identify features specific to GTB, including fimbril phimosis, tubal beads, and peritubal and periovarian adhesions 49,50]. PCR is the second most commonly used tool for GTB diagnosis (positivity rates of 58.7% and 86.4% in patients treated with ATT alone and IVF, respectively). Although laparoscopy can detect gross changes commonly observed in chronic cases, the subtle clinical manifestations of GTB may be overlooked during the early stages of infection [49,51]. Finally, bacillus cultures may also be used to identify GTB infections 49,52,53].
ATT is efficient in treating the early stage of GTB, but in the advance stage, IVF is the superior modality
In cases where there were no abnormalities in the laparoscopic or HSG findings, pregnancy could be achieved through spontaneous conception, with comparable results to those of IVF (48.4% vs. 49.2%) (Table 2) [54]. Bilateral salpingectomies are frequently performed before IVF in women with hydrosalpinx to increase the chances of conception and live delivery [55,56]. When tubal blockage is caused by genital TB, laparoscopic tubal surgery to treat infertility is contraindicated owing to increased risks such as severe bleeding, intestinal damage, and flare-up of tuberculosis and peritonitis [57,58]. Bacillus reactivation has been reported during surgery and found during HSG, laparoscopy, hysteroscopy, and laparotomy [54]. Persistent inflammation may lower pregnancy outcomes. Tubal corrective surgery may help achieve a higher pregnancy rate through spontaneous conception (29.3% vs. 5.1% in patients treated with surgery vs. those not treated with surgery, respectively). However, in IVF patients, tubal surgery might not be necessary. As shown in Table 5, comparable pregnancy rates were achieved in patients not treated with surgery (35.6% vs. 37.5% per patient and 13.9% vs. 16.6% per ET). However, a higher pregnancy loss rate was noted in patients who did not undergo tubal surgery (39.3%) than in those who underwent tubal surgery (26.5%). This result is in line with a meta-analysis by Harb et al. [59], where hydrosalpinx increased the risk of pregnancy loss, and surgical treatment may reduce this risk.
Pooled analysis and meta-analysis revealed higher pregnancy rates in patients who underwent IVF than in those receiving ATT only (37.9% vs. 23.8%; RR, 5.75; 95% CI, 1.31 to 25.29; P=0.02). The pregnancy rate did not differ between patients treated with IUI and those treated with ATT alone (18.1% vs. 23.8%; RR, 0.92; 95% CI, 0.65 to 1.31; P=0.65). These findings emphasize that IVF is the optimal method for the treatment of female infertility associated with advanced genital tuberculosis, as it bypasses the need for tubal fertilization, which in most cases is damaged by M.tb infection.
Based on our findings, we propose the following algorithm for the management of patients suspected of having GTB (Fig. 6). Patients presenting with infertility should undergo a thorough evaluation, including a detailed medical history, physical examination, assessment of prior TB exposure, and symptomatology. Subsequently, patients should undergo infertility workups, including USG, laparoscopy, and HSG.
Algorithm delineating the management approach for patients presenting with infertility and diagnosed with genital tuberculosis (GTB). TB, tuberculosis; USG, ultrasound sonography; HSG, hysterosalpingography; M.tb PCR, mycobacterium tuberculosis polymerase chain reaction; ATT, antitubercular therapy; IVF, in vitro fertilization.
Patients with definitive or suggestive GTB findings should immediately receive ATT for a minimum duration of 6 months. Conversely, patients with normal imaging results undergo confirmation of GTB through M.tb PCR testing, and those with positive results should initiate ATT accordingly. Upon completion of ATT, patients should be reassessed using USG, HSG, or laparoscopy.
Patients exhibiting normal imaging findings after ATT should be encouraged to pursue spontaneous conception. When spontaneous conception fails, patients should be offered IVF. Furthermore, patients presenting with tubal blockage, intrauterine synechia, or pelvic adhesions should be considered for corrective surgical intervention followed by IVF to facilitate pregnancy.
Limitations
This meta-analysis has several limitations. First, it was based on a very limited number of studies, which may have affected the statistical power. Second, high interstudy heterogeneity was noted owing to marked differences in patient inclusion criteria. A disadvantage of this study is that we were unable to analyze the recovery of fertility through tuberculosis treatment separately because of insufficient analysis of various causes of infertility. Because of the quality of the evidence, our results should be interpreted with caution.
Conclusion
Our analysis led us to conclude that in the early stages of GTB, characterized by the absence of noticeable damage to reproductive organs, ATT alone suffices to pursue pregnancy without the need for ART interventions. However, in cases of advanced-stage disease with substantial tubal pathologies, IVF has been demonstrated to be a superior modality for achieving higher pregnancy rates.
Moreover, early screening for latent or dormant FGTB through techniques such as endometrial PCR or histopathological testing may prove beneficial. Such screening measures can potentially improve pregnancy outcomes and prevent disease progression.
Supplementary Information
Notes
Conflict of interest
The authors have no conflict of interest.
Ethical approval
Not applicable.
Patient consent
Not applicable.
Funding information
The authors declare that (s)he has no relevant or material financial interests that relate to the research described in this paper. The authors received no financial support for the research, authorship, or publication of this article.