In vitro fertilization outcomes in previously cured tuberculosis patients: a retrospective study.

This study contributes clinically relevant evidence to fertility counseling and clinical decision-making regarding assisted reproductive technologies in women with a treated TB history. It helps clinicians better assess and manage reproductive treatment strategies in that population.

A retrospective study was conducted, involving total 35,428 female infertile patients who underwent IVF–ET at the Assisted Reproduction Center of Northwest Women’s and Children’s Hospital affiliated with Xi’an Jiaotong University from January 1, 2015, to September 30, 2022. The inclusion criteria included infertile women with a previously confirmed and fully treated TB history (pulmonary TB, pelvic TB, and other TB types) who underwent IVF or ICSI (intracytoplasmic sperm injection) protocols. Each patient was confirmed to have achieved clinical cure for TB from a specialized hospital. The exclusion criteria comprised active TB requiring or currently undergoing treatment, any form of untreated non-active TB (including latent TB), oocyte retrieval performed before complete TB resolution, intrauterine adhesion, thyroid dysfunction, acute severe systemic disease, and cancer. As part of routine pre-ART evaluation at our institution, PPD testing was performed for all patients, regardless of TB history, and was not used as an inclusion or exclusion criterion. In routine clinical practice, positive PPD results were followed by T-SPOT testing for further assessment. PPD and T-SPOT data were nearly complete. Only 12 patients without a TB history lacked PPD results. These missing data were not imputed, as PPD and T-SPOT were used solely for descriptive purposes and did not affect study eligibility or primary analyses. Following these criteria, 1386 cured TB patients were selected as the final TB group, and 33,951 patients without TB history served as the control group. The TB group was defined based on fresh embryo transfer cycles. Frozen–thaw cycles were not used to define the TB group but were derived from this cohort and included solely for secondary analyses. A subset of these patients subsequently underwent frozen–thaw cycles (1974 cases), which were included in a secondary comparative analysis with fresh cycles (Fig.  1 ). This study was approved by the Ethics Review Board of the Northwest Women’s and Children’s Hospital, Xi’an, China (No. 2023003). The written informed consents were obtained from all patients. Fig. 1 Flow chart of the study design. Patients were selected according to the inclusion and exclusion criteria Flow chart of the study design. Patients were selected according to the inclusion and exclusion criteria All patients accepted either an IVF or an ICSI protocol. Typically, fertilized embryos were transferred to the uterus following a culture period of 2–5 days, with surplus embryos undergoing cryopreservation for future utilization [ 15 ]. Serum human chorionic gonadotropin (hCG) levels were measured 14 day post-transfer. If positive, ultrasound would be performed about 28 day and 35 day post-transfer to ensure clinical pregnancy, respectively. The reproductive laboratory outcomes comprised fertilization rate, high-quality embryo rate, and blastocyst rate. The clinical outcomes included implantation rate, pregnancy rate, multiple pregnancy rate, miscarriage rate, ectopic pregnancy rate, live birth rate, birth weight of the baby, and pregnancy weight gain. The pregnancy rate and live birth rate were set as primary outcomes. Pregnancy confirmation was achieved through ultrasound visualization of fetal heart and bud. Live birth was defined as the delivery of at least one live baby after 28 weeks of gestation. Secondary outcomes encompassed reproductive laboratory outcomes, clinical outcomes except for pregnancy rate and live birth rate, and neonatal details. The analysis of data was processed by SPSS 25.0 statistical software. Baseline characteristics were assessed using t test. Categorical variables such as infertility types and causes were analyzed using chi-square tests. Given that some patients underwent multiple cycles, the pregnancy-related outcomes were analyzed by generalized estimating equations (GEE). P  < 0.05 was considered statistically significant.

The baseline characteristics of 1386 cured TB and 33,951 non-TB infertile patients using fresh cycles are summarized in Table  1 . The differences in age and BMI between TB and controls had no statistical significance ( P  > 0.05). The duration of infertility was longer in TB patients compared to controls ( P  = 0.006). Table 1 Baseline characteristics of TB and control patients Characteristics Female TB ( n  = 1386) Controls ( n  = 33,951) P Age (years) 31.07 ± 3.94 31.25 ± 4.41 0.099 BMI(kg/m 2) 22.03 ± 2.92 23.08 ± 3.43 0.237 Duration of infertility (years) 3.73 ± 2.78 3.52 ± 2.49 0.006 Sterility type (%)  < 0.001  Primary 790 (57.00) 17,058 (50.24)  Secondary 596 (43.00) 16,853 (49.64) Causes of infertility (%)  < 0.001  Female factors   Pelvic and fallopian tube factor 945 (68.18) 14,006 (41.25)   Ovulatory obstacle 41 (2.96) 2513 (7.40)   Endometriosis 7 (0.51) 352 (1.04)   Diminished ovarian reserve 11 (0.79) 336 (0.99)   Other female factors 159 (11.47) 5008 (14.75)  Male factors   Oligo-astheno-teratozoospermia 36 (2.60) 2999 (8.83)   Azoospermia 12 (0.87) 1499 (4.42)   Other male factors 54 (3.90) 2612 (7.69)  Mixed female and male factors 121 (8.73) 4626 (13.63) Pregnancy assistance program (%)  < 0.001  IVF 1184 (85.43) 25,045 (73.77)  ICSI 202 (14.57) 8906 (26.23) Baseline hormone  E2 102.23 ± 438.38 122.79 ± 548.36 0.092  LH 5.17 ± 5.29 5.41 ± 21.00 0.663  FSH 7.42 ± 2.88 7.87 ± 54.00 0.757  P 1.45 ± 3.86 1.25 ± 6.60 0.84  T 32.20 ± 22.83 33.52 ± 48.14 0.293  PRL 23.42 ± 55.90 27.72 ± 520.28 0.76 Antral follicle ( n ) 11.89 ± 5.36 12.17 ± 5.92 0.033 COH protocol (%)  < 0.001  Natural cycle 6 (0.43) 129 (0.38)  Long protocol 558 (40.26) 13,086 (38.54)  Short protocol 0 (0.00) 54 (0.16)  Ultralong protocol 353 (25.47) 9482 (27.93)  Ultrashort protocol 49 (3.54) 1743 (5.13)  Antagonist protocol 416 (30.01) 9379 (27.63)  Mini-stimulation protocol 4 (0.29) 78 (0.23) Semen source (%)  < 0.001  Fresh semen 1329 (95.89) 30,149 (88.80)  Frozen semen 28 (2.02) 2075 (6.11)  Surgery 29 (2.09) 1727 (5.09) Endometrial thickness on day of hCG (mm) 11.33 ± 2.29 11.66 ± 2.36  < 0.001 Male TB (%) 12 (0.87) 123 (0.36) 0.003 Number of embryo transferred ( n )  IVF 1.35 ± 0.48 1.25 ± 0.44  < 0.001  ICSI 1.52 ± 0.50 1.35 ± 0.48  < 0.001  Total 1.38 ± 0.49 1.28 ± 0.45  < 0.001 Baseline characteristics of TB and control patients Significant differences were observed in the infertility type, causes of infertility, and pregnancy assistance program between two groups ( P   0.05). The total number of antral follicles, serving as sensitive markers of ovarian reserve, was higher in control group than TB ( P  = 0.033). In terms of COH protocol, the long protocol was most frequently used in TB group. Similarly, in control group, the long protocol was predominant. Fresh semen was the primary source. TB group had a higher number of embryos transferred compared to controls in each protocol ( P  < 0.001). In addition, the control group exhibited a significantly greater endometrial thickness than TB ( P  < 0.001). The incidence of male TB in female TB group was higher than in controls ( P  = 0.003). The prevalence rate of TB in this study was 4.17% (1477/35428). Patients presented with various TB types (Supplementary Table 1). The outcomes were compared between TB and control groups of fresh cycles in IVF protocol, ICSI protocol and total patients (Table  2 ). Given the significant disparities between TB and controls, variables such as duration of infertility, total antral follicles, endometrial thickness on day of hCG, male TB, and number of embryo transferred were considered as confounding factors in statistical analysis. In the primary outcomes, no significant differences ( P  > 0.05) were observed in pregnancy rates between TB and control groups among IVF patients and total patient cohort. However, in ICSI patients, the pregnancy rate was higher in control group compared to TB ( P  = 0.027). Corresponding to that, the live birth rate also showed no significant difference ( P  > 0.05) between two groups in IVF patients and total patients. In ICSI patients, the live birth rate was higher in control group ( P  = 0.027). Table 2 Pregnancy outcomes between TB and control patients a Pregnancy outcomes Female TB ( n  = 1386) Controls ( n  = 33,951) P OR 95 CI Fertilization rate (%)  IVF 98.85% (9424/9534) 98.68% (191,826/194400) 0.112 1.218 (0.955,1.552)  ICSI 77.02% (1220/1584) 75.54% (54,352/71947) 0.585 1.139 (0.713,1.822) High-quality embryo rate (%)  IVF 58.06% (3742/6445) 37.25% (75,198/201886) 0.924 1.018 (0.703,1.474)  ICSI 58.20% (536/921) 52.31% (22,184/42408) 0.362 1.168 (0.837,1.631)  Total 58.08% (4278/7366) 39.86% (97,382/244294) 0.015 1.203 (1.037,1.395) Blastocyst rate (%)  IVF 71.34% (3916/5489) 71.02% (77,395/108981) 0.006 1.248 (1.065,1.463)  ICSI 67.80% (438/646) 60.06% (19,379/32264) 0.698 1.070 (0.760,1.506)  Total 70.97% (4354/6135) 68.51% (96,774/141245)  < 0.001 1.320 (1.140,1.527) Implantation rate (%)  IVF 47.98% (772/1609) 50.15% (17,311/34517) 0.643 0.995 (0.973,1.017)  ICSI 38.03% (116/305) 45.54% (6046/13277) 0.802 0.991 (0.922,1.065)  Total 46.39% (888/1914) 48.91% (23,377/47794) 0.689 0.996 (0.974,1.017) Pregnancy rate (%)   IVF 58.61% (694/1184) 60.93% (15,260/25045) 0.302 0.935 (0.824,1.062)   ICSI 49.50% (100/202) 57.99% (5165/8906) 0.027 0.719 (0.537,0.964)   Total 57.29% (794/1386) 60.16% (20,425/33951) 0.129 0.914 (0.814,1.027) Multiple pregnancy rate (%)  IVF 12.97% (90/694) 14.37% (2193/15260) 0.665 0.944 (0.727,1.225)  ICSI 14.00% (14/100) 16.86% (871/5165) 0.559 0.837 (0.459,1.523)  Total 13.10 (104/794) 15.00% (3064/20425) 0.677 0.949 (0.743,1.212) Miscarriage rate (%)  IVF 11.96% (83/694) 14.59% (2226/15260) 0.553 0.915 (0.682,1.228)  ICSI 17.00% (17/100) 13.55% (700/5165) 0.754 1.098 (0.611,1.975)  Total 12.59% (100/794) 14.33% (2926/20425) 0.083 0.818 (0.652,1.027) Ectopic pregnancy rate (%)  IVF 1.26% (10/794) 1.19% (181/15260) 0.934 1.000 (0.991,1.008)  ICSI 2.00% (2/100) 1.05% (54/5165) 0.393 1.915 (0.432,8.500)  Total 1.51% (12/794) 1.15% (235/20425) 0.678 0.998 (0.990,1.007) Live birth rate (%)  IVF 50.76% (601/1184) 51.30 (12,848/25045) 0.853 1.012 (0.892,1.148)   ICSI 40.10% (81/202) 49.47% (4406/8906) 0.027 0.703 (0.515,0.960)   Total 49.21% (682/1386) 50.82% (17,254/33951) 0.638 0.972 (0.865,1.093) Pregnancy weight gain(kg)  IVF 14.08 ± 4.71 13.35 ± 5.02 0.336 0.804 (0.515,1.254)  ICSI 12.32 ± 4.98 13.90 ± 5.06 0.425 0.954 (0.849,1.071)  Total 13.87 ± 4.77 13.79 ± 5.03 0.321 0.819 (0.551,1.215) Birth weight of baby(kg)  IVF 3.20 ± 0.52 3.18 ± 0.52 0.345 0.826 (0.556,1.228)  ICSI 3.12 ± 0.50 3.15 ± 0.57 0.785 0.985 (0.884,1.098)  Total 3.16 ± 0.43 3.19 ± 0.52 0.333 0.825 (0.560,1.217) a Primary outcomes have been formatted in bold for clarity Pregnancy outcomes between TB and control patients a a Primary outcomes have been formatted in bold for clarity Among other outcomes, the fertilization rate in IVF and ICSI patients had no significant difference ( P  > 0.05) between two groups. Similarly, the high-quality embryo rates in IVF patients and ICSI patients did not exhibit a significant difference ( P  > 0.05) between them. However, in total patients, the high-quality embryo rate was higher in TB group compared to controls ( P  = 0.015). The blastocyst rates in IVF patients ( P  = 0.006) and total patients ( P  < 0.001) were higher in TB group than controls, while no significant difference was observed between two groups in ICSI patients ( P  = 0.698). The implantation rate and multiple pregnancy rate did not differ significantly between TB and controls in each protocol ( P  > 0.05). In abnormal pregnancy outcomes, the miscarriage rate and ectopic pregnancy rate did not show statistical differences between TB and control groups in each protocol ( P  > 0.05). During follow-up, the pregnancy weight gain and birth weight of baby also had no significant difference between TB and controls in each protocol ( P  > 0.05). Some important pregnancy outcomes were compared between fresh and frozen–thaw cycles in cured TB patients (Table  3 ). The age ( P  < 0.001), BMI ( P  = 0.010), number of embryo transferred ( P  < 0.001), and endometrial thickness on day of hCG ( P  < 0.001) were all significantly different between fresh cycles and frozen–thaw cycles, and were identified as confounding factors. Upon adjusting them, the implantation rate in fresh cycles was lower than that in frozen–thaw cycles in IVF patients ( P  = 0.024), ICSI patients ( P  = 0.028) and total patients ( P  = 0.027). Correspondingly, the pregnancy rate ( P  < 0.05) and multiple pregnancy rate ( P  < 0.05) were lower in fresh cycles compared to frozen–thaw cycles in each protocol. The live birth rate in fresh cycles was lower than in frozen–thaw cycles in ICSI protocol ( P  = 0.011). Conversely, in IVF and total patients, the live birth rate in fresh cycles was higher than that in frozen–thaw cycles ( P   0.05) between fresh and frozen–thaw cycles in each protocol. Table 3 Primary outcomes between fresh cycles and frozen–thaw cycles in TB women b Fresh cycles Frozen–thaw cycles P OR 95 CI Age (years) 31.07 ± 3.94 31.58 ± 3.93  < 0.001 / / BMI (kg/m 2 ) 22.03 ± 2.92 21.77 ± 2.78 0.01 / / Number of embryo transferred ( n ) 1.38 ± 0.49 1.45 ± 0.51  < 0.001 / / Endometrial thickness on day of hCG (mm) 11.33 ± 2.89 10.09 ± 1.90  < 0.001 / / Male TB (%) 12 (0.87) 24 (1.22) 0.332 / / Implantation rate (%)  IVF 47.98% (772/1609) 49.42% (1199/2426) 0.024 0.091 (0.928,0.995)  ICSI 38.03% (116/305) 47.79% (205/429) 0.028 0.930 (0.847,1.024)  Total 46.39% (888/1914) 49.18% (1404/2855) 0.027 0.965 (0.934,0.996) Pregnancy rate (%)   IVF 58.61% (694/1184) 59.13% (997/1686) 0.001 0.751 (0.635,0.888)   ICSI 49.50% (100/202) 57.29% (165/288) 0.008 0.594 (0.404,0.875)   Total 57.29% (794/1386) 58.87% (1162/1974) 0.001 0.780 (0.671,0.908) Multiple pregnancy rate (%)  IVF 12.97% (90/694) 18.25% (182/997) 0.042 0.746 (0.537,1.038)  ICSI 14.00% (14/100) 22.42% (37/165) 0.023 0.392 (0.175,0.877)  Total 13.10% (104/794) 18.85% (219/1162)  < 0.001 0.036 (0.026,0.049) Live birth rate (%)   IVF 50.76% (601/1184) 49.35% (832/1686) 0.008 0.801 (0.679,0.944)   ICSI 40.10% (81/202) 48.26% (139/288) 0.011 0.596 (0.399,0.889)   Total 49.21% (682/1386) 49.19% (971/1974) 0.015 0.830 (0.714,0.965) Miscarriage rate (%)  IVF 11.96% (83/694) 15.25% (152/997) 0.083 1.071 (1.037,1.105)  ICSI 17.00% (17/100) 13.94% (23/165) 0.556 1.240 (0.606,2.539)  Total 12.59% (100/794) 15.06% (175/1162) 0.247 0.845 (0.636,1.124) Ectopic pregnancy rate (%)  IVF 1.44% (10/694) 1.40% (14/997) 0.503 1.341 (0.569,3.163)  ICSI 2.00% (2/100) 0.61% (1/165) 0.309 6.766 (0.170,269.186)  Total 1.51% (12/794) 1.29% (15/1162) 0.325 1.478 (0.678,3.222) Pregnancy weight gain(kg)  IVF 14.08 ± 4.71 13.89 ± 5.02 0.319 1.432 (0.707,2.902)  ICSI 12.32 ± 4.98 14.85 ± 5.96 0.051 0.103 (0.013,0.814)  Total 13.87 ± 4.77 13.98 ± 5.16 0.960 1.018 (0.513,2.019) Birth weight of baby(kg)  IVF 3.20 ± 0.52 3.15 ± 0.59 0.382 0.973 (0.915,1.035)  ICSI 3.12 ± 0.50 3.39 ± 0.39 0.655 0.965 (0.826,1.128)  Total 3.16 ± 0.43 3.18 ± 0.61 0.310 0.971 (0.917,1.021) b Primary outcomes have been formatted in bold for clarity Primary outcomes between fresh cycles and frozen–thaw cycles in TB women b b Primary outcomes have been formatted in bold for clarity Due to the limited sample size, TB lymphadenitis, TB of chest wall, bone TB, abdomen TB, thorax TB, urinary TB, nasal and ear TB, thyroid TB, endobronchial TB, and cerebral TB were categorized as the other types. Consequently, TB patients were stratified into three groups: pelvic TB, pulmonary TB, and other TB. Given that pelvic TB directly affected fallopian tubes and ovaries, which were essential reproductive organs, particular attention was given to the primary outcomes between pelvic TB and pulmonary TB as well as other TB (Table  4 ). In patients with pelvic TB and pulmonary TB, the age ( P  = 0.001), BMI ( P  = 0.002), and endometrial thickness on day of hCG ( P  = 0.005) were significantly different. Following the exclusion of the confounding factors with statistical significance, the pregnancy rate and live birth rate did not show significant difference ( P  > 0.05) between pelvic TB and pulmonary TB. When comparing pelvic TB to other TB types, only BMI exhibited a significant variance ( P  = 0.001) and was considered a confounding factor. Likewise, the pregnancy rate and live birth rate also did not differ significantly ( P  > 0.05) between pelvic TB and other types. Table 4 Primary outcomes in different types of TB c Pelvic TB Pulmonary TB other TB Pelvic TB vs pulmonary TB Pelvic TB vs other TB P OR 95 CI P OR 95 CI Age (years) 30.41 ± 4.12 31.38 ± 3.82 31.08 ± 3.95 0.001 / / 0.206 / / BMI(kg/m 2) 22.33 ± 2.90 21.68 ± 2.87 22.24 ± 2.94 0.002 / / 0.001 / / Number of embryo transferred ( n ) 1.37 ± 0.49 1.39 ± 0.49 1.38 ± 0.49 0.634 / / 0.839 / / Endometrial thickness on day of hCG (mm) 10.92 ± 2.33 11.39 ± 2.24 11.47 ± 2.30 0.005 / / 0.596 / / Male TB (%) 1 (0.37) 3 (0.53) 8 (1.46) 0.763 / / 0.163 / / Pregnancy rate (%) 61.19% (164/268) 56.24% (320/569) 56.47% (310/549) 0.401 0.875 (0.640,1.195) 0.233 0.829 (0.609,1.128) Live birth rate (%) 51.49% (138/268) 48.15% (274/569) 49.18% (270/549) 0.790 0.958 (0.700,1.311) 0.633 0.927 (0.681,1.264) c Primary outcomes have been formatted in bold for clarity Primary outcomes in different types of TB c c Primary outcomes have been formatted in bold for clarity

In this study, cured TB did not affect the primary assisted reproductive outcomes in infertile patients using IVF protocol. However, in ICSI protocol, TB had a negative impact on pregnancy outcomes, even with treatment. Frozen–thaw cycle had a worse final pregnancy outcome than fresh cycle in IVF patients, but a better final outcome in ICSI patients. If treated, there were no variations in pregnancy outcomes among different types of TB. Only in ICSI protocol did the pregnancy and live birth rates differ between control and TB groups, while evidence comparing IVF and ICSI separately in this context remains limited. Although direct clinical evidence was scarce, mechanistic considerations might help explain this observation. IVF involved co-incubation of oocytes and sperms, whereas ICSI involved direct injection of a single sperm into the oocyte cytoplasm [ 16 , 17 ], bypassing multiple natural sperm–oocyte selections and allowing sperm with subtle structural or DNA defects to fertilize the oocyte [ 18 – 20 ]. Consequently, embryos generated through ICSI—particularly from abnormal sperm—might be more dependent on the oocyte’s intrinsic capacity for damage recognition and repair, as well as on uterine support [ 21 ]. When oocyte or endometrial milieu was compromised—due to inflammation, endocrine dysregulation, or immune imbalance—the capacity to compensate for sperm-derived defects might be diminished. This might explain why ICSI outcomes were more sensitive to oocyte competence and uterine environment than conventional IVF. Consistently, women with abnormal oocytes or endometrium—including those with genital TB—tended to experience poorer pregnancy outcomes [ 13 , 22 – 25 ]. Chronic inflammation, cytokine changes or toxic substances induced by Mycobacterium tuberculosis might impair oocyte and endometrium quality [ 26 , 27 ]. Together, these considerations might help explain why differences in pregnancy outcomes between TB and control groups were observed only in ICSI. Furthermore, ICSI was primarily employed for male-factor infertility [ 28 ]. A high sperm DNA fragmentation negatively affected ICSI success, potentially amplified by the presence of oocyte dysmorphisms [ 21 ]. In couples with female TB history and male-factor infertility, the combined effects of female- and male-related impairments might contribute to the lower pregnancy rates observed in ICSI [ 21 , 29 – 31 ]. The apparent differences in sample sizes between ICSI and IVF groups might also have a minor impact. In addition, the miscarriage rate between TB and controls had no significant difference. Existing studies reported that the miscarriage rate was higher in TB patients than non-TB, or there was no difference between them. Sheng, Z. et al. [ 32 ] reported no statistically significant differences in miscarriage rates (early and late patterns) between women with treated prior-pulmonary TB and those without pulmonary TB, nor between treated and untreated prior-pulmonary TB groups. Yan, G.X. et al. [ 6 ] also showed no statistical difference in miscarriage rates between untreated prior pulmonary TB and non-TB patients. Lin et al. [ 24 ] reported a higher abortion rate in cured endometrial TB group. However, they enrolled only 184 TB patients receiving IVF protocol, limited by the small number and single protocol. Our results included several types of TB, with pulmonary TB accounting for the largest portion [ 6 ]. The effect of pulmonary TB on fertility was much smaller than that of reproductive TB in terms of anatomy. The inclusion of various types beyond reproductive TB could potentially lower the overall miscarriage rate. In addition, compared to controls, TB patients who had received effective management before IVF–ET were frequently under more intensive medical supervision. It assisted in the early identification and management of problems and reduced the risk of miscarriage. In practical application, we are more concerned about whether TB affected pregnancy outcomes when cured. Even after successful anti-TB treatment, Mycobacterium tuberculosis potentially affected the endometrial implantation environment. The damage caused by Mycobacterium tuberculosis to the endometrial blood flow and uterine cavity morphology was often irreversible. Endometrial damage incurred prior to treatment might not be completely restored [ 33 ]. Notably, in this study, the endometrial thickness of TB group was thinner than controls. To ensure the calculation accuracy, we added this confounding factor in the analysis. Dai, W., et al. [ 13 ] reported no difference in the pregnancy rate, abortion rate and live birth rate between genital TB and non-TB patients. The cumulative pregnancy rate in individuals with endometrial TB was significantly lower compared to tubal TB and non-TB. These overall trends aligned with our results. Singh, N., et al. [ 11 ] reported no difference in endometrial blood flow and pregnancy rate between patients with and without genital TB. Lin, M.M., et al. [ 24 ] reported that the cumulative pregnancy rate of cured endometrial TB was comparable to that of non-TB patients. The cumulative live birth rates were significantly lower, and the abortion rates were notably higher in TB group. However, the elimination of thinner endometrial thickness was not mentioned. Apart from genital TB, a retrospective cohort study of infertile women with prior pulmonary TB reported that clinical pregnancy and live birth rates were higher in treated patients than in untreated patients, and were comparable between treated pulmonary TB patients and non-TB controls [ 32 ]. After controlling for confounding factors, the implantation and pregnant rates in fresh cycles were lower than that in frozen–thaw cycles in each protocol. The high doses of hormone stimulation in fresh cycles sometimes affected the quality of endometrium, resulting in poor synchronization of endometrium with embryo. In a frozen–thaw cycle, embryo was usually transferred in a natural cycle without drug stimulation, or in a carefully prepared replacement cycle, which provided a more favorable environment in uterus, thereby supporting embryo implantation and the development of early pregnancy [ 34 – 37 ]. In addition, preimplantation genetic testing (PGT) could assess embryos before transfer to help avoid the transmission of chromosome and genetic diseases and reduce implantation failure and pregnancy loss. PGT tends to prioritize frozen transfer due to the flexibility schedule, the avoidance of effects of ovulation stimulating drugs, and the reduced risk of ovarian hyperstimulation syndrome [ 38 – 40 ]. The fresh transferred embryos had not undergone further genetic testing and possibly had chromosomal or genetic abnormalities that prevented the embryo from developing despite implantation. Frozen–thaw cycles could provide sufficient time for embryo detection, facilitating the selection of the most suitable embryo. For ICSI patients, the live birth rate in fresh cycles was also lower than in frozen–thaw cycles; but for IVF and all patients combined, it was slightly higher. This opposite phenomenon might be attributed to the increased occurrence of pregnancy complications following frozen–thaw cycles. Compared with fresh cycles, frozen–thaw cycles had a higher prevalence of hypertensive disorders of pregnancy, preterm prelabor rupture of membranes, bleeding disorders, cesarean section, post-term birth, macrosomia, and placenta accreta. These complications could potentially result in restricted fetal development and abnormal delivery, subsequently reducing the live birth rate. Although modern freezing techniques had greatly improved the survival of embryos, the process of freezing and thawing itself could influence the long-term developmental potential of certain embryos [ 15 , 41 – 45 ]. Moreover, ICSI protocol was more commonly used for PGT in practical applications due to its ability to avoid sperm contamination and optimize embryo quality. Therefore, more normal embryos might be selected in ICSI than IVF. The improvement of embryo itself had potentially increased the proportion of normal live births, resulting in higher live birth rate [ 39 , 46 , 47 ]. This study demonstrated that the primary pregnant outcomes of TB infection in pelvis, lungs, or other body parts did not show any significant differences as long as getting anti-TB treatment. It probably because that anti-TB drugs circulated through blood to various organs, effectively treating all parts of the human body, including reproductive organs [ 48 ]. Moreover, the prevalence rate of TB in this study (4.17%) exceeded the reported rate in China in 2023 (50 per 100,000) [ 49 ]. Another retrospective study revealed that the prevalence of prior pulmonary TB diagnosed by chest X-ray in infertile women was 10.4% (1487/14,254) [ 6 ]. Both indicated that infertile individuals exhibited a higher TB prevalence compared to general population. Non-TB patients were more likely to conceive naturally and avoided IVF–ET. Our data were derived from infertile patients seeking medical care at the hospital. The limitation of study site was more likely to enroll TB patients, resulting in the notably elevated TB prevalence. This study employed GEE to address the correlation between data in longitudinal analysis. The large enough sample size ensured more precise, reliable and generalizable results. Moreover, we subdivided two fertilization protocols to provide more comprehensive reference outcomes for TB patients. As for the limitations, the data were restricted to a single-center, without multi-center participation. Furthermore, it was a retrospective study. The TB histories of some patients were obtained orally without detailed medical records. Thus, the recall bias was inevitable. To mitigate bias in data collection, our medical personnel conducting oral history cross-verified the oral accounts through alternative sources such as drug utilization records, testimonies from other family members, or summaries of previous medical reports. In addition, the explanation for the observed differences in ICSI outcomes between TB and control groups relied primarily on mechanistic evidence, while direct clinical studies were limited. In the following study, we plan to conduct a prospective study in collaboration with multiple reproductive centers to closely monitor the treatment and subsequent pregnancy outcomes in TB patients.

Tuberculosis (TB) is a bacterial infection caused by Mycobacterium tuberculosis which invades several systems of human body [ 1 ]. In China, where the overall prevalence of TB is very high, female genital TB remains a significant cause of infertility. It potentially causes fallopian tube and endometrial lesions resulting in subfertility and pregnancy loss [ 2 – 4 ]. However, Mycobacterium tuberculosis in other systems was reported to affect the pregnancy and delivery outcomes, such as pulmonary TB [ 5 , 6 ]. Mycobacterium tuberculosis may potentially infect the genitals in a latent form, without imaging or laboratory abnormalities. It can spread to the female genital tract via blood or lymphatic routes, and may also be transmitted directly through infected pelvic organs, such as bladder and rectum [ 7 ]. Therefore, it is hypothesized that TB infection in any system has the potential to contribute to latent fertility loss. Infertility is defined as a disease of the male or female reproductive system and failure to conceive after 12 months of regular unprotected sexual intercourse [ 8 , 9 ]. In recent years, in vitro fertilization and embryo transfer (IVF–ET) has emerged as the most successful infertility treatment, commonly used in infertile patients with TB [ 6 , 10 ]. Though some studies have previously compared IVF–ET outcomes between patients with and without TB, they have predominantly focused on TB of the genital organs [ 11 – 14 ]. Except for genital TB, other types of TB account for a large proportion in infertile women. Moreover, the history of TB has caused great psychological pressure on patients to worry about the failure of pregnancy. It remains uncertain whether there are hidden effects on the reproductive system following previous TB treatment. Given that anti-TB drugs spread through the blood to all organs, we suppose that the resolution of Mycobacterium tuberculosis may have a positive effect on ovarian reserve and uterine microenvironment, improving pregnancy outcomes [ 9 ]. However, there were no clinical data to support this speculation. Hence, we aim to investigate whether different types of TB after treatment affect assisted reproductive outcomes in infertile women.

Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 17 KB) Supplementary file1 (DOCX 17 KB)