Clinical Outcomes of In Vitro Fertilization among Chinese Infertile Couples Treated for Syphilis Infection.

Syphilis is a sexually transmitted disease (STD) caused by Treponema pallidum , and it is primarily spread through sexual intercourse[ 1 , 2 ], blood transfusion[ 3 ] or maternal-neonatal transmission[ 4 ]. The incidence rate of syphilis has risen rapidly in China since 1978[ 5 , 6 ], and it has become the most common STD in economically developed regions[ 7 ]. Syphilis infection of the reproductive tract[ 8 ]can cause inflammatory disease and infertility[ 9 ]. Reproductive tract infection and inflammation cause 8% and 35% of all cases of male infertility, respectively[ 10 ]. In women, it can result in fallopian tube obstruction[ 11 , 12 ] and endometritis. Approximately 72.4 million people worldwide[ 13 ] have infertility problems. In vitro fertilization (IVF) may be the last resort for couples attempting to overcome infertility[ 14 , 15 ]. The relationship between syphilis and infertility has been rarely reported in the literature. One previous report[ 16 ] on the effects of syphilis on IVF outcome has shown lower fertilization and implantation rates due to infection of the male parent with latent syphilis compared with the rates in those without syphilis infection. However, women with syphilis were not included in the study. Therefore, we performed retrospective analysis of patients who received IVF/intra-cytoplasmic sperm injection (ICSI) to evaluate the effects of syphilis on IVF-embryo transfer (ET) outcome.

Syphilis infection reduces the clinical pregnancy rate after IVF/ICSI. The endometrial thickness and implantation rate of infertile couples treated for syphilis infection may be the most important factors affecting IVF outcome.

No significant differences were found in female age ( P = 0.99), duration of infertility ( P = 0.50), BMI ( P = 0.15), or the basal FSH ( P = 0.12), basal LH ( P = 0.59) or basal E 2 level ( P = 0.81) between the syphilis and control groups. Endometrial thickness on the hCG injection day was significantly higher for the syphilis group compared with the control group (16.9±5.4 vs. 13.0±4.7 mm, P <0.001). Syphilis infection likely affected endometrial thickness ( Table 1 ). Note: BMI = body mass index; FSH = follicle-stimulating hormone; LH = luteinizing hormone. a The independent samples t-test was conducted to compare the syphilis and control groups, and the values are presented as the mean ± standard deviation. b A significant difference was detected between the syphilis and control groups using the independent samples t-test ( P <0.05). The number of normal fertilizations (6.8±4.8 vs. 8.3±4.6, P = 0.004), number of oocytes with normal cleavage (6.3±4.7 vs. 8.1±4.6, P = 0.001) and implantation rate (24.2% vs. 34.4%, P = 0.003) significantly differed between the syphilis and control groups. The clinical pregnancy rate was significantly lower in the syphilis group compared with the control group (43.8% vs. 55.6%, P = 0.03), and the biochemical pregnancy rate was higher in the syphilis group (7.5% vs. 0.6%, P = 0.002) ( Table 2 ). Note: The quality screening criteria for high-quality embryos included the presence of 7–8 blastomeres of equal size with an accumulation of <20% cell debris by day 3 (Embryology, 2011). Normally fertilized oocytes were defined as those with two pronuclei (2PN) and 2 polar bodies at 16–20 hours after oocyte retrieval. The normal fertilization rate was calculated as follows: number of normally fertilized oocytes / (number of oocytes with one pronucleus + number with two pronuclei + number with multiple pronuclei + number with late cleavage). The implantation rate (IR) was defined as the number of gestational sacs per number of embryos transferred. The live birth rate per embryos transferred was defined as the total number of births per the number of embryos transferred that resulted in clinical pregnancy. The clinical pregnancy rate per embryo transferred was determined by the number of patients with a gestational sac in the uterus at 5 weeks after embryo transfer. Early miscarriage was defined as the natural termination of pregnancy before 12 weeks, even in the presence of a positive serum hCG test or an ultrasound-detected intrauterine gestational sac after embryo transfer. Biochemical pregnancy was defined as a pregnancy that did not clinically progress, accompanied by a β-hCG level of ≥25 U/L. a The independent samples t-test was performed to compare the syphilis and control groups, and the values are presented as the mean ± standard deviation. b A significant difference was detected between the syphilis and control groups using the independent samples t-test ( P <0.05). c The Mann-Whitney test was conducted to compare the non-normally distributed data, and the results are shown as the median (range). d Data were obtained using Pearson’s chi-square test. e A significant difference was detected between the syphilis and control groups using Pearson’s chi-square test ( P <0.05). f Data were obtained using Fisher’ Exact Test. No significant differences in age ( P = 0.25), the duration of infertility ( P = 0.23), basal FSH level ( P = 0.52), basal E 2 level ( P = 0.31), total dosage of Gn ( P = 0.79), number of retrieved oocytes ( P = 0.38), or early miscarriage rate ( P = 0.81) were found among the infected women, infected men, and infected couple subgroups. Fetal birth weight ( P = 0.006) significantly differed among the three subgroups. Furthermore, the LSD test revealed that the fetal birth weight was significantly lower for the offspring of the infected males compared with those of the infected females (2.7±0.4 vs. 3.0±0.4 kg, P = 0.01) and couples (2.7±0.4 vs. 3.1±0.5 kg, P = 0.007). The average gestational period was also shorter in the infected male group compared with the infected female (36.5±1.4 vs. 37.9±1.5 kg, P< 0.001) and couple groups (36.5±1.4 vs. 38.1±1.6 kg, P = 0.003). Fetal birth weight at delivery <37 weeks was significantly higher in the couple group compared the infected female (3.3±0.3 vs. 2.7±0.4 kg, P = 0.01) and infected male groups (3.3±0.3 vs. 2.4±0.2 kg, P = 0.001). No differences in the clinical pregnancy rate were detected among the subgroups (women: 43.1% vs. men: 44.9% vs. couples: 42.3%, P = 0.96) ( Table 3 ). Note: Multiple groups of quantitative data were compared using analysis of variance, and the LSD test was performed for pairwise comparisons between any two of the three groups. a The variance and LSD test were used to compare the three subgroups, and the values are presented as the mean ± standard deviation. b A significant difference was detected among the three subgroups by analysis of variance ( P <0.05). c Non-normally distributed data are presented as the median (range), as assessed by the Kruskal-Wallis test. d Data were obtained using Pearson’s chi-square test. f Data were obtained using Fisher’s Exact Test. * A significant difference was detected between the infected female and infected male groups. # A significant difference was detected between the infected female and infected couple groups. & A significant difference was detected between the infected male and infected couple groups. Logistic regression analysis was performed to evaluate the confounding factors for the clinical pregnancy rate. We found that the presence of syphilis (OR: 0.615, 95% CI: 0.382–0.991, P = 0.046), female age (OR: 0.895, 95% CI: 0.838–0.955, P = 0.001), and BMI (OR: 1.124, 95% CI: 1.023–1.235, P = 0.015) all influenced this rate ( Table 4 ). Note: The presence or absence of syphilis, female age, BMI, duration of infertility, basal FSH, LH and E 2 levels, total number of retrieved follicles, number of normally fertilized oocytes and number of fertilized oocytes with normal cleavage were considered as possible confounding factors and treated as continuous variables. The presence or absence of syphilis, female age and BMI were determined to be significant confounding factors. B: partial regression coefficient SE: standard error df: degree of freedom Exp(B): odds ratio

A positive history of syphilis infection can reduce the clinical pregnancy rate following IVF/ICSI. We have shown that this reduction in the clinical pregnancy rate involves a variety of factors, including endometrial thickness, the number of mature oocytes, the number of fertilized, normally cleaved oocytes, and the implantation rate. However, the mechanisms underlying the effects of syphilis on IVF success are currently unclear. Future studies should use larger sample sizes to better elucidate these mechanisms.

We reviewed the patients’ medical records to obtain information about their hospital and telephone follow-up assessments. The collected data were analyzed using SPSS 21.0. A total of 3,275 couples with no history of syphilis infection were included. We selected 160 patients from the 3,275 couples to form the control group. First, the 3,275 couples were divided into 20 groups based on age (range of 24–43 years), according to the age range of the syphilis group, and then random patient numbers were generated for each age group using SPSS 21.0 with the seed set to (1,100) and inclusion of four decimal places in the random numbers. According to the number of patients in each syphilis age group, 160 patients were extracted [ 27 ] from the randomly generated numbers in ascending order by stratified random sampling to form the control group. The t-test was used to perform between-group comparisons of the normally distributed quantitative datasets with homogeneous variances, and the results are presented as the mean ± standard deviation. The Mann-Whitney test was conducted to analyze the non-normally distributed datasets, and the results are presented as medians with ranges. Analysis of variance was applied to compare the means of the three quantitative subgroups when that data for each sample was normally distributed with a homogeneous variance. The least significant difference (LSD) test was used to perform pairwise comparisons between any two of the three groups, and the results are presented as the mean ± standard deviation. The Kruskal-Wallis test was conducted to assess non-normally distributed data among the three subgroups or those that did not exhibit homogeneous variances, and the results are presented as medians with ranges. The chi-square test was carried out to assess the qualitative data, and the results are reported as percentiles. Binary analyses were conducted using logistic regression, and factors were introduced into the regression equation using the Enter method. All results were considered significant (two-tailed) at P <0.05.

All patients in this study provided written informed consent for the use of their medical records. The Ethics Committee of the Sun Yat-Sen Memorial Hospital approved this study.The ethics committee approval of the Sun Yat-sen Memorial Hospital [2014] Record of ethics committee No. (02) Reproductive medicine centre: After censored by the medical ethics committee of the Sun Yat-sen Memorial Hospital, the clinical research project of your department “The effect of syphilis infection to outcomes of IVF: a retrospective case-control study” was approved to be carried out based on the retrospective study according with the requirements of medical ethics. Medical Ethics Committee 16 January 2014. In this retrospective case-control study, clinical data were obtained from 320 infertile couples presenting to Sun Yat-Sen Memorial Hospital between January 2008 and March 2014. The patients were divided into a syphilis group(n = 160) and a control group(n = 160)(age-matched at a ratio of 1:1). The syphilis group was divided into the following three subgroups: the infected male group (couples in which the male was anti-syphilis antibody-positive and the female was healthy, n = 69), the infected female group (couples in which the female was anti-syphilis antibody-positive and the male was healthy, n = 65), and the infected couple group (both partners were infected, n = 26). No interventional treatments were performed. Couples who failed to conceive after at least 1 year of intercourse without contraception were considered to be infertile. Infertility was caused by fallopian tube dysfunction and/or male-related factors (a diagnosis of asthenozoospermia and/or asthenoteratozoospermia or azoospermia according to the WHO Laboratory Manual for the Examination and Processing of Human Semen[ 17 , 18 ]). All patients underwent fresh ET during the first cycle and did not receive any medication for 3 months prior to IVF/ICSI. Patients were excluded if they had hysteromyoma, adenomyosis, ovarian tumor, endometriosis, polycystic ovarian syndrome (PCOS), hyperprolactinemia, thyroid disease, adrenal disease, diabetes, or a chromosomal aberration(determined by karyotype). Women with a reproductive tract abnormality or uterine malformation and those with a sperm or egg donor were also excluded. Ovarian function was examined, and peripheral blood samples were collected on days 2–4 of menstruation to examine the basic follicle-stimulating hormone (bFSH), basic luteinizing hormone (bLH), and estrogen (E 2 ) levels. Ultrasonography was performed to determine the antral follicle count (AFC). Routine semen and sperm morphology tests were conducted[ 15 ]. All participants received laboratory serologic examinations (the rapid plasma reagin [RPR] and/or Treponema pallidum particle agglutination [TPPA] tests) to test for the presence of syphilis infection prior to the IVF/ICSI treatment. All serum samples were first assessed using a non- Treponema pallidum antigen serologic (RPR) test to screen for syphilis, and then a Treponema pallidum antigen serologic (TPPA) test was conducted to confirm the syphilis-positive samples identified by the screening test[ 19 , 20 , 21 ]. For the RPR test, cardiolipin is used as an antigen to detect anti-cardiolipin antibodies in the serum[ 22 ]. All serum samples for the RPR test were progressively diluted (1:1 to 1:32) in normal saline for titer determination[ 23 ]. In the TPPA test, gelatin granules sensitized to Treponema pallidum bind to anti- Treponema pallidum antibodies in human serum. The combination of the RPR and TPPA tests is widely used[ 24 ] for the maximal detection of syphilis during all phases of the disease with adequate sensitivity and specificity. The patients were divided into the syphilis and control groups based on the laboratory serologic results. The patients in the syphilis group with a history of syphilis treatment (according to medical records) but no history of re-exposure did not receive syphilis treatment in this study[ 20 ]. An anti-syphilis treatment (penicillin G 20 ) was administered to the patients without a history of syphilis infection and with positive RPR or TPPA results[ 20 ]. A 4-fold (2 dilutions) or greater decrease in the anti- Treponema pallidum antibody titer was indicative of treatment efficacy. This titer remained stable over a long period of time in the few patients who received the standard anti-syphilis treatment, a phenomenon known as “fixed serology”, if the patients without clinical signs of relapse were not treated with anti-syphilis therapy. The patients received IVF/ICSI treatment at one month after the disappearance of their clinical syphilis symptoms or after their Treponema pallidum antigen serologic test results became negative. Some patients remained positive for Treponema pallidum -specific antibody after the anti-syphilis treatment. Couples were included in the control group if both patients showed negative RPR and TPPA results. The mid-luteal long protocol was employed as follows [ 25 ]: a 1.87 mg dose of long-acting GnRH-a (Diphereline; Ipsen Pharma Biotech, Signes, France) was administered during the prior mid-luteal menstrual cycle to promote pituitary downregulation, as determined byblood E 2 <50 ng/L, LH <5 IU/L, FSH <5 IU/L, and ultrasound-detected vaginal endometrial thickness <5 mm. Gonadotropin (Gn) was administered at 14 days after downregulation to promote follicle development. The Gn dosage was determined based on patient age, the basal FSH level, BMI, and the AFC, and 150–300 IU/d recombinant human FSH (Gonal-F; Merck Serono, Geneva, Switzerland) was administered. During ovarian hyperstimulation, the blood E 2 , progesterone, and LH levels were measured, and follicle development was monitored by transvaginal ultrasound (ALOKA 3500, Japan). When three follicles with ≥16 mm diameters, two follicles with ≥17 mm diameters, or one follicle with a ≥18 mm diameter was detected on ultrasound, 10,000 IU human chorionic gonadotropin (hCG) (Livzon, Zhuhai, China) was administered on that evening. Ultrasound-guided transvaginal oocyte retrieval was conducted at 34 to 36 hours after hCG injection, and a sperm sample was concurrently collected and subjected to density gradient centrifugation[ 15 ]. The fertilization regimen and traditional IVF and ICSI fertilization protocols used were selected based on the concentration and motility of the washed sperm. No special ovulation or sperm washing procedures were performed for the antibody-positive patients without a recent syphilis reinfection. Embryo morphology evaluations were conducted by monitoring the pronuclei for 17±1 hours after fertilization. Embryo cleavage and development were monitored for 44±1 hours and 68±1 hours after fertilization, respectively[ 26 ]. Optimal embryos that met the quality screening criteria did not exhibit multiple pronuclei; rather, they showed 4–5 blastomeres of homogenous sizes with an accumulation of <5% cell debris by day 2 and 7–9 blastomeres of homogenous sizes with an accumulation of <5% cell debris by day 3. Ultrasound-guidedET was conducted on day 3 after oocyte retrieval. When available, two embryos were transferred into the women who were <35 years old, and three embryos were transferred into those who were ≥35 years old. Luteal support was achieved via daily intramuscular injections of progesterone (60 mg/day) (Progesterone Injection, Xian Ju Pharmaceutical Co., Zhejiang, China) following oocyte retrieval for up to 14 days after ET. Serum β-hCG was tested on day 14 after ET, and progesterone administration was continued for 12 weeks after diagnosis of pregnancy. Ultrasonography was conducted at 35 days after ET, and clinical pregnancy was diagnosed via the presence of an intrauterine gestational sac, fetal bud, and fetal heartbeat; otherwise, biochemical diagnosis was performed.

(DOC) Click here for additional data file. (DOC) Click here for additional data file. (DOC) Click here for additional data file.