Relationship between semen regurgitation and pregnancy rates with intrauterine insemination.

From a total of 2221 IUI cycles, 264 were excluded leaving 1957 cycles available for analysis ( Figure 1 ). These cycles were performed on 660 women (median cycles, 3; interquartile range, 3), with an average age of 31.9 ± 4.9 SD. Regurgitation was noted in 715 of 1957 procedures (36.5%). Baseline characteristics of women including age, BMI, TMC, method of sperm preparation, type of medications used for controlled ovarian stimulation, duration of infertility, infertility diagnosis, race/ethnicity and number of cycles per couple were not statistically different between the two groups who had regurgitation and those who did not ( Table 1 ). Overall, in unadjusted analyses, there was no statistical or clinically meaningful difference in clinical pregnancy (10.5% versus 10.0%, RR= 0.99, 95% CI [0.73–1.35], Table 2 ) or live birth rates (6.3% versus 6.8%, RR=0.82 (0.53–1.26) when comparing cycles with regurgitation to those without regurgitation. None of the covariates examined met the criteria for confounding. For comparison with the crude estimates, results are also reported when adjusted for age, race/ethnicity, BMI, TMC, method of sperm preparation, years of infertility, medications and infertility diagnosis; however, results were unchanged ( Table 2 ). Pregnancy rates per cycle also did not differ in the presence or absence of regurgitation (15.0% versus 15.4%, respectively, unadjusted RR=0.97, 95% CI [0.75–1.24]). ( Table 2 ). Findings were consistent after excluding patients treated with gonadotropins (data not shown). When stratified by TMC (10 million), the presence of regurgitation did not affect the likelihood of pregnancy or clinical pregnancy and live birth rates ( supplemental Table 1 ). Sensitivity analyses of live birth outcomes also demonstrated lack of associations with regurgitation overall ( Supplemental Table 2 ) and by TMC strata ( Supplemental Table 3 ) when the clinical pregnancies with unconfirmed deliveries were retained in the analysis and treated as if all did not result in live births (i.e., worst case scenario). Assessments of the best case scenario, where clinical pregnancies with unconfirmed deliveries were assumed to result in live births, equate to the results presented for clinical pregnancy outcomes and also reflect no association with regurgitation.

Institutional Review Board approval was obtained prior to conducting this retrospective chart review. All IUI cycles performed between July 2007 and May 2012 were considered for inclusion. Exclusion criteria were a) two separate IUI procedures in the same treatment cycle, b) partner collected two semen samples the same day for IUI, c) partner-reported sample spill during collection or transportation of the sample, d) retrograde ejaculation collection for IUI, e) missing pregnancy outcome, f) presence or absence of regurgitation was undocumented or g) missing covariate data. Charts were reviewed for multiple factors which may be associated with pregnancy rates including age, body mass index (BMI), race/ethnicity, TMC, method of sperm preparation, IUI difficulty, and duration of infertility, infertility diagnoses and ovarian stimulation medications. Regurgitation was recorded in the procedure note by the provider as “present” or “absent.” The IUI sperm sample was prepared with either density gradient, wash only, or thaw only (in the case of frozen prewashed sperm). After liquefaction of fresh semen samples, the volume, count, and motility were recorded and TMC was calculated. Specimens with a TMC of >20 × 10 6 sperms were prepared using single phase density gradient in which semen was layered on a bed of pre-warmed 90% gradient solution (Sperm Care, In Vitro Care, Inc. Frederick, MD, USA) in 1 or 2 vials based on semen volume. Centrifugation was performed at 400 × g for 20 minutes or until a pellet formed. The supernatant was removed and the pellet was re-suspended in 3.0 ml of fresh sperm washing medium (SWM; HTF Hepes + 5.0 mg/ml HSA, In Vitro Care Inc. Frederick, MD, USA) and centrifuged for 10 minutes at 400 × g. Supernatant was again removed and the pellet re-suspended in 0.5 ml of SWM, and mixed thoroughly. Seven microliters of the sperm suspension were used to determine the count, motility, and calculate the TMC post-semen processing. After the completion of this process, the specimen was placed in an incubator at 37°C until IUI. Specimens with an initial TMC ≤ 20 × 10 6 were subjected to “wash only” preparation method. Specimens in this category were diluted with SWM in a 2:1 ratio (SWM/Semen) and mixed thoroughly. Specimens were then centrifuged at 400 × g for 10 minutes. The supernatant was removed, the pellet re-suspended in 3.0 ml of SWM, and the wash step repeated. The supernatant was again removed and the pellet re-suspended in 0.5 ml of SWM and mixed thoroughly. The TMC was determined post-processing and the specimen was placed in an incubator at 37°C until IUI. For prewashed frozen sperm, the sample was thawed, the TMC determined, and the sample placed in the incubator until the IUI procedure. For frozen sperm that was unwashed, the sample was prepared using either density gradient or wash only as described above. IUI was performed approximately 36 hours after subcutaneous hCG injection or the day after a positive ovulation predictor kit (testing was performed between noon and 3:00 pm). A speculum was placed and a flexible catheter was passed through the cervix and into the uterine cavity while observing with abdominal ultrasound. A soft insemination catheter was most commonly used (Soft-Pass Coaxial Insemination Catheter, Cook Medical, Bloomington, IN USA). If the IUI proved difficult, then a rigid memory catheter was used (Cook Medical). The difficulty of the IUI was documented by the provider (easy, moderately difficult or difficult) as well as components of the IUI technique (a bend in the catheter, regurgitation, bleeding from the cervical os, and type of the IUI catheter used) necessary to complete the procedure. Following the procedure, the patients rested on the exam table in the supine position for 5 to 10 minutes. They were instructed to check a home pregnancy test 15 days later if they did not begin menstruation. If the patient had a positive home pregnancy test, a quantitative hCG (qhCG) level was obtained. The qhCG was repeated 2–4 days later. An ultrasound was scheduled at approximately 7 weeks gestation. Most patients had a repeat obstetrical ultrasound two to three weeks later at 9 to 10 weeks gestational age. Our primary outcome was live birth delivery past 24 weeks gestation. The secondary outcomes were clinical pregnancy, defined by documentation of fetal heart beat with ultrasound and positive serum pregnancy test, defined as a serum qhCG > 10 mIU/mL fifteen days following IUI. Given the delivery outcomes of 74 clinical pregnancies were unconfirmed and excluded from analyses of live births, we conducted sensitivity analyses to examine the potential range of results after assuming that all clinical pregnancies with unconfirmed deliveries did not result in a live birth. We used Chi-square tests and Wilcoxon Rank Sum tests to evaluate the distribution of baseline patient characteristics and pregnancy results by presence of regurgitation. We calculated risk ratios (RR) and 95% confidence intervals (95% CI) using a generalized estimating equation method to estimate Poisson regression models with robust standard errors to account for within-cluster correlation. Informative cluster size, which may occur when the number of IUI cycles per couple is influenced by previous treatment outcomes, was addressed by fitting a cluster-weighted model that weighted the GEE score equation by the inverse of the number of IUI cycles completed for each couple. ( 8 , 9 ) Factors known or suspected to be associated with outcomes of IUI treatment were individually evaluated as confounders using a criterion of a 15% change in the adjusted RR compared to the crude value. The covariates examined included post-prep total motile sperm count (≤5, >5–10, >10–20, >20–30 and ≥30 million), female age (≤34, 35–39, ≥40 years), race/ethnicity (white, black, Hispanic, Asian and American Indian), body mass index (BMI) (<18.5, 18.5–24.9, 25–29.9, 30–39.9, and ≥40 kg/m 2 ), duration of infertility (≥3 vs <3 years), medication for controlled ovarian stimulation (COS) (clomiphene/letrozole, gonadotropins, or none), infertility diagnosis (endometriosis, ovulatory, tubal, other or unexplained) and sperm preparation (simple wash, density gradient and thawed only). We also examined potential confounding by insemination difficulty due to bleeding (yes, no), pain or cramping (yes, no), catheter bend (yes, no), or obstruction (yes, no), amount of mucus (minimum, moderate, abundant), catheter type (soft, memory), catheter bend (none, minimum, moderate/severe), and visualization (good, fair, poor). Unadjusted and adjusted models were repeated after excluding patients treated with gonadotropins. Stratified analyses were used to examine potential differences in the association between regurgitation and IUI outcomes by total motile sperm count.

Regurgitation is a common finding during the IUI procedure. We were reassured to find that overall the presence of regurgitation did not decrease the likelihood of pregnancy, contrary to our hypothesis. This is an important finding as currently no previous studies have evaluated regurgitation as a factor for pregnancy rates following IUI. Prospective studies of IUI success should include documentation of presence or absence of regurgitation at the time of IUI to confirm our findings and to further compare live birth rates.

The presence of semen regurgitation at the time of IUI was not associated with lower clinical pregnancy or live birth rates. Regurgitation is a relatively common occurrence during IUI (36.5% in our study) and it is surprising that it has not been evaluated as a factor related to pregnancy outcomes in previous studies. Factors previously associated with pregnancy rates in IUI cycles include the woman’s age, the length of infertility, infertility diagnosis, semen parameters on initial analysis and after wash, TMC after wash for IUI, fertility medication and the number of mature follicles ( 9 – 11 ). The TMC inseminated has been consistently cited in the literature as the index most predictive of conception in IUI cycles ( 12 , 13 ). We hypothesized that if regurgitation occurred at the time of IUI then we decreased the TMC available to achieve pregnancy. According to Merviel et al, the couple with the best chance of achieving pregnancy after IUI is a woman under 30 and a man with TMC of 5 million spermatozoa ( 9 ). In Merviel’s study, the cumulative clinical pregnancy rate per couple after multiple IUI (2.6 ± 1.6 IUI cycles (range, 1–9)) cycles was 28.5% when the TMC was lower than 5 million and was significantly higher (44.3%) when the TMC was above 5 million (p< 0.05). The median TMC inseminated in our study was 13.0 million (IQR=18.7). Therefore, if some of this specimen is lost through regurgitation, the remaining TMC was likely still above the 5 million threshold to achieve pregnancy. This is the first study to evaluate the effect of sperm regurgitation on human pregnancy outcomes during IUI procedure. One strength of our study is that all IUI cycles were performed under abdominal ultrasound guidance, thereby ensuring the tip of the catheter was located with the uterine cavity prior to injecting the sample. In most clinics, IUI is not performed under ultrasound guidance. We acknowledge that ultrasound guidance was not found to be associated with IUI outcomes ( 14 ). However, in our study the use of ultrasound guidance provides reassurance that regurgitation was not due to incomplete passage of the catheter through the internal os. Additional strengths of our study are the large number of cycles evaluated and the statistical approach that accounts for multiple IUI cycles per couple when number of cycles is influenced by previous treatment outcomes. Our evaluation also assessed several IUI and patient characteristics for potential confounding. Most importantly, we used live birth as our primary outcome rather than positive pregnancy test. Using clinical pregnancy rates and live birth rates more accurately demonstrates IUI success rates from the patient’s perspective. Limitations of our study include the inability to control for the provider performing the IUI, as differing techniques among providers could have affected the rates of regurgitation and/or pregnancy outcomes. Subgroup analyses are also limited by the smaller number of pregnancy events occurring within groups by TMC.

Intrauterine insemination (IUI) is a commonly performed procedure for treating infertility due to its relative simplicity and low cost. IUI is generally attempted before proceeding to more expensive and invasive assisted reproductive techniques, such as in vitro fertilization (IVF) with or without intracytoplasmic sperm injection (ICSI). Indications for IUI include mild to moderate male factor, unexplained infertility, donor sperm insemination, endometriosis, inability to achieve pregnancy by vaginal intercourse due to organic or psychosexual issues, viral disorders such as HIV to decrease the risk of transmission, and cervical factor infertility( 1 ). Published pregnancy rates following IUI have varied from one study to another likely due to patient selection criteria, infertility diagnoses, ovarian stimulation protocol, number of ovulatory follicles following ovarian stimulation, number of cycles performed, sperm parameters, and sperm preparation for IUI ( 2 – 4 ). Previously published clinical studies have identified numerous factors associated with increased IUI pregnancy rates, including period of abstinence, sperm normal morphology, sperm count, volume, and motility which together are used to calculate the post-prep total motile count (TMC) of sperm [TMC=volume (ml) × count (million/ml) × percent motility] ( 5 – 7 ). During the IUI procedure a plastic catheter is inserted into the uterine cavity and a concentrated sample of sperm is injected. During injection or upon removal of the catheter, occasionally backflow of the sample from the cervix is noted, known as regurgitation. Assuming that regurgitation could lead to a decrease in the TMC inseminated, it is possible that regurgitation could lead to lower pregnancy rates in the given cycle. The relationship between IUI sample regurgitation and treatment outcomes have been poorly investigated. The only study to acknowledge regurgitation in humans related to pregnancy outcomes observed with different regurgitation rates using two different insemination catheters, but found no significant difference in pregnancy outcomes ( 8 ). The objective of this study was to determine if the presence of regurgitation during IUI affected pregnancy rates. Our hypothesis was that the presence of regurgitation would be negatively associated with fecundity in females undergoing IUI.