- Split View
-
Views
-
Cite
Cite
J van Rijswijk, M R Caanen, V Mijatovic, C G Vergouw, P M van de Ven, C B Lambalk, R Schats, Immobilization or mobilization after IUI: an RCT, Human Reproduction, Volume 32, Issue 11, November 2017, Pages 2218–2224, https://doi.org/10.1093/humrep/dex302
- Share Icon Share
Abstract
Does 15 min of immobilization after IUI improve pregnancy rates?
Immobilization for 15 min after IUI does not improve pregnancy rates.
Prior RCTs report a beneficial effect of supine immobilization for 15 min following IUI compared to immediate mobilization, however, these studies can be criticized. Given the importance for the logistics in daily practice and the lack of biological plausibility we planned a replication study prior to potential implementation of this procedure.
A single centre RCT, based in an academic setting in the Netherlands, was performed. Participants were randomly assigned for 15 min of supine immobilization following IUI for a maximum of six cycles compared to the standard procedure of immediate mobilization following IUI. Participants and caregivers were not blinded to group assignment. An independent researcher used computer-generated tables to allocate treatments. Stratification occurred to the indication of IUI (unexplained or mild male subfertility). Revelation of allocation took place just before the insemination by the caregiver. The primary outcome was ongoing pregnancy rate per couple.
A total of 498 couples diagnosed with unexplained or mild male subfertility and an indication for treatment with IUI were approached and randomized in the study, of which 244 participants were assigned to 15 min of supine immobilization and 254 participants to immediate mobilization.
Participant characteristics were comparable between the groups, and 236 participants were analysed in the immobilization group, versus 245 in the mobilization group. The ongoing pregnancy rate per couple was not found to be superior in the immobilization group (one-sided P-value = 0.97) with 76/236 ongoing pregnancies (32.2%) being accomplished in the immobilization and 98/245 ongoing pregnancies (40.0%) in the immediate mobilization group (relative risk 0.81; 95% CI [0.63, 1.02], risk difference: −7.8%, 95% CI [−16.4%, 0.8%]). No difference was found in miscarriage rate, multiple gestation rate, live birth rate and time to pregnancy between the groups.
Owing to discontinuation of the planned treatment not all participants reached six IUI cycles or an ongoing pregnancy. However, this is as expected in IUI treatment and mirrors clinical practice. These participants were equally distributed across the two groups. Women with tubal pathology and endocrine disorders were excluded for this trial, and this might narrow generalizability.
This study shows no positive effect of 15 min of immobilization following IUI on pregnancy rates. Based on available evidence today, including our study, a possible beneficial effect of supine immobilization after IUI is at least doubtful and straightforward implementation does not seem to be justified.
No funding was received. All authors have nothing to disclose.
Dutch Trial Register NTR 2418.
20 July 2010.
11 August 2010.
Introduction
Subfertility, defined as the inability to conceive within 12 months of unprotected intercourse, affects 1 in 10 couples that strive for pregnancy (Beurskens et al., 1995). IUI is an established treatment for couples with mild male or idiopathic subfertility. Several aspects of the treatment, such as immediate mobilization or supine immobilization after the insemination, have been investigated in order to find ways to improve pregnancy outcomes. Two RCTs (Saleh et al., 2000) and (Custers et al., 2009) reported a beneficial effect of immobilization for 10–15 min after IUI on pregnancy rate compared to immediate mobilization. Saleh et al. (2000) reported a positive effect on the pregnancy rate of immobilization for 10 min in a small and unbalanced research group of 95 couples (40 versus 55 couples). Custers et al. (2009) found higher pregnancy rates for bedrest for 15 min compared to immediate mobilization in a larger multicentre RCT of 391 couples. However, critical analysis of the literature (Ledger, 2009) stressed serious weaknesses of this study, in particular the lower pregnancy rate than expected and the variation in stimulation protocols among the centres. Recently, Orief et al. (2015) compared 5, 10 and 20 min of immobilization (and not immediate mobilization) in only one IUI treatment cycle and found a higher clinical pregnancy rate 28 days after insemination for immobilization of 10 and 20 min compared to 5 min, but no significant difference between 10 and 20 min of immobilization. No data were reported on the ongoing pregnancy rate.
Above all, a most puzzling question is the underlying fertility enhancing mechanism of immobilization following IUI, which remains unclear. Spermatozoa can reach the Fallopian tube within 5 min after intravaginal insemination (Settlage et al., 1973). Furthermore, they can survive for days in the cervical crypts (Suarez and Pacey, 2006).
In view of the aforementioned deficiencies, we conducted a single centre RCT in an attempt to replicate the results of the previously mentioned studies and confirm the increase in ongoing pregnancies after immobilization of 15 min following a maximum of six IUI treatment cycles compared to immediate mobilization.
Materials and Methods
Study overview
The ‘IUI Up or Down’ study was approved by the ethics committee and institutional review board of the VU University Medical Centre, Amsterdam, The Netherlands (METc VUmc 2010/90). The trial was registered as IUI Up or Down in the Dutch Trial Registry (NTR 2418).
Study participants
Participants were recruited at the VU University Medical Centre, Amsterdam, the Netherlands. All couples with unexplained or mild male subfertility and an indication for treatment with IUI were eligible for enrolment in the study. All couples should have undergone a fertility work-up including at least a hysterosalpingography (HSG). Unexplained subfertility was diagnosed when ovulation was confirmed with a biphasic basal body temperature chart and/or a midluteal progesterone; an HSG and/or laparoscopy confirmed bilateral tubal patency; and a semen analysis after processing during fertility work-up gave a total motile sperm count (TMC) of >10 million spermatozoa. In case of mild male factor, the TMC was <10 million and >2 million spermatozoa during fertility work-up. Couples were excluded when women required an insemination with cryo-preserved semen or semen retrieved after flushing of the bladder; women were diagnosed with endocrine disorders such as polycystic ovary syndrome (PCOS) or premature ovarian insufficiency (characterized by basal FSH > 12 IU IU/l); and/or women had a medical history of trachelectomy because of cervical cancer.
Randomization and study intervention
Eligible couples were informed about the study by their gynaecologist or fertility physician and received written information. After giving written informed consent, the participants were randomly assigned in a 1:1 ratio to six cycles of IUI followed by immediate mobilization or six cycles of IUI followed by 15 min of supine immobilization before the first insemination. The randomization was performed by an independent researcher using computer-generated random table numbers, with a block size of 20. Stratification occurred according to the indication for IUI, unexplained subfertility or a mild male factor. A sealed envelope containing the allocated treatment was attached to the medical file of the participant and opened immediately before the first insemination by the gynaecologist or fertility physician performing the IUI. Participants and caregivers were not blinded to group assignment. If couples decided not to participate in this study, IUI was performed in accordance with the standard procedure, which was immediate mobilization following insemination.
All participants underwent the IUI procedure according to local protocol. A baseline transvaginal ultrasound (TVU) was performed on Day 3 of the menstrual cycle to rule out pre-existing ovarian cyst(s). A second TVU was performed approximately on Day 10 of the cycle to monitor follicular growth. Subsequent TVUs were performed at 1- to 2-day intervals until the mean diameter of the dominant follicle reached ≥18 mm. An injection of 10 000 IU hCG (Pregnyl®, Organon) was given to trigger ovulation. The insemination was carried out 40–42 h after the administration of hCG.
The first three IUI treatments were carried out in natural cycles followed by three treatments in stimulated cycles. This was the local standard procedure assuming a strategy to prevent, to some extent, multiple pregnancies. In the stimulated cycles, participants used recombinant or urinary FSH (Gonal F®, Serano Benelux, The Hague, Netherlands; Menopur®, Ferring, Hoofddorp, Netherlands) from Day 3 onwards. If more than three follicles of >18 mm, or 5 of >14 mm diameter, or an oestradiol serum concentration of >3000 pmol/l were present, the treatment was cancelled.
Semen processing
A sperm sample was produced predominantly in the clinic by masturbation. Spoken and written instructions about the collection (and in case of production at home, the transport) of the semen sample were given in advance. The time between semen production and processing was up to 1 h. Semen processing started after liquefaction with a centrifugation step using a density-gradient medium (70% Puresperm®, Nidacon and 30% human tubal fluid (HTF) hepes with 4 mg/ml human serum albumin (HSA; Albuman, Sanquin, Gynotec). Subsequently, two washing steps were performed in HTF (Gynotec) with 4 mg/ml HSA: one directly after the density-gradient step and one prior to insemination. Between the two washing steps, the semen was stored at room temperature in an ultra violet light repellent box in 5% CO2. A Makler chamber was used to count the number of spermatozoa.
Objectives and outcomes
The primary outcome of the study was ongoing pregnancy rate per couple, defined as an intrauterine pregnancy showing a normally developing foetus with a positive heartbeat seen by TVU 10 weeks following IUI. The secondary outcomes included multiple gestation rate per ongoing pregnancy, miscarriage rate per couple, live birth rate and time to pregnancy. A miscarriage was defined as the presence of non-vitality on ultrasound or spontaneous loss of pregnancy within 12 weeks gestation. Pregnancy was confirmed by serum concentration of β-hCG if no menstruation occurred 14 days after insemination. Lifestyle factors, such as BMI and smoking habits, were recorded.
Sample size calculation
In VU University Medical Centre the ongoing pregnancy rate with IUI treatment is 8% per cycle, independent of cycle number. Couples may undergo IUI for a maximum of six cycles unless pregnancy occurs, and the drop out percentage is ~5%.
The cumulative ongoing pregnancy rate according to life table analysis is around 48%. The alternative hypothesis in our study was that the claimed beneficial effect of immobilization by the previous studies is true (Saleh et al., 2000 and Custers et al., 2009). The previous study by Custers et al. (2009) suggested a potential increase of 2–3% per cycle of the occurrence of an ongoing pregnancy. With a maximum of six treatment cycles, a potential of 60% for the cumulative ongoing pregnancy rate may be yielded. In order to achieve 80% power to detect an absolute increase of 12% in cumulative ongoing pregnancy rate due to immobilization compared to immediate mobilization at a one-sided significance level of 5%, 229 couples needed to be included per study arm. Given an inclusion and completion rate of 90%, ~500 couples need to be approached. An interim analysis was not performed, as the study compares two regular interventions and no serious adverse events were anticipated.
Statistical analysis
Categorical data were reported as frequencies and percentages. Normally distributed continuous variables were summarized as the mean with SD, and non-normally distributed continuous variables were reported as median with interquartile range (IQR). All data were analysed according to the intention-to-treat principle. The primary analysis was performed by a Z-test comparing the proportion of ongoing pregnancies in the immobilization group to that in the immediate mobilization group at a one-sided significant level of 5%, and rejecting only in case of a significantly higher proportion of pregnancies in the immobilization group. For comparison of secondary outcomes a two-sided significance level of 5% was used. Chi-square test was used to test for a difference in secondary binary outcomes between the arms. Continuous outcomes were compared between arms using an independent T-test or Mann–Whitney U-test as appropriate. A discrete-time-event-survival analysis was conducted to compare the number of treatment cycles to ongoing pregnancy between the arms. A logistic regression model was used for the discrete-time survival analysis with the number of the treatment cycles and treatment arm as a categorical predictor and ongoing pregnancy as the outcome. Relative Risks (RR) and risk differences (RD) were calculated for the primary and other binary outcome measures together with their 95% CIs. All RR and odds ratios (OR) presented are for the immobilization arm relative to immediate mobilization arm.
All statistical analyses were performed in SPSS version 22.0 (IBM Corp., NY, USA).
Results
Study participants
Between August 2010 and May 2014, 498 couples with idiopathic or mild male subfertility and an indication for IUI were randomized in this study. Of these, 244 participants were assigned to 15 min of supine immobilization, and 254 participants to immediate mobilization. Two participants in the mobilization group did not undergo insemination owing to poor TMC after processing before the IUI (<0.5 million spermatozoa). Eight couples in the immobilization group and nine couples in the mobilization group did not meet the inclusion and exclusion criteria after reassessment, and they were excluded from analyses (Fig. 1). In the immobilization group, 89 couples did not reach a total of six IUI cycles or an ongoing pregnancy versus 80 couples in the mobilization group. Reasons for discontinuing treatment were advanced start of IVF, a naturally conceived pregnancy during the course of the study, and relationship problems or moving house. The follow-up rate with regard to the primary outcome, ongoing pregnancy following an insemination, was 100% in both groups.
The baseline characteristics were comparable between the two groups (Table I). The median female age at randomization was 35.1 years (IQR 31.9–37.9) in the immobilization group and 35.5 years (IQR 32.3–38.4) in the mobilization group. In 47 couples of the immobilization group, subfertility was based on a mild male factor versus 54 couples in the mobilization group. The median number of IUI cycles in both groups was four (IQR 3–6). The total amount of IUI cycles in the immobilization group was 950 (539 (56.7%) natural cycles) versus 984 IUI cycles in the mobilization group (581 (59.0%) natural cycles). The distribution of TMC values in the immobilization group (median 18.0, IQR 6.58–40.0) was comparable with the distribution in the mobilization group (median 16.9, IQR 6.6–39.0). Multifollicular growth was comparable between the groups as well.
Descriptive characteristic . | Immobilization* . | Mobilization* . |
---|---|---|
Number of participants | 236 | 245 |
Female age at randomization (years) | 35.1 (31.9–37.9) | 35.5 (32.3–38.4) |
Maternal BMI (kg/m2) | 22.6 (20.3–25.9) | 22.1 (20.3–24.8) |
Maternal ethnicity | ||
Caucasian | 173 (73.3) | 169 (69.0) |
Indian, Pakistani | 3 (1.3) | 8 (3.3) |
African (sub-Sahara) | 3 (1.3) | 3 (1.2) |
North-African | 10 (4.2) | 16 (6.5) |
Asian | 8 (3.4) | 14 (5.7) |
Surinam | 1 (0.4) | 3 (1.2) |
Turkish | 4 (1.7) | 5 (2.0) |
Other | 18 (7,6) | 8 (3.3) |
Unknown | 16 (6.8) | 19 (7.8) |
Maternal smoking (cigarettes/day) | ||
Smoker | 38 (16.1) | 35 (14.3) |
Non-smoker | 189 (80.1) | 202 (82.4) |
Unknown | 9 (3.8) | 8 (3.3) |
Subfertility | ||
Primary | 134 (56.8) | 129 (52.7) |
Secondary | 101 (42.8) | 114 (46.5) |
Duration of subfertility (years) | 2.4 (1.8–3.6) | 2.1 (1.6–3.0) |
Infertility based on | ||
Idiopathic | 189 (80.1) | 191 (78.0) |
Mild male factor | 47 (19.9) | 54 (22.0) |
Number of cycles per couple | 4 (3–6) | 4 (3–6) |
Characteristics cycles | ||
Number of cycles | 950 (49.1) | 984 (50.9) |
Spontaneous cycle | 539 (56.7) | 581 (59.0) |
Use of COH | 411 (43.3) | 403 (41.0) |
TMC | 18.0 (6.8–40.0) | 16.9 (6.6–39.0) |
Follicle growth | ||
1 follicle >14 mm | 734 (77.3) | 761 (77.3) |
>1 follicle >14 mm | 210 (22.1) | 217 (22.1) |
Unknown | 6 (0.6) | 6 (0.6) |
Descriptive characteristic . | Immobilization* . | Mobilization* . |
---|---|---|
Number of participants | 236 | 245 |
Female age at randomization (years) | 35.1 (31.9–37.9) | 35.5 (32.3–38.4) |
Maternal BMI (kg/m2) | 22.6 (20.3–25.9) | 22.1 (20.3–24.8) |
Maternal ethnicity | ||
Caucasian | 173 (73.3) | 169 (69.0) |
Indian, Pakistani | 3 (1.3) | 8 (3.3) |
African (sub-Sahara) | 3 (1.3) | 3 (1.2) |
North-African | 10 (4.2) | 16 (6.5) |
Asian | 8 (3.4) | 14 (5.7) |
Surinam | 1 (0.4) | 3 (1.2) |
Turkish | 4 (1.7) | 5 (2.0) |
Other | 18 (7,6) | 8 (3.3) |
Unknown | 16 (6.8) | 19 (7.8) |
Maternal smoking (cigarettes/day) | ||
Smoker | 38 (16.1) | 35 (14.3) |
Non-smoker | 189 (80.1) | 202 (82.4) |
Unknown | 9 (3.8) | 8 (3.3) |
Subfertility | ||
Primary | 134 (56.8) | 129 (52.7) |
Secondary | 101 (42.8) | 114 (46.5) |
Duration of subfertility (years) | 2.4 (1.8–3.6) | 2.1 (1.6–3.0) |
Infertility based on | ||
Idiopathic | 189 (80.1) | 191 (78.0) |
Mild male factor | 47 (19.9) | 54 (22.0) |
Number of cycles per couple | 4 (3–6) | 4 (3–6) |
Characteristics cycles | ||
Number of cycles | 950 (49.1) | 984 (50.9) |
Spontaneous cycle | 539 (56.7) | 581 (59.0) |
Use of COH | 411 (43.3) | 403 (41.0) |
TMC | 18.0 (6.8–40.0) | 16.9 (6.6–39.0) |
Follicle growth | ||
1 follicle >14 mm | 734 (77.3) | 761 (77.3) |
>1 follicle >14 mm | 210 (22.1) | 217 (22.1) |
Unknown | 6 (0.6) | 6 (0.6) |
*Values are n with percentage or median values with interquartile range.
COH, controlled ovarian hyperstimulation; TMC, total motile sperm count.
Descriptive characteristic . | Immobilization* . | Mobilization* . |
---|---|---|
Number of participants | 236 | 245 |
Female age at randomization (years) | 35.1 (31.9–37.9) | 35.5 (32.3–38.4) |
Maternal BMI (kg/m2) | 22.6 (20.3–25.9) | 22.1 (20.3–24.8) |
Maternal ethnicity | ||
Caucasian | 173 (73.3) | 169 (69.0) |
Indian, Pakistani | 3 (1.3) | 8 (3.3) |
African (sub-Sahara) | 3 (1.3) | 3 (1.2) |
North-African | 10 (4.2) | 16 (6.5) |
Asian | 8 (3.4) | 14 (5.7) |
Surinam | 1 (0.4) | 3 (1.2) |
Turkish | 4 (1.7) | 5 (2.0) |
Other | 18 (7,6) | 8 (3.3) |
Unknown | 16 (6.8) | 19 (7.8) |
Maternal smoking (cigarettes/day) | ||
Smoker | 38 (16.1) | 35 (14.3) |
Non-smoker | 189 (80.1) | 202 (82.4) |
Unknown | 9 (3.8) | 8 (3.3) |
Subfertility | ||
Primary | 134 (56.8) | 129 (52.7) |
Secondary | 101 (42.8) | 114 (46.5) |
Duration of subfertility (years) | 2.4 (1.8–3.6) | 2.1 (1.6–3.0) |
Infertility based on | ||
Idiopathic | 189 (80.1) | 191 (78.0) |
Mild male factor | 47 (19.9) | 54 (22.0) |
Number of cycles per couple | 4 (3–6) | 4 (3–6) |
Characteristics cycles | ||
Number of cycles | 950 (49.1) | 984 (50.9) |
Spontaneous cycle | 539 (56.7) | 581 (59.0) |
Use of COH | 411 (43.3) | 403 (41.0) |
TMC | 18.0 (6.8–40.0) | 16.9 (6.6–39.0) |
Follicle growth | ||
1 follicle >14 mm | 734 (77.3) | 761 (77.3) |
>1 follicle >14 mm | 210 (22.1) | 217 (22.1) |
Unknown | 6 (0.6) | 6 (0.6) |
Descriptive characteristic . | Immobilization* . | Mobilization* . |
---|---|---|
Number of participants | 236 | 245 |
Female age at randomization (years) | 35.1 (31.9–37.9) | 35.5 (32.3–38.4) |
Maternal BMI (kg/m2) | 22.6 (20.3–25.9) | 22.1 (20.3–24.8) |
Maternal ethnicity | ||
Caucasian | 173 (73.3) | 169 (69.0) |
Indian, Pakistani | 3 (1.3) | 8 (3.3) |
African (sub-Sahara) | 3 (1.3) | 3 (1.2) |
North-African | 10 (4.2) | 16 (6.5) |
Asian | 8 (3.4) | 14 (5.7) |
Surinam | 1 (0.4) | 3 (1.2) |
Turkish | 4 (1.7) | 5 (2.0) |
Other | 18 (7,6) | 8 (3.3) |
Unknown | 16 (6.8) | 19 (7.8) |
Maternal smoking (cigarettes/day) | ||
Smoker | 38 (16.1) | 35 (14.3) |
Non-smoker | 189 (80.1) | 202 (82.4) |
Unknown | 9 (3.8) | 8 (3.3) |
Subfertility | ||
Primary | 134 (56.8) | 129 (52.7) |
Secondary | 101 (42.8) | 114 (46.5) |
Duration of subfertility (years) | 2.4 (1.8–3.6) | 2.1 (1.6–3.0) |
Infertility based on | ||
Idiopathic | 189 (80.1) | 191 (78.0) |
Mild male factor | 47 (19.9) | 54 (22.0) |
Number of cycles per couple | 4 (3–6) | 4 (3–6) |
Characteristics cycles | ||
Number of cycles | 950 (49.1) | 984 (50.9) |
Spontaneous cycle | 539 (56.7) | 581 (59.0) |
Use of COH | 411 (43.3) | 403 (41.0) |
TMC | 18.0 (6.8–40.0) | 16.9 (6.6–39.0) |
Follicle growth | ||
1 follicle >14 mm | 734 (77.3) | 761 (77.3) |
>1 follicle >14 mm | 210 (22.1) | 217 (22.1) |
Unknown | 6 (0.6) | 6 (0.6) |
*Values are n with percentage or median values with interquartile range.
COH, controlled ovarian hyperstimulation; TMC, total motile sperm count.
Outcomes
No increase in the proportion of ongoing pregnancies was found when 15 min of supine immobilization followed IUI rather than immediate mobilization, with ongoing pregnancies occurring in 76/236 (32.2%) couples in the immobilization group and 98/245 (40.0%) in the mobilization group (RR 0.81; 95% CI [0.63,1.02], RD: −7.8%, 95% CI [−16.4%, 0.8%]) (Table II). The one-sided P-value for the alternative hypothesis that immobilization resulted in a higher proportion of pregnancies was 0.97. Note that in case a two-sided alternative hypothesis had been specified, rather than the one-sided hypothesis specified in the protocol, the conclusion would have been that the proportion of pregnancies did not differ between the groups (P = 0.075).
Outcome per couple . | Immobilization . | Mobilization . | All . | RR . | P-value . |
---|---|---|---|---|---|
Number of participants ITT1 | 236 | 245 | 481 | ||
Ongoing pregnancy2 | 76 (32.2) | 98 (40.0) | 174 (36.2) | 0.81 (0.63–1.02) | 0.973 |
Couples with at least one miscarriage | 24 (10.2) | 30 (12.2) | 54 (11.2) | 0.83 (0.50–1.38) | 0.474 |
Multiple gestation/ongoing pregnancy5 | 6 (7.9) | 4 (4.1) | 10 (5.7) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 93 (39.4) | 121 (49.4) | 214 (44.5) | 0.80 (0.65–0.98) | 0.034 |
Live births (≥24+0) | 73 (30.9) | 92 (37.6) | 165 (34.3) | 0.82 (0.64–1.06) | 0.134 |
Foetal loss 12–24+0 | 1 (1.3) | 4 (4.1) | 5 (2.9) | 0.32 (0.04–2.83) | 0.314 |
Missing6 | 2 (2.6) | 2 (2.0) | 4 (2.3) | ||
Number of babies | 81 | 97 | 178 | ||
Number of participants PPA7 | 147 | 163 | 310 | ||
Ongoing pregnancy | 69 (46.9) | 90 (55.2) | 159 (51.3) | 0.85 (0.68–1.06) | 0.943 |
Couples with at least one miscarriage | 21 (14.3) | 24 (14.7) | 45 (14.5) | 0.97 (0.56–1.67) | 0.914 |
Multiple gestation/ongoing pregnancy | 6 (8.7) | 4 (4.4) | 10 (6.3) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 83 (56.5) | 107 (65.6) | 190 (61.3) | 0.86 (0.72–1.03) | 0.104 |
Live births (≥24+0) | 67 (45.6) | 85 (52.1) | 152 (49.0) | 0.87 (0.69–1.10) | 0.254 |
Missing6 | 1 | 1 | 2 | ||
Foetal loss 12–24+0 | 1 (1.4) | 4 (4.4) | 5 (3.1) | 0.33 (0.04–2.85) | 0.314 |
Number of babies | 75 | 90 |
Outcome per couple . | Immobilization . | Mobilization . | All . | RR . | P-value . |
---|---|---|---|---|---|
Number of participants ITT1 | 236 | 245 | 481 | ||
Ongoing pregnancy2 | 76 (32.2) | 98 (40.0) | 174 (36.2) | 0.81 (0.63–1.02) | 0.973 |
Couples with at least one miscarriage | 24 (10.2) | 30 (12.2) | 54 (11.2) | 0.83 (0.50–1.38) | 0.474 |
Multiple gestation/ongoing pregnancy5 | 6 (7.9) | 4 (4.1) | 10 (5.7) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 93 (39.4) | 121 (49.4) | 214 (44.5) | 0.80 (0.65–0.98) | 0.034 |
Live births (≥24+0) | 73 (30.9) | 92 (37.6) | 165 (34.3) | 0.82 (0.64–1.06) | 0.134 |
Foetal loss 12–24+0 | 1 (1.3) | 4 (4.1) | 5 (2.9) | 0.32 (0.04–2.83) | 0.314 |
Missing6 | 2 (2.6) | 2 (2.0) | 4 (2.3) | ||
Number of babies | 81 | 97 | 178 | ||
Number of participants PPA7 | 147 | 163 | 310 | ||
Ongoing pregnancy | 69 (46.9) | 90 (55.2) | 159 (51.3) | 0.85 (0.68–1.06) | 0.943 |
Couples with at least one miscarriage | 21 (14.3) | 24 (14.7) | 45 (14.5) | 0.97 (0.56–1.67) | 0.914 |
Multiple gestation/ongoing pregnancy | 6 (8.7) | 4 (4.4) | 10 (6.3) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 83 (56.5) | 107 (65.6) | 190 (61.3) | 0.86 (0.72–1.03) | 0.104 |
Live births (≥24+0) | 67 (45.6) | 85 (52.1) | 152 (49.0) | 0.87 (0.69–1.10) | 0.254 |
Missing6 | 1 | 1 | 2 | ||
Foetal loss 12–24+0 | 1 (1.4) | 4 (4.4) | 5 (3.1) | 0.33 (0.04–2.85) | 0.314 |
Number of babies | 75 | 90 |
1ITT, Intention-to-treat analysis. 2Ongoing pregnancy including natural conceived pregnancies during the course of the study (seven immobilization versus eight mobilization P-value 0.85). 3Z-test for proportions. 4Chi-square test. 5One quadruplet in the immobilization group, one triplet in the mobilization group. 6The number of ongoing pregnancies of which the pregnancy outcome is unknown. 7PPA, per protocol analysis. The number of participants who reached a total of six IUI cycles or an ongoing pregnancy.
Values are n with percentage or relative risk (RR) with 95% CI.
Outcome per couple . | Immobilization . | Mobilization . | All . | RR . | P-value . |
---|---|---|---|---|---|
Number of participants ITT1 | 236 | 245 | 481 | ||
Ongoing pregnancy2 | 76 (32.2) | 98 (40.0) | 174 (36.2) | 0.81 (0.63–1.02) | 0.973 |
Couples with at least one miscarriage | 24 (10.2) | 30 (12.2) | 54 (11.2) | 0.83 (0.50–1.38) | 0.474 |
Multiple gestation/ongoing pregnancy5 | 6 (7.9) | 4 (4.1) | 10 (5.7) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 93 (39.4) | 121 (49.4) | 214 (44.5) | 0.80 (0.65–0.98) | 0.034 |
Live births (≥24+0) | 73 (30.9) | 92 (37.6) | 165 (34.3) | 0.82 (0.64–1.06) | 0.134 |
Foetal loss 12–24+0 | 1 (1.3) | 4 (4.1) | 5 (2.9) | 0.32 (0.04–2.83) | 0.314 |
Missing6 | 2 (2.6) | 2 (2.0) | 4 (2.3) | ||
Number of babies | 81 | 97 | 178 | ||
Number of participants PPA7 | 147 | 163 | 310 | ||
Ongoing pregnancy | 69 (46.9) | 90 (55.2) | 159 (51.3) | 0.85 (0.68–1.06) | 0.943 |
Couples with at least one miscarriage | 21 (14.3) | 24 (14.7) | 45 (14.5) | 0.97 (0.56–1.67) | 0.914 |
Multiple gestation/ongoing pregnancy | 6 (8.7) | 4 (4.4) | 10 (6.3) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 83 (56.5) | 107 (65.6) | 190 (61.3) | 0.86 (0.72–1.03) | 0.104 |
Live births (≥24+0) | 67 (45.6) | 85 (52.1) | 152 (49.0) | 0.87 (0.69–1.10) | 0.254 |
Missing6 | 1 | 1 | 2 | ||
Foetal loss 12–24+0 | 1 (1.4) | 4 (4.4) | 5 (3.1) | 0.33 (0.04–2.85) | 0.314 |
Number of babies | 75 | 90 |
Outcome per couple . | Immobilization . | Mobilization . | All . | RR . | P-value . |
---|---|---|---|---|---|
Number of participants ITT1 | 236 | 245 | 481 | ||
Ongoing pregnancy2 | 76 (32.2) | 98 (40.0) | 174 (36.2) | 0.81 (0.63–1.02) | 0.973 |
Couples with at least one miscarriage | 24 (10.2) | 30 (12.2) | 54 (11.2) | 0.83 (0.50–1.38) | 0.474 |
Multiple gestation/ongoing pregnancy5 | 6 (7.9) | 4 (4.1) | 10 (5.7) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 93 (39.4) | 121 (49.4) | 214 (44.5) | 0.80 (0.65–0.98) | 0.034 |
Live births (≥24+0) | 73 (30.9) | 92 (37.6) | 165 (34.3) | 0.82 (0.64–1.06) | 0.134 |
Foetal loss 12–24+0 | 1 (1.3) | 4 (4.1) | 5 (2.9) | 0.32 (0.04–2.83) | 0.314 |
Missing6 | 2 (2.6) | 2 (2.0) | 4 (2.3) | ||
Number of babies | 81 | 97 | 178 | ||
Number of participants PPA7 | 147 | 163 | 310 | ||
Ongoing pregnancy | 69 (46.9) | 90 (55.2) | 159 (51.3) | 0.85 (0.68–1.06) | 0.943 |
Couples with at least one miscarriage | 21 (14.3) | 24 (14.7) | 45 (14.5) | 0.97 (0.56–1.67) | 0.914 |
Multiple gestation/ongoing pregnancy | 6 (8.7) | 4 (4.4) | 10 (6.3) | 1.93 (0.57–6.61) | 0.294 |
Couples with at least one clinical pregnancy | 83 (56.5) | 107 (65.6) | 190 (61.3) | 0.86 (0.72–1.03) | 0.104 |
Live births (≥24+0) | 67 (45.6) | 85 (52.1) | 152 (49.0) | 0.87 (0.69–1.10) | 0.254 |
Missing6 | 1 | 1 | 2 | ||
Foetal loss 12–24+0 | 1 (1.4) | 4 (4.4) | 5 (3.1) | 0.33 (0.04–2.85) | 0.314 |
Number of babies | 75 | 90 |
1ITT, Intention-to-treat analysis. 2Ongoing pregnancy including natural conceived pregnancies during the course of the study (seven immobilization versus eight mobilization P-value 0.85). 3Z-test for proportions. 4Chi-square test. 5One quadruplet in the immobilization group, one triplet in the mobilization group. 6The number of ongoing pregnancies of which the pregnancy outcome is unknown. 7PPA, per protocol analysis. The number of participants who reached a total of six IUI cycles or an ongoing pregnancy.
Values are n with percentage or relative risk (RR) with 95% CI.
In the immobilization group, seven naturally conceived ongoing pregnancies occurred during the course of the study, compared to eight in the mobilization group. The per protocol analysis also showed no significant increase in ongoing pregnancies as a result of immobilization (69/147 (46.9%) versus 90/163 (55.2%) RR 0.85; CI 95% [0.68, 1.06], RD −8.3% 95% CI [−19.4%, 2.8%], one-sided P-value 0.94). In the immobilization group, 30 of the 69 ongoing pregnancies were established after an IUI treatment in a natural cycle (43.5%). In the mobilization group, 43 of the 90 ongoing pregnancies were established after an IUI treatment in a natural cycle (47.8%).
Regarding secondary outcomes, there was no significant difference in multiple gestation rate per ongoing pregnancy between the groups (6/76 (7.9%) in the immobilization group versus 4/98 (4.1%) in the mobilization group, RR 1.93; 95% CI [0.57,6.61], RD 3.8% 95% CI [−3.4%, 11.0%], two-sided P-value 0.29). One quadruplet occurred in the immobilization group versus one triplet in the mobilization group. Miscarriages occurred in 24/236 (10.2%) of the couples in the immobilization group compared to 30/245 (12.2%) in the mobilization group (RR 0.83; 95% CI [0.50,1.38], RD -2.1%, 95% CI [−7.7%, 3.6%], two-sided P-value 0.47).
A live birth beyond 24 weeks gestational age occurred in 73 (30.9%) of the couples in the immobilization group versus 92 (37.6%) in the mobilization group, this was not significantly different (RR 0.82, 95% CI [0.64,1.06], RD −6.6% 95% CI [−15.1%, 1.8%], two-sided P-value 0.13).
The discrete-time-event-survival analysis showed no significant difference in the number of cycles to ongoing pregnancy between the mobilization and immobilization group (OR 0.77; 95% CI [0.56, 1.06], two-sided P-value 0.11) (Fig. 2). The cumulative percentage of ongoing pregnancy rate per cycle is presented in a Kaplan–Meier survival curve (Supplementary Figure S1).
No adverse events have occurred.
Discussion
This study, evaluating ongoing pregnancy rates between supine immobilization for 15 min and immediate mobilization following IUI, showed no benefit of immobilization on ongoing pregnancy rate over immediate mobilization. Multiple pregnancies, miscarriages, live birth rates and time to ongoing pregnancy were comparable between the two groups.
Our results are not in line with previous studies (Saleh et al., 2000; Custers et al., 2009) showing significantly higher ongoing pregnancy rates after immobilization following IUI.
Although intuitively supine immobilization may be of benefit to conception, the biological plausibility of this remains obscure. What we know from earlier studies is that sperm cells can reach the Fallopian tube within 5 min after intravaginal insemination. Sperm migration through the cervical canal is accomplished by the spermatozoon itself, but above the internal os sperm ascent is the result of both active swimming and passive transport by uterine contractions (Settlage et al., 1973). It is unclear whether gravity affects this movement, let alone immobilization of 15 min. Besides, spermatozoa can survive for days in the cervical crypts, retaining their motility and viability (Suarez and Pacey, 2006). In a different situation, namely with embryo transfer during IVF treatment, immobilization does not seem to affect pregnancy rates or the position of the transferred embryo (Sharif et al., 1998; Purcell et al., 2007; Lambers et al., 2009). Finally, there is insufficient proof that a prolonged supine position after intercourse promotes the chances of pregnancy.
Given this questionable plausibility, several matters concerning the previous studies reporting an effect of immobilization merit attention. One remarkable aspect of the other large RCT (Custers et al., 2009) was that in a recently published follow-up study, the supine immobilization group showed a persisting significantly higher pregnancy rate after subsequent fertility treatment or natural conception over a period of 3 years (Scholten et al., 2014), suggesting that some differences in unrecorded and/or unobserved baseline characteristics between the compared groups may have been present despite adequate randomization. This is an unfortunate but impossible to completely rule out problem with otherwise optimally designed RCTs. Further differences between the current study and the other reported RCTs are the single centre design, which circumvented unwanted effects of between-centre differences caused by variation of stimulation strategies and laboratory practices (Ledger, 2009), and our evaluation of the effect of (the generally considered as standard) six IUI cycles instead of three (Saleh et al., 2000; Custers et al., 2009). The post hoc per protocol analysis of our data after three IUI cycles also showed no increase in ongoing pregnancy rate when supine immobilization followed IUI compared to immediate mobilization: (34/147 (23.1%) versus 58/163 (35.6%) respectively, RR 0.61; 95% CI 0.41–0.89, one-sided P-value 0.99). The post hoc intention-to-treat analysis of our data after three IUI cycles showed a similar result (37/236 (15.7%) versus 59/245 (24.1%), RR 0.65; 95% CI 0.45–0.94, one-sided P-value 0.99).
Strengths and limitations
The current adequately powered RCT was designed to confirm previous findings and avoid certain short-comings of these studies. In this particular case, a monocentre approach seemed optimal and furthermore cumulative ongoing pregnancy rates after maximally six cycles would agree more with common practice.
Not all participants reached six IUI cycles or an ongoing pregnancy because of discontinuation of the planned treatment (for social reasons, treatment alteration). However, this is as expected in IUI treatment and it mirrors clinical practice. These participants were equally distributed across the two groups and the mean number of cycles per couple did not differ.
The study was not blinded for participants and physicians. Obviously, blinding is impossible for mobilization or immobilization after IUI. A further limitation is that we studied subfertile women without tubal pathology, endocrinological disorders such as PCOS or premature ovarian insufficiency, and women not needing insemination with cryo-preserved semen or semen retrieved after flushing of the bladder. This might narrow the generalizability of our findings to these couples.
Furthermore, we carried out the first three IUI cycles unmedicated and the next three with injectables (our local standard protocol during the course of the study), which is slightly different from standard care at the moment i.e. IUI with oral agents for 3–4 cycles. (Diamond et al., 2015; Reindollar et al., 2010).
Some may regard the choice of a one-sided alternative hypothesis for our primary outcome as a short-coming of our study, given that two-sided hypotheses are preferred in RCTs which are only regarded ethical under clinical equipoise. Our one-sided alternative hypothesis as it was described in the original protocol was chosen in such a way that rejection of the null hypothesis supported the claim made by Custers et al. (2009). An alternative option would have been to use a two-sided testing procedure and evaluate whether conclusions of this trial were in line with this previous study. By doing so we again did not find a difference between the groups and clearly not a benefical effect of immobilization. Rather, there seems to be a trend towards a beneficial effect of immediate mobilization in the course of the IUI treatment cycles (Fig. 2 and Figure Supplementary S1) but formally (i.e.) we could not come to this conclusion, as the sample size is based on six cycles. Nevertheless, if this is a real effect then patients will benefit if immediate mobilization were to replace ineffective supine immobilization as the standard operation procedure post IUI as a consequence of the findings of our study.
Post-coital body position and fertility has always been a matter for lively discussion. However, although tempting, we stress that our findings cannot simply be extrapolated to natural conception.
Conclusion
Based on available evidence today, including our study, a possible beneficial effect of supine immobilization after IUI is at least doubtful. It has substantial impact on the work-flow and clinic occupancy. During the time that one IUI treatment is carried out in an examination room in ‘an immobilization setting’, two IUI treatments can be performed in the same room in ‘a mobilization setting’. Consequently, this has financial implications. Given this uncertainty, whether or not to implement this procedure should await the results of a meta-analysis, preferably of individual patient data from the currently executed trials.
This work was presented as an oral presentation at the 32ndAnnual Meeting of the European Society of Human Reproduction and Embryology 2016, Helsinki, Finland.
Supplementary data
Supplementary data are available at Human Reproduction online
Acknowledgements
First, we thank all couples who participated in the study. Also, we thank all contributors of the fertility clinic of the VU University Medical Centre for their excellent logistical assistance in this study. We gratefully acknowledge Y. Ammi for helping with the data collection.
Authors’ roles
R.S. and C.B.L. designed the study. R.S., J.v.R., M.R.C., C.B.L., C.G.V. and V.M. included couples. J.v.R. and M.R.C. co-ordinated this randomized controlled trial and collected the data. J.v.R., M.R.C. and P.M.v.d.V. conducted the analyses, under supervision of R.S. and C.B.L. All authors had full access to all of the data in the study, they critically revised and approved this version of the manuscript for publication.
Funding
No funding was received.
Conflict of interest
All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.