The effect of early administration of rectal progesterone in IVF/ICSI twin pregnancies on the preterm birth rate: a randomized trial.

BACKGROUND: The rate of multiple pregnancies in IVF/ICSI ranges from 20 to 30%. The incidence of preterm birth in multiple pregnancies is as high as 60% and is even higher in pregnancies conceived after IVF & ICSI. The effect of progesterone on prevention of preterm birth in twins is controversial. Our group has proven a positive effect in reduction of preterm birth, by starting progesterone from the mid-trimester, in exclusively IVF/ICSI singleton pregnancies but not twins. The purpose of our current study was to explore the effect of earlier administration of natural progesterone, in IVF/ICSI twin pregnancies starting at 11-14 weeks for prevention of preterm birth. METHODS: This is a double-blind, placebo controlled, single center, randomized clinical trial. Women with dichorionic twin gestations, having an IVF/ICSI trial were randomized to receive natural rectal progesterone (800 mg daily) vs placebo, starting early from 11 to 14 weeks. They also received oral and vaginal antimicrobial agents as part of our routine treatment for vaginitis and urinary tract infection. They were randomized regardless of cervical length and had no previous history of preterm birth or known Mullerian anomalies. The primary outcome was spontaneous preterm birth rate before 37 weeks. The secondary outcome was; spontaneous preterm birth before 34, 32, 28 weeks and neonatal outcome. RESULTS: A total of 203 women were randomized to both groups, final analysis included 199 women as 4 were lost to follow up. The base line characteristics as well as gestational age at delivery were not significantly different between the study and the placebo group (34.7 ± 3.6 vs 34.5 ± 4.5, P = 0.626). Progesterone administration was not associated with a significant decrease in the spontaneous preterm birth rates before 37 weeks (73.5% vs 68%, P = 0.551), before 34 (20.6% vs 21.6%, P = 0.649), before 32 (8.8% vs 12.4%, P = 0.46) & before 28 (4.9% vs 3.1%, P = 0.555) weeks. CONCLUSIONS: Rectal natural progesterone starting from the first trimester in IVF/ICSI twin pregnancies did not reduce spontaneous preterm birth. TRIAL REGISTRATION: The trial was registered on 31 January 2014 at www.ISRCTN.com, number 69810120.

Increased insulin resistance in men with unexplained infertility.

This prospective case-control study aimed to test the presence of insulin resistance (IR) in men with unexplained infertility. We included two groups: the study group including 160 infertile men with unexplained oligozoospermia (sperm count <10 × 106/ml) and normal hormonal profile, and the control group of 79 men with proven fertility within the preceding year. A fasting blood test measured IR, FSH, LH, total cholesterol, low-density lipoprotein, high-density lipoprotein and triglycerides. Insulin level was significantly higher in the study group (13.67 ± 10.44) compared with the control group (5.46 ± 3.15), P < 0.0001, and IR was significantly higher in the study group, P < 0.0001. FSH was significantly (P < 0.0001) higher in the study group (4.71 ± 2.57) than the control group (3.15 ± 1.92). LH was significantly higher in the study group (4.98 ± 2.41) compared with the control group (3.15 ± 1.12), P < 0.0001. Total cholesterol was significantly higher in the study group (198.29 ± 37.52) than the control group (182.45 ± 35.92), P < 0.05. In conclusion, IR in men with unexplained infertility may be a cause of reproductive and metabolic abnormalities. The benefit of insulin-sensitizing agents for these patients should be tested.

GnRH agonist plus vaginal progesterone for luteal phase support in ICSI cycles: a randomized study.

In this prospective randomized study, the effect of daily gonadotrophin-releasing hormone agonist (GnRHa) in the luteal phase on IVF and intracytoplasmic sperm injection (ICSI) outcomes was assessed. Women (n = 446) were counselled for IVF-ICSI, and randomized on the day of embryo transfer to group 1 (daily 0.1 mg subcutaneous GnRHa until day of beta-HCG) (n = 224) and group 2 (stopped GnRHa on day of HCG injection) (n = 222). Both groups received daily vaginal progesterone suppositories. Primary outcome was clinical pregnancy rate. Secondary outcome was ongoing pregnancy rate beyond 20 weeks. Mean age, oestradiol on day of HCG, number of oocytes retrieved, number of embryos transferred, and clinical and ongoing pregnancy rates were 28.9 ± 4.5 years, 2401 ± 746 pg/mL; 13.5 ± 6.0 oocytes; 2.6 ± 0.6 embryos, and 36.2% and 30.4% consecutively in group 1 compared with 29.7 ± 4.7 years, 2483 ± 867 pg/mL, 13.7 ± 5.5 oocytes, 2.7 ± 0.6 embryos, 30.6% pregnancy rate, and 25.7% ongoing pregnancy rate in group 2. No significant difference was found between the groups. Subcutaneous GnRHa during the luteal phase of long GnRHa protocol cycles does not increase clinical or ongoing pregnancy rates after IVF-ICSI.

Phase IV, open-label, randomized study of low-dose recombinant human follicle-stimulating hormone protocols for ovulation induction.

BACKGROUND: This Phase IV, open-label, multicentre, randomized study (MEnTOR) compared two low-dose recombinant human follicle-stimulating hormone (r-hFSH) protocols for ovulation induction. METHODS: This study was conducted in six Middle Eastern countries between March 2009 and March 2011. Eligible women (18-37 years), with World Health Organization Group II anovulatory infertility, were randomized to receive r-hFSH (starting daily dose: 75 IU) as a chronic low-dose (CLD) (37.5 IU dose increase on Day 14) or low-dose (LD) (37.5 IU dose increase on Day 7) protocol if no follicles were ≥ 10 mm. The maximum r-hFSH daily dose permitted was 225 IU/day. The total length of ovarian stimulation could not exceed 35 days, unless ultrasound assessment suggested imminent follicular growth and maturation. Patients underwent only one treatment cycle. Primary endpoint: incidence of mono-follicular development. Secondary endpoints included: stimulation duration and rates of bi-follicular development; human chorionic gonadotrophin administration rate; clinical pregnancy; and cycle cancellation (owing to inadequate response). Adverse events (AEs) were recorded. The primary efficacy analysis was performed using data from all patients who received at least one dose of correct study medication, had at least one efficacy assessment, and no protocol violations at treatment start (CLD group, n=122; LD group, n=125). RESULTS: Mono-follicular development rates (primary endpoint) were similar in both groups (CLD: 56.6% [69/122] versus LD: 55.2% [69/125], p=0.93; primary efficacy analysis population). Similarly, there were no significant differences between groups in bi-follicular development, clinical pregnancy or cycle cancellation (inadequate response) rates. In patients who received human chorionic gonadotrophin injections, the mean duration of stimulation was 13.7 days in the CLD group and 12.9 days in the LD group. Clinical pregnancy rates for those patients who received an hCG injection were similar in both groups (CLD: 20.2% [19/94] versus LD: 19.8% [18/91], p=0.94; primary efficacy analysis population). Most AEs were mild in severity. Only one case of ovarian hyperstimulation syndrome was reported (mild; CLD group). CONCLUSIONS: Efficacy and safety outcomes were similar for the two protocols.

The use of vaginal natural progesterone for prevention of preterm birth in IVF/ICSI pregnancies.

The aim of this study was to evaluate the effect of vaginal natural progesterone on the prevention of preterm birth in IVF/intracytoplasmic sperm injection (ICSI) pregnancies. A single-centre prospective placebo-controlled randomized study was performed. A total of 313 IVF/ICSI pregnant patients were randomized into two groups for either treatment with daily 400 mg vaginal natural progesterone or placebo, starting from mid-trimester up to 37 weeks or delivery. Amongst the patients, there were 215 singleton and 91 twin pregnancies. There was no significant difference in risk of preterm birth among all patients (OR 0.672, 95% CI 0.42-1.0. There was a significantly lower preterm birth rate in singleton pregnancies in the natural progesterone arm (OR 0.53, 95% CI 0.28-0.97) and no significant difference between both arms in twin pregnancies (OR 0.735, 95% CI 0.36-2). In conclusion, the administration of 400 mg vaginal natural progesterone from mid trimester reduced the incidence of preterm birth in singleton, but not in twin, IVF/ICSI pregnancies.

Intra-venous fluids for the prevention of severe ovarian hyperstimulation syndrome.

BACKGROUND: Ovarian hyperstimulation syndrome (OHSS) is a serious and potentially fatal complication of ovarian stimulation, which affects 1% to 14% of all in vitro fertilisation (IVF) or intracytoplasmic sperm injection (ICSI) cycles. A number of clinical studies with conflicting results have reported on the use of intravenous fluids such as albumin, hydroxyethyl starch, Haemaccel® and dextran as a possible way for preventing the severe form of OHSS. OBJECTIVES: To review the effectiveness and safety of administration of intravenous fluids such as albumin, hydroxyethyl starch, Haemaccel® and dextran in the prevention of severe ovarian hyperstimulation syndrome (OHSS) in IVF or ICSI treatment cycles. SEARCH STRATEGY: We searched the Cochrane Menstrual Disorders and Subfertility Group Specialised Register of controlled trials, the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, to third quarter 2010), MEDLINE (1950 to November 2010), EMBASE (1980 to November 2010) and The National Research Register (to November 2010). The citation lists of relevant publications, review articles, abstracts of scientific meetings and included studies were also searched. The authors were contacted to provide or clarify data that were unclear from the trial reports. SELECTION CRITERIA: Randomised controlled trials (RCTs) which compared the effects of intravenous fluids with placebo or no treatment for the prevention of severe OHSS in high risk women undergoing IVF or ICSI treatment cycles. DATA COLLECTION AND ANALYSIS: Two review authors independently scanned the abstracts, identified relevant papers, assessed inclusion of trials and trial quality and extracted relevant data. Validity was assessed in terms of method of randomisation, allocation concealment and outcomes. Where possible, data were pooled for analysis. A separate analysis of studies was performed for human albumin and hydroxyethyl starch versus placebo or no treatment. Other potential intravenous fluids have been identified, such as Haemaccel and dextran, however no randomised controlled studies on their applicability could be found. MAIN RESULTS: Nine RCTs involving 1660 (human albumin vs placebo) and 487 (HES vs placebo) randomised women, have been included in this review. There was a borderline statistically significant decrease in the incidence of severe OHSS with administration of human albumin (8 RCTs, OR 0.67, 95% CI 0.45 to 0.99).There was a statistically significant decrease in severe OHSS incidence with administration of hydroxyethyl starch (3 RCTs, OR 0.12, 95% CI 0.04 to 0.40). There was no evidence of statistical difference in the pregnancy rate between both groups of treatment. AUTHORS' CONCLUSIONS: There is limited evidence of benefit from intra-venous albumin administration at the time of oocyte retrieval in the prevention or reduction of the incidence of severe OHSS in high risk women undergoing IVF or ICSI treatment cycles. Hydroxyethyl starch markedly decreases the incidence of severe OHSS.

Recombinant versus urinary human chorionic gonadotrophin for final oocyte maturation triggering in IVF and ICSI cycles.

BACKGROUND: For the last few decades urinary human chorionic gonadotrophin (hCG) has been used to induce final oocyte maturation triggering in in vitro fertilization (IVF) and intra-cytoplasmic sperm injection (ICSI) cycles. Recombinant technology has allowed the production of two drugs that can be used for the same purpose, to mimic the endogenous luteinizing hormone (LH) surge. This allows commercial production to be adjusted according to market requirements; the removal of all urinary contaminants; and the safe subcutaneous administration of a compound with less batch-to-batch variation. However, prior to a change in practice the effectiveness of the recombinant drugs should be known compared to the currently used urinary human chorionic gonadotrophin (uhCG). OBJECTIVES: To assess the efficacy and safety of subcutaneous recombinant hCG (rhCG) and high dose recombinant LH (rLH) compared with intramuscular uhCG for inducing final oocyte maturation triggering in IVF and ICSI cycles. SEARCH STRATEGY: We searched the Cochrane Menstrual Disorders and Subfertility Group Trials Register (January 2010), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2010), MEDLINE (1966 to January 2010) and EMBASE (1980 to January 2010). SELECTION CRITERIA: Two review authors independently scanned titles and abstracts and selected those that appeared relevant for collection of the full paper. Only truly randomised controlled trials comparing rhCG and rLH with urinary hCG for final oocyte maturation triggering in IVF and ICSI cycles for treatment of infertility in normo-gonadotropic women were included. DATA COLLECTION AND ANALYSIS: Assessment for inclusion or exclusion, quality assessment and data extraction were performed independently by two authors. Discrepancies were discussed in the presence of a third author and consensus reached. Quality assessment included method of randomisation, allocation concealment, blinding of participants and assessors, reporting of a power calculation and intention-to-treat analysis. MAIN RESULTS: Fourteen RCTs (n = 2306) were identified; 11 compared rhCG with uhCG and three compared rhLH with uhCG. There was no evidence of a statistically significant difference between rhCG and uhCG regarding the ongoing pregnancy or live birth rate (6 RCTs: OR 1.04, 95% CI 0.79 to 1.37; P = 0.83, I(2) = 0%). There was no significant difference in the incidence of ovarian hyperstimulation syndrome (OHSS) between rhCG and uhCG (3 RCTs: OR 1.5, 95% CI 0.37 to 4.1; P = 0.37, I(2) = 0%). There was no evidence of statistically significant difference between rhLH and uhCG regarding the ongoing pregnancy or live birth rate (OR 0.94, 95% CI 0.50 to 1.76) and incidence of OHSS (OR 0.82, 95% CI 0.39 to 1.69). These results leave open the possibility of strong differences in favour of either treatment for both ongoing pregnancy and OHSS. AUTHORS' CONCLUSIONS: We conclude that there is no evidence of difference between rhCG or rhLH and uhCG in achieving final follicular maturation in IVF, with equivalent pregnancy rates and OHSS incidence. According to these findings uHCG is still the best choice for final oocyte maturation triggering in IVF and ICSI treatment cycles.

Gonadotrophin-releasing hormone antagonists for assisted reproductive technology.

BACKGROUND: Gonadotrophin-releasing hormone (GnRH) antagonists can be used to prevent a luteinizing hormone (LH) surge during controlled ovarian hyperstimulation (COH) without the hypo-estrogenic side-effects, flare-up, or long down-regulation period associated with agonists. The antagonists directly and rapidly inhibit gonadotropin release within several hours through competitive binding to pituitary GnRH receptors. This property allows their use at any time during the follicular phase. Several different regimes have been described including multiple-dose fixed (0.25 mg daily from day six to seven of stimulation), multiple-dose flexible (0.25 mg daily when leading follicle is 14 to 15 mm), and single-dose (single administration of 3 mg on day 7 to 8 of stimulation) protocols, with or without the addition of an oral contraceptive pill. Further, women receiving antagonists have been shown to have a lower incidence of ovarian hyperstimulation syndrome (OHSS). Assuming comparable clinical outcomes for the antagonist and agonist protocols, these benefits would justify a change from the standard long agonist protocol to antagonist regimens. This is an update of a Cochrane review first published in 2001, and previously updated in 2006. OBJECTIVES: To evaluate the effectiveness and safety of gonadotrophin-releasing hormone (GnRH) antagonists with the standard long protocol of GnRH agonists for controlled ovarian hyperstimulation in assisted conception cycle SEARCH STRATEGY: We performed electronic searches of major databases, for example Cochrane Menstrual Disorders and Subfertility Group Specialised Register, CENTRAL, MEDLINE, EMBASE (from 1987 to April 2010); and handsearched bibliographies of relevant publications and reviews, and abstracts of major scientific meetings, for example the European Society of Human Reproduction and Embryology (ESHRE) and American Society for Reproductive Medicine (ASRM). A date limited search of Cochrane Menstrual Disorders and Subfertility Group Specialised Register, CENTRAL from April 2010 to April 2011 was run. Eighteen studies have been entered into the Classification pending references section of this update. These studies will be appraised for inclusion or exclusion in the next update of this review, due April 2012. SELECTION CRITERIA: Two review authors independently screened the relevant citations for randomised controlled trials (RCTs) comparing different agonist versus antagonist protocols in women undergoing IVF or ICSI. DATA COLLECTION AND ANALYSIS: Two review authors independently assessed trial risk of bias and extracted data. If relevant data were missing or unclear, the authors were contacted for clarification. MAIN RESULTS: Forty-five RCTs (n = 7511) comparing the antagonist to the long agonist protocols fulfilled the inclusion criteria. There was no evidence of a statistically significant difference in rates of live-births (9 RCTs; odds ratio (OR) 0.86, 95% CI 0.69 to 1.08) or ongoing pregnancy (28 RCTs; OR 0.87, 95% CI 0.77 to 1.00). There was a statistically significant lower incidence of OHSS in the GnRH antagonist group (29 RCTs; OR 0.43, 95% CI 0.33 to 0.57). AUTHORS' CONCLUSIONS: The use of antagonist compared with long GnRH agonist protocols was associated with a large reduction in OHSS and there was no evidence of a difference in live-birth rates.

Prospective randomized study for hydrotubation versus no hydrotubation before intrauterine insemination in unexplained infertility.

The purpose of the study was to investigate the value of hydrotubation before intrauterine insemination (IUI). In 228 patients with the diagnosis of unexplained infertility, ovarian stimulation was performed before IUI, using 100mg of clomiphine citrate for 5 days from day 3 of the cycle and one ampoule of human menopausal gonadotrophin for 5 days from day 6 of the cycle. Folliculometry and determination of LH concentration in urine were performed daily until LH became positive, then randomization for hydrotubation before IUI versus no hydrotubation was performed. Fifteen patients were cancelled from the study due to poor response or stimulation of three or more follicles. A total of 213 patients were randomized as follows: 103 patients undergoing hydrotubation using 50 ml of saline and 110 patients with no hydrotubation. IUI was performed the following day and ongoing pregnancy occurred in 13 patients (12.6%) in the hydrotubation group and nine patients (8.2%) in the non-hydrotubation group with no significant difference (OR 1.66; 95% CI 0.62-4.63). In conclusion, hydrotubation before IUI does not improve pregnancy rate.

Improving the patient's experience of IVF/ICSI: a proposal for an ovarian stimulation protocol with GnRH antagonist co-treatment.

Patients undergoing IVF/ICSI frequently experience substantial treatment burden, risk and psychological distress. These three related elements contribute to a negative patient experience that can lead to treatment discontinuation if pregnancy is not achieved. One approach to minimize these factors is the use of protocols designed to achieve high term, singleton birth rates per IVF treatment started, while improving the patient's welfare. Gonadotrophin-releasing hormone (GnRH) antagonists may be suitable for inclusion in such a protocol. In clinical trial data and meta-analyses, treatment with these agents is associated with similar live birth rates but reduced treatment burden (duration and side effects) and less risk of ovarian stimulation syndrome, compared with GnRH agonist long protocols. GnRH antagonists may also be associated with reduced psychological distress compared with agonists, but so far, the evidence for this is inconclusive. To facilitate the implementation of treatments that optimize the patient's experience, a simple GnRH antagonist protocol for use in predicted normal responders is proposed.

Symposium: Update on prediction and management of OHSS. Prevention of OHSS.

Ovarian hyperstimulation syndrome (OHSS) is a major complication of ovulation induction. As the treatment of the syndrome is currently empirical, prevention is the most important aspect of its management. Identification of patients vulnerable to developing OHSS by taking a history of previous OHSS and polycystic ovarian syndrome is the first step in prevention. The use of mild stimulation protocols with small doses of gonadotrophin is also important. As gonadotrophin-releasing hormone (GnRH) antagonist protocol is associated with a lower risk of OHSS, antagonist could be the protocol of choice in high-risk patients. Withholding human chorionic gonadotrophin (HCG) and continuation of GnRH agonist will abort the syndrome but at the expense of loss of the cycle. Coasting, which involves stoppage of gonadotrophins until oestradiol drops to a low concentration before HCG injection, is an effective technique but it does not completely prevent OHSS. Intravenous albumin is useful in the prevention when given at time of oocyte retrieval. Cryopreservation of all embryos will reduce late-onset OHSS but not early-onset OHSS. In-vitro maturation of oocytes will avoid ovarian stimulation and totally prevent OHSS. Triggering ovulation with a lower dose of HCG is effective in reducing the incidence of OHSS. There are possible roles for metformin and dopamine agonist for prevention of OHSS.

Ultrasound cervical measurement and prediction of spontaneous preterm birth in ICSI pregnancies: a prospective controlled study.

A prospective controlled study was performed in which transvaginal ultrasound measurement of cervical length was compared in 222 twin ICSI pregnancies, 122 singleton ICSI pregnancies and 51 spontaneous singleton pregnancies. Preterm birth was defined as

Luteal support in reproduction: when, what and how?

PURPOSE OF REVIEW: Luteal phase support (LPS) is an integral part of the IVF cycles treated by gonadotropin-releasing hormone analogues. There is a worldwide controversy concerning the type of hormones used for LPS, its dose, duration, when to start and when to stop. This review will cover original as well as recent data on this topic. RECENT FINDINGS: There is a consensus in the literature among IVF centers that LPS is necessary for IVF cycles. Human chorionic gonadotropin is less commonly used than progesterone for LPS because of ovarian hyperstimulation syndrome risk. Several studies suggested that intramuscular progesterone is superior to vaginal progesterone for LPS; however, the majority of centers use vaginal progesterone to avoid side effects of intramuscular injection. There is no difference in pregnancy rate whether LPS is started on day of human chorionic gonadotropin, oocyte retrieval or embryo transfer. There is a strong evidence that LPS should be stopped either on the day of pregnancy test or the first ultrasound (6-7 weeks pregnancy). There is no evidence that addition of estrogen will improve pregnancy rate. SUMMARY: Progesterone is the preferred option for LPS. It should start within 2 days from triggering ovulation and should end on day of beta human chorionic gonadotropin or the day of the first ultrasound (6-7 weeks pregnancy).

Highly purified hMG achieves better pregnancy rates in IVF cycles but not ICSI cycles compared with recombinant FSH: a meta-analysis.

OBJECTIVE: Human menopausal gonadotropin (hMG) was demonstrated to be superior to recombinant FSH (rFSH) regarding clinical outcomes. It is not clear whether this change in the evidence was due to the introduction of highly purified (HP) hMG. DESIGN: Systematic review of properly randomised trials comparing HP-hMG vs. rFSH in women undergoing in vitro fertilisation (IVF) and/or intracytoplasmic sperm injection (ICSI). A meticulous search was performed using electronic databases and hand searches of the literature. RESULTS: Six trials (2371 participants) were included. Pooling of the trials demonstrated that the probability of clinical pregnancy following HP-hMG administration was higher than rFSH and reached borderline significance (odd ratio (O.R) = 1.21, 95% confidence interval (CI) = 1.00 to 1.45), but the ongoing pregnancy/live-birth rate was not statistically different between the two drugs, although it showed strong trends towards improvement with HP-hMG (O.R = 1.19, 95% CI = 0.98 to 1.44). Subgroup analysis comparing both drugs in IVF cycles demonstrated a statistically significant better ongoing pregnancy/live-birth rate in favour of HP-hMG (O.R = 1.31, 95% CI = 1.02 to 1.68). On the other hand, there was almost an equal ongoing pregnancy/live-birth rate in ICSI cycles (OR = 0.98, 95% CI = 0.7 to 1.36). CONCLUSIONS: HP-hMG should be preferred over rFSH in women undergoing assisted reproduction, especially if IVF is the intended method of fertilisation.

Characterization of embryonic cells obtained from multifetal reduction.

The multifetal reduction (MFR) procedure is usually reserved for high-order multiple pregnancies, and aspirated tissues are typically discarded. In this study, cells obtained from MFR tissue (termed multifetal reduction embryonic cells (MFR-ECs)), were characterized in vitro by genotypic and phenotypic analyses and tested in vivo by injection under the kidney capsule of nude mice. MFR-ECs were highly proliferative in culture and showed a normal karyotype by microarray CGH. Immunohistochemical analysis at day zero showed positive focal staining for desmin, S-100 protein, synaptophysin and chromogranin. Histology examination showed a mixture of cells from the three germ layers at different stages of differentiation. Markers of these stages included important developmental transcription factors, such as beta three-tubulin (ectoderm), paired box 6 (ectoderm) and alpha-smooth muscle actin (mesoderm). Quantitative polymerase chain reaction (qPCR) showed down-regulation of the mRNAs of cancer-related genes such as TP53. In vivo transplantation in nude mice showed a typical hyaline cartilage plate and no teratoma formation. Thus, MFR-ECs represent a rich, unique source for studying stem cell development, embryogenesis and cell differentiation.

Prospective randomized study comparing luteal phase support for ICSI patients up to the first ultrasound compared with an additional three weeks.

BACKGROUND: There is a consensus that administration of progesterone to women after IVF for luteal phase support (LPS) is associated with a higher ongoing pregnancy rate. However there are few studies, including only one randomized study, which have examined the optimal duration of LPS. METHODS: A questionnaire concerning details of LPS was returned from 21 leading IVF centres. We then randomized 257 women, who were pregnant after ICSI on day of first ultrasound, into two groups: to continue LPS for three more weeks or to stop on the day of ultrasound. RESULTS: The duration of LPS in the questionnaire varied from the day of positive pregnancy test up to 12 weeks of pregnancy in different centres. In the randomized study, 132 patients in Group A continued LPS for 3 weeks after first ultrasound, whereas 125 patients in Group B stopped LPS on day of first ultrasound. After confirming pulsations, the miscarriage rate up to 20 weeks of gestation was 4.6% (6/132) in group A and 4.8% (6/125) in group B [odds ratios (OR) = 0.94; 95% confidence intervals (CI) = 0.3-3.1]. Bleeding episodes were 15.9% in Group A compared with 20.8% in group B (OR = 0.72; 95% CI = 0.38-1.36). CONCLUSIONS: There is no international consensus about the duration of LPS; our single-centre randomized trial did not support extending the LPS beyond the day of first ultrasound demonstrating echoes and pulsations. Trials registry number-ISRCTN: 88722916.