Crossover or parallel design in infertility trials? The discussion continues

Intra-uterine insemination for unexplained subfertility

BACKGROUND

Intra-uterine insemination (IUI) is a widely-used fertility treatment for couples with unexplained subfertility. Although IUI is less invasive and less expensive than in vitro fertilisation (IVF), the safety of IUI in combination with ovarian hyperstimulation (OH) is debated. The main concern about IUI treatment with OH is the increase in multiple pregnancy rates.

OBJECTIVES

To determine whether, for couples with unexplained subfertility, the live birth rate is improved following IUI treatment with or without OH compared to timed intercourse (TI) or expectant management with or without OH, or following IUI treatment with OH compared to IUI in a natural cycle.

SEARCH METHODS

We searched the Cochrane Gynaecology and Fertility (CGF) Group trials register, CENTRAL, MEDLINE, Embase, PsycINFO, CINAHL and two trials registers up to 17 October 2019, together with reference checking and contact with study authors for missing or unpublished data.

SELECTION CRITERIA

Randomised controlled trials (RCTs) comparing IUI with TI or expectant management, both in stimulated or natural cycles, or IUI in stimulated cycles with IUI in natural cycles in couples with unexplained subfertility.

DATA COLLECTION AND ANALYSIS

Two review authors independently performed study selection, quality assessment and data extraction. Primary review outcomes were live birth rate and multiple pregnancy rate.

MAIN RESULTS

We include 15 trials with 2068 women. The evidence was of very low to moderate quality. The main limitation was very serious imprecision. IUI in a natural cycle versus timed intercourse or expectant management in a natural cycle It is uncertain whether treatment with IUI in a natural cycle improves live birth rate compared to treatment with expectant management in a natural cycle (odds ratio (OR) 1.60, 95% confidence interval (CI) 0.92 to 2.78; 1 RCT, 334 women; low-quality evidence). If we assume the chance of a live birth with expectant management in a natural cycle to be 16%, that of IUI in a natural cycle would be between 15% and 34%. It is uncertain whether treatment with IUI in a natural cycle reduces multiple pregnancy rates compared to control (OR 0.50, 95% CI 0.04 to 5.53; 1 RCT, 334 women; low-quality evidence). IUI in a stimulated cycle versus timed intercourse or expectant management in a stimulated cycle It is uncertain whether treatment with IUI in a stimulated cycle improves live birth rates compared to treatment with TI in a stimulated cycle (OR 1.59, 95% CI 0.88 to 2.88; 2 RCTs, 208 women; I² = 72%; low-quality evidence). If we assume the chance of achieving a live birth with TI in a stimulated cycle was 26%, the chance with IUI in a stimulated cycle would be between 23% and 50%. It is uncertain whether treatment with IUI in a stimulated cycle reduces multiple pregnancy rates compared to control (OR 1.46, 95% CI 0.55 to 3.87; 4 RCTs, 316 women; I² = 0%; low-quality evidence). IUI in a stimulated cycle versus timed intercourse or expectant management in a natural cycle In couples with a low prediction score of natural conception, treatment with IUI combined with clomiphene citrate or letrozole probably results in a higher live birth rate compared to treatment with expectant management in a natural cycle (OR 4.48, 95% CI 2.00 to 10.01; 1 RCT; 201 women; moderate-quality evidence). If we assume the chance of a live birth with expectant management in a natural cycle was 9%, the chance of a live birth with IUI in a stimulated cycle would be between 17% and 50%. It is uncertain whether treatment with IUI in a stimulated cycle results in a lower multiple pregnancy rate compared to control (OR 3.01, 95% CI 0.47 to 19.28; 2 RCTs, 454 women; I² = 0%; low-quality evidence). IUI in a natural cycle versus timed intercourse or expectant management in a stimulated cycle Treatment with IUI in a natural cycle probably results in a higher cumulative live birth rate compared to treatment with expectant management in a stimulated cycle (OR 1.95, 95% CI 1.10 to 3.44; 1 RCT, 342 women: moderate-quality evidence). If we assume the chance of a live birth with expectant management in a stimulated cycle was 13%, the chance of a live birth with IUI in a natural cycle would be between 14% and 34%. It is uncertain whether treatment with IUI in a natural cycle results in a lower multiple pregnancy rate compared to control (OR 1.05, 95% CI 0.07 to 16.90; 1 RCT, 342 women; low-quality evidence). IUI in a stimulated cycle versus IUI in a natural cycle Treatment with IUI in a stimulated cycle may result in a higher cumulative live birth rate compared to treatment with IUI in a natural cycle (OR 2.07, 95% CI 1.22 to 3.50; 4 RCTs, 396 women; I² = 0%; low-quality evidence). If we assume the chance of a live birth with IUI in a natural cycle was 14%, the chance of a live birth with IUI in a stimulated cycle would be between 17% and 36%. It is uncertain whether treatment with IUI in a stimulated cycle results in a higher multiple pregnancy rate compared to control (OR 3.00, 95% CI 0.11 to 78.27; 2 RCTs, 65 women; low-quality evidence).

AUTHORS' CONCLUSIONS

Due to insufficient data, it is uncertain whether treatment with IUI with or without OH compared to timed intercourse or expectant management with or without OH improves cumulative live birth rates with acceptable multiple pregnancy rates in couples with unexplained subfertility. However, treatment with IUI with OH probably results in a higher cumulative live birth rate compared to expectant management without OH in couples with a low prediction score of natural conception. Similarly, treatment with IUI in a natural cycle probably results in a higher cumulative live birth rate compared to treatment with timed intercourse with OH. Treatment with IUI in a stimulated cycle may result in a higher cumulative live birth rate compared to treatment with IUI in a natural cycle.

Assisted reproductive technologies for male subfertility

BACKGROUND

Intra-uterine insemination (IUI), in vitro fertilisation (IVF) and intracytoplasmic sperm injection (ICSI) are frequently used fertility treatments for couples with male subfertility. The use of these treatments has been subject of discussion. Knowledge on the effectiveness of fertility treatments for male subfertility with different grades of severity is limited. Possibly, couples are exposed to unnecessary or ineffective treatments on a large scale.

OBJECTIVES

To evaluate the effectiveness and safety of different fertility treatments (expectant management, timed intercourse (TI), IUI, IVF and ICSI) for couples whose subfertility appears to be due to abnormal sperm parameters.

SEARCH METHODS

We searched for all publications that described randomised controlled trials (RCTs) of the treatment for male subfertility. We searched the Cochrane Menstrual Disorders and Subfertility Group Specialised Register, the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, PsycINFO and the National Research Register from inception to 14 April 2015, and web-based trial registers from January 1985 to April 2015. We applied no language restrictions. We checked all references in the identified trials and background papers and contacted authors to identify relevant published and unpublished data.

SELECTION CRITERIA

We included RCTs comparing different treatment options for male subfertility. These were expectant management, TI (with or without ovarian hyperstimulation (OH)), IUI (with or without OH), IVF and ICSI. We included only couples with abnormal sperm parameters.

DATA COLLECTION AND ANALYSIS

Two review authors independently selected the studies, extracted data and assessed risk of bias. They resolved disagreements by discussion with the rest of the review authors. We performed statistical analyses in accordance with the guidelines for statistical analysis developed by The Cochrane Collaboration. The quality of the evidence was rated using the GRADE methods. Primary outcomes were live birth and ovarian hyperstimulation syndrome (OHSS) per couple randomised.

MAIN RESULTS

The review included 10 RCTs (757 couples). The quality of the evidence was low or very low for all comparisons. The main limitations in the evidence were failure to describe study methods, serious imprecision and inconsistency. IUI versus TI (five RCTs)Two RCTs compared IUI with TI in natural cycles. There were no data on live birth or OHSS. We found no evidence of a difference in pregnancy rates (2 RCTs, 62 couples: odds ratio (OR) 4.57, 95% confidence interval (CI) 0.21 to 102, very low quality evidence; there were no events in one of the studies).Three RCTs compared IUI with TI both in cycles with OH. We found no evidence of a difference in live birth rates (1 RCT, 81 couples: OR 0.89, 95% CI 0.30 to 2.59; low quality evidence) or pregnancy rates (3 RCTs, 202 couples: OR 1.51, 95% CI 0.74 to 3.07; I(2) = 11%, very low quality evidence). One RCT reported data on OHSS. None of the 62 women had OHSS.One RCT compared IUI in cycles with OH with TI in natural cycles. We found no evidence of a difference in live birth rates (1 RCT, 44 couples: OR 3.14, 95% CI 0.12 to 81.35; very low quality evidence). Data on OHSS were not available. IUI in cycles with OH versus IUI in natural cycles (five RCTs)We found no evidence of a difference in live birth rates (3 RCTs, 346 couples: OR 1.34, 95% CI 0.77 to 2.33; I(2) = 0%, very low quality evidence) and pregnancy rates (4 RCTs, 399 couples: OR 1.68, 95% CI 1.00 to 2.82; I(2) = 0%, very low quality evidence). There were no data on OHSS. IVF versus IUI in natural cycles or cycles with OH (two RCTs)We found no evidence of a difference in live birth rates between IVF versus IUI in natural cycles (1 RCT, 53 couples: OR 0.77, 95% CI 0.25 to 2.35; low quality evidence) or IVF versus IUI in cycles with OH (2 RCTs, 86 couples: OR 1.03, 95% CI 0.43 to 2.45; I(2) = 0%, very low quality evidence). One RCT reported data on OHSS. None of the women had OHSS.Overall, we found no evidence of a difference between any of the groups in rates of live birth, pregnancy or adverse events (multiple pregnancy, miscarriage). However, most of the evidence was very low quality.There were no studies on IUI in natural cycles versus TI in stimulated cycles, IVF versus TI, ICSI versus TI, ICSI versus IUI (with OH) or ICSI versus IVF.

AUTHORS' CONCLUSIONS

We found insufficient evidence to determine whether there was any difference in safety and effectiveness between different treatments for male subfertility. More research is needed.

Intra-uterine insemination for unexplained subfertility

BACKGROUND

Intra-uterine insemination (IUI) is a widely used fertility treatment for couples with unexplained subfertility. Although IUI is less invasive and less expensive thAppendixan in vitro fertilisation (IVF), the safety of IUI in combination with ovarian hyperstimulation (OH) is debated. The main concern about IUI treatment with OH is the increase in multiple pregnancy rate. This is an update of a Cochrane review (Veltman-Verhulst 2012) originally published in 2006 and updated in 2012.

OBJECTIVES

To determine whether, for couples with unexplained subfertility, IUI improves the live birth rate compared with timed intercourse (TI), or expectant management, both with and without ovarian hyperstimulation (OH).

SEARCH METHODS

We searched the Cochrane Gynaecology and Fertility (formerly Cochrane Menstrual Disorders and Subfertility Group) Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, inception to Issue 11, 2015), Ovid MEDLINE, Ovid EMBASE, PsycINFO and trial registers, all from inception to December 2015 and reference lists of articles. Authors of identified studies were contacted for missing or unpublished data. The evidence is current to December 2015.

SELECTION CRITERIA

Truly randomised controlled trial (RCT) comparisons of IUI versus TI, in natural or stimulated cycles. Only couples with unexplained subfertility were included.

DATA COLLECTION AND ANALYSIS

Two review authors independently performed study selection, quality assessment and data extraction. We extracted outcomes, and pooled data and, where possible, we carried out subgroup and sensitivity analyses.

MAIN RESULTS

We included 14 trials including 1867 women. IUI versus TI or expectant management both in natural cycleLive birth rate (all cycles)There was no evidence of a difference in cumulative live births between the two groups (Odds Ratio (OR) 1.60, 95% confidence interval (CI) 0.92 to 2.78; 1 RCT; n = 334; moderate quality evidence). The evidence suggested that if the chance of a live birth in TI was assumed to be 16%, that of IUI would be between 15% and 34%.Multiple pregnancy rateThere was no evidence of a difference in multiple pregnancy rate between the two treatment groups (OR 0.50, 95% CI 0.04 to 5.53; 1 RCT; n = 334; moderate quality evidence). IUI versus TI or expectant management both in stimulated cycleLive birth rate (all cycles)There was no evidence of a difference between the two treatment groups (OR 1.59, 95% CI 0.88 to 2.88; 2 RCTs; n = 208; I(2) = 72%; moderate quality evidence). The evidence suggested that if the chance of achieving a live birth in TI was assumed to be 26%, the chance of a live birth with IUI would be between 23% and 50%.Multiple pregnancy rateThere was no evidence of a difference in multiple pregnancy rates between the two treatment groups (OR 1.46, 95% CI 0.55 to 3.87; 4 RCTs, n = 316; I(2) = 0%; low quality evidence). IUI in a natural cycle versus IUI in a stimulated cycle Live birth rate (all cycles)An increase in live birth rate was found for women who were treated with IUI in a stimulated cycle compared with those who underwent IUI in natural cycle (OR 0.48, 95% CI 0.29 to 0.82; 4 RCTs, n = 396; I(2) = 0%; moderate quality evidence). The evidence suggested that if the chance of a live birth in IUI in a stimulated cycle was assumed to be 25%, the chance of a live birth in IUI in a natural cycle would be between 9% and 21%.Multiple pregnancy rateThere was no evidence of a difference in multiple pregnancy rate between the two treatment groups (OR 0.33, 95% CI 0.01 to 8.70; 2 RCTs; n = 65; low quality evidence). IUI in a stimulated cycle versus TI or expectant management in a natural cycleLive birth rate (all cycles)There was no evidence of a difference in live birth rate between the two treatment groups (OR 0.82, 95% CI 0.45 to 1.49; 1 RCT; n = 253; moderate quality evidence). The evidence suggested that if the chance of a live birth in TI or expectant management in a natural cycle was assumed to be 24%, the chance of a live birth in IUI in a stimulated cycle would be between 12% and 32%.Multiple pregnancy rateThere was no evidence of a difference in multiple pregnancy rate between the two treatment groups (OR 2.00, 95% CI 0.18 to 22.34; 2 RCTs; n = 304; moderate quality evidence). IUI in natural cycle versus TI or expectant management in stimulated cycle Live birth rate (all cycles)There was evidence of an increase in live births for IUI (OR 1.95, 95% CI 1.10 to 3.44; 1 RCT, n = 342; moderate quality evidence). The evidence suggested that if the chance of a live birth in TI in a stimulated cycle was assumed to be 13%, the chance of a live birth in IUI in a natural cycle would be between 14% and 34%.Multiple pregnancy rateThere was no evidence of a difference in multiple pregnancy rate between the groups (OR 1.05, 95% CI 0.07 to 16.90; 1 RCT; n = 342; moderate quality evidence).The quality of the evidence was assessed using GRADE methods. Quality ranged from low to moderate, the main limitation being imprecision in the findings for both live birth and multiple pregnancy..

AUTHORS' CONCLUSIONS

This systematic review did not find conclusive evidence of a difference in live birth or multiple pregnancy in most of the comparisons for couples with unexplained subfertility treated with intra-uterine insemination (IUI) when compared with timed intercourse (TI), both with and without ovarian hyperstimulation (OH). There were insufficient studies to allow for pooling of data on the important outcome measures for each of the comparisons.

Intra-uterine insemination for unexplained subfertility

BACKGROUND

Intra-uterine insemination (IUI) is a widely used fertility treatment for couples with unexplained subfertility. Although IUI is less invasive and less expensive than in vitro fertilisation (IVF), the safety of IUI in combination with ovarian hyperstimulation (OH) is debated. The main concern about IUI treatment with OH is the increase in multiple pregnancy rate.

OBJECTIVES

To determine whether, for couples with unexplained subfertility, IUI improves the live birth rate compared with timed intercourse (TI), both with and without ovarian hyperstimulation (OH).

SEARCH METHODS

We searched the Cochrane Menstrual Disorders and Subfertility Group Trials Register (searched July 2011), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2011, Issue 7), MEDLINE (1966 to July 2011), EMBASE (1980 to July 2011), PsycINFO (1806 to July 2011), SCIsearch and reference lists of articles. Authors of identified studies were contacted for missing or unpublished data.

SELECTION CRITERIA

Truly randomised controlled trials (RCTs) with at least one of the following comparisons were included: IUI versus TI, both in a natural cycle; IUI versus TI, both in a stimulated cycle; IUI in a natural cycle versus IUI in a stimulated cycle; IUI with OH versus TI in a natural cycle; IUI in a natural cycle versus TI with OH. Only couples with unexplained subfertility were included.

DATA COLLECTION AND ANALYSIS

Quality assessment and data extraction were performed independently by two review authors. Outcomes were extracted and the data were pooled. Subgroup and sensitivity analyses were done where possible.

MAIN RESULTS

One trial compared IUI in a natural cycle with expectant management and showed no evidence of increased live births (334 women: odds ratio (OR) 1.60, 95% confidence interval (CI) 0.92 to 2.8). In the six trials where IUI was compared with TI, both in stimulated cycles, there was evidence of an increased chance of pregnancy after IUI (six RCTs, 517 women: OR 1.68, 95% CI 1.13 to 2.50). A significant increase in live birth rate was found for women where IUI with OH was compared with IUI in a natural cycle (four RCTs, 396 women: OR 2.07, 95% CI 1.22 to 3.50). However the trials provided insufficient data to investigate the impact of IUI with or without OH on several important outcomes including live births, multiple pregnancies, miscarriage and risk of ovarian hyperstimulation. There was no evidence of a difference in pregnancy rate for IUI with OH compared with TI in a natural cycle (two RCTs, total 304 women: data not pooled). The final comparison of IUI in natural cycle to TI with OH showed a marginal, significant increase in live births for IUI (one RCT, 342 women: OR 1.95, 95% CI 1.10 to 3.44).

AUTHORS' CONCLUSIONS

There is evidence that IUI with OH increases the live birth rate compared to IUI alone. The likelihood of pregnancy was also increased for treatment with IUI compared to TI in stimulated cycles. One adequately powered multicentre trial showed no evidence of effect of IUI in natural cycles compared with expectant management. There is insufficient data on multiple pregnancies and other adverse events for treatment with OH. Therefore couples should be fully informed about the risks of IUI and OH as well as alternative treatment options.

Planning and analysis of cross-over trials in infertility

The use of cross-over trials in investigating treatments for infertility is discussed. A simple possible approach to analyzing such trials using the Mantel-Haenszel procedure is explained. A more flexible approach based on the normal-binomial mixture model of Ezzet and Whitehead is examined. It is shown how this may be implemented in various statistical packages by applying it to two real examples of trials in infertility. An approach that may be used to compare designs via simulation is explained briefly. It is concluded that provided that cross-over trials in infertility are regarded as parallel group trials with extra information rather than as cross-over trials with missing information, their use may be more promising than that has previously been concluded.

Intra-uterine insemination for unexplained subfertility

BACKGROUND

Intra-uterine insemination (IUI) is a widely used fertility treatment for couples with unexplained subfertility. Although IUI is less invasive and less expensive than in vitro fertilisation (IVF), the safety of IUI in combination with ovarian hyperstimulation (OH) is debated. The main concern about IUI treatment with OH is the increase in multiple pregnancy rates.

OBJECTIVES

To determine whether for couples with unexplained subfertility IUI improves the live birth rate compared with timed intercourse (TI), both with and without ovarian hyperstimulation.

SEARCH STRATEGY

We searched the Cochrane Menstrual Disorder and Subfertility Group Trials Register (searched March 2005), the Cochrane Central Register of Controlled Trials (The Cochrane Library 2005, Issue 4), MEDLINE (1966 to November 2005), EMBASE (1980 to November 2005), SCIsearch and reference lists of articles. Authors of identified studies were contacted for missing or unpublished data.

SELECTION CRITERIA

Truly randomised controlled trials (RCTs) with at least one of the following comparisons were included: --IUI versus TI, both in a natural cycle; --IUI versus TI, both in a stimulated cycle; --IUI in a natural cycle versus IUI in a stimulated cycle; --IUI with OH versus TI in natural cycle; --IUI in a natural cycle versus TI with OH. Only couples with unexplained subfertility were included.

DATA COLLECTION AND ANALYSIS

Quality assessment and data extraction were performed independently by two review authors. Outcomes were extracted and the data were pooled. Subgroup analyses and sensitivity analyses were done where possible.

MAIN RESULTS

In the six trials where IUI was compared with TI, both in stimulated cycles, there was evidence of an increased chance of pregnancy (six RCTs, 517 women: OR 1.68, 95% CI 1.13 to 2.50). A significant increase in pregnancy rate was also found for women where IUI with OH was compared with IUI in a natural cycle (three RCTs, 415 women: OR 2.33, 95% CI 1.46 to 3.71). However, the trials provided insufficient data to investigate the impact of IUI with or without OH on several important outcomes including live birth, multiple pregnancies, miscarriage and risk of ovarian hyperstimulation. There was no evidence of a difference in pregnancy rate for IUI with OH compared with TI in a natural cycle (one RCT, 51 women: OR 4.05, 95% CI 0.39 to 41.87). No RCTs were found for the other two comparisons.

AUTHORS' CONCLUSIONS

There is evidence that IUI with OH increases the live birth rate compared to IUI alone. The likelihood of pregnancy was also increased for treatment with IUI compared to TI both in stimulated cycles. There is insufficient data on multiple pregnancies and other adverse events for treatment with OH. Therefore, couples should be fully informed about the risks of IUI and OH as well as alternative treatment options.

The place of the crossover design in infertility trials: a maximum likelihood approach

BACKGROUND

For some years, there has been a debate as to the place of the crossover trial in assisted reproduction technology (ART). We aimed to investigate whether crossover and parallel designs result in different estimates of treatment effects.

METHODS

We carried out computer simulation of cohorts of patients undergoing either intra-uterine insemination (IUI) or IVF under both parallel and crossover designs, under scenarios involving censoring and carryover effects. Results of the simulation were analysed using a maximum likelihood approach.

RESULTS

No relevant difference was found between the designs. The crossover design resulted in slightly more pregnancies than the parallel design. Carryover effects may slightly distort the estimates of treatment effects. Crossover and parallel designs will produce essentially the same statistical estimates of treatment effect and percentage of pregnancies. The crossover design is an acceptable design in infertility research provided the data are analysed correctly.

Systematic review of the treatment of ovulatory infertility with clomiphene citrate and intrauterine insemination

BACKGROUND

Controlled ovarian hyperstimulation (COH) with clomiphene citrate (CC) combined with intrauterine insemination (IUI) is often used as treatment for ovulatory infertility which includes unexplained, male, cervical, endometriosis, and tubal infertility.

AIMS

To review the effectiveness of CC and IUI in ovulatory infertility.

METHODS

Systematic review of pertinent randomised controlled trials (RCT) using the bibliographic databases MEDLINE and EMBASE. References of selected articles identified were hand-searched for additional relevant citations.

RESULTS

Six published RCT were included in the overall review. Meta-analysis demonstrated a higher cycle pregnancy rate (CPR) with CC and IUI compared to timed intercourse in the natural cycle (P < 0.001 and odds ratio = 4.6, 95% CI = 1.9-11.3). Treatment with gonadotrophins and IUI results in a higher CPR compared to CC and IUI (P = 0.005 and odds ratio = 2.9, 95% CI = 1.3-6.2). Further RCT are required comparing CC and IUI with IUI or CC alone before one can make firm conclusions.

CONCLUSIONS

Clomiphene citrate combined with IUI is more effective than timed intercourse in the natural cycle at achieving pregnancy in couples with ovulatory infertility. However, treatment with gonadotrophins and IUI is superior to CC and IUI.

Fallopian tube sperm perfusion versus intrauterine insemination in unexplained infertility: a randomized, prospective, cross-over trial

OBJECTIVE

To compare the results of fallopian tube sperm perfusion (FSP) versus standard intrauterine insemination (IUI) in patients with unexplained infertility undergoing controlled ovarian hyperstimulation (COH).

DESIGN

Randomized, prospective, cross-over study.

SETTING

Reproductive medicine unit of a university hospital.

PATIENT(S)

Fifty-six couples with unexplained infertility.

INTERVENTION(S)

COH was induced by recombinant FSH and monitored by serial transvaginal ultrasound. On the day of hCG administration during the first treatment cycle, patients were randomized to either IUI or FSP. Thereafter, in case no pregnancy was achieved, patients went on being treated with FSP and IUI in alternate cycles. A maximum of four treatment cycles per couple was performed. FSP was performed using a pediatric Foley's catheter inseminating 4 mL of sperm-enriched suspension; a Kremer-Delafontaine catheter delivering 0.5 mL of sperm suspension was used for IUI.

MAIN OUTCOME MEASURE(S)

Clinical pregnancy rate per cycle.

RESULT(S)

One hundred twenty-seven cycles (58 FSP, 69 IUI) were performed. The clinical pregnancy rate per cycle was 21.7% for IUI and 8.6% for FSP, respectively. No major adverse effects were recorded for either technique.

CONCLUSION(S)

After COH, FSP is less effective than IUI in couples with unexplained infertility.

The effect of metformin on ovarian stimulation and in vitro fertilization in insulin-resistant women with polycystic ovary syndrome: an open-label randomized cross-over trial

Metformin effectively restores insulin sensitivity in insulin-resistant women with polycystic ovary syndrome (PCOS). We examined whether metformin, given prior to and during ovarian stimulation for in vitro fertilization (IVF), altered follicle stimulating hormone (FSH) requirement and increased the number of collected oocytes in these women. Seventeen insulin-resistant women with PCOS were recruited to our IVF unit to receive two consecutive cycles of ovarian stimulation with or without metformin co-treatment, the order of treatments being randomized using a table of random numbers. Metformin treatment (1500 mg/day) started 3 weeks before downregulation with buserelin acetate and was continued throughout ovarian stimulation with human recombinant FSH. Nine women completed both cycles, the results of eight women being excluded because of pregnancy after the first cycle (n = 4) or because the protocol of the study was not followed (n = 4). Mean total FSH dose was 2301 IU (range 1500-6563 IU) in metformin cycles and 2174 IU (range 1200-3900 IU) in parallel control cycles, while the mean number of collected oocytes was 8.6 (range 2-28) and 4.6 (range 1-16), respectively. Bayesian analysis showed probabilities of 0.05 that metformin reduces FSH requirement by at least 10%, and of 0.61 that at least 10% more oocytes are collected after metformin co-treatment. Co-administration of metformin is therefore likely to increase the number of oocytes collected after ovarian stimulation in insulin-resistant women with PCOS but is unlikely to reduce the requirement for FSH.

Meta-analyses involving cross-over trials: methodological issues

BACKGROUND

Meta-analysis of randomized controlled trials (RCTs) is usually based on trials where patients are randomized individually into two different, parallel, treatment groups. This paper concentrates on RCTs of a different design-two-period, two-treatment cross-over trials.

METHODS

The characteristics of these trials are outlined, with detailed examples of methods for analysis for both continuous and binary data. These case studies are then extended into the context of a meta-analysis. The Cochrane Library was surveyed to assess current practice for synthesis.

RESULTS

Methods are described for continuous and binary data for use both when the necessary paired data are given and also when they need to be calculated or imputed, and some suggestions are provided to help people wishing to synthesize data from cross-over trials into meta-analyses. The survey suggested that about 8% of the trials in the Cochrane library were cross-over trials and 18% of the reviews referred to such trials, although there was no consistent approach to their inclusion into the reviews.

CONCLUSIONS

Methods do exist for including valuable information from two-period, two-treatment cross-over trials into quantitative reviews. However, poor reporting of cross-over trials will often impede attempts to perform a meta-analysis using the available methods.

The alternating-sequence design (or multiple-period crossover) trial for evaluating treatment efficacy in infertility

OBJECTIVE

To determine whether a constant-sequence or an alternating-sequence design is better for the evaluation of infertility treatment efficacy when multiple cycles of treatment are undertaken.

DESIGN

A simulation exercise using analytical methods.

SETTING

University medical center.

PATIENT(S)

A hypothetical, heterogeneous population of infertile patients participating in a randomized trial comparing an experimental treatment, with effectiveness of 2.0, to no treatment.

INTERVENTION(S)

Comparison of a constant-sequence design in which the subject receives the same intervention or the alternating-sequence design in which experimental and control treatments are crossed over after each successive cycle.

MAIN OUTCOME MEASURE(S)

Relative risks of pregnancy per cycle and overall after a maximum of five cycles of treatment.

RESULT(S)

With both designs, the pregnancy rates in experimental and control groups showed a consistent decrease with each successive cycle. The overall effectiveness in the constant-sequence design was underestimated at 1.83, whereas in the alternating-sequence design it was overestimated at 2.06. However, by restricting the analysis in the latter design only to the odd-numbered cycles, the relative risk was precisely correct at 2.00.

CONCLUSION(S)

When multiple cycles of treatment are undertaken to evaluate the efficacy of infertility therapy, the alternating-sequence design with restriction of the analysis to only the odd-numbered treatment cycles provides an unbiased estimation of the treatment effect.