Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Review - Intervention

原因不明の不妊症に対する人工授精

Authors' declarations of interest

Version published: 03 March 2020 Version history

https://doi.org/10.1002/14651858.CD001838.pub6
Collapse all Expand all

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trials registers up to 17 October 2019, together with reference checking and contact with study authors for missing or unpublished data.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.

PICOs

Population
Intervention
Comparison
Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

原因不明の不妊症に対する人工授精

レビューの論点

人工授精（排卵誘発剤の併用あり/なし）による治療は、排卵誘発剤と併用または単独で行われた、タイミング療法や待機療法（診察せずに経過をみるだけ）と比較して、原因不明の不妊症カップルの出生率を改善するか。また、排卵誘発剤を併用した人工授精は、併用しない人工授精と比べて出生率を改善するか。

背景

人工授精は、原因不明の不妊症カップルによく用いられる治療法である。人工授精を行う月経周期では、男性のパートナーの精子を調整し、排卵する時期に子宮に直接注入する。人工授精を行う周期では、排卵する卵の数を増やすために排卵を促す薬（排卵誘発剤）を使うことがある。.しかしながら、これらの薬物には副作用があり、また多胎妊娠のリスクが高まる。待機療法や排卵にあわせて性交をとるタイミング療法でも、妊娠率は上昇し、出生につながる。このレビューを参考にすることで、原因がわかっていない不妊症の治療を開始するカップルの意思決定を強化したい。

研究特性

2068人の女性を含む15のランダム化比較試験を特定した。排卵誘発剤を併用もしくは単独の人工授精で治療を受けた女性と、排卵誘発剤を併用もしくは単独のタイミング療法か待機療法に割り当てられた女性を比較した。また、排卵誘発剤を併用して人工授精を受けた女性と単独で人工授精を受けた女性を比較した。主要なアウトカムは、出生率と多胎の妊娠率とした。他のアウトカムとして、妊娠率、流産率および他の有害事象を含めた。エビデンスは2019年10月現在のものである。

主要な結果

ほとんどの治療群では、いずれかの人工授精が別の治療方法と比較して、許容可能な多胎妊娠率におさまりつつ累積出生率(一定の治療期間を終える時点での合計出生率)を改善するかどうかは不明である。しかし、自然妊娠の可能性が低いカップルでは、排卵誘発剤を併用した人工授精が、待機療法と比較して、おそらく累積出生率を改善するだろうという、中程度の質のエビデンスを認めた。同様に、人工授精単独での治療は、おそらく排卵誘発剤を併用した待機療法と比較して、累積出生率を改善するという中程度の質のエビデンスがあった。

エビデンスの質

エビデンスの質は、累積出生率については低度から中等度、多胎妊娠率については低度であった。エビデンスに関する主な限界は、サンプルサイズが小さく、数少ない研究をまとめたことによる、調査結果の精度不足である。

Authors' conclusions

available in

Implications for practice

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, or treatment with IUI with OH compared to IUI in a natural cycle, improves cumulative live birth rates with acceptable multiple pregnancy rates in couples with unexplained subfertility. However, there is moderate‐quality evidence that 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, there is moderate‐quality evidence that treatment with IUI in a natural cycle probably results in a higher cumulative live birth rate compared to treatment with timed intercourse with OH.

Implications for research

None of the comparisons examined in this review were adequately addressed, due to insufficient data; questions still remain as to whether live birth rates can be improved while keeping the risk of multiple pregnancy and other adverse events to acceptable levels. There is therefore a need for better‐designed randomised controlled trials which evaluate the effects of different IUI modalities on important outcome measures. Furthermore, future trials should endeavour to report live birth and other outcome data in forms which enable appropriate analyses.

Summary of findings

Open in table viewer
Summary of findings for the main comparison. IUI in a natural cycle compared to TI or expectant management in a natural cycle for unexplained subfertility

IUI compared to TI or expectant management both in natural cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI
Comparison: TI or expectant management both in natural cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management both in natural cycle

IUI

Live birth rates per couple (all cycles)

156 per 1000

228 per 1000
(145 to 339)

OR 1.60
(0.92 to 2.78)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Multiple pregnancy rate per couple

12 per 1000

6 per 1000
(0 to 63)

OR 0.50
(0.04 to 5.53)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Pregnancy rate per couple (all cycles)

162 per 1000

228 per 1000
(145 to 338)

OR 1.53
(0.88 to 2.64)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Ovarian hyperstimulation syndrome rate per woman ‐ not reported

See comment

See comment

‐

‐

See comment

No data reported on outcome

Miscarriage rate per couple

54 per 1000

42 per 1000
(16 to 107)

OR 0.77
(0.28 to 2.11)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 05.06
(0.24 to 1106.21)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Open in table viewer
Summary of findings 2. IUI in a stimulated cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility

IUI compared to TI or expectant management both in stimulated cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI
Comparison: TI or expectant management both in stimulated cycles

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management both in stimulated cycle

IUI

Live birth rate per couple (all cycles)

255 per 1000

352 per 1000
(231 to 496)

OR 1.59
(0.88 to 2.88)

208
(2 studies)

⊕⊕⊝⊝
lowa

‐

Multiple pregnancy rate per couple

43 per 1000

62 per 1000
(24 to 148)

OR 1.46
(0.55 to 3.87)

316
(4 studies)

⊕⊕⊝⊝
lowa

‐

Pregnancy rate per couple (all cycles)

234 per 1000

340 per 1000
(258 to 436)

OR 1.69
(1.14 to 2.53)

517
(6 studies)

⊕⊕⊝⊝
lowb,c

‐

Ovarian hyperstimulation syndrome rate per woman

Not estimable (no events in control group)

OR 2.75
(0.11 to 69.83)

68
(1 study)

⊕⊝⊝⊝
very lowa,b

‐

Miscarriage rate per couple

57 per 1000

91 per 1000
(33 to 228)

OR 1.66
(0.56 to 4.88)

208
(2 studies)

⊕⊕⊝⊝
lowa

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 3.06
(0.12 to 76.95)

100
(1 study)

⊕⊕⊝⊝
lowa

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.
bDowngraded by one level for serious risk of bias: most domains of risk of bias were assessed as 'unclear'.
cDowngraded by one level for serious imprecision: small sample size.

Open in table viewer
Summary of findings 3. IUI in a stimulated cycle compared to TI or expectant management in a natural cycle for unexplained subfertility

IUI in stimulated cycle compared to TI or expectant management in natural cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI in stimulated cycle
Comparison: TI or expectant management in natural cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management in a natural cycle

IUI in a stimulated cycle

Live birth rate per couple (all cycles)‐ clomiphene citrate or letrozole

90 per 1000

307 per 1000
(165 to 497)

OR 4.48
(2 to 10.01)

201
(1 study)

⊕⊕⊕⊝
moderatea

‐

Live birth rate per couple (all cycles)‐ clomiphene citrate or gonadotropins

238 per 1000

204 per 1000
(123 to 318)

OR 0.82
(0.45 to 1.49)

253
(1 study)

⊕⊕⊝⊝
lowb

‐

Multiple pregnancy rate per couple

4 per 1000

13 per 1000
(2 to 79)

OR 3.01
(0.47 to 19.28)

454
(2 studies)

⊕⊕⊝⊝
lowb

‐

Pregnancy rate per couple (all cycles)‐ clomiphene citrate or letrozole

110 per 1000

366 per 1000
(215 to 549)

OR 4.68
(2.22 to 9.86)

201
(1 study)

⊕⊕⊕⊝
moderatea

‐

Ovarian hyperstimulation syndrome rate per woman

See comment

See comment

See comment

See comment

See comment

No events in intervention or control groups

Miscarriage rate per couple

31 per 1000

84 per 1000
(36 to 183)

OR 2.87
(1.18 to 7.01)

454
(2 studies)

⊕⊕⊕⊝
moderatea

‐

Ectopic pregnancy rate per couple

Not estimable (no events in intervention group)

‐

OR 9.28
(0.49 to 174.6)

201
(1 study)

⊕⊕⊝⊝
lowb

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by one level for serious imprecision: small sample size with a low event rate.
bDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Open in table viewer
Summary of findings 4. IUI in a natural cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility

IUI in natural cycle compared to TI or expectant management in stimulated cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI in natural cycle
Comparison: TI or expectant management in stimulated cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management in stimulated cycle

IUI in natural cycle

Live birth rate per couple (all cycles)

131 per 1000

227 per 1000
(142 to 341)

OR 1.95
(1.10 to 3.44)

342
(1 study)

⊕⊕⊕⊝
moderatea

‐

Multiple pregnancy rate per couple

6 per 1000

6 per 1000
(0 to 88)

OR 1.05
(0.07 to 16.90)

342
(1 study)

⊕⊕⊝⊝
lowb

‐

Pregnancy rate per couple (all cycles)

143 per 1000

228 per 1000
(144 to 339)

OR 1.77
(1.01 to 3.08)

342
(1 study)

⊕⊕⊕⊝
moderatea

‐

Ovarian hyperstimulation syndrome rate per woman ‐ not reported

See comment

See comment

‐

‐

See comment

No data reported on outcome

Miscarriage rater per couple

46 per 1000

42 per 1000
(15 to 111)

OR 0.91
(0.32 to 2.58)

342
(1 study)

⊕⊕⊝⊝
lowb

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 5.30
(0.25 to 111.26)

342
(1 study)

⊕⊕⊝⊝
lowb

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by one level for serious imprecision: small sample size with a low event rate.
bDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Open in table viewer
Summary of findings 5. IUI in a stimulated cycle compared to IUI in a natural cycle for unexplained subfertility

IUI in stimulated cycle compared to IUI in natural cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI in a stimulated cycle
Comparison: IUI in a natural cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

IUI in a natural cycle

IUI in a stimulated cycle

Live birth rare per couple (all cycles)

139 per 1000

250 per 1000
(165 to 361)

OR 2.07
(1.22 to 3.50)

396
(4 studies)

⊕⊕⊝⊝
lowa,b

‐

Multiple pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 3.00
(0.11 to 78.27)

39
(1 study)

⊕⊕⊝⊝
lowc

‐

Pregnancy rate per couple (all cycles)

63 per 1000

302 per 1000
(36 to 831)

OR 6.43
(0.56 to 73.35)

26
(1 study)

⊕⊕⊝⊝
lowc

‐

Ovarian hyperstimulation syndrome rate per woman ‐ not measured

See comment

See comment

Not estimable

‐

See comment

No events in intervention and control groups

Miscarriage rate per couple

Not estimable (no events in control group)

‐

OR 5.21
(0.19 to 141.08)

26
(1 study)

⊕⊕⊝⊝
lowc

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 6.48
(0.33 to 127.09)

211
(1 study)

⊕⊝⊝⊝
very lowb,c

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by one level for serious imprecision: small sample size with a low event rate.
bDowngraded by one level for serious risk of bias: sequence generation and allocation concealment rated as 'unclear' for the largest/larger study.
cDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Background

available in

Description of the condition

Of all couples presenting with fertility problems, about 25% have no cause that can be identified (NICE 2013). Couples are classified as having unexplained subfertility when they have tried to conceive for at least one year and the fertility work‐up has shown patent fallopian tubes, an ovulatory menstrual cycle and a normal semen analysis.

Description of the intervention

Intra‐uterine insemination (IUI) is a commonly‐used treatment in couples with unexplained subfertility. IUI is a relatively simple procedure in which semen is 'washed' in the laboratory and inserted in the uterine cavity using a small catheter at the time of ovulation. IUI can be performed with or without drugs for ovarian hyperstimulation (OH). For correct timing of the insemination, cycle monitoring is performed. This is usually done by ultrasound assessment of follicle growth or by monitoring the pre‐ovulatory luteinising hormone rise in blood or urine. In hyperstimulated cycles, ovulation is often induced by an injection of human chorionic gonadotropin (hCG), which improves timing possibilities. In contrast to IUI procedure, with expectant management couples either receive cycle monitoring for correct timing of sexual intercourse, i.e. timed intercourse (TI), or no intervention at all.

How the intervention might work

The rationale for performing IUI is that the motile spermatozoa, which are morphologically normal, can be concentrated in a small volume and placed directly into the uterus close to the released oocyte. In this way the cervix, which also acts as a reservoir for sperm, is bypassed. Accurate timing of the insemination is therefore of great importance. IUI can be performed with or without ovarian hyperstimulation (OH). The role of IUI in fertility treatment is often debated, in particular whether or not it is superior to TI, and whether or not OH should be used at the same time (Cohlen 2005; Hughes 2003; Stewart 2003). Commonly‐used drugs for ovarian hyperstimulation are clomiphene citrate (CC) and aromatase inhibitors, such as letrozole, both of which are oral treatments, and gonadotropins which are administered by subcutaneous injection. The aim of OH is to increase the number of oocytes available for fertilisation and to enhance accurate timing.

The use of OH in fertility treatment for unexplained subfertility is associated with benefits as well as increased risk of multiple pregnancies. When Hughes 1997 published a meta‐analysis indicating that the average fecundability is approximately five‐fold higher for treatment with IUI and OH, the Royal College of Obstetricians and Gynaecologists (RCOG 1998) concluded accordingly that "OH with IUI is an effective treatment for couples with unexplained infertility". However, major concerns were raised about the incidence of multiple pregnancies, and OH became less popular. These concerns have resulted in an adjustment of the advice for treatment of couples with unexplained subfertility. The NICE fertility guideline states that "ovarian hyperstimulation should not be offered to women with unexplained subfertility" (NICE 2013).

The increase in multiple pregnancies is a logical consequence of stimulated growth of multiple follicles. The incidence of multiple pregnancies after treatment with OH and IUI varies between 10% and 40%, and the overall contribution of this treatment to multiple births is estimated to be around 30% (Fauser 2005). However, recent studies have reported lower incidence of multiple pregnancies with OH. For example, a recent randomised controlled trial has reported multiple pregnancy rates of between 1.4% and 2.2% with OH (Danhof 2018); rates of between 3% and 10% have also been reported in another trial (Diamond 2015). The question is whether this multiple pregnancy rate is acceptable or whether it can be reduced to acceptable numbers. Recently, more evidence has been collected showing that mild ovarian hyperstimulation with strict cancellation criteria reduces the risk of achieving multiple pregnancies to approximately 10%, without compromising pregnancy rates (ESHRE 2006; Ragni 2006; Van Rumste 2006). Because maternal and neonatal morbidity and mortality rates are significantly increased in multiple pregnancies (Fauser 2005; Ombelet 2005), caregivers should take extra care to keep the multiple pregnancy rate to a minimum. Couples should be well‐informed by their physicians, especially as many couples wish to conceive twins (Ryan 2004) and prefer a higher pregnancy chance over safety.

Some authors have stated that treatment with OH results in an unacceptably high incidence of high‐order multiple pregnancies (Gleicher 2000; Nan 1994) and treatment with IUI in natural cycles should be preferred (Fauser 2005; Goverde 2005). Others have reported that the risk of multiple pregnancies could be reduced, with strict monitoring of the people undergoing treatment (Dickey 2005; Tur 2005). Te Velde 1999 concluded that IUI with OH is an appropriate treatment option if done with a mild stimulation protocol, careful cycle monitoring and with strict cancellation criteria. It is, however, still not known to what extent multiple pregnancies can be avoided if these criteria are met. Besides, the use of strict cycle cancellation criteria might result in a reduced overall pregnancy rate. Several trials using mild stimulation protocols for IUI have been published, showing promising results of acceptable pregnancy rates with very low multiple pregnancy rates (Balasch 2004). As IVF allows better control over reducing the risk of a multiple pregnancy (Gleicher 2000), and IVF with single‐embryo transfer is becoming more accepted, it has been argued that IVF is a safer treatment option than IUI with OH.

Why it is important to do this review

Although ovarian hyperstimulation seems to result in higher pregnancy rates, it also increases the risk of multiple pregnancies and ovarian hyperstimulation syndrome (OHSS). High risk of multiple pregnancies, for instance, increases the risk of neonatal morbidity and mortality and maternal morbidity (Guzick 1999; Ombelet 2006). Fertility guidelines recommend IUI without OH for couples with unexplained subfertility because of the increased risk of multiple pregnancies and OHSS associated with OH (NICE 2013). This systematic review, originally published in 2006, and updated in 2012 (Veltman‐Verhulst 2012) and 2016 (Veltman‐Verhulst 2016), was therefore undertaken to assess the evidence on the benefits and adverse events associated with IUI with or without OH, compared to timed intercourse or expectant management with or without OH, or with IUI with OH compared to IUI without OH, for couples with unexplained subfertility.

Objectives

available in

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.

Methods

available in

Criteria for considering studies for this review

Types of studies

Published and unpublished randomised controlled trials (RCTs) were eligible for inclusion. We excluded non‐randomised studies (e.g. studies with evidence of inadequate sequence generation such as alternate days, patient numbers) as they are associated with a high risk of bias. We attempted to contact the author of the study if the randomisation or allocation method was unclear.

We excluded trials that did not report separate data for women with unexplained subfertility or where such data were not obtainable from the authors. We assessed the trial design (cross‐over or parallel) and included cross‐over trials if we could extract pre‐cross‐over data.

Types of participants

I. Couples with unexplained subfertility, defined as follows.

  1. Normal ovulatory status (determined by either biphasic basal body temperature chart, normal luteal progesterone, in‐phase endometrial biopsy or ovulation detected with ultrasound).

  2. Tubal patency (determined by hysterosalpingography or laparoscopy, or both).

  3. A normal semen sample according to World Health Organization (WHO) criteria current at the time of the trial.

II. Couples who had tried to conceive for at least one year.

Participants excluded were: couples with a known cause of infertility including a moderate male factor, moderate to severe endometriosis (according to the American Society for Reproductive Medicine (ASRM) classification), tubal disease, or a cervical factor.

We contacted study authors to obtain data of couples with unexplained infertility if groups with mixed infertility causes were studied. If we could not extract relevant data separately for included participants, we excluded the study.

We excluded trials of participants with mild to moderate endometriosis only.

Types of interventions

Trials with at least one of the following comparisons:

  1. Intra‐uterine insemination (IUI) in a natural cycle versus timed intercourse (TI) or expectant management in a natural cycle;

  2. IUI in a stimulated cycle versus TI or expectant management in a stimulated cycle;

  3. IUI in a stimulated cycle versus TI or expectant management in a natural cycle;

  4. IUI in a natural cycle versus TI or expectant management in a stimulated cycle;

  5. IUI in a stimulated cycle versus IUI in a natural cycle.

Ovarian hyperstimulation (OH) was achieved with either clomiphene citrate, aromatase inhibitors (letrozole) or gonadotropins.
We included expectant management as a variant of timed intercourse.

We excluded the following interventions:

  1. intra‐cervical insemination, because we consider this to be a different treatment modality (Ripps 1994) and it is the topic of another Cochrane Review (Kop 2018);

  2. donor insemination.

Types of outcome measures

Primary outcomes

  1. Live birth rate per couple: all cycles. Live birth is defined as delivery of a live foetus after 20 completed weeks of gestational age;

  2. Multiple pregnancy rate per couple. Multiple pregnancies confirmed by ultrasound, with or without selective reduction, were recorded.

Secondary outcomes

  1. Pregnancy rate per couple: all cycles. Pregnancy includes clinical pregnancy, defined by the presence of an intra‐uterine gestational sac or foetal heartbeat visualised by an ultrasound scan before 12 weeks, and/or ongoing pregnancy, defined as a pregnancy extending beyond 12 weeks of gestation, confirmed by ultrasound or delivery.

Other adverse events:

  1. Moderate or severe ovarian hyperstimulation syndrome (OHSS), rate per woman;

  2. Miscarriage rate per couple;

  3. Ectopic pregnancy rate per couple.

We excluded pregnancies confirmed only by detection of hCG in serum or urine (biochemical pregnancies). When pregnancy was not further defined, and remained unclear even after contacting the authors, we assumed the pregnancy to be clinical.

Search methods for identification of studies

We searched for all reports which describe (or might describe) randomised controlled trials of IUI with or without OH, in couples with unexplained infertility. The original search was performed in 2005 and was last updated in October 2019, in consultation with the Gynaecology and Fertility Group (CGF) Information Specialist.

Electronic searches

We searched:

  1. Cochrane Gynaecology and Fertility Specialised Register; PROCITE platform (searched 17 October 2019) (Appendix 1);

  2. CENTRAL; via the Cochrane Register of Studies Online (CRSO); web platform (searched 17 October 2019) (Appendix 2);

  3. MEDLINE; OVID platform (searched from 1946 to 17 October 2019) (Appendix 3);

  4. Embase; OVID platform (searched from 1980 to 17 October 2019) (Appendix 4);

  5. PsycINFO; OVID platform (searched from 1806 to 17 October 2019) (Appendix 5);

  6. CINAHL; EBSCO platform (searched from 1961 to October 2019) (Appendix 6).

We combined the MEDLINE search with the Cochrane highly sensitive search strategy for identifying randomised trials in MEDLINE: sensitivity‐ and precision‐maximising version (2008 revision; Lefebvre 2011). We combined the Embase and PsycINFO searches with trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN) ( www.sign.ac.uk/methodology/filters.html#random).

Other electronic sources of trials included:

  1. Trial registers for ongoing and registered trials (up to October 2019):

    1. www.clinicaltrials.gov (a service of the US National Institutes of Health);

    2. www.who.int/trialsearch/Default.aspx (The World Health Organization International Trials Registry Platform search portal) Note: it is now mandatory for Cochrane Reviews to include searches of trial registers;

  2. Database of Abstracts of Reviews of Effects (DARE) in the Cochrane Library at onlinelibrary.wiley.com/o/cochrane/cochrane_cldare_articles_fs.html (for reference lists from relevant non‐Cochrane reviews) (up to October 2019);

  3. Web of Knowledge at wokinfo.com/ (another source of trials and conference abstracts) (up to October 2019);

  4. OpenGrey at www.opengrey.eu/ for unpublished literature from Europe (up to October 2019);

  5. LILACS database at regional.bvsalud.org/php/index.php?lang=en (for trials from the Portuguese‐ and Spanish‐speaking world) (up to October 2019);

  6. PubMed and Google Scholar (for recent trials not yet indexed in MEDLINE) (up to October 2019);

  7. Epistemonikos database at www.epistemonikos.org/en for systematic reviews that are useful for reference‐checking for trials (up to October 2019);

  8. European Society of Human Reproduction and Embryology (ESHRE) and American Society for Reproductive Medicine (ASRM) 2019 conference abstracts.

Searching other resources

We handsearched reference lists of articles retrieved by the search and contacted experts in the field to obtain additional data. We also handsearched relevant journals and conference abstracts that were not covered in the Cochrane Gynaecology and Fertility register, in liaison with the Information Specialist.

Data collection and analysis

Selection of studies

After an initial screening of titles and abstracts produced by the search, we retrieved the full texts of all potentially eligible studies. Two review authors (ROA and JDA) independently examined these full‐text articles for compliance with the inclusion criteria and selected studies eligible for inclusion in the 2020 update. We contacted study investigators as required, to clarify study eligibility. We resolved any disagreements about study eligibility by discussion. We documented the selection process with a 'PRISMA' flow chart (Liberati 2009).

Data extraction and management

Two review authors (ROA and JDA) independently extracted data from eligible studies using a data extraction form designed and pilot‐tested by the review authors, resolving any disagreements by discussion. Data extracted included study characteristics and outcome data. Where studies had multiple publications the review authors collated multiple reports of the same study, so that each study, rather than each report, was the unit of interest in the review, and such studies had a single study ID with multiple references. We contacted study investigators for further data on methods or results, or both, as required.

Assessment of risk of bias in included studies

Two review authors (ROA and JDA) independently assessed the included studies for risks of bias using the Cochrane 'Risk of bias' assessment tool (Higgins 2017) to assess: selection bias (random sequence generation and allocation concealment); performance bias (blinding of participants and personnel); detection bias (blinding of outcome assessors); attrition bias (incomplete outcome data); reporting bias (selective reporting); and other potential bias. We resolved disagreements by discussion or by involving a third review author. We described all judgements fully in the 'Risk of bias' table for each included study and incorporated our judgements into the interpretation of the review findings.

Measures of treatment effect

Only dichotomous data were reported in this review and we used the numbers of events in the control and intervention groups of each study to calculate Mantel‐Haenszel odds ratios (ORs). We reversed the direction of effect of individual studies, where required, to ensure consistency across trials. We presented 95% confidence intervals (CIs) for all outcomes. We assessed whether the estimates calculated in the review for individual studies were compatible in each case with the estimates reported in the study publications.

Unit of analysis issues

We analysed data per randomised couple or woman, because per‐treatment‐cycle data may lead to biased results (Dias 2008). In the case of a cross‐over trial, we only analysed data prior to cross‐over. For studies where data did not allow analysis (e.g. per‐cycle data) we contacted study authors for per‐woman data. Where we could not obtain appropriate data after contact with authors, we excluded such data (per‐cycle) from meta‐analyses. We counted multiple live births (e.g. twins or triplets) as one live birth event.

Dealing with missing data

We analysed the data on an intention‐to‐treat basis as far as possible. In the case of missing data we contacted authors of the published trials and included the newly‐obtained data in the analysis. However, where the study authors did not provide additional data, we assumed that no live births occurred in participants without a reported outcome. We assumed that women who dropped out or were excluded after randomisation were not pregnant. Women who were excluded because they conceived before receiving treatment were included as a success in the allocated group in the ITT analysis. For other outcomes, we analysed only the available data.

Assessment of heterogeneity

We considered whether the clinical and methodological characteristics of the included studies were sufficiently similar for meta‐analysis to provide a clinically meaningful summary. We assessed statistical heterogeneity using the I2 statistic. We took an I2 statistic measurement greater than 50% to indicate substantial heterogeneity (Deeks 2017).

Assessment of reporting biases

In view of the difficulty of detecting and correcting for publication bias and other reporting biases, we minimised their potential impact by ensuring a comprehensive search for eligible studies and by being alert for duplication of data. Where there were 10 or more studies in an analysis, we had intended to use a funnel plot to explore the possibility of small‐study effects (a tendency for estimates of the intervention effect to be more beneficial in smaller studies) (Sterne 2017).

Data synthesis

Where the studies were sufficiently similar, we combined the data using a fixed‐effect model in the following comparisons:

  1. Intra‐uterine insemination (IUI) in a natural cycle versus timed intercourse (TI) or expectant management in a natural cycle;

  2. IUI in a stimulated cycle versus TI or expectant management in a stimulated cycle;

  3. IUI in a stimulated cycle versus TI or expectant management in a natural cycle;

  4. IUI in a natural cycle versus TI or expectant management in a stimulated cycle;

  5. IUI in a stimulated cycle versus IUI in a natural cycle.

An increase in the odds of a particular outcome, which may be beneficial (e.g. live birth) or detrimental (e.g. adverse events), was displayed graphically in the meta‐analyses to the right of the centre‐line and a decrease in the odds of an outcome to the left of the centre‐line.

When pre‐cross‐over data were available, we included cross‐over trials in the analysis and pooled them with parallel trials.

Subgroup analysis and investigation of heterogeneity

Where data were available, we planned to conduct subgroup analyses to determine the separate evidence on the number of treatment cycles for live birth, and methods of ovarian hyperstimulation. However, there were insufficient data to perform subgroup analyses for the number of treatment cycles in any the comparisons, and for methods of ovarian hyperstimulation in some of the comparisons. We therefore did not report the findings by subgroups for any of the comparisons, except where data could not be pooled in meta‐analyses due to the presence of high heterogeneity. Where a visual scan of the forest plots or the results of statistical tests indicated substantial heterogeneity, we explored possible explanations in the text or in subgroup analyses.

Sensitivity analysis

Where appropriate, we performed sensitivity analyses for the review's primary outcomes (live birth and multiple pregnancy) to determine whether the results were robust to decisions made during the review process. These analyses considered the effects of excluding the following studies:

  1. Studies that did not clearly describe adequate procedures for random sequence generation and allocation concealment;

  2. Studies with a unit‐of‐analysis error (such as those in which cross‐over data were analysed as if they derived from parallel studies).

Overall quality of the body of evidence: 'Summary of findings' tables

We prepared 'Summary of findings' tables using GRADEpro software (GRADEpro GDT 2015) and Cochrane methods. These tables evaluate the overall quality of the body of evidence for the review outcomes (live birth, multiple pregnancy, pregnancy, OHSS, miscarriage, ectopic pregnancy) for the main review comparisons:

  1. IUI versus TI or expectant management both in a natural cycle;

  2. IUI in a stimulated cycle versus TI or expectant management in a stimulated cycle;

  3. IUI in a stimulated cycle versus TI or expectant management in a natural cycle;

  4. IUI in a natural cycle versus TI or expectant management in a stimulated cycle; and

  5. IUI in a stimulated cycle versus IUI in a natural cycle).

We assessed the quality of the evidence using GRADE criteria: risk of bias, consistency of effect, indirectness, imprecision and publication bias (Schünemann 2017). Two review authors working independently made judgements about evidence quality (high, moderate, low or very low), with disagreements resolved by discussion. We justified, documented, and incorporated our judgements into the reporting of results for each outcome.

Results

Description of studies

Results of the search

For the 2020 update, the searches identified 925 records (most of which were also identified during the previous updates) after removal of duplicates. We retrieved three full‐text articles, one of which was eligible for inclusion (Farquhar 2018). Fourteen studies were identified from the previous updates; we therefore include a total of 15 studies (21 references) in the review. See Figure 1 for details of the screening and selection process and Characteristics of included studies for studies included in the review.

Figure 1
Study flow diagram.

Study flow diagram.

We tried to contact all study authors to retrieve unpublished details. Responses from three study authors resulted in the exclusion of the following trials: Nulsen 1993; Prentice 1995; Serhal 1988. Seven study authors (Agarwal 2004; Arici 1994; Bhattacharya 2008; Guzick 1999; Melis 1995; Murdoch 1991; Steures 2006a) provided unpublished information or data, which we used in this analysis. Some authors could not provide us with the requested data, while others did not respond to our requests.

Included studies

Design

We included 15 RCTs. All trials were published in journals (Janko 1998 was published as an abstract only) and were available in English. The trials were carried out in different countries: UK (Bhattacharya 2008; Chung 1995; Murdoch 1991), USA (Arici 1994; Deaton 1990; Guzick 1999), Italy (Arcaini 1996; Melis 1995), the Netherlands (Goverde 2000;Steures 2006a), India (Agarwal 2004), New Zealand (Farquhar 2018), Slovakia (Janko 1998), Sweden (Karlstrom 1993), and multiple European countries (Crosignani 1991).

Twelve were parallel‐group studies (Agarwal 2004; Arcaini 1996; Bhattacharya 2008; Chung 1995; Farquhar 2018; Goverde 2000;Guzick 1999; Janko 1998; Karlstrom 1993; Melis 1995; Murdoch 1991; Steures 2006a), and three were cross‐over studies (Arici 1994; Crosignani 1991; Deaton 1990).

Power calculation

Bhattacharya 2008, Farquhar 2018, Goverde 2000 and Steures 2006a were the only studies which performed a power calculation. All four studies reached the targeted inclusion number to obtain enough power (80% to 90% with 5% level of significance). Further details about the included trials are provided in the Characteristics of included studies table.

Participants

The 15 trials comprised a total of 2068 women with unexplained subfertility.

The ages of the women were mentioned in most trials as either a mean ± standard deviation (SD) or a median and range. One trial (Janko 1998) did not report the women's ages. The overall age reported in the studies was similar. The mean or median age was between 30 years and 34 years (with comparable SDs). The maximum age of the participants was stated in six studies, only one of whom was above the age of 40 years (Arici 1994).

The duration of subfertility was given in 11 trials and reported as a mean duration or range. Five studies only included couples with subfertility for more than at least three years. The overall subfertility duration ranged from one year to 15 years.

Two studies used a prediction score of validated Hunault model (Hunault 2004), including women with a prediction score of natural conception of less than 30% (Farquhar 2018) or between 30% and 40% (Steures 2006a).

Types of subfertility

The definition of unexplained subfertility was similar between studies. Six trials enrolled participants with unexplained subfertility only. Five trials also included participants with male factor subfertility. In those studies the data for unexplained subfertility were either reported separately or obtained from the authors. One study selected couples with unexplained subfertility and an intermediate prognosis (Steures 2006a). Five studies reported the inclusion of women with either surgically‐corrected endometriosis (Deaton 1990), mild or stage II treated endometriosis (Guzick 1999) or minimal/mild endometriosis( Bhattacharya 2008; Farquhar 2018; Karlstrom 1993), which we considered to be unexplained subfertility. Melis 1995 specifically excluded participants if minor disorders such as minimal endometriosis were found in the investigation. Although our protocol stipulated only women with minimal and mild endometriosis, we decided to include Deaton 1990, despite the inclusion of three participants (out of 51 in total) with moderate endometriosis.

All studies reported a thorough fertility investigation, including a laparoscopy. A semen analysis was performed at least once in all studies. In nine studies the semen quality was reported according to the WHO criteria. Two studies (Arcaini 1996; Janko 1998) did not specify the criteria for a normal semen analysis. Chung 1995 used a sperm count per ejaculate instead of per ml. The data in Guzick 1999 were based only on a normal sperm count and a normal motility according to Kruger criteria.

Primary or secondary subfertility

Nine trials contained a mixed population of couples who had never achieved a pregnancy (primary subfertility) and those who had previously been pregnant (secondary subfertility). The remaining trials did not give any description for inclusion of people with secondary subfertility.

Previous treatment

Couples who have previously had failed fertility treatment have a lower probability of conception in subsequent treatment attempts. It is therefore important in fertility trials to report if couples have undergone previous treatment. Of the 15 included studies only two trials included couples who had previously had unsuccessful fertility treatment (Farquhar 2018; Melis 1995). Five trials did not include previously‐treated participants (Agarwal 2004; Arici 1994; Guzick 1999; Karlstrom 1993; Murdoch 1991) and the remaining trials did not provide information about previous treatment.

Interventions
Number of trials included per comparison

  1. IUI in a natural cycle versus expectant management in a natural cycle: one trial (Bhattacharya 2008)..

  2. IUI in a stimulated cycle versus TI in a stimulated cycle: seven trials (Agarwal 2004; Arcaini 1996; Chung 1995; Crosignani 1991; Janko 1998; Karlstrom 1993; Melis 199

  3. IUI in a stimulated cycle versus TI or expectant management in a natural cycle: three trials (Deaton 1990; Farquhar 2018; Steures 2006a).

  4. IUI in a natural cycle versus expectant management in a stimulated cycle: one trial (Bhattacharya 2008).

  5. IUI in a stimulated cycle versus IUI in natural cycle: four trials (Arici 1994; Goverde 2000; Guzick 1999; Murdoch 1991)

Agarwal 2004, although included in the review, was excluded from the primary analysis. This Indian study had a high dropout percentage (37%) in the treatment group which caused severely unbalanced groups. The main reason for dropout was financial constraints, so this introduces a considerable bias. In three of the included studies (Bhattacharya 2008; Farquhar 2018; Steures 2006a) expectant management was performed instead of TI.

Treatment

The treatment methods varied substantially between studies. Seven studies used gonadotropins for ovarian hyperstimulation. Arcaini 1996 offered both gonadotropins and clomiphene citrate, which resulted in a high‐dose hyperstimulation. Five studies used clomiphene only, one study used clomiphene and letrozole (Farquhar 2018) and Crosignani 1991 did not report the method of ovarian hyperstimulation. The different fertility centres in this multicentre trial used different treatments. More details on drug dose and method can be found in the prognostic factor table (Appendix 7) and the Characteristics of included studies table. Additional gonadotropin‐releasing hormone agonist (GnRHa) was used by Chung 1995 and Murdoch 1991. All studies used human chorionic gonadotropin (hCG) (5000 to 10,000 IU) for triggering ovulation. Chung 1995 also provided hCG in the post‐ovulatory phase.

The timing of IUI was similar among the studies. Follicle development was usually monitored by ultrasound scan (USS) and serum estradiol levels (serum‐E2). The hCG was given when the dominant follicles reached a mean diameter of 16 mm to 18 mm. Insemination was performed 30 hours to 48 hours after hCG administration. Arcaini 1996 performed a double insemination at 24 and 48 hours, and in Murdoch 1991 insemination took place on alternate days until ovulation was confirmed. Follicular development in natural cycles was monitored by ultrasound or luteinising hormone (LH) urine tests, and intercourse was advised at 12 hours to 40 hours after the hCG or LH surge. Couples were mostly advised to have intercourse more than once.

In the studies with expectant management instead of TI (Bhattacharya 2008; Farquhar 2018; Steures 2006a), couples were given general advice about the need for regular intercourse.

The number of cycles in included studies ranged from one to eight.

Cancellation criteria

The most serious adverse effects of ovarian hyperstimulation are multiple pregnancies and ovarian hyperstimulation syndrome (OHSS). These risks can both be reduced by the cancellation of the treatment cycle if excessive follicle stimulation occurs. It is important that fertility trials report the cancellation criteria they applied. Firstly, this ensures that participants were not exposed to a higher risk of multiple pregnancy or OHSS to increase the pregnancy rate, and secondly, it reduces the bias introduced by cancellation of treatment in initially randomised groups.

Eleven studies described criteria for cancellation of the treatment cycle. Insemination or hCG administration did not take place if the cancellation criteria were met. Six studies used serum‐E2 levels to determine over‐ or under‐stimulation as well as a maximum of dominant follicles (four follicles of a maximum 16 mm diameter). Arcaini 1996 accepted a maximum of six dominant follicles. Four studies did not describe any cancellation criteria.

Outcomes

Ten trials reported live birth, our primary outcome of interest. The other studies reported pregnancy as the main outcome. Pregnancy was confirmed by ultrasound in nine trials. In Guzick 1999 pregnancy was confirmed by two hCG measurements or live birth. Others did not report the method of pregnancy confirmation. The reported pregnancies were mostly clinical. The multiple pregnancy rate was mentioned in 13 trials, miscarriage in 11, ectopic pregnancy in 11, and OHSS in 10 trials. These events were often reported as total numbers or as post‐cross‐over data and therefore often could not be used in the meta‐analysis.

Excluded studies

For the 2020 update, we excluded two studies in addition to those previously excluded. One was excluded because it was a prospective cohort study (Van Eekelen 2019) while the other study (Zolghadri 2012) was excluded because it included a mix of women with unexplained infertility and polycystic ovarian syndrome, with no separate data for the two groups. We contacted the study authors for data on women with unexplained infertility but we did not receive any response. We had excluded 23 studies during the previous updates (see details below); we therefore excluded a total of 25 studies from this review update. See Characteristics of excluded studies.

For the 2016 update, 10 studies were excluded (Aanesen 2014; Check 2013; Barros Delgadillo 2008; Barros‐Delgadillo 2010; Kabouk 2010; Leanza 2014a; Leanza 2014b; Peeraer 2013; Wadhwa 2013; Xu 2014). Four studies were not RCTs (Check 2013; Barros Delgadillo 2008; Leanza 2014a; Leanza 2014b), four studies did not include a comparison of interest to this review (Barros‐Delgadillo 2010; Kabouk 2010; Peeraer 2013; Wadhwa 2013), one study was a cohort study (Aanesen 2014) and another study was ineligible as it investigated donor sperm (Xu 2014).

For the 2012 update, 13 studies were excluded (Aboulghar 1993; Doyle 1991; Evans 1991; Gregoriou 1995; Ho 1998; Kirby 1991; Martinez 1990; Martinez 1991; Nulsen 1993; Prentice 1995; Serhal 1988; Tummon 1997; Zikopoulos 1993). Two studies were found not to be randomised studies (Aboulghar 1993; Serhal 1988); inadequate method of randomisation was the reason for exclusion of another two trials (Nulsen 1993; Prentice 1995); one study (Tummon 1997) included women with endometriosis only and thus did not focus on unexplained subfertility; Martinez 1990 reported biochemically‐confirmed pregnancies only and was therefore excluded; Ho 1998 did not report separate data for couples with unexplained subfertility; six studies (Doyle 1991; Evans 1991; Gregoriou 1995; Kirby 1991; Martinez 1991; Zikopoulos 1993) reported pre‐ or post‐cross‐over 'per‐cycle' only, instead of 'per randomised woman', and therefore could not be included. We contacted authors of these studies to obtain relevant data such as pre‐cross‐over 'per randomised woman' data but we did not receive any response from the authors.

Studies awaiting classification

There are no studies awaiting classification. However if pre‐cross‐over data of the excluded studies become available we will reconsider inclusion and report the studies in an update of this review.

Ongoing studies

We identified one study which was ongoing and appeared to meet the inclusion criteria (NCT03455426).

Risk of bias in included studies

See the ’Risk of bias’ graph (Figure 2) and ’Risk of bias’ summary (Figure 3).

Figure 2
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

In nine of the included studies (Agarwal 2004; Arici 1994; Bhattacharya 2008; Chung 1995; Farquhar 2018; Goverde 2000; Melis 1995; Murdoch 1991; Steures 2006a), we considered the methods used for sequence generation and allocation concealment to be adequate and we therefore rated them as being at low risk of bias. In the remaining six trials (Arcaini 1996; Crosignani 1991; Deaton 1990; Guzick 1999; Janko 1998; Karlstrom 1993), the methods used for random sequence generation and allocation concealment were not sufficiently described to draw a conclusive judgement, so we rated the risk of bias as unclear.

Blinding

We assessed all the included studies as being at unclear risk for performance bias, as none of them reported blinding due to the nature of the interventions. For detection bias, we rated all the included studies as being at low risk; we considered that non‐blinding was unlikely to affect the outcomes of interest, as they were objectively assessed.

Incomplete outcome data

We used an intention‐to‐treat (ITT) analysis when possible. In three trials an ITT analysis was not possible (Crosignani 1991; Deaton 1990; Karlstrom 1993), as the trials only reported the number of participants analysed.

In Murdoch 1991 one woman became pregnant spontaneously between treatment cycles. This pregnancy resulted in a live birth and was entered as such in the analysis. Goverde 2000 also reported spontaneous pregnancies that occurred between treatment cycles. Because it was unclear in which group these pregnancies occurred, we could not use them in the ITT analysis.

Seven of the 15 included trials clearly mentioned the number of dropouts and the reasons for dropping out (Arici 1994; Deaton 1990; Farquhar 2018; Goverde 2000; Guzick 1999; Melis 1995; Steures 2006a). Murdoch 1991 reported the number of dropouts but did not give any information on reasons for dropping out. Bhattacharya 2008 had a dropout rate of less than 1%. The studies with the highest dropout rates were Arcaini 1996 (dropout of 20.6%) and Agarwal 2004 (19%). In Agarwal the couples mainly left the study for financial reasons, which resulted in an unevenly‐distributed dropout rate of 37% in the treatment group as compared to 1% in the control group. The dropout rate usually increased in studies with a longer follow‐up period. Because this review included trials with different durations, it was difficult to compare the dropout rates. We rated nine of the included studies as being at low risk of bias, three as unclear and another three as high risk of bias.

Selective reporting

There was a risk of selective reporting in some of the studies included in this review. Live birth data were not reported in five studies (Arcaini 1996; Crosignani 1991; Deaton 1990; Janko 1998; Karlstrom 1993). Adverse events were often not reported by group but as a study total, which could not be included in the analysis. Multiple pregnancy rates were not reported in two trials (Crosignani 1991; Janko 1998). We rated nine of the included studies as being at low risk of bias, five as being at unclear risk and one as being at high risk. We did not construct a funnel plot for any of the analyses, as none of them included up to 10 studies.

Other potential sources of bias

To reduce bias introduced by a cross‐over study design, we included pre‐cross‐over data only. Three studies used a cross‐over design (Arici 1994; Crosignani 1991; Deaton 1990). In this design participants were initially randomised to the treatment or control group but then crossed over to the other group after a certain number of treatment cycles. The duration of these studies varied from two to eight treatment cycles per couple. In two studies (Arici 1994; Crosignani 1991) the participants crossed over after one treatment cycle. In Deaton 1990 participants crossed over after four cycles.

We rated nine of the included studies as being at low risk of bias because baseline demographic characteristics of participants between the two treatment groups were similar. We rated the remaining six studies at unclear risk for this domain because there was insufficient information to make a conclusive judgement on the baseline demographic characteristics of participants.

Effects of interventions

See: Summary of findings for the main comparison IUI in a natural cycle compared to TI or expectant management in a natural cycle for unexplained subfertility; Summary of findings 2 IUI in a stimulated cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility; Summary of findings 3 IUI in a stimulated cycle compared to TI or expectant management in a natural cycle for unexplained subfertility; Summary of findings 4 IUI in a natural cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility; Summary of findings 5 IUI in a stimulated cycle compared to IUI in a natural cycle for unexplained subfertility

This section describes the results of the meta‐analyses and sensitivity analyses.

Comparison 1. IUI in a natural cycle versus timed intercourse (TI) or expectant management in a natural cycle

The results from this comparison were all obtained from Bhattacharya 2008. Data for the unexplained subfertility group only were provided by the trial author.

1.1 Live birth rate per couple (all cycles)

Analysis 1.1
It is uncertain whether treatment with IUI in a natural cycle results in a higher cumulative live birth rate compared to treatment with expectant management in a natural cycle (OR 1.60, 95% CI 0.92 to 2.78; 1 RCT, 334 women; low‐quality evidence). The evidence suggests that if we assume the chance of a live birth using expectant management to be 16%, that of IUI would be between 15% and 34%.

1.2 Multiple pregnancy rate per couple

Analysis 1.2

It is uncertain whether treatment with IUI in a natural cycle results in a lower multiple pregnancy rate compared to treatment with expectant management in a natural cycle (OR 0.50, 95% CI 0.04 to 5.53; 1 RCT, 334 women; low‐quality evidence). The evidence suggests that if we assume the risk of a multiple pregnancy using expectant management to be 1%, the risk using IUI would be between 0% and 6%.

We did not perform sensitivity analyses as only one trial with low risk of selection bas was included in the analysis for both primary outcomes.

1.3 Pregnancy rate per couple (all cycles)

Analysis 1.3
Compared to expectant management in a natural cycle, it is uncertain whether treatment with IUI in a natural cycle results in a higher cumulative pregnancy rate (OR 1.53, 95% CI 0.88 to 2.64; 1 RCT, 334 women; low‐quality evidence). The evidence suggests that if we assume the chance of a pregnancy using expectant management to be 16%, that of IUI would be between 15% and 34%.

Other adverse events
Moderate or severe ovarian hyperstimulation syndrome rate per woman

Data on OHSS were not reported.

1.4 Miscarriage rate per couple

Analysis 1.4

It is uncertain whether treatment with IUI in a natural cycle results in a lower miscarriage rate compared to treatment with expectant management in a natural cycle. Sixteen miscarriages were reported in a total of 334 couples, seven in the IUI group and nine in the expectant management group (OR 0.77, 95% CI 0.28 to 2.11; 1 RCT, 334 women; low‐quality evidence). The evidence suggests that if we assume the risk of a miscarriage using expectant management to be 5%, that of IUI would be between 2% and 11%.

1.5 Ectopic pregnancy rate per couple

Analysis 1.5
Compared to expectant management in a natural cycle, it is uncertain whether treatment with IUI in a natural cycle results in a lower ectopic pregnancy rate. Two ectopic pregnancies were reported in a total of 334 couples, with both occurring in the IUI group (OR 5.06, 95% CI 0.24 to 106.21; 1 RCT, 334 women; low‐quality evidence).

Comparison 2. IUI in a stimulated cycle versus TI or expectant management in a stimulated cycle

2.1 Live birth rate per couple (all cycles)

Analysis 2.1
Only two of the six trials included in the analysis reported live birth rates (Chung 1995; Melis 1995). It is uncertain whether treatment with IUI in a stimulated cycle results in a higher cumulative live birth rate 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). The evidence suggests that if we assume the chance of a live birth using TI to be 26%, that of IUI would be between 23% and 50%. Statistical heterogeneity was detected (P = 0.06, I2 = 72%) between the two studies, with inconsistency in the directions of effect estimates. This may be explained by the fact that all participants in Melis 1995 had previously received fertility treatment.

We did not perform sensitivity analysis excluding studies at unclear or high selection bias as both trials were at low risk of selection bias

2.2 Multiple pregnancy rate per couple

Analysis 2.2
Four studies reported their multiple pregnancies per treatment arm (Arcaini 1996; Chung 1995; Karlstrom 1993; Melis 1995), with a total of 17 multiple pregnancies in 316 couples. Arcaini 1996, Chung 1995 and Karlstrom 1993 reported one high‐order multiple pregnancy each. The studies reported 11 multiple pregnancies in the IUI group and six in the TI group (representing 13.5% of the total number of pregnancies in these studies). Pooling the data from these studies showed that it is uncertain whether treatment with IUI in a stimulated cycle results in a lower multiple pregnancy rate compared to treatment with TI in a stimulated cycle (OR 1.46, 95% CI 0.55 to 3.87; 4 RCTs, 316 women; I2 = 0%; low‐quality evidence). The evidence suggests that if we assume the risk of a multiple pregnancy using TI to be 4%, the risk using IUI would be between 2% and 15%.

2.3 Pregnancy rate per couple (all cycles)

Analysis 2.3
Six trials reported pregnancy rates per couple. There were 517 women included in this analysis, reporting 149 cumulative pregnancies. The result showed that treatment with IUI in a stimulated cycle may result in a higher cumulative pregnancy rate compared to treatment with TI in a stimulated cycle (OR 1.69, 95% CI 1.14 to 2.53; 6 RCTs, 517 women; I2 = 8%; low‐quality evidence). This suggests that if we assume the chance of a pregnancy with TI to be 23%, the chance of a pregnancy in women using IUI would be between 26% and 44%.

Other adverse events
2.4 Moderate to severe ovarian hyperstimulation syndrome rate per woman

Analysis 2.4
One case of OHSS was reported in 36 couples who were treated with IUI, while none was reported in 32 couples who received TI. It is, however, uncertain whether treatment with IUI in a stimulated cycle results in a lower OHSS rate compared to treatment with TI in a stimulated cycle (OR 2.75, 95% CI 0.11 to 69.83; 1 RCT, 68 women; very low‐quality evidence).

2.5 Miscarriage rate per couple

Analysis 2.5
Twenty‐seven miscarriages were reported in total. Fifteen were reported by treatment arm, nine in the IUI group and six in the TI group. It is uncertain whether treatment with IUI in a stimulated cycle results in a lower miscarriage rate compared to treatment with TI in a stimulated cycle (OR 1.66, 95% CI 0.56 to 4.88; I2 = 0%; 2 RCTs, 208 women; low‐quality evidence). The evidence suggests that if we assume the risk of miscarriage using TI to be 6%, the risk using IUI would be between 3% and 23%.

2.6 Ectopic pregnancy rate per couple

Analysis 2.6
One case of ectopic pregnancy, occurring in the IUI group, was reported by one study of 100 couples. It is, however, uncertain whether treatment with IUI in a stimulated cycle results in a higher ectopic pregnancy rate compared to treatment with TI in a stimulated cycle (OR 3.06, 95% CI 0.12 to 76.95; 1 RCT, 100 women; low‐quality evidence).

Comparison 3. IUI in a stimulated cycle versus TI or expectant management in a natural cycle

3.1 Live birth rate per couple (all cycles)

Analysis 3.1; Figure 4
Data on this outcome were reported by two studies (Farquhar 2018; Steures 2006a). We did not pool data from the two studies, because of extreme statistical heterogeneity (I2 = 91%). The presence of extreme heterogeneity could be due to the use of different population groups and OH drugs by the two studies: Farquhar 2018 investigated women with a low chance of natural conception (prediction score of natural conception less than 30%) using clomiphene citrate or letrozole, while Steures 2006a included couples with an intermediate chance of spontaneous conception (prediction score of natural conception between 30% and 40%), using clomiphene citrate or gonadotropins.

Figure 4
Forest plot of comparison 3: IUI in stimulated cycle versus TI or expectant management in natural cycle; outcome 3.1 Live birth rate per couple.

Forest plot of comparison 3: IUI in stimulated cycle versus TI or expectant management in natural cycle; outcome 3.1 Live birth rate per couple.

3.1.1 Clomiphene citrate or letrozole (Farquhar 2018)

In couples with a low prediction score of natural conception, treatment with IUI in a stimulated cycle probably results in a higher cumulative 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). The evidence suggests that if we assume the chance of a live birth with expectant management in a natural cycle to be 9%, the chance of a live birth with IUI in a stimulated cycle would be between 17% and 50%.

3.1.2 Clomiphene citrate or gonadotropins (Steures 2006a)

It is uncertain whether treatment with IUI in a stimulated cycle results in a higher cumulative live birth rate compared to treatment with expectant management in a natural cycle (OR 0.82, 95% CI 0.45 to 1.49; 1 RCT, 253 women; low‐quality evidence). The evidence suggests that if we assume the chance of a live birth with expectant management in a natural cycle to be 24%, the chance of a live birth with IUI in a stimulated cycle would be between 12% and 32%.

3.2 Multiple pregnancy rate per couple

Analysis 3.2

Data on this outcome were reported by Farquhar 2018 and Steures 2006a. It is uncertain whether treatment with IUI in a stimulated cycle results in a lower multiple pregnancy rate compared to treatment with expectant management in a natural cycle (OR 3.01, 95% CI 0.47 to 19.28; 2 RCTs, 454 women; I2 = 0%; low‐quality evidence). The evidence suggests that if we assume the risk of a multiple pregnancy with expectant management in a natural cycle to be 0%, the risk of a multiple pregnancy with IUI in a stimulated cycle would be between 0% and 8%.

We did not perform sensitivity analyses excluding studies at unclear or high selection bias, as both trials were at low risk of selection bias for the primary outcomes.

3.3 Pregnancy rate per couple (all cycles)

Analysis 3.3

Data on this outcome were reported by three studies (Deaton 1990; Farquhar 2018; Steures 2006a). We did not pool data from the three studies, because of extreme statistical heterogeneity (I2 = 86%). The presence of significant heterogeneity might be due to the use of different population groups and OH drugs by the three studies: Deaton 1990 investigated women with mild or moderate endometriosis, using clomiphene citrate only; Farquhar 2018 investigated women with a low chance of spontaneous conception (prediction score of natural conception less than 30%), using clomiphene citrate or letrozole; Steures 2006a included couples with an intermediate chance of spontaneous conception (prediction score of natural conception between 30% and 40%). using clomiphene citrate or gonadotropins.

3.3.1 Clomiphene citrate only (Deaton 1990)

It is uncertain whether treatment with IUI in a stimulated cycle results in a higher cumulative pregnancy rate compared to treatment with TI in a natural cycle (OR 3.20, 95% CI 0.82 to 12.50; 1 RCT, 51 women; low‐quality evidence).

3.3.2 Clomiphene citrate or letrozole (Farquhar 2018)

Treatment with IUI in a stimulated cycle probably results in a higher cumulative pregnancy rate compared to treatment with expectant management in a natural cycle (OR 4.68, 95% CI 2.22 to 9.86; 1 RCT, 201 women; moderate‐quality evidence). The evidence suggests that if we assume the chance of a pregnancy with expectant management in a natural cycle to be 11%, the chance of a pregnancy with IUI in a stimulated cycle would be between 22% and 55%.

3.3.3 Clomiphene citrate or gonadotropins (Steures 2006a)

It is uncertain whether treatment with IUI in a stimulated cycle results in a higher cumulative pregnancy rate compared to treatment with expectant management in a natural cycle (OR 0.80, 95% CI 0.45 to 1.42; 1 RCT, 253 women, low‐quality evidence).

Other adverse events
Moderate or severe ovarian hyperstimulation syndrome rate per woman

Two studies (Deaton 1990; Farquhar 2018) reported on OHSS in this comparison; there were no cases of OHSS in either study.

3.4 Miscarriage rate per couple

Analysis 3.4

Data on this outcome were reported by Farquhar 2018 and Steures 2006a. Treatment with IUI in a stimulated cycle probably results in a higher miscarriage rate compared to treatment with expectant management in a natural cycle (OR 2.87, 95% CI 1.18 to 7.01; 2 RCTs, 454 women; I2 = 0%; moderate‐quality evidence). The evidence suggests that if we assume the risk of miscarriage with expectant management in a natural cycle to be 3%, the risk of miscarriage with IUI in a stimulated cycle would be between 4% and 18%.

3.5 Ectopic pregnancy rate per couple

Analysis 3.5

Data on this outcome were reported only by Farquhar 2018. It is uncertain whether treatment with IUI in a stimulated cycle results in a lower ectopic pregnancy rate compared to treatment with expectant management in a natural cycle (OR 9.28, 95% CI 0.49 to 174.60; 1 RCT, 201 women; low‐quality evidence)

Comparison 4. IUI in a natural cycle versus TI or expectant management in a stimulated cycle

Bhattacharya 2008 studied this comparison with IUI in a natural cycle compared to TI in a clomiphene citrate‐stimulated cycle.

4.1 Live birth rate per couple (all cycles)

Analysis 4.1

Treatment with IUI in a natural cycle probably results in a higher cumulative live birth rate compared to expectant management in a stimulated cycle (OR 1.95, 95% CI 1.10 to 3.44; 1 RCT, 342 women: moderate‐quality evidence). The evidence suggests that if we assume the chance of a live birth with expectant management in a stimulated cycle to be 13%, the chance of a live birth with IUI in a natural cycle would be between 14% and 34%.

4.2 Multiple pregnancy per couple

Analysis 4.2

It is uncertain whether treatment with IUI in a natural cycle results in a lower multiple pregnancy rate compared to treatment with expectant management in a stimulated cycle (OR 1.05, 95% CI 0.07 to 16.90; 1 RCT, 342 women; low‐quality evidence). The evidence suggests that if we assume the risk of a multiple pregnancy in expectant management in a stimulated cycle to be 1%, the risk of a multiple pregnancy with IUI in a natural cycle would be between 0% and 9%.

We did not perform sensitivity analysis for the main outcomes, as we included only one study with low risk of selection bias in this comparison.

4.3 Pregnancy rate per couple (all cycles)

Treatment with IUI in a natural cycle probably results in a higher cumulative pregnancy rate compared to treatment with expectant management in a stimulated cycle (OR 1.77, 95% CI 1.01 to 3.08; 1 RCT, 342 women; moderate‐quality evidence). The evidence suggests that if we assume the chance of a pregnancy with expectant management in a stimulated cycle to be 14%, the chance of a pregnancy with IUI in a natural cycle would be between 14% and 34%.

Other adverse events
Moderate or severe ovarian hyperstimulation syndrome rate per woman

OHSS was not reported.

4.4 Miscarriage rate per couple

Analysis 4.4
It is uncertain whether treatment with IUI in a natural cycle results in a lower miscarriage rate compared to treatment with expectant management in a stimulated cycle (OR 0.91, 95% CI 0.32 to 2.58; 1 RCT, 342 women; low‐quality evidence). The evidence suggests that if we assume the risk of a miscarriage with expectant management in a stimulated cycle to be 5%, the risk of a miscarriage with IUI in a natural cycle would be between 2% and 11%.

4.5 Ectopic pregnancy

Analysis 4.5
Although two ectopic pregnancies occurred in the IUI group, it is uncertain whether treatment with IUI in a natural cycle results in a lower ectopic pregnancy rate compared to treatment with expectant management in a stimulated cycle (OR 5.30, 95% CI 0.25 to 111.26; 1 RCT, 342 women; low‐quality evidence).

Comparison 5. IUI in a stimulated cycle versus IUI in a natural cycle

5.1 Live birth rate per couple (all cycles)

Analysis 5.1
Three trials reported the number of live births per treatment arm (Arici 1994; Goverde 2000; Murdoch 1991). We obtained the live birth data from Guzick 1999 after contacting the study authors. Treatement 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; I2 = 0%; 4 RCTs, 396 women; low‐quality evidence). The evidence suggests that if we assume the chance of a live birth with IUI in a natural cycle to be 14%, the chance of a live birth with IUI in a stimulated cycle would be between 17% and 36%. In a sensitivity analysis, we obtained a similar result by excluding Arici 1994, a cross‐over trial (OR 2.02, 95% CI 1.18 to 3.45; I2 = 0%; 3 RCTs, 370 women); however, it is uncertain whether treatment with IUI in a stimulated cycle results in a higher cumulative live birth rate compared to treatment with IUI in a natural cycle when we excluded Guzick 1999, a study with unclear risk of bias for random sequence generation and allocation concealment (OR 1.69, 95% CI 0.83 to 3.44; I2 = 0%; 3 RCTs, 185 women).

5.2 Multiple pregnancy rate per couple

Analysis 5.2
It is uncertain whether treatment with IUI in a stimulated cycle results in a higher multiple pregnancy rate compared to treatment with IUI in a natural cycle (OR 3.00, 95% CI 0.11 to 78.27; 1 RCT, 39; women; low‐quality evidence) (Figure 5).

Figure 5
Forest plot of comparison 5: IUI in stimulated cycle versus IUI in a natural cycle; outcome 5.2 Multiple pregnancy rate per couple.

Forest plot of comparison 5: IUI in stimulated cycle versus IUI in a natural cycle; outcome 5.2 Multiple pregnancy rate per couple.

5.3 Pregnancy rate per couple (all cycles)

Analysis 5.3
It is uncertain whether treatment with IUI in a stimulated cycle results in a higher cumulative pregnancy rate compared to treatment with IUI in a natural cycle (OR 6.43, 95% CI 0.56 to 73.35; 1 RCT, 26 women; low‐quality evidence). The evidence suggests that if we assume the chance of a pregnancy with IUI in a natural cycle to be 6%, the chance of a pregnancy with IUI in a stimulated cycle would be between 4% and 83%.

Other adverse events
Moderate or severe ovarian hyperstimulation syndrome rate per woman

There were no reported cases of OHSS.

5.4 Miscarriage rate per couple

Analysis 5.4

One miscarriage in the stimulated group was reported by Arici 1994. Guzick 1999 reported a total miscarriage rate of approximately 24% in all couples undergoing IUI treatment. It is uncertain whether treatment with IUI in a stimulated cycle results in a higher miscarriage rate compared to treatment with IUI in a natural cycle (OR 5.21, 95% CI 0.19 to 141.08; 1 RCT, 26 women; low‐quality evidence).

5.5 Ectopic pregnancy rate per couple

Analysis 5.5
It is uncertain whether treatment with IUI in a stimulated cycle results in a higher ectopic pregnancy rate compared to treatment with IUI in a natural cycle (OR 6.48, 95% CI 0.33 to 127.09; 1 RCTs, 211 women; very low‐quality evidence).

Discussion

available in

Summary of main results

This is an update of a Cochrane Review which assessed whether the live birth rate is improved with intrauterine insemination (IUI), with or without ovarian hyperstimulation (OH), compared to timed intercourse (TI) or expectant management with or without OH, or with IUI with OH compared to IUI in a natural cycle, for couples with unexplained subfertility. Fifteen studies met the eligibility criteria for inclusion; they investigated different IUI treatment modalities in five comparisons involving a total of 2068 women. Primary outcomes reported were cumulative live birth rates and multiple pregnancy rates. Secondary outcomes reported were cumulative pregnancy rates and other adverse events, including moderate to severe ovarian hyperstimulation syndrome (OHSS) rates, miscarriage rates, and ectopic pregnancy rates.The five comparisons identified in the included studies are classified into two broad groups as follows:

1. IUI with or without OH versus TI or expectant management with or without OH

We identified four comparisons in this group:

  1. IUI in a natural cycle versus TI or expectant management in a natural cycle;

  2. IUI in a stimulated cycle versus TI or expectant management in a stimulated cycle;

  3. IUI in a stimulated cycle versus TI or expectant management in a natural cycle; and

  4. IUI in a natural cycle versus TI or expectant management in a stimulated cycle.

For live birth, it is uncertain whether treatment with IUI in a natural cycle results in a higher cumulative live birth rate compared to treatment with expectant management in a natural cycle. We obtained a similar result with treatment with IUI in a stimulated cycle compared to TI in a stimulated cycle. However, there is moderate‐quality evidence that treatment with IUI in a stimulated cycle is likely to result in a higher cumulative live birth rate compared to treatment with expectant management in a natural cycle in couples with a low prediction score of natural conception. The evidence suggests that if we assume the chance of a live birth with expectant management in a natural cycle to be 9%, the chance of a live birth with IUI in a stimulated cycle would be between 17% and 50%. Similarly, there is moderate‐quality evidence that treatment with IUI in a natural cycle is likely to result in a higher cumulative live birth rate compared to expectant management in a stimulated cycle. The evidence suggests that if we assume the chance of a live birth with expectant management in a stimulated cycle to be 13%, the chance of a live birth with IUI in a natural cycle would be between 14% and 34%.

For multiple pregnancy, it is uncertain whether treatment with IUI with or without OH results in lower multiple pregnancy rates compared to treatment with TI or expectant management, with or without OH. For other adverse events such as OHSS rates, miscarriage rates and ectopic pregnancy rates, there were insufficient data to adequately assess the impact of these treatment modalities; data on these outcomes were either not reported or reported in single studies with small sample sizes.

2. IUI in a stimulated cycle versus IUI in a natural cycle

For live birth, there is low‐quality evidence that 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. The evidence suggests that if we assume the chance of a live birth with IUI in a natural cycle to be 14%, the chance of a live birth with IUI in a stimulated cycle would be between 17% and 36%. We tested the robustness of this finding in a sensitivity analysis by excluding a study with unclear risk of bias for random sequence generation and allocation concealment; the result showed that without this trial there was probably little or no difference in live birth rates between the two treatment groups.

For multiple pregnancy, it is uncertain whether treatment with IUI in a stimulated cycle results in a lower multiple pregnancy rate compared with treatment with IUI in a natural cycle. For other adverse events such as OHSS rates, miscarriage rates and ectopic pregnancy rates, there were insufficient data to adequately assess the impact of this treatment modality; data on these outcomes were mostly reported in single studies with small sample sizes.

Overall completeness and applicability of evidence

Most of the comparisons identified in this review were investigated in single trials, thereby making it difficult to combine data in meta‐analyses. Where we could pool data in meta‐analyses, the overall effect estimates usually had wide confidence intervals due to small sample sizes with low numbers of events. Furthermore, we were not able to assess all the stipulated outcomes for each comparison. There were not enough data available to adequately assess the effects of the various interventions on live birth, multiple pregnancy and other adverse events. Many studies were not included in the review as they reported per‐cycle data only. One of the areas of particular concern in fertility trials is what statisticians refer to as the 'unit of analysis' error (Vail 2003). It is methodologically incorrect to report data per cycle when it is women or couples who are randomised, because many of the women would have undergone more than one treatment cycle (Dias 2008; Johnson 2003; Vail 2003). Yet pregnancy rate per cycle is a commonly‐reported outcome in fertility trials and reviews. We could also not include many studies with a cross‐over design in the review because they did not report pre‐cross‐over data. We could not include post‐cross‐over data because couples would have received both treatment modalities. There have been many discussions about whether a cross‐over design is appropriate for fertility trials, mainly because participants drop out after treatment success, which results in a selected participant population post‐cross‐over. For this reason, it has been suggested that the cross‐over design has no place in infertility trials (Daya 1993). A cross‐over design could result in an overestimation of the treatment effect (Khan 1996; Norman 2000). Whether this overestimation could be statistically corrected for or whether it is clinically relevant remains under debate (Cohlen 1998; McDonnell 2004; Vail 2003).

Adverse events were often not mentioned, or mentioned by group instead of by treatment modality. We included studies mentioning 'no events' in the meta‐analyses, which might result in underestimation of the adverse events. On the other hand, these adverse events are mostly attributed to OH and not to IUI. It seems therefore unlikely that any significant difference in adverse events would be found when OH with TI is compared to OH with IUI. We could not determine the separate evidence for the number of treatment cycles and methods of ovarian hyperstimulation, due to insufficient studies. In addition, we did not examine the effects of important prognostic factors such as age, duration of subfertility and previous treatment in subgroup analyses, although there would have been insufficient data to do these analyses if we had decided to do so. The extent to which some of the findings which show evidence of a difference between the intervention groups are applicable to couples with different prognoses and baseline fecundity is therefore unclear.

The more recent studies adopted expectant management as a control treatment for IUI instead of timed intercourse (Bhattacharya 2008; Farquhar 2018; Steures 2006a). Because timing of intercourse (TI) interferes with the natural coital habits of a couple, expectant management has been proposed as the more appropriate comparison treatment for IUI (Wilcox 1995). A recent meta‐analysis, however, shows no significant difference in the pregnancy rate between studies comparing IUI versus TI and studies with IUI versus expectant management (Snick 2008). Whether it is appropriate to pool studies including TI and expectant management in a meta‐analysis remains unclear. We did not, however, pool the data from Bhattacharya 2008, Farquhar 2018 and Steures 2006a with data of trials applying TI because these trials comprised comparisons that were not subject to many other trials.

Quality of the evidence

Most of the studies included in this review were of moderately‐good quality in relation to risk of bias. We excluded studies with a cross‐over design which did not report pre‐cross‐over data. Most of the included studies clearly described their methods of sequence generation and allocation concealment. None of the included studies reported blinding, due to the nature of the interventions; however, lack of blinding may not have any effect on the outcomes of interest, as they were objectively assessed. For most comparisons, effect estimates were associated with imprecision due to small sample sizes and wide confidence intervals. We assessed the quality of the evidence for the review’s five comparisons; the quality of the evidence for our primary outcomes (cumulative live birth and multiple pregnancy) ranged from low to moderate. The main limitations in the body of evidence were very serious imprecision due to small sample sizes with low event rates and effect estimates with wide confidence intervals, and inadequate reporting of adverse events. We could not examine the presence of publication or reporting bias in funnel plots for any of the comparisons, as none of the analyses included 10 or more studies. We therefore did not downgrade the quality of the evidence for publication bias.

Potential biases in the review process

To prevent selection bias, the included studies were independently selected and data extracted by two review authors (ROA and SVV for the 2020 update). We resolved disagreements by discussion, keeping the protocol as our guideline to reduce bias. We consulted a statistician (Andy Vail) for the data extraction of complicated cross‐over trials. We have extracted and calculated the events per randomised couple for this meta‐analysis, which resulted in the exclusion of six studies reporting only post‐cross‐over or per‐cycle data. As a consequence, the analysis is statistically less biased, although a selection bias may have occurred. We could not construct a funnel plot for any of the outcomes to check for the presence of publication bias as none of the outcomes was reported by 10 or more studies. We have, however, minimised the effect of publication bias by undertaking a comprehensive search for all eligible studies.

Agreements and disagreements with other studies or reviews

Previous review articles and meta‐analyses (e.g. Aboulghar 2003; Balasch 2004; Cohlen 2005; Costello 2004; Hughes 1997; Zeyneloglu 1998) have all assessed the effectiveness of IUI and ovarian hyperstimulation, but they have all calculated pregnancy rates per cycle, which makes it difficult to compare their findings with those of this review which included per‐couple data only. Per‐cycle data produce biased results that may exaggerate treatment results (Dias 2008). A Cochrane Review and meta‐analysis published in 2019 (Wang 2019) evaluated the effectiveness and safety of different approaches to clinical management (expectant management, ovarian stimulation, intra‐uterine insemination, ovarian stimulation‐intra‐uterine insemination, and in vitro‐fertilisation/intracytoplasmic sperm injection) in couples with unexplained infertility. There was insufficient evidence of differences in live births between intra‐uterine insemination or ovarian stimulation‐intra‐uterine insemination compared to expectant management (OR 1.21, 95% CI 0.61 to 2.43; low‐certainty evidence; OR 1.61, 95% CI 0.88 to 2.94; low‐certainty evidence respectively). These findings are similar to our findings on live births in the corresponding comparisons in our review.

Study flow diagram.
Figures and Tables -
Figure 1

Study flow diagram.

Navigate to figure in ReviewOpen in new tab

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Figures and Tables -
Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Navigate to figure in ReviewOpen in new tab

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Figures and Tables -
Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Navigate to figure in ReviewOpen in new tab

Forest plot of comparison 3: IUI in stimulated cycle versus TI or expectant management in natural cycle; outcome 3.1 Live birth rate per couple.
Figures and Tables -
Figure 4

Forest plot of comparison 3: IUI in stimulated cycle versus TI or expectant management in natural cycle; outcome 3.1 Live birth rate per couple.

Navigate to figure in ReviewOpen in new tab

Forest plot of comparison 5: IUI in stimulated cycle versus IUI in a natural cycle; outcome 5.2 Multiple pregnancy rate per couple.
Figures and Tables -
Figure 5

Forest plot of comparison 5: IUI in stimulated cycle versus IUI in a natural cycle; outcome 5.2 Multiple pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 1 Live birth rate per couple (all cycles).
Figures and Tables -
Analysis 1.1

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 1 Live birth rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 2 Multiple pregnancy rate per couple.
Figures and Tables -
Analysis 1.2

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 2 Multiple pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 3 Pregnancy rate per couple (all cycles).
Figures and Tables -
Analysis 1.3

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 3 Pregnancy rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 4 Miscarriage rate per couple.
Figures and Tables -
Analysis 1.4

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 4 Miscarriage rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 5 Ectopic pregnancy rate per couple.
Figures and Tables -
Analysis 1.5

Comparison 1 IUI versus TI or expectant management both in natural cycle, Outcome 5 Ectopic pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 1 Live birth rate per couple (all cycles).
Figures and Tables -
Analysis 2.1

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 1 Live birth rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 2 Multiple pregnancy rate per couple.
Figures and Tables -
Analysis 2.2

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 2 Multiple pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 3 Pregnancy rate per couple (all cycles).
Figures and Tables -
Analysis 2.3

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 3 Pregnancy rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 4 Moderate or severe ovarian hyperstimulation syndrome rate per woman.
Figures and Tables -
Analysis 2.4

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 4 Moderate or severe ovarian hyperstimulation syndrome rate per woman.

Navigate to figure in ReviewOpen in new tab

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 5 Miscarriage rate per couple.
Figures and Tables -
Analysis 2.5

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 5 Miscarriage rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 6 Ectopic pregnancy rate per couple.
Figures and Tables -
Analysis 2.6

Comparison 2 IUI versus TI or expectant management both in stimulated cycle, Outcome 6 Ectopic pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 1 Live birth rate per couple (all cycles).
Figures and Tables -
Analysis 3.1

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 1 Live birth rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 2 Multiple pregnancy rate per couple.
Figures and Tables -
Analysis 3.2

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 2 Multiple pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 3 Pregnancy rate per couple (all cycles).
Figures and Tables -
Analysis 3.3

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 3 Pregnancy rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 4 Miscarriage rate per couple.
Figures and Tables -
Analysis 3.4

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 4 Miscarriage rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 5 Ectopic pregnancy rate per couple.
Figures and Tables -
Analysis 3.5

Comparison 3 IUI in stimulated cycle versus TI or expectant management in natural cycle, Outcome 5 Ectopic pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 1 Live birth rate per couple (all cycles).
Figures and Tables -
Analysis 4.1

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 1 Live birth rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 2 Multiple pregnancy rate per couple.
Figures and Tables -
Analysis 4.2

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 2 Multiple pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 3 Pregnancy rate per couple (all cycles).
Figures and Tables -
Analysis 4.3

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 3 Pregnancy rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 4 Miscarriage rate per couple.
Figures and Tables -
Analysis 4.4

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 4 Miscarriage rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 5 Ectopic pregnancy rate per couple.
Figures and Tables -
Analysis 4.5

Comparison 4 IUI in natural cycle versus TI or expectant management in stimulated cycle, Outcome 5 Ectopic pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 1 Live birth rate per couple (all cycles).
Figures and Tables -
Analysis 5.1

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 1 Live birth rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 2 Multiple pregnancy rate per couple.
Figures and Tables -
Analysis 5.2

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 2 Multiple pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 3 Pregnancy rate per couple (all cycles).
Figures and Tables -
Analysis 5.3

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 3 Pregnancy rate per couple (all cycles).

Navigate to figure in ReviewOpen in new tab

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 4 Miscarriage rate per couple.
Figures and Tables -
Analysis 5.4

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 4 Miscarriage rate per couple.

Navigate to figure in ReviewOpen in new tab

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 5 Ectopic pregnancy rate per couple.
Figures and Tables -
Analysis 5.5

Comparison 5 IUI in stimulated cycle versus IUI in a natural cycle, Outcome 5 Ectopic pregnancy rate per couple.

Navigate to figure in ReviewOpen in new tab
Summary of findings for the main comparison. IUI in a natural cycle compared to TI or expectant management in a natural cycle for unexplained subfertility

IUI compared to TI or expectant management both in natural cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI
Comparison: TI or expectant management both in natural cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management both in natural cycle

IUI

Live birth rates per couple (all cycles)

156 per 1000

228 per 1000
(145 to 339)

OR 1.60
(0.92 to 2.78)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Multiple pregnancy rate per couple

12 per 1000

6 per 1000
(0 to 63)

OR 0.50
(0.04 to 5.53)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Pregnancy rate per couple (all cycles)

162 per 1000

228 per 1000
(145 to 338)

OR 1.53
(0.88 to 2.64)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Ovarian hyperstimulation syndrome rate per woman ‐ not reported

See comment

See comment

‐

‐

See comment

No data reported on outcome

Miscarriage rate per couple

54 per 1000

42 per 1000
(16 to 107)

OR 0.77
(0.28 to 2.11)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 05.06
(0.24 to 1106.21)

334
(1 study)

⊕⊕⊝⊝
lowa

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Figures and Tables -
Summary of findings for the main comparison. IUI in a natural cycle compared to TI or expectant management in a natural cycle for unexplained subfertility
Navigate to table in Review
Summary of findings 2. IUI in a stimulated cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility

IUI compared to TI or expectant management both in stimulated cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI
Comparison: TI or expectant management both in stimulated cycles

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management both in stimulated cycle

IUI

Live birth rate per couple (all cycles)

255 per 1000

352 per 1000
(231 to 496)

OR 1.59
(0.88 to 2.88)

208
(2 studies)

⊕⊕⊝⊝
lowa

‐

Multiple pregnancy rate per couple

43 per 1000

62 per 1000
(24 to 148)

OR 1.46
(0.55 to 3.87)

316
(4 studies)

⊕⊕⊝⊝
lowa

‐

Pregnancy rate per couple (all cycles)

234 per 1000

340 per 1000
(258 to 436)

OR 1.69
(1.14 to 2.53)

517
(6 studies)

⊕⊕⊝⊝
lowb,c

‐

Ovarian hyperstimulation syndrome rate per woman

Not estimable (no events in control group)

OR 2.75
(0.11 to 69.83)

68
(1 study)

⊕⊝⊝⊝
very lowa,b

‐

Miscarriage rate per couple

57 per 1000

91 per 1000
(33 to 228)

OR 1.66
(0.56 to 4.88)

208
(2 studies)

⊕⊕⊝⊝
lowa

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 3.06
(0.12 to 76.95)

100
(1 study)

⊕⊕⊝⊝
lowa

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.
bDowngraded by one level for serious risk of bias: most domains of risk of bias were assessed as 'unclear'.
cDowngraded by one level for serious imprecision: small sample size.

Figures and Tables -
Summary of findings 2. IUI in a stimulated cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility
Navigate to table in Review
Summary of findings 3. IUI in a stimulated cycle compared to TI or expectant management in a natural cycle for unexplained subfertility

IUI in stimulated cycle compared to TI or expectant management in natural cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI in stimulated cycle
Comparison: TI or expectant management in natural cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management in a natural cycle

IUI in a stimulated cycle

Live birth rate per couple (all cycles)‐ clomiphene citrate or letrozole

90 per 1000

307 per 1000
(165 to 497)

OR 4.48
(2 to 10.01)

201
(1 study)

⊕⊕⊕⊝
moderatea

‐

Live birth rate per couple (all cycles)‐ clomiphene citrate or gonadotropins

238 per 1000

204 per 1000
(123 to 318)

OR 0.82
(0.45 to 1.49)

253
(1 study)

⊕⊕⊝⊝
lowb

‐

Multiple pregnancy rate per couple

4 per 1000

13 per 1000
(2 to 79)

OR 3.01
(0.47 to 19.28)

454
(2 studies)

⊕⊕⊝⊝
lowb

‐

Pregnancy rate per couple (all cycles)‐ clomiphene citrate or letrozole

110 per 1000

366 per 1000
(215 to 549)

OR 4.68
(2.22 to 9.86)

201
(1 study)

⊕⊕⊕⊝
moderatea

‐

Ovarian hyperstimulation syndrome rate per woman

See comment

See comment

See comment

See comment

See comment

No events in intervention or control groups

Miscarriage rate per couple

31 per 1000

84 per 1000
(36 to 183)

OR 2.87
(1.18 to 7.01)

454
(2 studies)

⊕⊕⊕⊝
moderatea

‐

Ectopic pregnancy rate per couple

Not estimable (no events in intervention group)

‐

OR 9.28
(0.49 to 174.6)

201
(1 study)

⊕⊕⊝⊝
lowb

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by one level for serious imprecision: small sample size with a low event rate.
bDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Figures and Tables -
Summary of findings 3. IUI in a stimulated cycle compared to TI or expectant management in a natural cycle for unexplained subfertility
Navigate to table in Review
Summary of findings 4. IUI in a natural cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility

IUI in natural cycle compared to TI or expectant management in stimulated cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI in natural cycle
Comparison: TI or expectant management in stimulated cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

TI or expectant management in stimulated cycle

IUI in natural cycle

Live birth rate per couple (all cycles)

131 per 1000

227 per 1000
(142 to 341)

OR 1.95
(1.10 to 3.44)

342
(1 study)

⊕⊕⊕⊝
moderatea

‐

Multiple pregnancy rate per couple

6 per 1000

6 per 1000
(0 to 88)

OR 1.05
(0.07 to 16.90)

342
(1 study)

⊕⊕⊝⊝
lowb

‐

Pregnancy rate per couple (all cycles)

143 per 1000

228 per 1000
(144 to 339)

OR 1.77
(1.01 to 3.08)

342
(1 study)

⊕⊕⊕⊝
moderatea

‐

Ovarian hyperstimulation syndrome rate per woman ‐ not reported

See comment

See comment

‐

‐

See comment

No data reported on outcome

Miscarriage rater per couple

46 per 1000

42 per 1000
(15 to 111)

OR 0.91
(0.32 to 2.58)

342
(1 study)

⊕⊕⊝⊝
lowb

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 5.30
(0.25 to 111.26)

342
(1 study)

⊕⊕⊝⊝
lowb

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by one level for serious imprecision: small sample size with a low event rate.
bDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Figures and Tables -
Summary of findings 4. IUI in a natural cycle compared to TI or expectant management in a stimulated cycle for unexplained subfertility
Navigate to table in Review
Summary of findings 5. IUI in a stimulated cycle compared to IUI in a natural cycle for unexplained subfertility

IUI in stimulated cycle compared to IUI in natural cycle for unexplained subfertility

Patient or population: participants with unexplained subfertility
Settings: fertility clinic
Intervention: IUI in a stimulated cycle
Comparison: IUI in a natural cycle

Outcomes

Illustrative comparative risks* (95% CI)

Relative effect
(95% CI)

No of Participants
(studies)

Quality of the evidence
(GRADE)

Comments

Assumed risk

Corresponding risk

IUI in a natural cycle

IUI in a stimulated cycle

Live birth rare per couple (all cycles)

139 per 1000

250 per 1000
(165 to 361)

OR 2.07
(1.22 to 3.50)

396
(4 studies)

⊕⊕⊝⊝
lowa,b

‐

Multiple pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 3.00
(0.11 to 78.27)

39
(1 study)

⊕⊕⊝⊝
lowc

‐

Pregnancy rate per couple (all cycles)

63 per 1000

302 per 1000
(36 to 831)

OR 6.43
(0.56 to 73.35)

26
(1 study)

⊕⊕⊝⊝
lowc

‐

Ovarian hyperstimulation syndrome rate per woman ‐ not measured

See comment

See comment

Not estimable

‐

See comment

No events in intervention and control groups

Miscarriage rate per couple

Not estimable (no events in control group)

‐

OR 5.21
(0.19 to 141.08)

26
(1 study)

⊕⊕⊝⊝
lowc

‐

Ectopic pregnancy rate per couple

Not estimable (no events in control group)

‐

OR 6.48
(0.33 to 127.09)

211
(1 study)

⊕⊝⊝⊝
very lowb,c

‐

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; OR: Odds ratio;

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

aDowngraded by one level for serious imprecision: small sample size with a low event rate.
bDowngraded by one level for serious risk of bias: sequence generation and allocation concealment rated as 'unclear' for the largest/larger study.
cDowngraded by two levels for very serious imprecision: small sample size with a low event rate and effect estimate with a wide confidence interval.

Figures and Tables -
Summary of findings 5. IUI in a stimulated cycle compared to IUI in a natural cycle for unexplained subfertility
Navigate to table in Review
Comparison 1. IUI versus TI or expectant management both in natural cycle

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Live birth rate per couple (all cycles) Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

2 Multiple pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

3 Pregnancy rate per couple (all cycles) Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

4 Miscarriage rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

5 Ectopic pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected
Figures and Tables -
Comparison 1. IUI versus TI or expectant management both in natural cycle
Navigate to table in Review
Comparison 2. IUI versus TI or expectant management both in stimulated cycle

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Live birth rate per couple (all cycles) Show forest plot

2

208

Odds Ratio (M‐H, Fixed, 95% CI)

1.59 [0.88, 2.88]

1.1 Gonadotropins

2

208

Odds Ratio (M‐H, Fixed, 95% CI)

1.59 [0.88, 2.88]

2 Multiple pregnancy rate per couple Show forest plot

4

316

Odds Ratio (M‐H, Fixed, 95% CI)

1.46 [0.55, 3.87]

2.1 Clomiphene Citrate

1

40

Odds Ratio (M‐H, Fixed, 95% CI)

0.43 [0.02, 11.18]

2.2 Gonadotropins

2

208

Odds Ratio (M‐H, Fixed, 95% CI)

1.61 [0.44, 5.89]

2.3 Clomiphene Citrate and Gonadotropins

1

68

Odds Ratio (M‐H, Fixed, 95% CI)

1.88 [0.32, 11.00]

3 Pregnancy rate per couple (all cycles) Show forest plot

6

517

Odds Ratio (M‐H, Fixed, 95% CI)

1.69 [1.14, 2.53]

3.1 Clomiphene Citrate

1

40

Odds Ratio (M‐H, Fixed, 95% CI)

0.30 [0.03, 2.93]

3.2 Gonadotropins

4

319

Odds Ratio (M‐H, Fixed, 95% CI)

1.68 [1.03, 2.75]

3.3 Clomiphene Citrate and Gonadotropins

1

68

Odds Ratio (M‐H, Fixed, 95% CI)

2.62 [0.98, 6.98]

3.4 Clomiphene citrate OR Gonadotropins

1

90

Odds Ratio (M‐H, Fixed, 95% CI)

1.72 [0.50, 5.89]

4 Moderate or severe ovarian hyperstimulation syndrome rate per woman Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

4.1 Clomiphene Citrate and Gonadotropins

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

5 Miscarriage rate per couple Show forest plot

2

208

Odds Ratio (M‐H, Fixed, 95% CI)

1.66 [0.56, 4.88]

5.1 Gonadotropins

2

208

Odds Ratio (M‐H, Fixed, 95% CI)

1.66 [0.56, 4.88]

6 Ectopic pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

6.1 Gonadotropins

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]
Figures and Tables -
Comparison 2. IUI versus TI or expectant management both in stimulated cycle
Navigate to table in Review
Comparison 3. IUI in stimulated cycle versus TI or expectant management in natural cycle

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Live birth rate per couple (all cycles) Show forest plot

2

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

1.1 Clomiphene Citrate or Letrozole

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

1.2 Clomiphene Citrate or Gonadotropins

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

2 Multiple pregnancy rate per couple Show forest plot

2

454

Odds Ratio (M‐H, Fixed, 95% CI)

3.01 [0.47, 19.28]

2.1 Clomiphene Citrate or Letrozole

1

201

Odds Ratio (M‐H, Fixed, 95% CI)

5.05 [0.24, 106.53]

2.2 Clomiphene Citrate or Gonadotropins

1

253

Odds Ratio (M‐H, Fixed, 95% CI)

2.0 [0.18, 22.34]

3 Pregnancy rate per couple (all cycles) Show forest plot

3

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

3.1 Clomiphene Citrate

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

3.2 Clomiphene Citrate or Letrozole

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

3.3 Clomiphene Citrate or Gonadotropins

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

4 Miscarriage rate per couple Show forest plot

2

454

Odds Ratio (M‐H, Fixed, 95% CI)

2.87 [1.18, 7.01]

4.1 Clomiphene Citrate or Letrozole

1

201

Odds Ratio (M‐H, Fixed, 95% CI)

6.25 [0.74, 52.91]

4.2 Clomiphene Citrate or Gonadotropins

1

253

Odds Ratio (M‐H, Fixed, 95% CI)

2.28 [0.84, 6.20]

5 Ectopic pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected
Figures and Tables -
Comparison 3. IUI in stimulated cycle versus TI or expectant management in natural cycle
Navigate to table in Review
Comparison 4. IUI in natural cycle versus TI or expectant management in stimulated cycle

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Live birth rate per couple (all cycles) Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

2 Multiple pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

3 Pregnancy rate per couple (all cycles) Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

4 Miscarriage rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

5 Ectopic pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Random, 95% CI)

Totals not selected
Figures and Tables -
Comparison 4. IUI in natural cycle versus TI or expectant management in stimulated cycle
Navigate to table in Review
Comparison 5. IUI in stimulated cycle versus IUI in a natural cycle

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

1 Live birth rate per couple (all cycles) Show forest plot

4

396

Odds Ratio (M‐H, Fixed, 95% CI)

2.07 [1.22, 3.50]

1.1 Clomiphene Citrate

1

26

Odds Ratio (M‐H, Fixed, 95% CI)

3.75 [0.29, 47.99]

1.2 Gonadotropins

3

370

Odds Ratio (M‐H, Fixed, 95% CI)

2.02 [1.18, 3.45]

2 Multiple pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

2.1 Gonadotropins

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

3 Pregnancy rate per couple (all cycles) Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

3.1 Clomiphene Citrate

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

4 Miscarriage rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

4.1 Clomiphene Citrate

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

4.2 Gonadotropins

0

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]

5 Ectopic pregnancy rate per couple Show forest plot

1

Odds Ratio (M‐H, Fixed, 95% CI)

Totals not selected

5.1 Gonadotropins

1

Odds Ratio (M‐H, Fixed, 95% CI)

0.0 [0.0, 0.0]
Figures and Tables -
Comparison 5. IUI in stimulated cycle versus IUI in a natural cycle
Navigate to table in Review