Medically assisted reproduction in the context of time

(1)

Scholten, I.

Publication date

2015

Document Version

Final published version

Link to publication

Citation for published version (APA):

Scholten, I. (2015). Medically assisted reproduction in the context of time.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

4

The effectiveness of

intrauterine insemination:

a matched cohort study

Submitted Irma Scholten Maud van Zijl Inge M. Custers Monique Brandes Judith Gianotten Paul J.Q. van der Linden Peter G.A. Hompes Fulco van der Veen Ben W.J. Mol

(3)

ABSTRACT

Intrauterine insemination is often used as first line treatment in couples with unexplained subfertility and a poor prognosis on natural conception, although it’s effectiveness is not proven. We performed a retrospective matched cohort study in which we compared pregnancy rates in couples who voluntarily dropped out of treatment with IUI to couples who continued treatment with IUI. Couples were matched on hospital, age, duration of subfertility, primary or secondary subfertility and diagnosis. After three years, there were 41 pregnancies in the IUI group (144 couples, 28%) versus 18 pregnancies in the no treatment group (72 couples, 25%) (RR 1.1 (0.59-2.2)(p=0.4)). The cumulative pregnancy rate after three years was 40% in both groups, showing no difference in time to ongoing pregnancy (shared frailty model p=0.86). In couples with unexplained subfertility and a poor prognosis for natural conception, treatment with IUI does not to add to expectant management. There is need for a randomized clinical trial comparing IUI with expectant management in these couples.

(4)

The effectiveness of intrauterine insemination: a matched cohort study

53

4 INTRODUCTION

Intrauterine insemination (IUI), with or without mild ovarian stimulation, is often used as first-line treatment for couples with unexplained subfertility (1,2). However, the NICE guideline on fertility, published in 2013, advices not to routinely offer IUI to subfertile couples, but to proceed immediately to treatment with in vitro fertilization (IVF) after a period of expectant management. The authors of the guideline base their recommendation for this strategy on the lack of evidence for the effectiveness of IUI in couples with unexplained subfertility and mild male subfertility (3). The evidence underpinning this recommendation is however very scarce indeed. For couples with unexplained subfertility and an intermediate prognosis of natural conception according to the Hunault model, there is only one trial that showed that immediate start with IUI does not lead to more live births compared to expectant management for the same period (RR 0.85 (95% CI 0.63-1.1)) (4,5). For the effectiveness of IUI in couples with a poor prognosis on natural conception there is no evidence at all, as trials comparing immediate IUI to expectant management have not been performed in these couples.

Indeed, IUI is applied as a first line infertility treatment in many countries, a practice that is based on tradition as well as on the wish to act, rather than to wait for natural conception. In view of this, a randomized clinical trial (RCT) allocating couples that have been trying to conceive for several years to a no-treatment arm, is difficult to accept, both for couples as well as their doctors. We therefore chose the only design currently feasible and conducted a retrospective matched cohort study in which we compared ongoing pregnancy rates in couples with unexplained subfertility who dropped out of an IUI program after one or two cycles (no-treatment group) versus couples who had also undergone one or two cycles, but who continued treatment with IUI (IUI group). To capture the overall chance of an ongoing pregnancy, cumulative pregnancy rates after three years were used as primary outcome. As couples who do not conceive after IUI usually progress to treatment with IVF, analysis in the IUI group was restricted to the moment IVF was started to ensure the proper evaluation of the effectiveness of IUI. A separate analysis was performed including the period after IVF cycles were started, to assess the effect of adding IVF to IUI in these couples.

METHODS

Subjects

We performed a retrospective matched cohort study among couples with an indication for IUI who received at least one treatment cycle. The couples had been treated with IUI between January 2000 and December 2008 in the fertility clinics of one university hospital and four teaching hospitals in the Netherlands.

(5)

All couples had been trying to conceive for at least 12 months, and had undergone a fertility work-up. This consisted of a medical history, assessment of ovulation either by basal body temperature, ultrasound and/or mid-luteal serum progesterone, semen analysis and a post-coital test. Tubal patency of both tubes was confirmed by hysterosalpingography and/or laparoscopy (6).

First, we identified couples who discontinued treatment for non-medical reasons after having received one or two IUI cycles (no-treatment group). While we would prefer to include couples who did not receive treatment at all to test the effectiveness of IUI, we included couples who voluntarily stopped treatment after one or two cycles as a proxy. We did so, because couples who do not receive treatment at all are not registered at any point and therefore cannot be identified for a retrospective analysis. The identified couples were matched in a 1:2 ratio to couples who had also received one or two cycles of IUI, but continued IUI for at least another cycle after those first two cycles (IUI group). Couples in the no-treatment group were included from the moment they finished their last IUI cycle. The matched couples were included from the moment they finished the same amount of cycles (i.e. one or two cycles). Matching was done for hospital, female age, duration of subfertility, diagnosis and subfertility being primary of secondary. Duration of subfertility was defined as time in months from the start of trying to conceive to the date of first treatment. Diagnoses selected for this study were unexplained subfertility, mild male subfertility, and cervical factor subfertility. These three diagnoses were included as no absolute factor for infertility had been identified in these couples hampering natural conception. Mild male subfertility was defined as a Total Motile Count below 10*106_{/mL. Cervical factor subfertility} was diagnosed by means of at least one well-timed PCT in which no progressive motile spermatozoa were seen in five high-power fields at a magnification of 400 (TMC at least 10 million).

Continuous matching factors were divided in groups. Female age was categorized as <20, 20-24, 25-29, 30-34, 35-39 or >40 years. Duration of subfertility was categorized as 12-17, 18-23, 24-25-29, 30-35 and >35 months. When a perfect match was not possible, we accepted a mismatch on diagnosis and, if needed, also on duration of subfertility or female age with a maximum difference of one class.

Treatment procedure

According to the national guideline that was valid in the study period, treatment with IUI was indicated in couples with unexplained subfertility for at least three years, for couples with cervical factor subfertility and for couples with mild male subfertility. Each hospital performed IUI according to local protocol, yet these protocols were partly standardized according to recommendations of the national guideline. The national guideline advised to perform IUI with mild ovarian stimulation for couples with unexplained subfertility, while IUI without ovarian stimulation was advised in cervical factor and male subfertility. It was advised to perform at least six treatment cycles before progressing towards IVF (NVOG 1999). In general, both clomiphene citrated and FSH could be

(6)

55

4

used for stimulation, aiming for 2-3 follicles of at least 16mm. Ovulation was induced with a HCG injection or, in the natural cycle, detected by a rise in urinary LH level. Insemination with processed semen (0.2-0.5 ml) was performed 20-30 hours after the urinary LH peak or 38-42 hours after HCG injection. Luteal phase support was not given (7).

Follow up

All included women were followed for a maximum of three years after the first IUI or until an ongoing pregnancy leading to live birth occurred. We registered all treatment cycles, i.e. IUI or IVF/ ICSI and their stimulation regimens. The primary outcome was cumulative ongoing pregnancy rate after three years. An ongoing pregnancy was defined as a pregnancy confirmed at ultrasound at a gestational age of at least 10 weeks.

Data were collected through searching medical charts. When this information was insufficient, we contacted the couple, either by phone or questionnaire, to gain further information.

Analysis

The Cochrane’s Q test for non-parametrical matched data was used to compare the absolute number of pregnancies. Cumulative pregnancy rates were calculated by plotting Kaplan-Meier curves. We used a shared frailty model to compare cumulative pregnancy rates between groups, accounting for the matching of couples (8).

Time to pregnancy was censored at the moment a couple was lost-to-follow up, when active child wish ended, or when couples in the no treatment group decided to re-start treatment. Since we aimed to assess the effectiveness of IUI, time to pregnancy was censored in the IUI group at the moment a couple started treatment with IVF. An additional analysis was performed including these IVF cycles to assess the effect of IVF treatment on cumulative pregnancy rates.

Ethical approval

The current observational study with retrospective data collection was exempt from Institutional Review Board (IRB) approval, as under the Dutch Medical Research Involving Human Subjects Act, IRB approval is only required when patients are subjected to an intervention.

RESULTS

We identified 72 couples that fulfilled the inclusion criteria, and that discontinued IUI after one or two treatment cycles (no-treatment group). These couples were matched to 144 couples who continued IUI after having had 1 or 2 cycles (IUI group). In both groups, mean female age was 32.5 years. Median duration of subfertility was 32 months in the IUI group and 30 months in the no-treatment

(7)

group respectively. Unexplained subfertility was the most frequent diagnosis in both groups (68% versus 63%), while the majority of couples suffered from primary subfertility (62% versus 57%). The mean chance on natural conception within 12 months, calculated at baseline, was 23% in the IUI group versus 22% in the no-treatment group. (Table 1).

Table 1. Baseline characteristics

IUI group (n=144) No treatment (n=72) Age years (SD) 32.5 (4.5) 32.5 (4.5)

Duration of subfertility median in months 30 32

Primary subfertility % (n) 62% (n=89) 57% (n=41)

Diagnosis

Unexplained subfertility % (n) 69% (n=100) 65% (n=47)

Mild male factor % (n) 19% (n=28) 22% (n=16)

Cervical factor % (n) 11% (n=16) 13% (n=9)

Prognosis on natural conception % 23% 22%

After three years, there were 41 ongoing pregnancies (11 after natural conception, 30 after IUI, 1 unknown origin) in the IUI group (28%) versus 18 ongoing pregnancies, all after natural conception, in the no-treatment group (25%) (RR 1.1 (95% CI 0.59-2.2) (Cochrane Q p=0.4)). Of the pregnancies due to IUI, the majority was in a stimulated cycle (n=23, 77%). Three IUI pregnancies originated in a natural cycle (10%) and of four IUI pregnancies (13%), it was unknown whether they occurred in a natural or stimulated cycle. Thus, the ongoing pregnancy rates were 3/131 (2.3%) per natural cycle and 23/409 (5.6%) per stimulated cycle. There was one multiple pregnancy in the IUI group after a stimulated cycle versus no multiple pregnancies in the no-treatment group. The Kaplan-Meier curves showed cumulative pregnancy rates after three years of 40% in both the IUI group and the no-treatment group (shared frailty model: p=0.86). (Figure 1, Table 2)

In a secondary analysis we considered the cumulative pregnancy rates after three years including the period after IVF was started for couples in the IUI group. In this analysis there were 75 ongoing pregnancies,12 after natural conception, 30 after IUI, 32 after IVF and 1 unknown in the IUI group (52%) compared to the previously mentioned 18 pregnancies (25%) in the no-treatment group (RR 1.7 (95% 1.1-2.8)(Cochrane Q p<0.01)). The Kaplan-Meier curve showed a cumulative pregnancy rate of 60% in the IUI group (shared frailty model: p=0.05) (Figure 2, Table 2). Overall, there were 7 multiple pregnancies in the IUI group including the period after start IVF, 1 after IUI and 6 after IVF. Detailed information on the treatment cycles in the IUI group is shown in table 2. The majority of IUI cycles was stimulated (n=409, 76%).

(8)

57

4

Figure 1. Cumulative pregnancy rates after three years

Duration of follow up (months)

36 24

12 0

Cumulative ongoing pregnancy (%)

1,0 0,8 0,6 0,4 0,2 0,0 IUI group No treatment group Page 1

Table 2. Outcomes and treatment cycles (including IVF cycles)

IUI group n=144

No treatment n=72

Natural concepted pregnancies 12 18

IUI cycles (natural cycle) 131 na

Pregnancies after IUI (natural cycle) 3 na

Pregnancy rate per cycle 2.3% na

IUI cycles (stimulated cycle) 409 na

Pregnancies after IUI (stimulated cycle) 23 na

Pregnancy rate per cycle 5.6% na

IVF cycles 159 na

Pregnancies after IVF 32 na

Pregnancy rate per cycle 20% na

(9)

Figure 2. Cumulative pregnancy rates after three years, including IVF cycles

DISCUSSION

In this matched cohort study we found no indication that treatment with IUI was effective over no treatment after two failed IUI cycles, in couples with unexplained subfertility and a poor prognosis on natural conception. Only when in vitro fertilization (IVF) cycles were performed, treatment resulted in additional pregnancies over no-treatment. Our data question the effectiveness of IUI in couples with unexplained subfertility or mild male subfertility, even when the prognosis for natural conception is poor.

The strength of this study is that the chosen design -at this moment in time- provides the best evidence possible for the effectiveness of IUI in couples with poor chances for natural conception. We used matching on all known important prognostic factors for pregnancy to minimize selection bias (9,10). We applied strict censoring criteria after inclusion and selected a population with a poor prognosis on natural conception as shown by the baseline characteristics and the calculated chances of natural conception within a year, enhanced by the inclusion of couples who did not conceive after one or two cycles of IUI (5).

The drawback of this study is obviously its retrospective design in which we use one- or two failed IUI cycles as an entry criterion. As stated before, it is difficult to identify couples who received solely

Duration of follow up in months

36 24

12 0

Cumulative ongoing pregnancy (%)

1,0 0,8 0,6 0,4 0,2 0,0 IUI group No treatment group Page 1

(10)

59

4

expectant management, as they are not registered. One could comment that by using this proxy, we identify couples with an even worse prognosis as the majority of pregnancies after IUI occur in the first three cycles and therewith bias our results (11). However, this only strengthens this paper, as we are interested to test effectiveness of IUI in couples with poor prognosis. Not becoming pregnant after the first cycles of IUI reflects a group of patients with poor prognosis indeed.

The relevance of our study is that it questions the effectiveness of IUI. In our opinion, one cannot qualify these data as of low quality, while in the meantime continuing an IUI program without evidence of its effectiveness.

In our study, some couples with unexplained subfertility received IUI without stimulation, despite the knowledge that IUI without ovarian stimulation is an ineffective treatment (12,13). The Dutch guideline at that time left room for IUI without stimulation in these couples by acknowledging the higher burden of IUI with stimulation, which was advised to be taken into account when deciding on IUI with or without stimulation (7). Currently, diagnosing cervical factor subfertility is not regarded anymore as relevant, while it was diagnosed in 11% of couples in our study for whom IUI without stimulation was advised. Since the PCT is abandoned, these couples are now diagnosed as having unexplained subfertility and treated with IUI and ovarian stimulation anyhow (14).

The per cycle pregnancy rate for IUI cycles of 2.3% for unstimulated cycles and of 5.6% for stimulated cycles might be lower than success rates reported in other studies. Yet, the success rate of IUI is probably dependent on the prognosis of the couple and it might well be that the poor prognosis of the couples, we studied explains the low success rates.

In this study we used three year pregnancy rates to capture to overall potential of couples to conceive naturally as, although declining, there remains a change for natural conception each cycle (15). Showing comparable ongoing pregnancy rates after three years, the current study agrees with the NICE guideline that IUI should not be routinely offered. The NICE guideline suggests to consider immediate start of IVF treatment after two years of unsuccessful intercourse (3). However, a recently published trial, that compared IUI with mild stimulation to immediate IVF in couples with unexplained subfertility and a poor prognosis, found treatment with IVF comparable to treatment with IUI (16). Since no treatment was not evaluated in that trial, it does not allow conclusions on the effectiveness of IUI. The high conception rate in the non-treatment cycles in that study, however, justify a randomized clinical trial comparing IUI with mild stimulation to expectant management in couples unexplained subfertility with a poor prognosis for natural conception.

(11)

REFERENCES

1. Steures P, van der Steeg JW, Hompes PG, van der Veen F, Mol BW. Intrauterine insemination in The Netherlands. Reprod Biomed Online. 2007 Jan;14(1):110–6.

2. Kupka MS, Ferraretti AP, de Mouzon J, Erb K, D’Hooghe T, Castilla JA, et al. Assisted reproductive technology in Europe, 2010: results generated from European registers by ESHRE†. Hum Reprod. 2014 Jul 27;0(0):1–15. 3. National Institute for Health and Clinical Excellence. Assessment and treatment for people with fertility problems.

2013.

4. Steures P, van der Steeg JW, Hompes PGA, Habbema JDF, Eijkemans MJC, Broekmans FJ, et al. Intrauterine insemination with controlled ovarian hyperstimulation versus expectant management for couples with unexplained subfertility and an intermediate prognosis: a randomised clinical trial. Lancet. 2006 Jul 15;368(9531):216–21. 5. Hunault CC, Habbema JDF, Eijkemans MJC, Collins JA, Evers JLH, te Velde ER. Two new prediction rules for

spontaneous pregnancy leading to live birth among subfertile couples, based on the synthesis of three previous models. Hum Reprod. 2004 Sep;19(9):2019–26.

6. Dutch society of Obstetrics and Gynaecology. Orienterend Fertiliteitsonderzoek. 2004. 7. Dutch society of Obstetrics and Gynaecology. Guideline Intrauterine Insemination. 1999.

8. Gerster M, Madsen M, Andersen PK. Matched survival data in a co-twin control design. Lifetime Data Anal. 2014 Jan;20(1):38–50.

9. Austin PC. Comparing paired vs non-paired statistical methods of analyses when making inferences about absolute risk reductions in propensity-score matched samples. Stat Med. 2011 May 20;30(11):1292–301.

10. Leushuis E, van der Steeg JW, Steures P, Bossuyt PMM, Eijkemans MJC, van der Veen F, et al. Prediction models in reproductive medicine: a critical appraisal. Hum Reprod Update. 2009;15(5):537–52.

11. Custers IM, Steures P, Hompes P, Flierman P, van Kasteren Y, van Dop PA, et al. Intrauterine insemination: how many cycles should we perform? Hum Reprod. 2008 Apr;23(4):885–8.

12. Guzick DS, Carson SA, Coutifaris C, Overstreet JW, Factor-Litvak P, Steinkampf MP, et al. Efficacy of superovulation and intrauterine insemination in the treatment of infertility. National Cooperative Reproductive Medicine Network. N Engl J Med. 1999 Jan 21;340(3):177–83.

13. Bhattacharya S, Harrild K, Mollison J, Wordsworth S, Tay C, Harrold A, et al. Clomifene citrate or unstimulated intrauterine insemination compared with expectant management for unexplained infertility: pragmatic randomised controlled trial. BMJ. 2008 Jan;337:a716.

14. Scholten I, Moolenaar LM, Gianotten J, van der Veen F, Hompes PGA, Mol BWJ, et al. Long term outcome in subfertile couples with isolated cervical factor. Eur J Obstet Gynecol Reprod Biol. 2013 Oct;170(2):429–33. 15. Gnoth C, Godehardt D, Godehardt E, Frank-Herrmann P, Freundl G. Time to pregnancy: results of the German

prospective study and impact on the management of infertility. Hum Reprod. 2003 Sep 1;18(9):1959–66. 16. Bensdorp AJ, Tjon-Kon-Fat RI, Bossuyt PMM, Koks CAM, Oosterhuis GJE, Hoek A, et al. Prevention of multiple

pregnancies in couples with unexplained or mild male subfertility: randomised controlled trial of in vitro fertilisation with single embryo transfer or in vitro fertilisation in modified natural cycle compared with intrauterine inse. BMJ. 2015 Jan;350:g7771.