- Email: [email protected]
Patients' preferences for intrauterine insemination or in-vitro fertilization
RBMOnline - Vol 15. No 4. 2007 422-427 Reproductive BioMedicine Online; www.rbmonline.com/Article/2861 on web 31 July 2007
!RTICLE 0ATIENTS PREFERENCES FOR INTRAUTERINE INSEMINATION OR IN VITRO FERTILIZATION $R VAN 7EERT GRADUATED FROM MEDICAL SCHOOL IN AT THE 5NIVERSITY OF !MSTERDAM AND STARTED HER GYNAECOLOGICAL RESIDENCY IN )N SHE COMPLETED HER TRAINING AND SHE IS NOW A STAFF MEMBER OF THE GYNAECOLOGICAL TEAM AT THE -EDICAL #ENTRE !LKMAAR IN 4HE .ETHERLANDS $URING HER RESIDENCY SHE PARTICIPATED IN THE RESEARCH OF THE #ENTRE FOR 2EPRODUCTIVE -EDICINE AT THE !CADEMICAL -EDICAL #ENTRE IN !MSTERDAM 3HE HOPES TO lNISH HER THESIS ON MALE SUBFERTILITY AND !24 IN
$R VAN 7EERT *ANNE -EIJE VAN 7EERT *ANNEKE VAN DEN "ROEK *AN 7ILLEM VAN DER 3TEEG &ULCO VAN DER 6EEN 0AUL ! &LIERMAN "EN 7* -OL 0IETERNEL 3TEURES #ENTRE FOR 2EPRODUCTIVE -EDICINE !CADEMIC -EDICAL #ENTRE #ENTRE OF 2EPRODUCTIVE -EDICINE ROOM ( $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY -EIBERGDREEF !: !MSTERDAM 4HE .ETHERLANDS $EPARTMENT OF 0UBLIC (EALTH %RASMUS -EDICAL #ENTRE 2OTTERDAM 4HE .ETHERLANDS $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY /NZE ,IEVE 6ROUWE 'ASTHUIS !MSTERDAM 4HE .ETHERLANDS $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY -ÉXIMA -EDICAL #ENTRE 6ELDHOVEN 4HE .ETHERLANDS #ORRESPONDENCE 4EL &AX E MAIL JMVANWEERT AMCUVANL
!BSTRACT Patients’ preferences for intrauterine insemination (IUI) relative to IVF were assessed using trade-off interviews, and the number of IUI cycles they would undergo before changing to IVF. A total of 73 couples undergoing IUI with a total of 111 interviews were included. Scenarios were offered where pregnancy chance after IUI was varied against a fixed pregnancy rate after IVF. The impact of multiple pregnancy risk on the couple’s preference was also investigated. Interviews were held before starting IUI, after three or four IUI cycles and after six IUI cycles. With decreasing probability of ongoing pregnancy after IUI, an increasing number of couples switched their preference from IUI to IVF. This switch occurred after six cycles at a significantly higher (P = 0.01) mean cumulative pregnancy rate (53%) compared with other groups (31%). With increasing risk of multiple pregnancy, preference for IUI declined only slightly, with mean risks of 73, 78 and 83% of a multiple pregnancy for the three groups respectively. In conclusion, at baseline and after three cycles of IUI the majority of couples undergoing IUI preferred continuation of IUI over IVF. A clear shift in preference towards IVF occurred after six cycles. Risk of multiple pregnancy did not affect preference for IUI with ovarian stimulation. Keywords: intrauterine insemination, IVF, preference, subfertility
)NTRODUCTION
Intrauterine insemination (IUI) with or without ovarian stimulation is frequently used in the treatment of couples suffering from subfertility. Indications for IUI are mild male subfertility, cervical hostility or unexplained subfertility (Crosignani et al., 1994; Campana et al., 1996; Guzick et al., 1998, Khalil et al., 2001). Dys- or anovulation is a debatable indication for IUI as, in common practice, IUI is only initiated in those cases where the anovulation has been treated but a pregnancy has not occurred within 1e year or where other factors are present, e.g. mild male subfertility or cervical hostility. The overall pregnancy rates for IUI with ovarian stimulation vary between 9 and 36% per cycle (Cohlen et al., 2004; Steures et al., 2004).
When IUI fails, the next step is usually IVF. The pregnancy rates after IVF vary from 12 to 33% per cycle (Goverde et al., 2000; Pandian et al., 2002; Sharma et al., 2002). At present, it is unclear how many cycles of IUI should be given before proceeding to IVF. Some authors advocate that after four to six cycles of IUI the pregnancy rates decline (Campana et al., 1996; Khalil et al., 2001; Dickey et al., 2002; Duran et al., 2002; Steures et al. 2004), while others have suggested that there are still acceptable pregnancy rates achievable until the 10th cycle (Berg et al., 1997; Hendin et al., 2000). In the decision when to proceed to IVF after IUI there are some important medical and non-medical issues to consider. The
© 2007 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
priority of the clinician is obviously to select the appropriate patients for IUI to achieve acceptable pregnancy rates. This selection should ideally be based on the subfertility diagnosis followed by the use of prediction models (van Weert et al. 2005; Custers et al. 2007). Although both IUI with ovarian stimulation and IVF are associated with the risk of multiple pregnancies (Keith et al., 1999), IVF bears the risk of ovarian hyperstimulation syndrome (OHSS), which is virtually absent in IUI. The non-medical issues concern the higher costs of IVF and the more intensive treatment strategy with a higher drop-out rate relative to IUI (Land et al., 1997; Goverde et al., 2000). Nowadays, it is generally acknowledged that patients’ preferences should also be incorporated into medical decisionmaking (Kassirer, 1994). Several studies have shown that patients’ perspectives on the burden and benefits of therapy can differ from those of the health professionals (Devereaux et al., 2001; Steures et al., 2005). Therefore, the decision when to start and how long to continue IUI or proceeding to IVF should not only be based on the expected success rates of these treatments, but should also be based on the preferences of couples for either IUI or IVF. Patients’ preferences in reproductive medicine can be studied by a trade-off interview (Nieuwkerk et al., 1998; Bayram et al., 2005; Steures et al., 2005). In a previous study on patients’ preferences for expectant management versus IUI in unexplained subfertility, it was found that couples prefer IUI with or without ovarian stimulation when the probability of a treatment-independent pregnancy in the next 12 months is <50 and <40%, respectively. The risk of a multiple pregnancy did not affect their preference for IUI (Steures et al., 2005). The aim of the present study was to assess in patients undergoing IUI their preferences for IUI versus IVF and their evaluation of the risk of a multiple pregnancy, at different times during their IUI treatment and using the same trade-off design.
-ATERIALS AND METHODS The study was performed between July 2003 and September 2004 at the Academic Medical Centre (AMC) and the Onze Lieve Vrouwe Gasthuis (OLVG) in Amsterdam, The Netherlands. The study used the same methodology as that used in a previous study on patients’ preferences for either expectant management or IUI (Steures et al., 2005). Couples with unexplained subfertility, mild male subfertility or cervical hostility, who were starting or undergoing IUI, were invited to participate in this study, as these are the indications for IUI in the clinic used for the study. In case of additional anovulation, ovulation induction was performed prior to IUI. All women received mild ovarian hyperstimulation during IUI treatment with 75 IU FSH. A language barrier was a reason for exclusion, but as all couples entering a treatment programme in our clinic should master either Dutch or English this did not lead to any exclusions. The selection of couples depended upon the presence of the interviewer on non-fixed scheduled days. The interviewer was a recently graduated doctor with experience in the field of subfertility. Prior to the interview, baseline characteristics of these couples were abstracted from their medical files. These RBMOnline®
included maternal age, fertility treatment in the past, cycle history and presence and outcome of earlier pregnancies. The interviews took place at three different stages, i.e. when couples started IUI (baseline group), after the third or fourth cycle (midgroup), and after the sixth IUI cycle (end group). Couples were followed during their IUI cycles and, if possible, they were asked for a second and/or third interview. A structured interview was designed to assess patients’ preference for IUI relative to IVF conditional on the probability of a pregnancy after IUI. The interviews were conducted by the same person in a standardized manner with the aid of a standardized information sheet on IVF and IUI. Most interviews took place with both partners present. The aim of the interviewer was to elicit agreement upon the responses of both partners. Firstly, in all interviews detailed information was given about IVF. Ovarian hyperstimulation during IVF, oocyte retrieval and embryo transfer were explained. The pregnancy rate per IVF cycle in the interview was set at 20% per cycle. It was explained that one in two couples would be pregnant after three cycles. The possibility of single embryo transfer was not discussed. Subsequently, the interviewer provided structural information about the adverse effects of IVF. The risks and consequences of OHSS were explained with its diverse expressions such as nausea, pain, ascites, hospitalization, thrombo-embolism and, as an ultimate consequence, death. The possibility of severe OHSS in this interview was set at 2% per cycle (Papanikolaou et al., 2005). Furthermore, possible cancellation of an IVF cycle was discussed, e.g. in cases of threatening OHSS or in case of poor follicle development. Additionally, information on the risk of a multiple pregnancy and its possible implications, such as premature birth and its implications for future development of the child, hospitalisation (incubator), a lower birth weight and maternal complications like pre-eclampsia, were given. The possibility of a multiple pregnancy after IVF in this interview was set at 25% per ongoing pregnancy (Goverde et al., 2000; Andersen et al., 2005). Secondly, the interviewer provided information about IUI with ovarian stimulation, follicular growth monitoring, ovulation induction and insemination procedure. As the chance on an ongoing pregnancy after IUI was the variable in the trade-off interviews, no set pregnancy rate was mentioned. Furthermore, structural information was given on the adverse effects of IUI with ovarian stimulation. The possible cancellation of IUI was discussed, and the virtual absence of OHSS in our clinic was mentioned. Additionally, it was stated that, similar to IVF, the risk of a multiple pregnancy and its possible consequences existed. The chance of a multiple pregnancy after IUI with ovarian stimulation in this interview was set at 30% per ongoing pregnancy (Guzick et al., 1999; Gleicher et al., 2000; Goverde et al., 2000). The information given by the interviewer was similar to the counselling before the start of IUI in the clinic. In The Netherlands IVF is not an alternative for these couples but a second line therapy, so an extensive explanation of IVF was not part of the counselling process before the start of IUI. The interviewer asked the couple whether IUI with ovarian stimulation or IVF was preferred in a scenario with a 50% probability of a pregnancy after 1 year of IUI with ovarian stimulation. Depending on the preference of the couple, the probability of a pregnancy after IUI was subsequently decreased or increased with 5% steps until they switched their initial preference.
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
To determine the impact of the risk of a multiple pregnancy on their preference, couples were asked if they would continue IUI with ovarian stimulation in a scenario with 0% chance of having a multiple pregnancy after IUI with ovarian stimulation or would prefer IVF. If a couple chose IUI with ovarian stimulation, the probability of a multiple pregnancy was systematically increased with steps of 5% until the couple preferred not to continue IUI with ovarian stimulation and switched to IVF.
$ATA ANALYSIS Since one couple could have been interviewed more than once, each interview was analysed as a separate unit. The mean probability of a pregnancy after IUI with ovarian stimulation at which couples switched in their decision from IUI to IVF was calculated. These calculations were performed in three subgroup analyses i.e. the baseline group, the mid-group and the end group. Furthermore, differences in the mean thresholds between the groups were analysed using a one-way analysis of variance (ANOVA). To visualize the thresholds of the patients, the percentage of couples that preferred IVF for each respective probability of a pregnancy after IUI with ovarian stimulation was plotted. To evaluate the impact of multiple pregnancies, the mean thresholds at which couples would switch from IUI with ovarian stimulation to IVF was calculated and the mean thresholds compared between the groups in a one-way ANOVA. It was anticipated that the couples’ preferences might be influenced by their baseline characteristics. Therefore, the association between baseline characteristics on one hand and the threshold-preference for each couple on the other hand was assessed. The continuous baseline characteristics (maternal age and duration of subfertility) were assessed using Spearman correlation coefficients, whereas preference thresholds of categorical baseline characteristics, i.e. subfertility being primary or secondary, fertility treatments, having a child or not and having a regular or irregular cycle, were compared using Student’s t-tests.
2ESULTS In all, 73 couples were interviewed, of which 41 were interviewed once, 26 were interviewed twice, and six couples were interviewed three times. In total 111 interviews were taken, 33 interviews in the baseline group, 45 interviews in the mid-group (after three or four cycles), and 33 interviews in the end group (after six cycles).
0REFERENCES RELATED TO PROBABILITY OF PREGNANCY
The baseline characteristics are listed in Table 1. Overall, the mean maternal age was 34.4 years, the mean duration of subfertility was 3.7 years and anovulation was present in seven (10%) women. In total 31 couples (42%) suffered from secondary subfertility. From these couples 15 (21%) had at least one live born child. A total of 15 (21%) couples had a history of IUI in a previous episode of subfertility. The baseline characteristics of the three groups showed no statistical differences.
With decreasing probability of an ongoing pregnancy after IUI with ovarian stimulation, an increasing number of couples switched their preference from IUI to IVF (Figure 1). In the end group, couples switched their preference to IVF at higher pregnancy rates of IUI with ovarian stimulation as compared with the baseline group and mid-group. In the baseline-group 97% of couples opted for IUI with ovarian stimulation if the pregnancy rate with IUI in the next 12 months was set at 100%. In the mid-group this percentage was 98% and end group all but two couples (94%) would opt for IUI with ovarian stimulation with this scenario. In the baseline- and mid-groups three couples refused treatment with IVF (9 and 7%, respectively), even if they had no chance to conceive with IUI; in the end group this was only one couple (3%). The mean threshold for a pregnancy in the next 12 months with IUI at which couples switched their preference from IUI to IVF was 31% at baseline, 31% in the mid-group and 53% in the end group. ANOVA showed a significant difference in the means of the thresholds between all groups with an overall difference of 11% (P = 0.01). When analysing the groups separately, there was no statistical difference between the baseline group and the midgroup (difference 0.6%), whereas a significant difference was found between the mean of the thresholds in the baseline group and the end group (difference 21%, P = 0.002) and the midgroup and the mean of the threshold in end group (difference 21%, P < 0.01).
0REFERENCES RELATED TO PROBABILITY OF MULTIPLE PREGNANCY With an increasing probability of a multiple pregnancy after IUI with ovarian stimulation an increasing number of couples would switch their preference from IUI to IVF in the baseline and mid-groups (Figure 2). In the end-group, few couples were influenced in their preference for either IUI or IVF based on the probability of a multiple pregnancy. The majority of couples that would continue IUI with ovarian stimulation, even at a probability of a multiple pregnancy of 50%, would still do so even at a 100% probability of a multiple pregnancy. If the risk of a multiple pregnancy after IUI was set at 0%, 82% (27 couples) of the baseline group, 69% (31 couples) of the midgroup, and 42% (14 couples) of the end group would continue IUI. In the baseline group, 16 (59%) couples that had initially preferred IUI would continue IUI even at a 100% certainty of achieving a multiple pregnancy. In the mid-group and in the end group this was the case for 19 couples (61%) and 10 couples (71%), respectively. The mean threshold at which the couples in the baseline group with an initial preference for IUI switched their preference to IVF was at a risk of 73% on a multiple pregnancy, 78% in the mid-group, and 83% in the end group. ANOVA showed a significant difference in the means of the thresholds between all groups with an overall difference of 12% (P = 0.02). When analysing the groups separately, there was no statistical difference between the baseline group and the midgroup (difference 5%), whereas a significant difference was found between the mean of the thresholds in the baseline group and the end group (difference 12%, P = 0.03) and a borderline significant difference between the mid-group and the mean of the threshold in end group (difference 20%, P = 0.05).
RBMOnline®
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
Table 1. Baseline characteristics of patients questioned as to their preference for intrauterine insemination (IUI) or IVF.
Maternal age (years)a Duration of subfertility (years)a Anovulationb Secondary subfertilityb Live bornb,c History of IUIb
All couples (n = 73)
Baseline group (n = 33)
Mid-group (n = 45)
End group (n = 33)
34.4 (26.4–40.6) 3.7 (0.4–9.9) 7 (10) 31 (42) 15 (21) 15 (21)
33.8 (26.4–39.8) 3.0 (0.4–7.1) 4 (12) 16 (48) 8(24) 6 (18)
34.1 (27.0–40.5) 4.2 (1.0–9.9) 4 (9) 18 (40) 8 (18) 10 (22)
34.5 (27.5–40.6) 4.35 (1.11–9.5) 1 (3) 12 (36) 6 (18) 7 (21)
Baseline group = before the start of IUI; mid-group = after three or four cycles; end group = after six cycles. There were no statistically significant differences between the groups. a Ranges are given in parentheses. bPercentages are given in parentheses. cNumber of live-born children born to couples with secondary subfertility.
Figure 1. Patients’ preferences for intrauterine insemination (IUI) with ovarian stimulation as expressed as a function of the probability of a pregnancy after IUI with ovarian stimulation. COH = controlled ovarian stimulation.
Figure 2. Patients’ preferences for IUI with ovarian stimulation as a function of the probability of a multiple pregnancy.
RBMOnline®
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
No significant correlation was found between maternal age and duration of the subfertility and the thresholds of the couples. Also, there were no statistically significant differences in the mean of the thresholds of couples when divided in groups for primary or secondary subfertility, presence of anovulation, or history of IUI or not.
$ISCUSSION The provision of information before treatment is an essential aspect of fertility care in terms of patient satisfaction, preparation and anxiety reduction. It can lead to a marked reduction in nonattendance at clinic appointments for IVF (Malone, 2003). This study shows that the evaluation of IUI was different at different stages of a treatment programme. At the start and after three or four cycles of IUI, couples would on average expect a cumulative pregnancy rate of 31% in a year to continue with IUI. After six cycles of IUI couples would on average expect a statistically significantly higher cumulative pregnancy rate of 53% in a year to continue this treatment. Since a cumulative pregnancy rate of 53% after six unsuccessful IUI cycles seems unrealistic, it was concluded that very few couples would be inclined to continue IUI after six cycles in daily practice. Another notable finding was that more than 70% of couples that initially preferred IUI over IVF would continue this treatment at a 100% certainty of getting a multiple pregnancy, with 83% of couples after six cycles of IUI. The lack of a negative attitude towards multiple pregnancies in subfertile couples has been described before (Ryan et al., 2004; Steures et al., 2005). One could argue that these couples lack knowledge about the risks of a multiple pregnancy and are therefore not capable of making a reasoned decision for one or the other treatment (Ryan et al., 2004). Most couples with subfertility are however well informed about these risks and, even if these risks are systematically emphasized, were more concerned about perceived treatment failure than about iatrogenic complications, such as a multiple pregnancy (Murray et al., 2004). In the present study, structured interviews were performed in which the potential disadvantages of multiple pregnancies were emphasized systematically. In spite of these efforts, patients preferred treatment even at a considerable risk of a multiple pregnancy. The present study clearly demonstrates that structured information on the potential hazards of multiple pregnancies has little impact on patients’ preferences. Considering the fact that the association between pregnancy chance and risk of multiple pregnancy was not discuss with couples, the impact of the potential hazards of a multiple pregnancy might be even lower on patient preferences. This study, and a previous study by the same authors, on patients preferences for expectant management versus IUI unambiguously show that the evaluation of a multiple pregnancy by subfertile couples is independent of whether they have just finished the basic fertility work-up or undergoing fertility treatment (Steures et al., 2005).
There are some methodological issues to consider. Preference studies are prone to interpretation bias of patient and interviewer. First of all, preference measurements are subject to
instability over time because patients acquire more information on the treatment side effects and effectiveness during their course of treatment. Patients were interviewed at different moments in their treatment course to give more insight into this instability. Not all couples were interviewed at baseline, but as the interviews (at baseline, after three cycles and after six cycles) rather than the couples were used as separate units in the analysis, this has not led to selection bias. Secondly, there are instrument-related biases such as the presentation and interpretation of the mentioned risks. For instance, in a face-to-face interview there are more tools to inform the patient in a manner that fits her intelligence and background knowledge but at the same time such an intervention is prone to a subjective presentation of the facts by the interviewer. Possibly, this bias is present in this study because the instrument used was a face-to-face interview by a single interviewer. Thirdly, one could argue that the patients entered in this study had already been counselled to undergo IUI in the first place and would have been informed about the risks and success rates of IUI by the clinician entering them in IUI programme. Couples undergoing IUI could therefore only be couples that chose IUI over IVF anyway. This is however not the case in The Netherlands, where most patients can as a rule only undergo those treatments that have been advised by their clinician, using Dutch guidelines and indications. Fourthly, patient preferences are undoubtedly influenced by social and cultural differences. Specifically, the differences in reimbursement policy concerning artificial reproductive therapies between different countries make the results of this preference study not applicable to for instance a subfertility population in the United States. In The Netherlands most patients are adequately insured and will be reimbursed for most of their fertility treatments. This study portrays the preferences of patients without the influence of financial issues. Finally, confounding factors like multiparity and maternal age did not have a statistically significant impact on patient’s preferences but this could be due to the number of patients included. To reach definitive conclusions about the impact of confounding factors a larger study is needed. It was found that a few couples had a pre-set preference for either IUI or IVF (four couples for IUI and three couples for IVF). They showed little confidence in the other treatment option even when risks on side effects and complications were significantly increased or decreased. This was despite the fact that couples had received thorough information about these potential risks of treatments. The IUI success rate in The Netherlands is 9% per cycle and the contribution made by IUI to the number of multiple pregnancies in The Netherlands was much smaller than the contribution made by IVF (Steures et al., 2007). These data were not used in the counselling of couples as these data had not been published at the time of the study. If the results of this national survey are combined with the results of our preference study, it could be argued that if patient preferences were taken into account after counselling with these data, almost no one would prefer IUI over IVF (lower success rate, lower multiple pregnancy rate)! RBMOnline®
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
In summary, at the start of IUI with ovarian stimulation and after three or four cycles the majority of the patients wanted to continue this treatment. After six cycles most couples demanded such high cumulative pregnancy rates with IUI that continuation of IUI became unrealistic and IVF would be preferred. The risk of a multiple pregnancy hardly affected the preference for IUI with ovarian stimulation in relation to IVF. These data should be incorporated into clinical guidelines on IUI and IVF.
2EFERENCES Andersen AN, Gianaroli L, Felberbaum R et al. 2005 The European IVF-monitoring programme (EIM), European Society of Human Reproduction and Embryology (ESHRE). Assisted reproductive technology in Europe, 2001. Results generated from European registers by ESHRE. Human Reproduction 20, 1158–1176. Bayram N, van Wely M, van der Veen F et al. 2005 Treatment preferences and trade-offs for ovulation induction in clomiphene citrate-resistant patients with polycystic ovary syndrome. Fertility and Sterility 84, 420–425. Berg U, Brucker C, Berg FD1997 Effect of motile sperm count after swim-up on outcome of intrauterine insemination. Fertility and Sterility 67, 747–750. Campana A, Sakkas D, Stalberg A et al. 1996 Intrauterine insemination: evaluation of the results according to the woman’s age, sperm quality, total sperm count per insemination and life table analysis. Human Reproduction 11, 732–736. Cohlen BJ 2004 Should we continue performing intrauterine inseminations in the year 2004? Gynecologic and Obstetric Investigation 59, 3–13. Crosignani PG, Walters DE 1994 Clinical pregnancy and male subfertility; the ESHRE multicentre trial on the treatment of male subfertility. European Society of Human Reproduction and Embryology. Human Reproduction 9, 1112–1118. Custers IM, Steures P, van der Steeg JW et al. 2007 External validation of a prediction molde for ongoing pregnancy after IUI. Fertility and Sterility 88, 425–431. Devereaux PJ, Anderson DR, Gardner MJ et al. 2001 Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. British Medical Journal 323, 1218–1222. Dickey RP, Taylor SN, Lu PY et al. 2002 Effect of diagnosis, age, sperm quality, and number of preovulatory follicles on the outcome of multiple cycles of clomiphene citrate-intrauterine insemination. Fertility and Sterility 78, 1088–1095. Duran HE, Morshedi M, Kruger T, Oehninger S 2002 Intrauterine insemination: a systematic review on determinants of success. Human Reproduction Update 8, 373–384. Gleicher N, Oleske DM, Tur-Kaspa I et al. 2000 Reducing the risk of high-order multiple pregnancy after ovarian stimulation with gonadotropins. New England Journal of Medicine 343, 2–7. Goverde AJ, McDonnell J, Vermeiden JP et al. 2000 Intrauterine insemination or in-vitro fertilisation in idiopathic subfertility and male subfertility: a randomised trial and cost–effectiveness analysis. Lancet 355, 13–18. Guzick DS, Carson SA, Coutifaris C et al. 1999 Efficacy of superovulation and intrauterine insemination in the treatment of infertility. National Co-operative Reproductive Medicine Network. New England Journal of Medicine 340, 177–183. Guzick DS, Sullivan MW, Adamson GD et al. 1998 Efficacy of treatment for unexplained infertility. Fertility and Sterility 70, 207–213. Hendin BN, Falcone T, Hallak J et al. 2000 The effect of patient and semen characteristics on live birth rates following intrauterine insemination: a retrospective study. Journal of Assisted Reproduction and Genetics 17, 245–252. Kassirer JP1994 Incorporating patients’ preferences into medical decisions. New England Journal of Medicine 330, 1895–1896. Keith L, Oleszczuk JJ 1999 Iatrogenic multiple birth, multiple
RBMOnline®
pregnancy and assisted reproductive technologies. International Journal of Gynaecology and Obstetrics 64, 11–25. Khalil MR, Rasmussen PE, Erb K, Laursen SB et al. 2001 Homologous intrauterine insemination. An evaluation of prognostic factors based on a review of 2473 cycles. Acta Obstetricia et Gynecologica Scandinavica 80, 74–81. Land JA, Courtar DA, Evers JL 1997 Patient dropout in an assisted reproductive technology program: implications for pregnancy rates. Fertility and Sterility 68, 278–281. Malone C 2003 Provision of information to patients in an NHS IVF unit. Human Fertility (Camb) 6, 26–29. Murray S, Shetty A, Rattray A et al. 2004 A randomized comparison of alternative methods of information provision on the acceptability of elective single embryo transfer. Human Reproduction 19, 911–916. Nieuwkerk PT, Hajenius PJ, Van der Veen F et al. 1998 Systemic methotrexate therapy versus laparoscopic salpingostomy in tubal pregnancy. Part II. Patient preferences for systemic methotrexate. Fertility and Sterility 70, 518–522. Pandian Z, Bhattacharya S, Nikolaou D et al. 2002 In vitro fertilisation for unexplained subfertility. In: Cochrane Database System Review (2), CD003357. Papanikolaou EG, Tournaye H, Verpoest W et al. 2005 Early and late ovarian hyperstimulation syndrome: early pregnancy outcome and profile. Human Reproduction 20, 636–641. Ryan GL, Zhang SH, Dokras A et al. 2004 The desire of infertile patients for multiple births. Fertility and Sterility 81, 500–504. Sharma V, Allgar V, Rajkhowa M 2002 Factors influencing the cumulative conception rate and discontinuation of in vitro fertilization treatment for infertility. Fertility and Sterility 78, 40–46. Steures P, van der Steeg JW, Hompes PGA et al. 2007 Intrauterine insemination in The Netherlands. Reproductive BioMedicine Online 14, 110–116. Steures P, Berkhout JC, Hompes PG et al. 2005 Patients’ preferences in deciding between intrauterine insemination and expectant management. Human Reproduction 20, 752–755. Steures P, van der Steeg JW, Mol BW et al. 2004 Prediction of an ongoing pregnancy after intrauterine insemination. Fertility and Sterility 82, 45–51. van Weert JM, Repping S, van der Steeg JW et al. 2005 IUI in male subfertility: are we able to select the proper patients? Reproductive BioMedicine Online 11, 624–631. Received 22 March 2007; refereed 13 April 2007; accepted 22 June 2007.
!RTICLE 0ATIENTS PREFERENCES FOR INTRAUTERINE INSEMINATION OR IN VITRO FERTILIZATION $R VAN 7EERT GRADUATED FROM MEDICAL SCHOOL IN AT THE 5NIVERSITY OF !MSTERDAM AND STARTED HER GYNAECOLOGICAL RESIDENCY IN )N SHE COMPLETED HER TRAINING AND SHE IS NOW A STAFF MEMBER OF THE GYNAECOLOGICAL TEAM AT THE -EDICAL #ENTRE !LKMAAR IN 4HE .ETHERLANDS $URING HER RESIDENCY SHE PARTICIPATED IN THE RESEARCH OF THE #ENTRE FOR 2EPRODUCTIVE -EDICINE AT THE !CADEMICAL -EDICAL #ENTRE IN !MSTERDAM 3HE HOPES TO lNISH HER THESIS ON MALE SUBFERTILITY AND !24 IN
$R VAN 7EERT *ANNE -EIJE VAN 7EERT *ANNEKE VAN DEN "ROEK *AN 7ILLEM VAN DER 3TEEG &ULCO VAN DER 6EEN 0AUL ! &LIERMAN "EN 7* -OL 0IETERNEL 3TEURES #ENTRE FOR 2EPRODUCTIVE -EDICINE !CADEMIC -EDICAL #ENTRE #ENTRE OF 2EPRODUCTIVE -EDICINE ROOM ( $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY -EIBERGDREEF !: !MSTERDAM 4HE .ETHERLANDS $EPARTMENT OF 0UBLIC (EALTH %RASMUS -EDICAL #ENTRE 2OTTERDAM 4HE .ETHERLANDS $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY /NZE ,IEVE 6ROUWE 'ASTHUIS !MSTERDAM 4HE .ETHERLANDS $EPARTMENT OF /BSTETRICS AND 'YNAECOLOGY -ÉXIMA -EDICAL #ENTRE 6ELDHOVEN 4HE .ETHERLANDS #ORRESPONDENCE 4EL &AX E MAIL JMVANWEERT AMCUVANL
!BSTRACT Patients’ preferences for intrauterine insemination (IUI) relative to IVF were assessed using trade-off interviews, and the number of IUI cycles they would undergo before changing to IVF. A total of 73 couples undergoing IUI with a total of 111 interviews were included. Scenarios were offered where pregnancy chance after IUI was varied against a fixed pregnancy rate after IVF. The impact of multiple pregnancy risk on the couple’s preference was also investigated. Interviews were held before starting IUI, after three or four IUI cycles and after six IUI cycles. With decreasing probability of ongoing pregnancy after IUI, an increasing number of couples switched their preference from IUI to IVF. This switch occurred after six cycles at a significantly higher (P = 0.01) mean cumulative pregnancy rate (53%) compared with other groups (31%). With increasing risk of multiple pregnancy, preference for IUI declined only slightly, with mean risks of 73, 78 and 83% of a multiple pregnancy for the three groups respectively. In conclusion, at baseline and after three cycles of IUI the majority of couples undergoing IUI preferred continuation of IUI over IVF. A clear shift in preference towards IVF occurred after six cycles. Risk of multiple pregnancy did not affect preference for IUI with ovarian stimulation. Keywords: intrauterine insemination, IVF, preference, subfertility
)NTRODUCTION
Intrauterine insemination (IUI) with or without ovarian stimulation is frequently used in the treatment of couples suffering from subfertility. Indications for IUI are mild male subfertility, cervical hostility or unexplained subfertility (Crosignani et al., 1994; Campana et al., 1996; Guzick et al., 1998, Khalil et al., 2001). Dys- or anovulation is a debatable indication for IUI as, in common practice, IUI is only initiated in those cases where the anovulation has been treated but a pregnancy has not occurred within 1e year or where other factors are present, e.g. mild male subfertility or cervical hostility. The overall pregnancy rates for IUI with ovarian stimulation vary between 9 and 36% per cycle (Cohlen et al., 2004; Steures et al., 2004).
When IUI fails, the next step is usually IVF. The pregnancy rates after IVF vary from 12 to 33% per cycle (Goverde et al., 2000; Pandian et al., 2002; Sharma et al., 2002). At present, it is unclear how many cycles of IUI should be given before proceeding to IVF. Some authors advocate that after four to six cycles of IUI the pregnancy rates decline (Campana et al., 1996; Khalil et al., 2001; Dickey et al., 2002; Duran et al., 2002; Steures et al. 2004), while others have suggested that there are still acceptable pregnancy rates achievable until the 10th cycle (Berg et al., 1997; Hendin et al., 2000). In the decision when to proceed to IVF after IUI there are some important medical and non-medical issues to consider. The
© 2007 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
priority of the clinician is obviously to select the appropriate patients for IUI to achieve acceptable pregnancy rates. This selection should ideally be based on the subfertility diagnosis followed by the use of prediction models (van Weert et al. 2005; Custers et al. 2007). Although both IUI with ovarian stimulation and IVF are associated with the risk of multiple pregnancies (Keith et al., 1999), IVF bears the risk of ovarian hyperstimulation syndrome (OHSS), which is virtually absent in IUI. The non-medical issues concern the higher costs of IVF and the more intensive treatment strategy with a higher drop-out rate relative to IUI (Land et al., 1997; Goverde et al., 2000). Nowadays, it is generally acknowledged that patients’ preferences should also be incorporated into medical decisionmaking (Kassirer, 1994). Several studies have shown that patients’ perspectives on the burden and benefits of therapy can differ from those of the health professionals (Devereaux et al., 2001; Steures et al., 2005). Therefore, the decision when to start and how long to continue IUI or proceeding to IVF should not only be based on the expected success rates of these treatments, but should also be based on the preferences of couples for either IUI or IVF. Patients’ preferences in reproductive medicine can be studied by a trade-off interview (Nieuwkerk et al., 1998; Bayram et al., 2005; Steures et al., 2005). In a previous study on patients’ preferences for expectant management versus IUI in unexplained subfertility, it was found that couples prefer IUI with or without ovarian stimulation when the probability of a treatment-independent pregnancy in the next 12 months is <50 and <40%, respectively. The risk of a multiple pregnancy did not affect their preference for IUI (Steures et al., 2005). The aim of the present study was to assess in patients undergoing IUI their preferences for IUI versus IVF and their evaluation of the risk of a multiple pregnancy, at different times during their IUI treatment and using the same trade-off design.
-ATERIALS AND METHODS The study was performed between July 2003 and September 2004 at the Academic Medical Centre (AMC) and the Onze Lieve Vrouwe Gasthuis (OLVG) in Amsterdam, The Netherlands. The study used the same methodology as that used in a previous study on patients’ preferences for either expectant management or IUI (Steures et al., 2005). Couples with unexplained subfertility, mild male subfertility or cervical hostility, who were starting or undergoing IUI, were invited to participate in this study, as these are the indications for IUI in the clinic used for the study. In case of additional anovulation, ovulation induction was performed prior to IUI. All women received mild ovarian hyperstimulation during IUI treatment with 75 IU FSH. A language barrier was a reason for exclusion, but as all couples entering a treatment programme in our clinic should master either Dutch or English this did not lead to any exclusions. The selection of couples depended upon the presence of the interviewer on non-fixed scheduled days. The interviewer was a recently graduated doctor with experience in the field of subfertility. Prior to the interview, baseline characteristics of these couples were abstracted from their medical files. These RBMOnline®
included maternal age, fertility treatment in the past, cycle history and presence and outcome of earlier pregnancies. The interviews took place at three different stages, i.e. when couples started IUI (baseline group), after the third or fourth cycle (midgroup), and after the sixth IUI cycle (end group). Couples were followed during their IUI cycles and, if possible, they were asked for a second and/or third interview. A structured interview was designed to assess patients’ preference for IUI relative to IVF conditional on the probability of a pregnancy after IUI. The interviews were conducted by the same person in a standardized manner with the aid of a standardized information sheet on IVF and IUI. Most interviews took place with both partners present. The aim of the interviewer was to elicit agreement upon the responses of both partners. Firstly, in all interviews detailed information was given about IVF. Ovarian hyperstimulation during IVF, oocyte retrieval and embryo transfer were explained. The pregnancy rate per IVF cycle in the interview was set at 20% per cycle. It was explained that one in two couples would be pregnant after three cycles. The possibility of single embryo transfer was not discussed. Subsequently, the interviewer provided structural information about the adverse effects of IVF. The risks and consequences of OHSS were explained with its diverse expressions such as nausea, pain, ascites, hospitalization, thrombo-embolism and, as an ultimate consequence, death. The possibility of severe OHSS in this interview was set at 2% per cycle (Papanikolaou et al., 2005). Furthermore, possible cancellation of an IVF cycle was discussed, e.g. in cases of threatening OHSS or in case of poor follicle development. Additionally, information on the risk of a multiple pregnancy and its possible implications, such as premature birth and its implications for future development of the child, hospitalisation (incubator), a lower birth weight and maternal complications like pre-eclampsia, were given. The possibility of a multiple pregnancy after IVF in this interview was set at 25% per ongoing pregnancy (Goverde et al., 2000; Andersen et al., 2005). Secondly, the interviewer provided information about IUI with ovarian stimulation, follicular growth monitoring, ovulation induction and insemination procedure. As the chance on an ongoing pregnancy after IUI was the variable in the trade-off interviews, no set pregnancy rate was mentioned. Furthermore, structural information was given on the adverse effects of IUI with ovarian stimulation. The possible cancellation of IUI was discussed, and the virtual absence of OHSS in our clinic was mentioned. Additionally, it was stated that, similar to IVF, the risk of a multiple pregnancy and its possible consequences existed. The chance of a multiple pregnancy after IUI with ovarian stimulation in this interview was set at 30% per ongoing pregnancy (Guzick et al., 1999; Gleicher et al., 2000; Goverde et al., 2000). The information given by the interviewer was similar to the counselling before the start of IUI in the clinic. In The Netherlands IVF is not an alternative for these couples but a second line therapy, so an extensive explanation of IVF was not part of the counselling process before the start of IUI. The interviewer asked the couple whether IUI with ovarian stimulation or IVF was preferred in a scenario with a 50% probability of a pregnancy after 1 year of IUI with ovarian stimulation. Depending on the preference of the couple, the probability of a pregnancy after IUI was subsequently decreased or increased with 5% steps until they switched their initial preference.
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
To determine the impact of the risk of a multiple pregnancy on their preference, couples were asked if they would continue IUI with ovarian stimulation in a scenario with 0% chance of having a multiple pregnancy after IUI with ovarian stimulation or would prefer IVF. If a couple chose IUI with ovarian stimulation, the probability of a multiple pregnancy was systematically increased with steps of 5% until the couple preferred not to continue IUI with ovarian stimulation and switched to IVF.
$ATA ANALYSIS Since one couple could have been interviewed more than once, each interview was analysed as a separate unit. The mean probability of a pregnancy after IUI with ovarian stimulation at which couples switched in their decision from IUI to IVF was calculated. These calculations were performed in three subgroup analyses i.e. the baseline group, the mid-group and the end group. Furthermore, differences in the mean thresholds between the groups were analysed using a one-way analysis of variance (ANOVA). To visualize the thresholds of the patients, the percentage of couples that preferred IVF for each respective probability of a pregnancy after IUI with ovarian stimulation was plotted. To evaluate the impact of multiple pregnancies, the mean thresholds at which couples would switch from IUI with ovarian stimulation to IVF was calculated and the mean thresholds compared between the groups in a one-way ANOVA. It was anticipated that the couples’ preferences might be influenced by their baseline characteristics. Therefore, the association between baseline characteristics on one hand and the threshold-preference for each couple on the other hand was assessed. The continuous baseline characteristics (maternal age and duration of subfertility) were assessed using Spearman correlation coefficients, whereas preference thresholds of categorical baseline characteristics, i.e. subfertility being primary or secondary, fertility treatments, having a child or not and having a regular or irregular cycle, were compared using Student’s t-tests.
2ESULTS In all, 73 couples were interviewed, of which 41 were interviewed once, 26 were interviewed twice, and six couples were interviewed three times. In total 111 interviews were taken, 33 interviews in the baseline group, 45 interviews in the mid-group (after three or four cycles), and 33 interviews in the end group (after six cycles).
0REFERENCES RELATED TO PROBABILITY OF PREGNANCY
The baseline characteristics are listed in Table 1. Overall, the mean maternal age was 34.4 years, the mean duration of subfertility was 3.7 years and anovulation was present in seven (10%) women. In total 31 couples (42%) suffered from secondary subfertility. From these couples 15 (21%) had at least one live born child. A total of 15 (21%) couples had a history of IUI in a previous episode of subfertility. The baseline characteristics of the three groups showed no statistical differences.
With decreasing probability of an ongoing pregnancy after IUI with ovarian stimulation, an increasing number of couples switched their preference from IUI to IVF (Figure 1). In the end group, couples switched their preference to IVF at higher pregnancy rates of IUI with ovarian stimulation as compared with the baseline group and mid-group. In the baseline-group 97% of couples opted for IUI with ovarian stimulation if the pregnancy rate with IUI in the next 12 months was set at 100%. In the mid-group this percentage was 98% and end group all but two couples (94%) would opt for IUI with ovarian stimulation with this scenario. In the baseline- and mid-groups three couples refused treatment with IVF (9 and 7%, respectively), even if they had no chance to conceive with IUI; in the end group this was only one couple (3%). The mean threshold for a pregnancy in the next 12 months with IUI at which couples switched their preference from IUI to IVF was 31% at baseline, 31% in the mid-group and 53% in the end group. ANOVA showed a significant difference in the means of the thresholds between all groups with an overall difference of 11% (P = 0.01). When analysing the groups separately, there was no statistical difference between the baseline group and the midgroup (difference 0.6%), whereas a significant difference was found between the mean of the thresholds in the baseline group and the end group (difference 21%, P = 0.002) and the midgroup and the mean of the threshold in end group (difference 21%, P < 0.01).
0REFERENCES RELATED TO PROBABILITY OF MULTIPLE PREGNANCY With an increasing probability of a multiple pregnancy after IUI with ovarian stimulation an increasing number of couples would switch their preference from IUI to IVF in the baseline and mid-groups (Figure 2). In the end-group, few couples were influenced in their preference for either IUI or IVF based on the probability of a multiple pregnancy. The majority of couples that would continue IUI with ovarian stimulation, even at a probability of a multiple pregnancy of 50%, would still do so even at a 100% probability of a multiple pregnancy. If the risk of a multiple pregnancy after IUI was set at 0%, 82% (27 couples) of the baseline group, 69% (31 couples) of the midgroup, and 42% (14 couples) of the end group would continue IUI. In the baseline group, 16 (59%) couples that had initially preferred IUI would continue IUI even at a 100% certainty of achieving a multiple pregnancy. In the mid-group and in the end group this was the case for 19 couples (61%) and 10 couples (71%), respectively. The mean threshold at which the couples in the baseline group with an initial preference for IUI switched their preference to IVF was at a risk of 73% on a multiple pregnancy, 78% in the mid-group, and 83% in the end group. ANOVA showed a significant difference in the means of the thresholds between all groups with an overall difference of 12% (P = 0.02). When analysing the groups separately, there was no statistical difference between the baseline group and the midgroup (difference 5%), whereas a significant difference was found between the mean of the thresholds in the baseline group and the end group (difference 12%, P = 0.03) and a borderline significant difference between the mid-group and the mean of the threshold in end group (difference 20%, P = 0.05).
RBMOnline®
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
Table 1. Baseline characteristics of patients questioned as to their preference for intrauterine insemination (IUI) or IVF.
Maternal age (years)a Duration of subfertility (years)a Anovulationb Secondary subfertilityb Live bornb,c History of IUIb
All couples (n = 73)
Baseline group (n = 33)
Mid-group (n = 45)
End group (n = 33)
34.4 (26.4–40.6) 3.7 (0.4–9.9) 7 (10) 31 (42) 15 (21) 15 (21)
33.8 (26.4–39.8) 3.0 (0.4–7.1) 4 (12) 16 (48) 8(24) 6 (18)
34.1 (27.0–40.5) 4.2 (1.0–9.9) 4 (9) 18 (40) 8 (18) 10 (22)
34.5 (27.5–40.6) 4.35 (1.11–9.5) 1 (3) 12 (36) 6 (18) 7 (21)
Baseline group = before the start of IUI; mid-group = after three or four cycles; end group = after six cycles. There were no statistically significant differences between the groups. a Ranges are given in parentheses. bPercentages are given in parentheses. cNumber of live-born children born to couples with secondary subfertility.
Figure 1. Patients’ preferences for intrauterine insemination (IUI) with ovarian stimulation as expressed as a function of the probability of a pregnancy after IUI with ovarian stimulation. COH = controlled ovarian stimulation.
Figure 2. Patients’ preferences for IUI with ovarian stimulation as a function of the probability of a multiple pregnancy.
RBMOnline®
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
No significant correlation was found between maternal age and duration of the subfertility and the thresholds of the couples. Also, there were no statistically significant differences in the mean of the thresholds of couples when divided in groups for primary or secondary subfertility, presence of anovulation, or history of IUI or not.
$ISCUSSION The provision of information before treatment is an essential aspect of fertility care in terms of patient satisfaction, preparation and anxiety reduction. It can lead to a marked reduction in nonattendance at clinic appointments for IVF (Malone, 2003). This study shows that the evaluation of IUI was different at different stages of a treatment programme. At the start and after three or four cycles of IUI, couples would on average expect a cumulative pregnancy rate of 31% in a year to continue with IUI. After six cycles of IUI couples would on average expect a statistically significantly higher cumulative pregnancy rate of 53% in a year to continue this treatment. Since a cumulative pregnancy rate of 53% after six unsuccessful IUI cycles seems unrealistic, it was concluded that very few couples would be inclined to continue IUI after six cycles in daily practice. Another notable finding was that more than 70% of couples that initially preferred IUI over IVF would continue this treatment at a 100% certainty of getting a multiple pregnancy, with 83% of couples after six cycles of IUI. The lack of a negative attitude towards multiple pregnancies in subfertile couples has been described before (Ryan et al., 2004; Steures et al., 2005). One could argue that these couples lack knowledge about the risks of a multiple pregnancy and are therefore not capable of making a reasoned decision for one or the other treatment (Ryan et al., 2004). Most couples with subfertility are however well informed about these risks and, even if these risks are systematically emphasized, were more concerned about perceived treatment failure than about iatrogenic complications, such as a multiple pregnancy (Murray et al., 2004). In the present study, structured interviews were performed in which the potential disadvantages of multiple pregnancies were emphasized systematically. In spite of these efforts, patients preferred treatment even at a considerable risk of a multiple pregnancy. The present study clearly demonstrates that structured information on the potential hazards of multiple pregnancies has little impact on patients’ preferences. Considering the fact that the association between pregnancy chance and risk of multiple pregnancy was not discuss with couples, the impact of the potential hazards of a multiple pregnancy might be even lower on patient preferences. This study, and a previous study by the same authors, on patients preferences for expectant management versus IUI unambiguously show that the evaluation of a multiple pregnancy by subfertile couples is independent of whether they have just finished the basic fertility work-up or undergoing fertility treatment (Steures et al., 2005).
There are some methodological issues to consider. Preference studies are prone to interpretation bias of patient and interviewer. First of all, preference measurements are subject to
instability over time because patients acquire more information on the treatment side effects and effectiveness during their course of treatment. Patients were interviewed at different moments in their treatment course to give more insight into this instability. Not all couples were interviewed at baseline, but as the interviews (at baseline, after three cycles and after six cycles) rather than the couples were used as separate units in the analysis, this has not led to selection bias. Secondly, there are instrument-related biases such as the presentation and interpretation of the mentioned risks. For instance, in a face-to-face interview there are more tools to inform the patient in a manner that fits her intelligence and background knowledge but at the same time such an intervention is prone to a subjective presentation of the facts by the interviewer. Possibly, this bias is present in this study because the instrument used was a face-to-face interview by a single interviewer. Thirdly, one could argue that the patients entered in this study had already been counselled to undergo IUI in the first place and would have been informed about the risks and success rates of IUI by the clinician entering them in IUI programme. Couples undergoing IUI could therefore only be couples that chose IUI over IVF anyway. This is however not the case in The Netherlands, where most patients can as a rule only undergo those treatments that have been advised by their clinician, using Dutch guidelines and indications. Fourthly, patient preferences are undoubtedly influenced by social and cultural differences. Specifically, the differences in reimbursement policy concerning artificial reproductive therapies between different countries make the results of this preference study not applicable to for instance a subfertility population in the United States. In The Netherlands most patients are adequately insured and will be reimbursed for most of their fertility treatments. This study portrays the preferences of patients without the influence of financial issues. Finally, confounding factors like multiparity and maternal age did not have a statistically significant impact on patient’s preferences but this could be due to the number of patients included. To reach definitive conclusions about the impact of confounding factors a larger study is needed. It was found that a few couples had a pre-set preference for either IUI or IVF (four couples for IUI and three couples for IVF). They showed little confidence in the other treatment option even when risks on side effects and complications were significantly increased or decreased. This was despite the fact that couples had received thorough information about these potential risks of treatments. The IUI success rate in The Netherlands is 9% per cycle and the contribution made by IUI to the number of multiple pregnancies in The Netherlands was much smaller than the contribution made by IVF (Steures et al., 2007). These data were not used in the counselling of couples as these data had not been published at the time of the study. If the results of this national survey are combined with the results of our preference study, it could be argued that if patient preferences were taken into account after counselling with these data, almost no one would prefer IUI over IVF (lower success rate, lower multiple pregnancy rate)! RBMOnline®
Article - Patients’ preferences for intrauterine insemination or IVF - J-M van Weert et al.
In summary, at the start of IUI with ovarian stimulation and after three or four cycles the majority of the patients wanted to continue this treatment. After six cycles most couples demanded such high cumulative pregnancy rates with IUI that continuation of IUI became unrealistic and IVF would be preferred. The risk of a multiple pregnancy hardly affected the preference for IUI with ovarian stimulation in relation to IVF. These data should be incorporated into clinical guidelines on IUI and IVF.
2EFERENCES Andersen AN, Gianaroli L, Felberbaum R et al. 2005 The European IVF-monitoring programme (EIM), European Society of Human Reproduction and Embryology (ESHRE). Assisted reproductive technology in Europe, 2001. Results generated from European registers by ESHRE. Human Reproduction 20, 1158–1176. Bayram N, van Wely M, van der Veen F et al. 2005 Treatment preferences and trade-offs for ovulation induction in clomiphene citrate-resistant patients with polycystic ovary syndrome. Fertility and Sterility 84, 420–425. Berg U, Brucker C, Berg FD1997 Effect of motile sperm count after swim-up on outcome of intrauterine insemination. Fertility and Sterility 67, 747–750. Campana A, Sakkas D, Stalberg A et al. 1996 Intrauterine insemination: evaluation of the results according to the woman’s age, sperm quality, total sperm count per insemination and life table analysis. Human Reproduction 11, 732–736. Cohlen BJ 2004 Should we continue performing intrauterine inseminations in the year 2004? Gynecologic and Obstetric Investigation 59, 3–13. Crosignani PG, Walters DE 1994 Clinical pregnancy and male subfertility; the ESHRE multicentre trial on the treatment of male subfertility. European Society of Human Reproduction and Embryology. Human Reproduction 9, 1112–1118. Custers IM, Steures P, van der Steeg JW et al. 2007 External validation of a prediction molde for ongoing pregnancy after IUI. Fertility and Sterility 88, 425–431. Devereaux PJ, Anderson DR, Gardner MJ et al. 2001 Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. British Medical Journal 323, 1218–1222. Dickey RP, Taylor SN, Lu PY et al. 2002 Effect of diagnosis, age, sperm quality, and number of preovulatory follicles on the outcome of multiple cycles of clomiphene citrate-intrauterine insemination. Fertility and Sterility 78, 1088–1095. Duran HE, Morshedi M, Kruger T, Oehninger S 2002 Intrauterine insemination: a systematic review on determinants of success. Human Reproduction Update 8, 373–384. Gleicher N, Oleske DM, Tur-Kaspa I et al. 2000 Reducing the risk of high-order multiple pregnancy after ovarian stimulation with gonadotropins. New England Journal of Medicine 343, 2–7. Goverde AJ, McDonnell J, Vermeiden JP et al. 2000 Intrauterine insemination or in-vitro fertilisation in idiopathic subfertility and male subfertility: a randomised trial and cost–effectiveness analysis. Lancet 355, 13–18. Guzick DS, Carson SA, Coutifaris C et al. 1999 Efficacy of superovulation and intrauterine insemination in the treatment of infertility. National Co-operative Reproductive Medicine Network. New England Journal of Medicine 340, 177–183. Guzick DS, Sullivan MW, Adamson GD et al. 1998 Efficacy of treatment for unexplained infertility. Fertility and Sterility 70, 207–213. Hendin BN, Falcone T, Hallak J et al. 2000 The effect of patient and semen characteristics on live birth rates following intrauterine insemination: a retrospective study. Journal of Assisted Reproduction and Genetics 17, 245–252. Kassirer JP1994 Incorporating patients’ preferences into medical decisions. New England Journal of Medicine 330, 1895–1896. Keith L, Oleszczuk JJ 1999 Iatrogenic multiple birth, multiple
RBMOnline®
pregnancy and assisted reproductive technologies. International Journal of Gynaecology and Obstetrics 64, 11–25. Khalil MR, Rasmussen PE, Erb K, Laursen SB et al. 2001 Homologous intrauterine insemination. An evaluation of prognostic factors based on a review of 2473 cycles. Acta Obstetricia et Gynecologica Scandinavica 80, 74–81. Land JA, Courtar DA, Evers JL 1997 Patient dropout in an assisted reproductive technology program: implications for pregnancy rates. Fertility and Sterility 68, 278–281. Malone C 2003 Provision of information to patients in an NHS IVF unit. Human Fertility (Camb) 6, 26–29. Murray S, Shetty A, Rattray A et al. 2004 A randomized comparison of alternative methods of information provision on the acceptability of elective single embryo transfer. Human Reproduction 19, 911–916. Nieuwkerk PT, Hajenius PJ, Van der Veen F et al. 1998 Systemic methotrexate therapy versus laparoscopic salpingostomy in tubal pregnancy. Part II. Patient preferences for systemic methotrexate. Fertility and Sterility 70, 518–522. Pandian Z, Bhattacharya S, Nikolaou D et al. 2002 In vitro fertilisation for unexplained subfertility. In: Cochrane Database System Review (2), CD003357. Papanikolaou EG, Tournaye H, Verpoest W et al. 2005 Early and late ovarian hyperstimulation syndrome: early pregnancy outcome and profile. Human Reproduction 20, 636–641. Ryan GL, Zhang SH, Dokras A et al. 2004 The desire of infertile patients for multiple births. Fertility and Sterility 81, 500–504. Sharma V, Allgar V, Rajkhowa M 2002 Factors influencing the cumulative conception rate and discontinuation of in vitro fertilization treatment for infertility. Fertility and Sterility 78, 40–46. Steures P, van der Steeg JW, Hompes PGA et al. 2007 Intrauterine insemination in The Netherlands. Reproductive BioMedicine Online 14, 110–116. Steures P, Berkhout JC, Hompes PG et al. 2005 Patients’ preferences in deciding between intrauterine insemination and expectant management. Human Reproduction 20, 752–755. Steures P, van der Steeg JW, Mol BW et al. 2004 Prediction of an ongoing pregnancy after intrauterine insemination. Fertility and Sterility 82, 45–51. van Weert JM, Repping S, van der Steeg JW et al. 2005 IUI in male subfertility: are we able to select the proper patients? Reproductive BioMedicine Online 11, 624–631. Received 22 March 2007; refereed 13 April 2007; accepted 22 June 2007.