- Split View
-
Views
-
Cite
Cite
Jim X. Wang, Michael J. Davies, Robert J. Norman, Polycystic ovarian syndrome and the risk of spontaneous abortion following assisted reproductive technology treatment, Human Reproduction, Volume 16, Issue 12, 1 December 2001, Pages 2606–2609, https://doi.org/10.1093/humrep/16.12.2606
- Share Icon Share
Abstract
BACKGROUND: A high proportion of infertile patients have polycystic ovarian syndrome (PCOS) with a reportedly greater risk of spontaneous abortion. Because of the close link between PCOS and obesity and the independent association of obesity with poor pregnancy outcomes, it is important to distinguish the possible confounding effect of body mass index (BMI) or other variables from that of PCOS. This study aims to determine the effect of PCOS status on the risk of spontaneous abortion with adjustment for body mass and several other confounding factors in a large cohort of pregnant infertile women. METHODS: The patients (n = 1018) were treated in a tertiary infertility centre. Their PCOS status was determined by standard criteria and their BMI had been taken less than 1 year before the pregnancy. Patients whose PCOS status or BMI measurements were not assessed were excluded. Student's t-test or χ2 test were used to test the difference between the PCOS and non-PCOS groups while a multivariate logistical regression model was used to assess the effect of PCOS, BMI and other confounding factors. RESULTS: Overall, the incidence of PCOS was 37% in this cohort. The overall incidence of spontaneous abortion in the study population was 21%. Univariate analysis showed that women with PCOS had a significantly greater risk of spontaneous abortion compared with non-PCOS women (25 versus 18%, P < 0.01). However, using multivariate logistic regression analysis this effect was reduced to a non-significant level [odds ratio (OR) = 1.10, 95% confidence interval (CI) 0.85–1.36] after adjusting for obesity and patients/treatment combination factor, and to nil after adjusting for all confounding factors considered in this study (OR = 0.98, 95% CI 0.75–1.28). CONCLUSION: The results of this study suggest that the higher risk of spontaneous abortion observed in women with PCOS is likely to be due to their high prevalence of obesity and the type of treatment they receive.
Introduction
A high proportion of infertile patients have polycystic ovarian syndrome (PCOS). In addition to many metabolic and clinical symptoms (Solomon, 1999), it has also been reported that women with PCOS sequelae may suffer a greater risk of spontaneous abortion (Ruutiainen and Seppala, 1991; Hamilton Fairley et al., 1992; Balen et al., 1993; Tarlatzis et al., 1995). Because of the close link between PCOS and obesity and the association of obesity with poor pregnancy outcomes (Norman and Clark, 1998), it is necessary to distinguish the possible confounding effect of body mass index (BMI) or other variables from the effect of PCOS. This study aims to determine the independent effect of PCOS status on the risk of spontaneous abortion after the adjustment for body mass and other possible confounding factors in a large cohort of women who became pregnant following assisted reproductive technology treatment.
Materials and methods
The patients included in this cohort study were treated in the Reproductive Medicine Unit, Department of Obstetrics and Gynaecology, University of Adelaide, during the period of 1987–1999. Their PCOS status was determined by criteria described by Norman et al. (Norman et al., 1995). Patients with PCOS were defined as those who showed an increased concentration of serum testosterone above 2.5 nmol/l or elevated androstenedione, together with a low concentration of sex hormone binding globulin (SHBG) (normal range 20–100 nmol/l), in addition to characteristic ovarian morphology on ultrasound (defined as the presence of eight or more peripheral cysts less than 10 mm in diameter with increased stroma in one or both ovaries). BMI was measured <1 year before the pregnancy. Patients whose PCOS status or BMI was not assessed were excluded. Altogether 1018 pregnancies were included. The major types of infertility aetiology were tubal blockage (34%), male factor (35%), unexplained infertility (16%), endometriosis (9%), and others (6%). BMI (kg/m2) was used as a measure of body mass. The treatment related factors included were three treatment modalities, IVF, gamete intraFallopian tube transfer (GIFT) and intracytoplasmic sperm injection (ICSI) and concentration of oestradiol at oocyte retrieval, set at low (<2 nmol/l), medium (2–7.9 nmol/l) and high (≥8 nmol/l). Treatment protocols have been described elsewhere (Kerin et al., 1984; Sathanandan et al., 1989; Norman et al., 1991). Briefly, all cycles included in the analysis utilized a gonadotrophin agonist (leuprolide acetate; Abbott Pharmaceuticals, Australia) either as a pituitary luteal phase desensitizing protocol or occasionally a flare protocol. Gonadotrophins were introduced on day 2 of the induced menstrual bleed in the luteal desensitizing protocol. The ovarian response was monitored in a similar fashion to ovulation induction but the criteria for the administration of HCG was determined by the presence of at least two follicles >17 mm diameter.
Pregnancy was defined as the presence of an embryonic sac(s) by an ultrasound scan carried out 4–6 weeks after embryo transfer. Spontaneous abortion was defined as a pregnancy failing to reach 20 full weeks of gestation, excluding losses due to ectopic pregnancy or induced abortion. No patient was lost to follow-up. Either Student's t-test or χ2 test was used to test the difference between the PCOS and non-PCOS groups. A multivariate logistical regression model was used to assess the effect of PCOS, BMI and other confounding factors. A statistical significance level of P < 0.05 was used.
Results
Overall, the prevalence of PCOS was 37% in this cohort. As shown in Table I, there were significant differences between women with and without PCOS, in age, BMI and concentration of oestradiol. Amongst women with PCOS there was also a significantly greater proportion who were overweight/obese or had had a previous spontaneous abortion. More women with PCOS were treated by conventional IVF, thus significantly fewer were treated by ICSI (about 5% by GIFT in each group).
The overall incidence of spontaneous abortion in the study population was 21% (211/1018). The unadjusted rate of spontaneous abortion in relation to various confounding factors and the adjusted odds ratios(OR) resulting from multivariate logistic regression analysis are shown in Table II. The results of univariate analysis (χ2 test) showed that women with PCOS had significantly greater risk of spontaneous abortion than those without PCOS (25 versus 18%, P < 0.01). In addition, older age, a history of previous spontaneous abortion as well as obesity (BMI >30 kg/m2) all significantly increased the risk of spontaneous abortion (P < 0.01). The concentration of oestradiol at oocyte retrieval was also positively associated with the risk of spontaneous abortion (P < 0.01). Women treated by ICSI, which was less frequently used in women with PCOS in our treatment population, had significantly lower risks (P < 0.05) than those treated by IVF or GIFT.
Although in univariate analysis it was found that PCOS status had a significant effect on the risk of spontaneous abortion, multivariate logistic regression analysis showed that the effect of PCOS status was reduced to a lower, only marginally significant, level by adjusting for either obesity [OR = 1.37, 95% confidence interval (CI) 1.00–1.89] or patients/treatment combination factor (OR = 1.27, 95% CI 0.92–1.75) respectively. The effect of PCOS was further reduced to a non-significant level (OR = 1.10, 95% CI 0.85–1.36) after adjusting for these two factors simultaneously, and to nil after adjusting for all confounding factors considered in this study (OR = 0.98, 95% CI 0.75–1.28).
Discussion
It is well known that women with PCOS have a high prevalence of obesity (Pasquali et al., 1997; Norman and Clark, 1998; Solomon, 1999) which can detrimentally affect the pregnancy outcome (Norman and Clark, 1998). The results of this study indicated that the high risk of spontaneous abortion in women with PCOS was related to their high prevalence of obesity and the treatment type they received, as often dictated by their aetiology. The findings of the current study confirm that the incidence of spontaneous abortion increases with increasing BMI; an observation previously reported in women both with PCOS (Hamilton Fairley et al., 1992; Franks and Hamilton Fairley, 1994) and without (Bohrer and Kemmann, 1987; Hamilton Fairley et al., 1992; Fedorcsak et al., 2000). Obesity, therefore, needs to be considered as a confounding factor, given its higher prevalence in PCOS than non-PCOS women. Many of the previous studies suffered a limitation in their sample size and failed to adjust for confounding factors in assessing the effect of PCOS on pregnancy outcomes. The results of the present study suggested that effect of obesity was independent of PCOS status.
Infertility treatment covers a variety of treatment modalities for patients with different aetiology and involves different types or doses of drugs for ovarian stimulation. As shown here, the treatment type and stimulation levels could also have an effect on the risk of spontaneous abortion and act as confounding factors for the effect of PCOS in this study. The reduced risk of spontaneous abortion for patients receiving ICSI treatment has been reported previously (Orvieto et al., 2000). In the present study, the unadjusted effect may be partly due to age difference. Another reason could be the extent of complexity of infertility aetiology since ICSI was commonly used for severe male factor infertility, rather than for female factor infertility. The concentration of oestradiol at oocyte retrieval was usually the highest oestradiol concentration observed during the ovarian stimulation process, so it reflected the maximum body response to gonadotrophin stimulation and the maximum exposure of the uterus to oestradiol. It has been reported that that high oestradiol concentrations were detrimental to uterine receptivity (Valbuena et al., 1999) while reduced oestradiol concentrations could increase uterine receptivity (Simon et al., 1998). Women with PCOS have responded to ovarian stimulation more vigorously than those without PCOS in this study as indicated by their significantly higher oestradiol concentration. Higher oestradiol concentrations were linked with an increased risk of spontaneous abortion in this study. Oestradiol concentration may also act as a modifying factor in women with PCOS. The risk of spontaneous abortion was increased in women with PCOS at higher concentrations of oestradiol. The relative risk of spontaneous abortion by the concentration of oestradiol was 0.7 at <2 nmol/l, 1.5 at 2–7.9 nmol/l and 1.2 at ≥8 nmol/l. This finding implicates ovarian stimulation as a possible cause for the higher risk of spontaneous abortion in assisted reproductive technology pregnancies. Further confirmation of this finding is necessary. The identification of treatment related risk factors may increase our understanding of the cause of the high risk of spontaneous abortion amongst assisted reproductive technology pregnancies and help to reduce it.
In conclusion, this study suggests that the effect of PCOS on the risk of spontaneous abortion in pregnant women following assisted reproductive technology treatment could be due to their high prevalence of obesity and the type of treatment they received which, in turn, can be related to their infertility aetiology.
. | PCOS . | Non-PCOS . | P . |
---|---|---|---|
aExcluding obese or very obese women. | |||
bAbout 5% in each group were treated by gamete intraFallopian tube transfer (GIFT) and the remaining were treated by intracytoplasmic sperm injection (ICSI). | |||
Age (years) | 31.4 ± 3.8 | 32.7 ± 4.3 | < 0.0001 |
BMI (kg/m2) | 26.3 ± 5.6 | 24.3 ± 4.4 | < 0.0001 |
Maximal oestradiol (nmol/l) | 8.9 ± 4.3 | 6.9 ± 3.7 | < 0.0001 |
Percentage overweighta | 26% | 21% | 0.05 |
Percentage obese/very obese | 24% | 12% | < 0.0001 |
Percentage with previous spontaneous abortion | 32% | 22% | < 0.01 |
Percentage treated by IVFb | 55% | 41% | < 0.01 |
n | 373 | 645 |
. | PCOS . | Non-PCOS . | P . |
---|---|---|---|
aExcluding obese or very obese women. | |||
bAbout 5% in each group were treated by gamete intraFallopian tube transfer (GIFT) and the remaining were treated by intracytoplasmic sperm injection (ICSI). | |||
Age (years) | 31.4 ± 3.8 | 32.7 ± 4.3 | < 0.0001 |
BMI (kg/m2) | 26.3 ± 5.6 | 24.3 ± 4.4 | < 0.0001 |
Maximal oestradiol (nmol/l) | 8.9 ± 4.3 | 6.9 ± 3.7 | < 0.0001 |
Percentage overweighta | 26% | 21% | 0.05 |
Percentage obese/very obese | 24% | 12% | < 0.0001 |
Percentage with previous spontaneous abortion | 32% | 22% | < 0.01 |
Percentage treated by IVFb | 55% | 41% | < 0.01 |
n | 373 | 645 |
. | PCOS . | Non-PCOS . | P . |
---|---|---|---|
aExcluding obese or very obese women. | |||
bAbout 5% in each group were treated by gamete intraFallopian tube transfer (GIFT) and the remaining were treated by intracytoplasmic sperm injection (ICSI). | |||
Age (years) | 31.4 ± 3.8 | 32.7 ± 4.3 | < 0.0001 |
BMI (kg/m2) | 26.3 ± 5.6 | 24.3 ± 4.4 | < 0.0001 |
Maximal oestradiol (nmol/l) | 8.9 ± 4.3 | 6.9 ± 3.7 | < 0.0001 |
Percentage overweighta | 26% | 21% | 0.05 |
Percentage obese/very obese | 24% | 12% | < 0.0001 |
Percentage with previous spontaneous abortion | 32% | 22% | < 0.01 |
Percentage treated by IVFb | 55% | 41% | < 0.01 |
n | 373 | 645 |
. | PCOS . | Non-PCOS . | P . |
---|---|---|---|
aExcluding obese or very obese women. | |||
bAbout 5% in each group were treated by gamete intraFallopian tube transfer (GIFT) and the remaining were treated by intracytoplasmic sperm injection (ICSI). | |||
Age (years) | 31.4 ± 3.8 | 32.7 ± 4.3 | < 0.0001 |
BMI (kg/m2) | 26.3 ± 5.6 | 24.3 ± 4.4 | < 0.0001 |
Maximal oestradiol (nmol/l) | 8.9 ± 4.3 | 6.9 ± 3.7 | < 0.0001 |
Percentage overweighta | 26% | 21% | 0.05 |
Percentage obese/very obese | 24% | 12% | < 0.0001 |
Percentage with previous spontaneous abortion | 32% | 22% | < 0.01 |
Percentage treated by IVFb | 55% | 41% | < 0.01 |
n | 373 | 645 |
. | Factors . | n . | Spontaneous abortion (%) abortion (%) . | OR . | 95% CI . |
---|---|---|---|---|---|
aWhen modelled as ordinal variables, there were significant trend effects (P < 0.05). | |||||
PCOS | No | 645 | 18 | 1 | |
Yes | 373 | 25 | 0.98 | 0.65–1.49 | |
Age(years)a | <30 | 315 | 17 | 1 | |
30–34.9 | 456 | 20 | 1.28 | 0.86–1.91 | |
35–39.9 | 209 | 23 | 1.46 | 0.90–2.36 | |
≥40 | 38 | 34 | 2.70 | 1.20–6.05 | |
BMI(kg/m2)a | <20 | 112 | 10 | 0.51 | 0.26–1.00 |
20–24.9 | 509 | 19 | 1 | ||
25–29.9 | 231 | 23 | 1.20 | 0.80–1.79 | |
30–34.9 | 116 | 32 | 1.79 | 1.16–2.75 | |
≥35 | 50 | 26 | 1.59 | 0.96–2.50 | |
Previous spontaneous abortion | None | 754 | 18 | 1 | |
One or more | 264 | 27 | 1.32 | 1.06–1.64 | |
Treatment and aetiology | Conventional IVF | 470 | 25 | 1 | |
ICSI for male factor only | 332 | 13 | 0.52 | 0.34–0.80 | |
ICSI for other aetiology | 56 | 16 | 0.68 | 0.31–1.48 | |
GIFT | 160 | 25 | 0.93 | 0.59–1.47 | |
Oestradiol at oocyte retrieval | <2 nmol/l | 41 | 20 | 1 | |
2–7.9 nmol/l | 546 | 17 | 1.01 | 0.45–2.25 | |
>8 nmol/l | 421 | 26 | 1.73 | 1.05–2.84 |
. | Factors . | n . | Spontaneous abortion (%) abortion (%) . | OR . | 95% CI . |
---|---|---|---|---|---|
aWhen modelled as ordinal variables, there were significant trend effects (P < 0.05). | |||||
PCOS | No | 645 | 18 | 1 | |
Yes | 373 | 25 | 0.98 | 0.65–1.49 | |
Age(years)a | <30 | 315 | 17 | 1 | |
30–34.9 | 456 | 20 | 1.28 | 0.86–1.91 | |
35–39.9 | 209 | 23 | 1.46 | 0.90–2.36 | |
≥40 | 38 | 34 | 2.70 | 1.20–6.05 | |
BMI(kg/m2)a | <20 | 112 | 10 | 0.51 | 0.26–1.00 |
20–24.9 | 509 | 19 | 1 | ||
25–29.9 | 231 | 23 | 1.20 | 0.80–1.79 | |
30–34.9 | 116 | 32 | 1.79 | 1.16–2.75 | |
≥35 | 50 | 26 | 1.59 | 0.96–2.50 | |
Previous spontaneous abortion | None | 754 | 18 | 1 | |
One or more | 264 | 27 | 1.32 | 1.06–1.64 | |
Treatment and aetiology | Conventional IVF | 470 | 25 | 1 | |
ICSI for male factor only | 332 | 13 | 0.52 | 0.34–0.80 | |
ICSI for other aetiology | 56 | 16 | 0.68 | 0.31–1.48 | |
GIFT | 160 | 25 | 0.93 | 0.59–1.47 | |
Oestradiol at oocyte retrieval | <2 nmol/l | 41 | 20 | 1 | |
2–7.9 nmol/l | 546 | 17 | 1.01 | 0.45–2.25 | |
>8 nmol/l | 421 | 26 | 1.73 | 1.05–2.84 |
. | Factors . | n . | Spontaneous abortion (%) abortion (%) . | OR . | 95% CI . |
---|---|---|---|---|---|
aWhen modelled as ordinal variables, there were significant trend effects (P < 0.05). | |||||
PCOS | No | 645 | 18 | 1 | |
Yes | 373 | 25 | 0.98 | 0.65–1.49 | |
Age(years)a | <30 | 315 | 17 | 1 | |
30–34.9 | 456 | 20 | 1.28 | 0.86–1.91 | |
35–39.9 | 209 | 23 | 1.46 | 0.90–2.36 | |
≥40 | 38 | 34 | 2.70 | 1.20–6.05 | |
BMI(kg/m2)a | <20 | 112 | 10 | 0.51 | 0.26–1.00 |
20–24.9 | 509 | 19 | 1 | ||
25–29.9 | 231 | 23 | 1.20 | 0.80–1.79 | |
30–34.9 | 116 | 32 | 1.79 | 1.16–2.75 | |
≥35 | 50 | 26 | 1.59 | 0.96–2.50 | |
Previous spontaneous abortion | None | 754 | 18 | 1 | |
One or more | 264 | 27 | 1.32 | 1.06–1.64 | |
Treatment and aetiology | Conventional IVF | 470 | 25 | 1 | |
ICSI for male factor only | 332 | 13 | 0.52 | 0.34–0.80 | |
ICSI for other aetiology | 56 | 16 | 0.68 | 0.31–1.48 | |
GIFT | 160 | 25 | 0.93 | 0.59–1.47 | |
Oestradiol at oocyte retrieval | <2 nmol/l | 41 | 20 | 1 | |
2–7.9 nmol/l | 546 | 17 | 1.01 | 0.45–2.25 | |
>8 nmol/l | 421 | 26 | 1.73 | 1.05–2.84 |
. | Factors . | n . | Spontaneous abortion (%) abortion (%) . | OR . | 95% CI . |
---|---|---|---|---|---|
aWhen modelled as ordinal variables, there were significant trend effects (P < 0.05). | |||||
PCOS | No | 645 | 18 | 1 | |
Yes | 373 | 25 | 0.98 | 0.65–1.49 | |
Age(years)a | <30 | 315 | 17 | 1 | |
30–34.9 | 456 | 20 | 1.28 | 0.86–1.91 | |
35–39.9 | 209 | 23 | 1.46 | 0.90–2.36 | |
≥40 | 38 | 34 | 2.70 | 1.20–6.05 | |
BMI(kg/m2)a | <20 | 112 | 10 | 0.51 | 0.26–1.00 |
20–24.9 | 509 | 19 | 1 | ||
25–29.9 | 231 | 23 | 1.20 | 0.80–1.79 | |
30–34.9 | 116 | 32 | 1.79 | 1.16–2.75 | |
≥35 | 50 | 26 | 1.59 | 0.96–2.50 | |
Previous spontaneous abortion | None | 754 | 18 | 1 | |
One or more | 264 | 27 | 1.32 | 1.06–1.64 | |
Treatment and aetiology | Conventional IVF | 470 | 25 | 1 | |
ICSI for male factor only | 332 | 13 | 0.52 | 0.34–0.80 | |
ICSI for other aetiology | 56 | 16 | 0.68 | 0.31–1.48 | |
GIFT | 160 | 25 | 0.93 | 0.59–1.47 | |
Oestradiol at oocyte retrieval | <2 nmol/l | 41 | 20 | 1 | |
2–7.9 nmol/l | 546 | 17 | 1.01 | 0.45–2.25 | |
>8 nmol/l | 421 | 26 | 1.73 | 1.05–2.84 |
To whom correspondence should be addressed. E-mail: jim.wang@adelaide.edu.au
Ms Barbara Godfrey and other staff of Reproductive Medicine Unit are acknowledged for their contribution to the data collection as part of their routine work.
References
Balen, A.H., Tan, S.L., MacDougall, J. and Jacobs, H.S. (
Bohrer, M. and Kemmann, E. (
Fedorcsak, P., Storeng, R., Dale, P.O. et al. (
Franks, S. and Hamilton Fairley, D. (
Hamilton Fairley, D., Kiddy, D., Watson, H. et al. (
Kerin, J.F., Warnes, G.M., Quinn, P. et al. (
Norman, R.J. and Clark, A.M. (
Norman, R.J., Masters, S.C., Hague, W. et al. (
Norman, R.J., Warnes, G.M., Wang, X. et al. (
Orvieto, R., Ben-Rafael, Z., Ashkenazi, J. et al. (
Pasquali, R., Casimirri, F. and Vicennati, V. (
Ruutiainen, K. and Seppala, M. (
Sathanandan, M., Warnes, G.M., Kirby, C.A. et al. (
Simon, C., Garcia Velasco, J.J., Valbuena, D. et al. (
Solomon, C.G. (
Tarlatzis, B.C., Grimbizis, G., Pournaropoulos, F. et al. (