1 Danish Epidemiology Science Centre, University of Århus, Denmark and 2 Epidemiology Branch, NIEHS, Research Triangle Park, NC, USA
3 To whom correspondence should be addressed at: Danish Epidemiology Science Centre, Department of Epidemiology and Social Medicine, Building 260, University of Århus, Vennelyst Boulevard 6, 8000 Århus C, Denmark. e-mail: ob{at}soci.au.dk
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: birthweight/Caesarean section/infertility/preterm delivery/time to pregnancy
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Infertility can be caused by disruption in any of several reproductive processes. Some of these problems, such as reproductive tract infections, hormonal disorders or poor placentation, could also plausibly contribute to adverse birth outcomes. Despite the high proportion of babies born to infertile couples, surprisingly little research has considered the relationship between infertility and pregnancy outcome. If infertility is a risk factor for poor pregnancy outcome, identifying such pregnancies as high risk so that they can be closely monitored might reduce adverse events.
In this paper we use data on live births within a large Danish pregnancy cohort to examine whether infertile couples, identified as those taking >1 year to conceive, have an elevated risk of preterm delivery, decreased birthweight, or delivery by Caesarean section. We adjust for several possible risk factors that might affect both infertility and birth outcome, and we compare infertile couples who conceived with and without treatment.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We had 64 167 records of responders to the first interview between December 12, 1997 and July 31, 2001. As shown in Figure 1, we excluded pregnancies ending in other than singleton live births as well as those with critical missing data. We focused on live births because both preterm delivery and reduced growth may have long-term consequences. We used two sources of information on gestational age: one was based on the womans estimated date of delivery provided at the first interview, and the other was gestational age from the birth record. Because we needed good quality information on gestational age at delivery, we identified records with data from both sources (58 656, 99% of the available records). Of these, we included 45 519 (77.6%) that had the same estimated week of delivery reported in both instances and 10 387 (17.7%) that differed by 1 week. When sources differed, we used the gestation reported in the birth record. We thus had 55 906 births for analysis. Data on mothers age at delivery, sex of the baby, Caesarean sections and birth outcomes were taken from the hospital birth record obtained from the Danish National Board of Health. The other analysis variables were derived from the first interview. Median time of interview among women in this analysis was the 16th week, and 95% of interviews were completed by the 25th week.
|
The social status variable was based on the self-reported job title of the current or most recent job (if the women worked in the last 6 months). If participants were in school, even if working part time, the social status was based on the type of education. We included in the high social status all those in management and requiring higher education, generally 4 years beyond the compulsory level of 9 years. Office workers, service workers, skilled manual workers, and women in the military constituted the middle category, and unskilled workers were classified in the lowest social class. Job titles had been categorized according to Denmarks job classification scheme (Danmarks Statistiks Fagklassifikation, 1996
).
We estimated the association between a TTP of >1 year and dichotomous birth outcomes (preterm delivery and very preterm delivery, which was defined as a gestation ending before the completion of the 34th week, planned Caesarean section and emergency Caesarean section) using logistic regression models. We modelled in one variable with eight levels all five categories of TTP (right away, which was the reference category, 12 months, 35 month, 612 months and >12 months) and the three categories of unplanned pregnancy (partly planned, conceived despite contraception and not planned) We stratified all analyses by parity (primiparas versus multiparas), and we adjusted for maternal age at delivery, pre-pregnancy body mass index (BMI, kg/m2), smoking in pregnancy, social status of the mother and sex of the baby. Menstrual cycle characteristics may influence the estimate of both TTP and gestational age and thus we adjusted for menarche and cycle length/regularity in the models where gestational age was not included as a covariate. All variables were categorized as reported in Table I and Table II. When examining multiparas, we further adjusted for parity (2nd birth versus 3rd or higher). The decision on which factors to include was made a priori.
|
|
We examined the association between TTP and birthweight by fitting linear regression models using covariates as described above, but we excluded menarche and menstrual cycle characteristics and included gestational age at birth (in 15 categories) to explore whether the growth of the baby was affected independently of gestational age. We also estimated the association between a TTP >1 year and low birthweight at term, restricting the analyses to births occurring from week 37 onward. We used logistic regression and adjusted for the same covariates as in the birthweight models but with the narrowed gestational age range.
For all outcomes we estimated the association with a TTP >1 year among all women and we repeated the analyses excluding all women who reported having received infertility treatment, regardless of the duration of their TTP. For the sake of brevity, we only present tables including the estimates for a TTP >1 year.
There might be adverse effects of infertility treatment or differences in the severity of infertility between women who conceive with and without treatment. We therefore estimated the odds ratio (OR) of adverse birth outcomes associated with treatment restricting the analysis to women taking >1 year to conceive. We adjusted for the same covariates as in the TTP analyses and stratified on parity.
Couples planning a pregnancy may include an excess of infertile couples, so we repeated all analyses using as a reference category the combination of couples who conceived right away, the part planners, those who conceived despite birth control, and the non-planners.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Table III shows estimates for the association between a TTP of >1 year and the outcomes of interest. Among both primiparas and multiparas, infertile women had an increased risk of all adverse study outcomes, although the crude estimates were higher than the adjusted ones. Notably, the adjusted ORs of preterm birth were significantly elevated even after excluding couples reporting infertility treatment.
|
Table IV shows the relationship between treatment and the same outcomes shown in Table III among women with a TTP >1 year. Among primiparas there were no substantial differences between treated and untreated. However, among multiparas, the risks were consistently elevated for the treated women with the exception of mean birthweight, which was very similar for treated and untreated women.
|
When we used a reference category of the combination of couples who conceived right away, the part planners, those who conceived despite birth control, and the non-planners the results were virtually the same (data not shown).
Finally, we reintroduced into the analyses women who had been excluded because of missing data in pre-pregnancy BMI, smoking, or the estimate of the expected date of delivery, or because of discrepancies in the two values of gestational age (n = 4278). Missing values for BMI and smoking were coded in separate categories in the models. Compared with the adjusted estimates among all women in Table III, the differences concerned only the estimates of a delivery before 34 completed weeks, as primiparas had a weaker estimate (OR: 1.36; 95% confidence interval: 0.96, 1.91), and multiparas a stronger one (1.68; 1.09, 2.59).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Adverse birth outcomes have previously been described for singleton babies born after IVF (MRC Working Party, 1990; Tan et al., 1992
; Olivennes et al., 1993
; Wang et al., 1994
, 2002; Dhont et al., 1999
; Koudstaal et al., 2000
). Less is known about the general population of infertile couples, though increased adverse birth outcomes have been reported (Bhalla et al., 1992
; Olivennes et al., 1993
; Joffe et al., 1994
; Sundstrom et al., 1997
; Henriksen et al., 1997
).
Our data show that infertile couples are at significantly elevated risk of adverse birth outcomes, and the risks cannot be attributed solely to the effects of infertility treatment. A long waiting time to pregnancy is a proxy for conditions in both the male and female, and a number of these conditions may share one or more causal paths with the adverse birth outcomes. When we adjusted for factors such as age and smoking, the associations between infertility and birth outcomes decreased in magnitude, indicating that these factors may explain some of the association. However, significant associations still remained, probably due to further overlapping aetiological paths for which we could not adjust. For example, Chlamydia infection is known to cause infertility (Toye et al., 1993; Land and Evers, 2002
) and is also associated with preterm delivery (Johns Hopkins Study, 1989
; Kovacs et al., 1998
; Andrews et al., 2000
). Other possible common causes of infertility and preterm birth are prenatal diethylstilbestrol exposure (Senekjian et al., 1988
), solvent exposure (Lindbohm 1995
; Sallmen et al., 1995
) and psychological stress (Copper et al., 1996
).
We conducted a series of exclusion analyses to try to identify possible mediators in the TTPpreterm birth relationship. A long waiting time was associated with pre-eclampsia in a previous study conducted on women enrolled in the Danish National Birth Cohort (Basso et al., 2003), but the association with preterm delivery reported in the present study is not a consequence of pre-eclampsia; the association was essentially unchanged when we excluded pre-eclamptic pregnancies. If abnormal hormonal factors were the mediator, the association might be weaker if women with irregular menstrual cycles were excluded, but the effect was essentially unchanged. Nor did excluding women with a high BMI change the association. There was also no evidence that age-related changes in ovarian and uterine function were important mediators, because when older women were excluded, the magnitude of the association actually increased somewhat, though not significantly.
We had anticipated that couples undergoing infertility treatment would show higher risks than untreated infertile couples because treated couples may have more severe infertility than those who conceive without treatment, and the treatment procedures may have adverse effects. For example, ovulation induction results in increased endometrial thickness, and this might be detrimental to implantation and placentation (Weissman et al., 1999). Our observation that treated and untreated primiparas have similar risks provides some reassurance that treatments per se may not be a major factor in adverse birth outcomes. However, the underlying causes of infertility are likely to differ between treated and untreated individuals. Without taking these differences into account, the effect of treatment cannot be definitively evaluated.
Our study is limited in other ways beyond our inability to ascertain causes of infertility. The data on infertility, treatment and most of the covariates were obtained by self-report. Since TTP was categorized in the questionnaire, all couples who took >1 year to conceive were in one group, so we were unable to control for duration of infertility. Finally, the relatively low participation rate in the cohort could be a source of bias. Unplanned pregnancies may be under-represented. This could create a spurious association if fertile women with a higher risk of adverse birth outcome were systematically under-represented. An examination of selection in the cohort within one geographical area (Northern Jutland: Nøhr, 2001) revealed somewhat higher rates of deliveries before 32 weeks in the general population compared with the cohort (1.0 versus 0.6%), but similar overall preterm birth rates (5.0 versus 4.4%). Thus, at least our findings regarding preterm birth are unlikely to be explained by selection bias. We also evaluated bias from unplanned pregnancies (Weinberg at al., 1994
; Olsen et al., 1998
) within our study by including partly planned and unplanned pregnancies within the reference group with which the infertile group were compared, and by adding back the women who had been excluded because of missing data or differences in the two sources of gestational age. In both analyses the findings were essentially unchanged.
The study had numerous strengths. Our sample was large, which allowed us to evaluate effects for primiparas and multiparas separately. Data on waiting time to pregnancy and planning status were collected before the outcome, so knowledge of the outcome did not influence reporting. Infertility treatment is free of charge in Denmark, at least for a limited number of attempts, so treatment is not limited to high-income couples. We had data on several potential risk factors, and results were adjusted for these. Our data on gestational age were reliable, and this increased the accuracy of our determination of preterm births and allowed us to adjust appropriately for gestational age when looking for effects on birthweight.
Adverse birth outcomes are important causes of infant and maternal morbidity. With >10% of babies born to infertile couples, it is important to consider this group as potentially high risk when providing prenatal care. In addition, research aimed at understanding the biological basis of the association between infertility and adverse birth outcomes is needed. Such research may suggest new approaches for preventing adverse events.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Basso, O., Weinberg, C.R., Baird, D.D., Wilcox, A.J. and Olsen, J. (2003) Subfecundity as a correlate of preeclampsia: a study within the Danish National Birth Cohort. Am. J. Epidemiol., 157, 195202.
Bhalla, A.K., Sarala, G. and Dhaliwal, L. (1992) Pregnancy following infertility. Aust. NZ Obstet. Gynecol., 32, 249251.
Copper, R.L., Goldenberg, R.L., Das, A. et al. (1996) The preterm prediction study: maternal stress is associated with spontaneous preterm birth at less than 35 weeks gestation. Am. J. Obstet. Gynecol., 175, 12861292.[ISI][Medline]
Danmarks Statistiks Fagklassifikation (1996) 1. udgave. Danmarks Statistik, Copenhagen.
Dhont, M., De Sutter, P., Ruyssinck, G., Martens, G. and Bekaert, A. (1999) Perinatal outcome of pregnancies after assisted reproduction: a case-control study. Am. J. Obstet. Gynecol., 181, 688695.[ISI][Medline]
Henriksen, T.B., Baird, D.D., Olsen, J., Hedegaard, M., Secher, N.J. and Wilcox, A.J. (1997) Time to pregnancy and preterm delivery. Obstet. Gynecol., 89, 594599.
Johns Hopkins Study of Cervicitis and Adverse Pregnancy Outcome (1989) Association of Chlamydia trachomatis and Mycoplasma hominis with intrauterine growth retardation and preterm delivery. Am. J. Epidemiol., 129, 12471257.[Abstract]
Joffe, M. and Li, Z. (1994) Association of time to pregnancy and the outcome of pregnancy. Fertil. Steril., 62, 7175.[ISI][Medline]
Juul, S., Karmaus, W., Olsen, J. and the European Infertility and Subfecundity Study Group (1999) Regional differences in waiting time to pregnancy: pregnancy-based surveys from Denmark, France, Germany, Italy and Sweden. Hum. Reprod., 14, 12501254.
Koudstaal, J., Braat, D.D., Bruinse, H.W., Naaktgeboren, N., Vermeiden, J.P. and Visser, G.H. (2000) Obstetric outcome of singleton pregnancies after IVF: a matched control study in four Dutch university hospitals. Hum. Reprod., 15, 18191825.
Kovacs, L., Nagy, E., Berbik, I., Meszaros, G., Deak, J. and Nyari, T. (1998) The frequency and the role of Chlamydia trachomatis infection in premature labor. Int. J. Gynaecol. Obstet., 62, 4754.[CrossRef][ISI][Medline]
Land, J.A. and Evers, J.L. (2002) Chlamydia infection and subfertility. Best Pract. Res. Clin. Obstet. Gynecol., 16, 901912.[CrossRef]
Lindbohm, M.L. (1995) Effects of parental exposure to solvents on pregnancy outcome. J. Occup. Environ. Med., 37, 908914.[ISI][Medline]
MRC Working Party on Children Conceived by In Vitro Fertilisation (1990) Births in Great Britain resulting from assisted conception, 197887. Br. Med. J., 300, 12291233.[ISI][Medline]
Nøhr, E.A. (2001) Kan selektionen til "Bedre Sundhed for Mor og Barn" forårsage bias af forskningsresultaterne? (Can selection into the cohort "Better Health for Mother and Child" cause bias in research estimates?). Thesis, Aarhus University, Aarhus, Denmark.
Olivennes, F., Rufat, P., Andre, B., Pourade, A., Quiros, M.C. and Frydman, R. (1993) The increased risk of complication observed in singleton pregnancies resulting from in-vitro fertilization (IVF) does not seem to be related to the IVF method itself. Hum. Reprod., 8, 12971300.[Abstract]
Olsen, J., Juul, S. and Basso, O. (1998) Measuring time to pregnancy. Methodological issues to consider. Hum. Reprod., 13, 17511756.
Olsen, J., Melbye, M., Olsen, S.F., Sorensen, T.I., Aaby, P., Andersen, A.M., Taxbol, D., Hansen, K.D., Juhl, M., Schow, T.B. et al. (2001) The Danish National Birth Cohortits background, structure and aim. Scand. J. Public Health, 29, 300307.[CrossRef][ISI][Medline]
Reubinoff, B.E., Samueloff, A., Ben-Haim, M., Friedler, S., Schenker, J.G. and Lewin, A. (1997) Is the obstetric outcome of in vitro fertilized singleton gestations different from natural ones? A controlled study. Fertil. Steril., 67, 10771083.[CrossRef][ISI][Medline]
Sallmen, M., Lindbohm, M.L., Kyyronen, P., Nykyri, E., Anttila, A., Taskinen, H. and Hemminki, K. (1995) Reduced fertility among women exposed to organic solvents. Am. J. Ind. Med., 27, 699713.[ISI][Medline]
Senekjian, E.K., Potkul, R.K., Frey, K. and Herbst, A.L. (1988) Infertility among daughters either exposed or not exposed to diethylstilbestrol. Am. J. Obstet. Gynecol., 158, 493498.[ISI][Medline]
Sundstrom, I., Ildgruben, A. and Hogberg, U. (1997) Treatment-related and treatment-independent deliveries among infertile couples, a long-term follow-up. Acta Obstet. Gynecol. Scand., 76, 238243.[ISI][Medline]
Tan, S.L., Doyle, P., Campbell, S., Beral, V., Rizk, B., Brinsden, P., Mason, B. and Edwards, R.G. (1992) Obstetric outcome of in vitro fertilization pregnancies compared with normally conceived pregnancies. Am. J. Obstet. Gynecol., 167, 778784.[ISI][Medline]
Toth, M., Witkin, S.S., Ledger, W. and Thaler, H. (1988) The role of infection in the etiology of preterm birth. Obstet. Gynecol., 71, 723726.[Abstract]
Toye, B., Laferriere, C., Claman, P., Jessamine, P. and Peeling, R. (1993) Association between antibody to the chlamydial heat-shock protein and tubal infertility. J. Infect. Dis., 168, 12361240.[ISI][Medline]
Wang, J.X., Clark, A.M., Kirby, C.A., Philipson, G., Petrucco, O., Anderson, G. and Matthews, C.D. (1994) The obstetric outcome of singleton pregnancies following in-vitro fertilization/gamete intra-fallopian transfer. Hum. Reprod., 9, 141146.[Abstract]
Wang, J.X., Norman, R.J. and Kristiansson, P. (2002) The effect of various infertility treatments on the risk of preterm birth. Hum. Reprod., 17, 945949.
Weinberg, C.R., Baird, D.D. and Wilcox, A.J. (1994) Sources of bias in studies of time to pregnancy. Stat. Med., 13, 671681.[ISI][Medline]
Weissman, A., Gotlieb, L. and Casper, R.F. (1999) The detrimental effect of increased endometrial thickness on implantation and pregnancy rates and outcome in an in vitro fertilization program. Fertil. Steril., 71, 147149.[CrossRef][ISI][Medline]
Williams, M.A., Goldman, M.B., Mittendorf, R. and Monson, R.R. (1991) Subfertility and the risk of low birth weight. Fertil. Steril., 56, 668671.[ISI][Medline]
Submitted on April 30, 2003; accepted on July 22, 2003.