Human Reproduction, Vol. 18, No. 2, 437-440,
February 2003
© 2003 European Society of Human Reproduction and Embryology
Embryo reduction and birth weight discordance in dichorionic twins
Department of Obstetrics and Gynecology, Hôpital Antoine Béclère, Assistance Publique Hopitaux de Paris, Université Paris XI, 92140 Clamart, France
1 To whom correspondence should be addressed. e-mail: francois.audibert{at}abc.ap-hop-paris.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
BACKGROUND: Twin birth weight discordance is associated with a poor perinatal outcome. The aim of this study was to analyse the risk factors of growth discordance among dichorionic twin pregnancies. METHODS: A cohort of 346 dichorionic twin pregnancies delivered at one perinatal centre between January 1996 and December 1999 was analysed. Two groups were created, according to the presence or absence of intra-pair birth weight discordance (n = 75 and 271 respectively). Birth weight discordance was defined as a difference of
20% of the weight of the heavier twin. The two groups were compared by uni- and multivariate analysis, with regard to the woman's characteristics, risk factors for growth restriction or discordance, and outcome of pregnancy. RESULTS: Pregnancies with birth weight discordance had a poor outcome compared with pregnancies without discordance, with a 4-fold increase in neonatal mortality. The rate of iatrogenic embryo reduction was significantly higher in discordant pregnancies (14.7 versus 6.6%, P = 0.03). The risk of birth weight discordance was increased with a larger starting number of embryos before reduction [20.2% (64/317), 28.6% (6/21), 57.1% (4/7) and 100% (1/1) respectively, for an initial number of two (no reduction), three, four, and five embryos, P = 0.02]. In multivariate analysis, embryo reduction was the only significant risk factor for the occurrence of birth weight discordance [adjusted odds ratio (OR) = 2.3 (1.0–5.2)]. CONCLUSIONS: Birth weight discordance carries a poor perinatal outcome. Embryo reduction is an independent risk factor for birth weight discordance in dichorionic twins. This finding emphasises the need for better control of assisted reproductive technology in order to avoid high-order multiple pregnancies.
Key words: birth weight discordance/embryo reduction/growth restriction/twin birth
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Twin births account for an increasing percentage of low birth weight infants, preterm births, and perinatal mortality (Petterson et al., 1998
; Kogan et al., 2000). The rate of twin births has dramatically increased in the last decades (Martin and Park, 1999), mainly as a result of the development of assisted reproductive technology (ART) (Bergh et al., 1999). In addition, the use of ART increases the risk of low birth weight in singletons and twins (Schieve et al., 2002). It has been shown that birth weight discordance between twins is a risk factor for poor perinatal outcome (Fraser et al., 1994; Blickstein et al., 2000; Cooperstock et al., 2000). Indeed, discordant growth has been associated with a 6.5-fold increased risk for fetal death and 2.5-fold increased risk for perinatal mortality in twins (O’Brien et al., 1986). Moreover, the risk of very preterm birth (<32 weeks) is dramatically increased in the presence of birth weight discordance (Cooperstock et al., 2000). This increase is mainly due to obstetric interventions triggered by the fear of fetal death of a compromised twin (Hollier et al., 1999), thus exposing the other twin to the complications of prematurity. The underlying mechanisms of discordance may be very different in monochorionic and dichorionic twins. In monochorionic pregnancies, the presence of placental anastomosis and specific complications such as twin–to–twin transfusion syndrome (Ville et al., 1995) or abnormal umbilical cord insertion (Hanley et al., 2002) account for the majority of growth discordances (D’Alton and Simpson, 1995). In contrast, specific causes of growth discordance in dichorionic twins have been poorly studied. The purpose of this study was to investigate the prenatal factors associated with the occurrence of birth weight discordance in dichorionic twins. |
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
We studied a cohort of all twin dichorionic pregnancies delivered in our tertiary perinatal centre at
24 weeks gestation between January 1996 and December 1999. Gestational age was determined by a first trimester ultrasound in all cases. Chorionicity was determined by first trimester ultrasound in all cases and all placentas were examined by a pathologist. This examination confirmed that all cases were dichorionic pregnancies, with chorion interposition between the two amnionic layers. Embryo reduction was proposed to women with high-order multiple pregnancies. Among these, there were initially 21 triple pregnancies, seven quadruple pregnancies, and one quintuple pregnancy. All procedures of embryo reduction were performed between 10 and 12 weeks gestation, using a transabdominal intracardiac injection of potassium chloride under ultrasound guidance.
Data were obtained from the obstetric and perinatal databases for all pregnancies, and ultrasound reports and individual charts were reviewed for all cases with birth weight discordance or embryo reduction. Overall, 358 sets of dichorionic twins were included. Six cases were excluded because of intra-uterine fetal death at 4 weeks before birth. Indeed, birth weight discordance was obviously present in these cases, and we have considered that keeping these cases would have biased the definition of our study groups (weight discordance could be a cause or a consequence of fetal death). None of these excluded cases had undergone embryo reduction procedure. Another six cases were excluded because of missing data, leaving 346 dichorionic pregnancies for the final study group.
Birth weight discordance, expressed as a percentage, was determined as 100(A-B)/A, where A was the birth weight of the heavier twin and B was the birth weight of the lighter twin (Cooperstock et al., 2000). Small for gestational age (SGA) newborns were defined by a birth weight <10th percentile for gestational age (Leroy and Lefort, 1971; Keen and Pearse, 1985). Discordant sets of twins were defined by a birth weight discordance >20%.
Statistical analysis was performed using Stata 7.0 software (Stata Corporation, College Station, TX, USA). Univariate analysis was performed by 
2-test for categorical variables and by two-tailed t-test for numerical variables. Multivariate analysis was done by stepwise forward logistic regression, with a P value of 
0.25 for entry into the model. The following variables were either previously demonstrated to be associated to growth restriction or were found to be associated with discordance in the univariate analysis and were therefore included in the model: maternal age >35 years, nulliparity, ovulation induction, IVF (including ICSI and oocyte donation), embryo reduction, fetal or neonatal malformation. A P-value < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The mean maternal age of the group was 31.9 ± 4.4 years and 66% of women were nulliparous. The majority of pregnancies (62%) were obtained by ART (22% ovulation induction, 28% IVF, 10% ICSI, and 2% by oocyte donation). An embryo reduction was performed in 29 pregnancies (8.4%). Birth weight discordance was found in 75 sets of twins (21.7%).
Table I compares the characteristics of pregnancies, according to the presence or absence of birth weight discordance. Among discordant twins, the rate of pregnancies obtained by ART was higher than among concordant twins (72 versus 60%, P = 0.053). The rate of embryo reduction was significantly higher in discordant twins (14.7 versus 6.6%, P = 0.03). Moreover, the risk of birth weight discordance increased with greater initial number of embryos [20.2% (64/317), 28.6% (6/21), 57.1% (4/7) and 100% (1/1) respectively for an initial number of two (no reduction), three, four, and five embryos, P = 0.02]. Data on outcome of pregnancies according to birth weight discordance are given in Table II. Perinatal mortality and morbidity were significantly increased in the presence of discordance. Birth weight discordance was associated with a four-fold increased risk of neonatal mortality, and a two-fold increased risk of neonatal intensive care unit (NICU) admission. When comparing reduced versus non-reduced pregnancies, we found a significantly higher incidence of birth weight discordance, but no difference was noted in the incidence of small for gestational age new-borns, in one or both twins. Moreover, birth weights did not differ significantly between reduced and non-reduced pregnancies (Table III). We calculated the ‘number needed to harm’ with the formula 1/(0.038–0.02) = 5.5, which means that, based on our results, 5–6 procedures of reductions might give rise to one more discordant twin set. In multivariate analysis, embryo reduction was the only significant and independent risk factor for the occurrence of birth weight discordance (Table IV).
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
We found a significant and independent association between birth weight discordance and embryo reduction in dichorionic twin pregnancies. Few studies have focused on the association between birth weight discordance in dichorionic twins and embryo reduction. In a large multicentre study, Evans and co-workers noted that birth weight discordance increased with greater starting number before reduction (Evans et al., 2001
), but this issue was not compared with ‘primary’ twin pregnancies. In a case–control study, Silver and colleagues found that twin gestations resulting from multifetal reduction were at increased risk for preterm birth, fetal growth restriction, and discordancy when compared with non-reduced twins conceived as a result of fertility therapy (Silver et al., 1997). This finding is in conflict with another study comparing the outcome of 441 pregnancies reduced to twins with 136 non-reduced dichorionic twins (Torok et al., 1998), showing no association between embryo reduction and intrauterine growth restriction or birth weight discordance, except for very high order pregnancies (starting fetal number of 5).
Several limitations of this study must be acknowledged: this is a retrospective study conducted in a single tertiary perinatal centre, and our population may not reflect the general twin population, due to a high rate of ART, and to a high rate of pregnancies referred because of complications (this bias concerns our whole population, not only the subset of those twins born after embryo reduction). However, the proportion of discordant twin pregnancies in our population is consistent with other reports using the same definition of birth weight discordance. Indeed, Hanley et al. 2002 found a discordance in 61/341 (17.9%) of dichorionic pregnancies in a recent study. In the same way, Torok et al. note that 26/136 (19.1%) of non-reduced dichorionic pregnancies showed a birth weight discordance of 20% (Torok et al., 1998). All these studies might be concerned by the same selection bias but, to our knowledge, there is no available population-based data about birth weight discordance in dichorionic twins.
We focused our attention towards growth discordance rather than intrauterine growth restriction, because discordance raises a difficult challenge in choosing the appropriate time for delivery. Indeed, the decision to deliver twins prematurely when only one twin appears compromised may result in unnecessary prematurity and subsequent morbidity for the other twin. While several studies emphasise the clinical importance of growth discordance in twins (Hollier et al., 1999; Cooperstock et al., 2000), others conclude that prematurity and low-birth-weight, not birth weight discordance, are the greatest threat to the new-born twin (Patterson and Wood, 1990; Fraser et al., 1994).
It is not clear whether the association we found between embryo reduction and growth discordance is causative; however, we believe that we controlled for the major bias that could generate such an association. The mechanisms by which birth weight discordance might be caused by embryo reduction remain unclear and warrant further research. The hypothesis of ‘uterine imprinting’ determined early in gestation is proposed by Evans et al. 2001. This early imprinting might persist after the reduction, thus limiting the functional placental territory dedicated to the adjacent fetus. Another possible cause of growth restriction in one twin is the local inflammatory response that could be developed at the contact of the reduced sac. However, there are currently no studies to confirm this hypothesis.
We conclude that multifetal pregnancy reduction should now be considered an independent epidemiological risk factor for birth weight discordance in twins, a condition associated with a poor perinatal outcome. Therefore, ultrasound monitoring of twin pregnancies resulting from reduction should be encouraged. In addition, further studies examining the outcome of reduced pregnancies should include the issue of discordance in their results. Finally, our findings add further concern about infertility therapy risks, and emphasise the need for preventing high-order multiple pregnancies.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Bergh, T., Ericson, A., Hillensjo, T., Nygren, K.G. and Wennerholm, U.B. (1999) Deliveries and children born after in-vitro fertilisation in Sweden 1982–1995: a retrospective cohort study. Lancet, 354, 1579–1585.[CrossRef][Web of Science][Medline]
Blickstein, I., Goldman, R.D. and Mazkereth, R. (2000) Risk for one or two very low birth weight twins: a population study. Obstet. Gynecol., 96, 400–402.[Medline]
Cooperstock, M.S., Tummaru, R., Bakewell, J. and Schramm, W. (2000) Twin birth weight discordance and risk of preterm birth. Am. J. Obstet. Gynecol., 183, 63–67.[Medline]
D’Alton, M.E. and Simpson, L.L. (1995) Syndromes in twins. Semin. Perinatol., 19, 375–386.[Medline]
Evans, M.I., Berkowitz, R.L., Wapner, R.J., Carpenter, R.J., Goldberg, J.D., Ayoub, M.A., Horenstein, J., Dommergues, M., Brambati, B., Nicolaides, K.H., et al. (2001) Improvement in outcomes of multifetal pregnancy reduction with increased experience. Am. J. Obstet. Gynecol., 184, 97–103.[CrossRef][Web of Science][Medline]
Fraser, D., Picard, R., Picard, E. and Leiberman, J.R. (1994). Birth weight discordance, intrauterine growth retardation and perinatal outcomes in twins. J. Reprod. Med., 39, 504–508.[Medline]
Hanley, M.L., Ananth, C.V., Shen-Schwarz, S., Smulian, J.C., Lai, Y.L. and Vintzileos, A.M. (2002) Placental cord insertion and birth weight discordancy in twin gestations. Obstet. Gynecol., 99, 477–482.[CrossRef][Web of Science][Medline]
Hollier, L.M., McIntire, D.D. and Leveno, K.J. (1999) Outcome of twin pregnancies according to intrapair birth weight differences. Obstet. Gynecol., 94, 1006–1010.[Medline]
Keen, D.V. and Pearse, R.G. (1985) Birthweight between 14 and 42 weeks’ gestation. Arch. Dis. Child., 60, 440–446.
Kogan, M.D., Alexander, G.R., Kotelchuck, M., MacDorman, M.F., Buekens, P., Martin, J.A. and Papiernik, E. (2000) Trends in twin birth outcomes and prenatal care utilization in the United States, 1981–1997. JAMA, 284, 335–341.
Leroy, B. and Lefort, F. (1971) Poids et taille des nouveau-nés à la naissance. Rev. Fr. Gynecol. Obstet., 66, 391–396.[Medline]
Martin, J. and Park, M. (1999) Trends in twin and triplet births. Natl Vital Stat. Rep., 47, 1–16.
O’Brien, W.F., Knuppel, R.A., Scerbo, J.C. and Rattan, P.K. (1986) Birth weight in twins: an analysis of discordancy and growth retardation. Obstet. Gynecol., 67, 483–486.[Web of Science][Medline]
Patterson, R.M. and Wood, R.C. (1990). What is twin birthweight discordance? Am. J. Perinatol., 7, 217–219.[Medline]
Petterson, B., Blair, E., Watson, L. and Stanley, F. (1998) Adverse outcome after multiple pregnancy. Baillieres Clin. Obstet. Gynecol., 12, 1–17.[CrossRef][Web of Science][Medline]
Schieve, L., Meikle, S., Ferre, C., Peterson, H., Jeng, G. and Wilcox, L. (2002) Low and very low birth weight in infants conceived with use of assisted reproductive technology. N. Engl. J. Med., 346, 731–737.
Silver, R.K., Helfand, B.T., Russell, T.L., Ragin, A., Sholl, J.S. and MacGregor, S.N. (1997) Multifetal reduction increases the risk of preterm delivery and fetal growth restriction in twins: a case-control study. Fertil. Steril., 67, 30–33.[CrossRef][Web of Science][Medline]
Torok, O., Lapinski, R., Salafia, C.M., Bernasko, J. and Berkowitz, R.L. (1998) Multifetal pregnancy reduction is not associated with an increased risk of intrauterine growth restriction, except for very-high-order multiples. Am. J. Obstet. Gynecol., 179, 221–225.[CrossRef][Web of Science][Medline]
Ville, Y., Hyett, J., Hecher, K. and Nicolaides, K. (1995). Preliminary experience with endoscopic laser surgery for severe twin-twin transfusion syndrome. N. Engl. J. Med., 332, 224–227.
Submitted on August 29, 2002; accepted on October 23, 2002.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||