- Email: [email protected]
Delivery of the term twin
Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 18, No. 4, pp. 625–630, 2004 doi:10.1016/j.bpobgyn.2004.04.010 available online at http://www.sciencedirect.com
8 Delivery of the term twin J. F. R. Barrett*
MBBCH, MD, FRCOG FRCSC
Specialist in Maternal Fetal Medicine and Director of Multiple Births, Sunnybrook and Women’s College Hospital; Associate Professor, University of Toronto Sunnybrook and Women’s College Hospital, University of Toronto, 60 Grosvenor Street, Toronto, Ont., Canada M5S 1B6
The ever-increasing incidence of twin pregnancies world wide, together with the increasing trend to caesarean delivery, has resulted in intense scrutiny of the most appropriate method of twin delivery. The term twin has an increased risk of twin mortality compared to term singletons and this might be a result of the increase risk of labour and delivery compared to that of singletons. There are three ways to address this from the literature. The first is to compare outcome for the second twin versus the first twin, and to compare these outcomes in those twins delivered vaginally compared to those delivered by lower section caesarean section (LSCS). The second is to compare outcomes for twins delivered vaginally and for those delivered by caesarean section (CS). These data show higher rates of adverse perinatal outcome for the twin at or near term if delivery is vaginal versus CS. The third method is to compare outcomes for twins delivered by planned vaginal birth (VB; actual VB plus emergency CS) versus planned CS. This chapter will review this data thus outline an ongoing randomized controlled trial—the Twin Birth Study. Key words: twin delivery; term twin; vaginal delivery; caesarean section; randomized controlled trial.
The incidence of twins worldwide continues to increase.1 Attempts at reducing the incidence of higher-order multiples, such as triplets or more, have met with some success in countries that have legislated against multiple embryo reimplantation during in vitro fertilization (IVF) cycles. However, even in these tightly controlled cycles twin pregnancies occur at a rate 10-fold that of normal cycle conception.2 In addition, and unfortunately, many countries (including the author’s) still do not have effective control over, or even accurate information of, practices within fertility centers, and significant numbers of higher-order multiples are still conceived. Although not without some controversy and ethical considerations, these higher-order multiples are often reduced. However, the final number reduced to is usually two3, thus further increasing the population of twins that will ultimately need delivering. Finally, in Western countries * Tel.: þ 1-416-323-7551; Fax: þ1-416-323-6319. E-mail address: [email protected] (J.F.R. Barrett). 1521-6934/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.
626 J. F. R. Barrett
the number of spontaneously conceived twins is increasing consequent on the population choosing to conceive at later maternal ages.4 The end result of all of this is reflected in the author’s hospital, a university center in Toronto, Canada, in which 6% of all deliveries were twins in 2001– 2002, compared to 2.3% just 4 years ago. There is evidence that this population of predominately IVF/assisted reproductive technology (ART) conceptions have an increased number of complications during pregnancy than the spontaneously conceived population.5 Therefore, a thorough understanding of the twin at more advanced gestation, especially delivery, is coming under intense scrutiny and demands detailed attention. The Term Breech Trial demonstrated fetal benefit for the delivery of the term singleton breech presentation by caesarian section (CS) without a major increase in maternal risk.6 These findings are being extrapolated to many other at-risk situations, especially twins, in which the second twin is often non-vertex. In Chapter 7 of Ballie`re Tindall’s Operative delivery (edited by Baskett and Arulkumaran and published in 2001), we presented the ‘how to of twin delivery’. This current chapter focuses on the specific risks of the term twin and the data that is currently guiding both research and some changes in practice in this area.
WHAT IS THE RISK OF ADVERSE PERINATAL OUTCOME IN TWINS AT 32 WEEKS OR MORE? The risk of death for twins has decreased over time in Canada but still continues to be high.7 Table 1 shows not only that the risk is appreciable but that over the past 10 years this risk has not decreased at the same rate in more mature twins (34 – 37 weeks) as in aged 32 – 34 weeks. This implies that whereas our neonatal colleagues are improving the results of more premature infants, the obstetricians are not making similar inroads in fetal loss. The next piece of sobering information for obstetricians is that there is a higher risk of death among twin fetuses that are . 2500 g at birth than among singletons of the same birth weight. Kielly reviewed the data on 16 831 multiple births from the New York City Department of Health’s computerized vital records for the period 1978 – 1984.8 The neonatal mortality rate for twins versus singletons at 2501 – 3000 and $ 3001 g was 4.3/1000 versus 3.8/1000 (RR ¼ 1.12) and 7.4/1000 versus 2.2/1000 (RR ¼ 3.32), respectively. Furthermore, even more cause for alarm is the suggestion that the intrapartum fetal death rate for twins is higher than singletons. In the same study, Kielly reported that the intrapartum death rate for twins at $ 2501 g was 1.22/1000 versus Table 1. Gestational age-specific risk of stillbirth and infant death among twin births in Canada. 1985–1987 Gestational age (weeks) 32 –33 34 –36 37 –41
1994–1996
Stillbirth rate
Infant mortality rate
Stillbirth rate
Infant mortality rate
2.1/1000 4.3/1000 7.7/1000
30.5/1000 13.1/1000 7.5/1000
2.4/1000 2.7/1000 4.5/1000
20.8/1000 9.9/1000 4.8/1000
Rates of stillbirth are expressed as per 1000 fetuses at risk; infant death rates are expressed as per 1000 live births.
Delivery of the term twin 627
0.34/1000, in singletons(RR ¼ 3.54, 95% CI ¼ 1.82, 6.88). Other studies have confirmed this higher risk of fetal and neonatal death in twins versus singletons if the pregnancy is at or near term or above 2500 g in birth weight.9 – 16 Neonatal seizures, respiratory morbidity and low Apgar scores at 1 and 5 minutes have also been shown to be higher for twins than for the singleton infant at birth weights . 1500 and . 3000 g.14 – 16 There are two possible reasons for the data discussed above. The increased risk of term twin mortality compared to singletons might be a result of: (i) an increase in the risk of stillbirth; or (ii) the increased risk of labour and delivery compared to that of singletons.
THE RISK OF STILLBIRTH IN TERM TWINS There is now overwhelming cohort and epidemiological data demonstrating the increase risk of stillbirth in twins more than 37– 38 weeks gestation compared to that of singletons.17 – 22 In the absence of a randomized controlled trial (RCT) but extrapolating from the RCT addressing the management of post-term singletons, many authorities, including the International Society for Twin Studies and the SOGC, now recommend delivery before the end of the 38th week of gestation.23 Failing that, very close fetal surveillance, such as twice weekly BPP, should be undertaken. A word of caution: One should not extrapolate this data to delivery by low section caesarean section (LSCS) too early. There is evidence that the risk of neonatal respiratory problems increases if elective CS is undertaken before 38 weeks.24 Thus the ideal time to schedule an elective CS is at 38 weeks gestation. If there is uncertainty about the gestational age, consideration can be given to confirming fetal maturity by checking the lecitin/sphingomyelin (L/S) ratio in the amniotic fluid or by managing the pregnancy expectantly using serial fetal monitoring (twice-weekly nonstress and/or biophysical profile tests) until one is confident that the fetuses are mature.
WHAT IS THE EVIDENCE THAT A POLICY OF PLANNED CS MIGHT BE BENEFICIAL FOR TWINS AT OR NEAR TERM? There are three ways of addressing this from the literature. The first is to compare the outcome for the second twin versus first twin and then to compare these outcomes in those twins delivered vaginally with those delivered by LSCS. In a recent study of 1305 twin pairs delivered between 1988 and 1999 in Nova Scotia, in which second-born twins were compared to the first-born twins at $ 1500 g birth weight, the risk of adverse perinatal outcome (intrapartum fetal death, neonatal death, moderate –severe respiratory distress syndrome, asphyxia, trauma and complications of prematurity) was significantly increased (RR ¼ 2.1, 95% CI ¼ 1.4, 3.1) for second-born twins.15 There is also evidence that the second twin is at greater risk of adverse perinatal outcome than the first twin if delivery is vaginal; the same has not been shown if delivery is by CS. Arnold and colleagues undertook a matched case-control study of preterm twin pairs.14 The risk of respiratory distress syndrome was increased for the second twin compared to the first if delivery was vaginal (OR ¼ 14.2, 95% CI ¼ 2.5, 81.1) but not if delivery was by CS (OR ¼ 0.90, 95% CI ¼ 0, 17.8). The second method is to compare outcomes for twins delivered vaginally versus those delivered by CS. These data also show higher rates of adverse perinatal outcome
628 J. F. R. Barrett
for the twin at or near term if delivery is vaginal rather than by CS. In the Kiely review8, for twins in vertex presentation weighing . 3000 g at birth, the neonatal mortality rate was 12.3/1000 versus 2.9/1000 (RR ¼ 4.22) if delivery was vaginal rather than by CS. These data are possibly strongly affected by selection bias and therefore the third method is to compare outcomes for twins delivered by planned vaginal birth (VB; actual VB plus emergency CS) versus planned CS. There has been only one RCTof planned CS versus planned VB for twins, in which 60 pairs of twins were enrolled.25 There were no perinatal deaths or cases of serious neonatal morbidity in either group. The sample size was too small to answer the question of the better approach to delivery. A Cochrane Review incorporating this one trial has recommended that a larger RCT be undertaken.26 Because of the limited information from RCTs, we undertook a systematic review of studies that compared the policies of planned VB and planned CS for the delivery of twins weighing at least 1500 g or reaching at least 32 weeks gestation.27 The metaanalysis did not find significant differences between the two approaches to delivery in terms of mortality or neonatal morbidity, although low Apgar score at 5 minutes was reduced with a policy of CS. This finding, however, was confined to the twins in which twin A presented as a breech. Since this analysis was undertaken, a further cohort study of 2890 pairs of twins $ 36 weeks found that there were no deaths of either twin in those delivered by planned CS ðn ¼ 454Þ but that in those undergoing planned VB ðn ¼ 2436Þ there were no deaths of the first twin but nine second-twin deaths.28
WHERE DO WE GO FROM HERE? It seems that many of physicians are in equipoise on this most fundamental of all aspects of twin research: What is the best way to deliver twins? In 2001, Hutton undertook a survey of Canadian practitioners to determine their views toward different delivery options for twins.29 Most respondents indicated that for twins at 32 or more weeks gestation in which twin A was vertex, they would usually recommend a planned VB, with the recommendation of planned VB being as high as 100% for the vertex/vertex combination at term to as low as 78% for the vertex/ footling breech combination at 32 – 36 weeks. However, respondents to the survey were not convinced that planned VB was the best approach to delivery, as 64% indicated they would be willing to enroll their patients with twin pregnancies in a well-designed RCT comparing planned VB with planned CS. The interest in a large twin-delivery trial was greatest for twins at term (55%) and for twins presenting vertex/ non-vertex (58%). However, 48% were willing to enroll women with twins at 32 –36 weeks gestation and 42% were willing to enroll twins presenting vertex/vertex. The high number of physicians willing to enroll vertex/vertex twins (22) probably reflects the fact that: (i) up to 20% of vertex second twins will change presentation spontaneously after twin A is delivered30; (ii) Kiely found better outcomes for the twin in vertex presentation if delivery was by cesarean versus vaginal8; (iii) it is the view of practitioners experienced in the management of labour and the delivery of twins that a substantial number of those presenting vertex/vertex will present with serious acute intrapartum problems following the delivery of twin A (e.g. conversion to transverse lie, cord prolapse, prolonged interval delivery of twin B), which might lead to emergency CS, perinatal death and neonatal morbidity; and (iv) if there are benefits to avoiding labour, both twins—regardless of presentation—should benefit.
Delivery of the term twin 629
Many policies in obstetrics have been accepted as the standard of care without adequate evidence to support them. Once a policy of clinical management has been accepted and implemented into practice, it is very difficult to undertake research that is designed to determine the effectiveness of the practice. Although the Term Breech Trial emphasized the relative safety of a policy of planned CS for the mother6, the recently updated Cochrane review has found a higher risk of serious maternal morbidity following a policy of planned CS if the fetus is a singleton breech; the longer-term impact of a policy of planned CS for the mother is not known.31 The focus among practitioners has moved away from keeping CS rates low and towards supporting maternal choice for method of delivery. When the Term Breech Trial was conducted, practice had already shifted towards planned CS. Recruitment to this study was therefore confined to a minority of practitioners who had maintained their skills and confidence in vaginal breech delivery. We believed that a large RCT of planned CS for twins should be conducted before practice changed. Thankfully, the Canadian Institutes of Health Research agreed and in Spring 2003 the Twin Birth Study was funded. This international multicentered RCTwill unroll and follow 2400 twins and their mothers, who have been randomized to deliver vaginally or by LSCS between 32 and 38 weeks gestation. At the time that this article comes to print, we will hopefully be well underway trying to answer another of the great enigmas of obstetrics.
Practice points † twin delivery is an increasing dilemma because of the large numbers of twins resulting from the assisted reproductive technologies † the term twin is subject to greater risks of antepartum stillbirth and intrapartum mortality and morbidity than the term singleton † retrospective studies have shown an advantage of delivery by caesarean section for the term twin † meta-analysis of the studies comparing planned vaginal birth to planned caesarean section do not show an advantage to delivery by caesarean section, but the number of studies are small and underpowered. A large RCT—the Twin Birth Study—is underway to solve this dilemma
Research agenda † for as many centers as possible to enroll in the ongoing RCT † to investigate the reasons behind the increased incidence of term twin stillbirth † to evaluate the effect of chorionicity on the delivery of twins
REFERENCES 1. Barrett J & Bocking A. The SOGC consensus statement: management of twin pregnancies Part 2. SOGC 2000; 22(7): 623. 2. Wimalasundera RC, Trew G & Fisk NM. Reducing the incidence of twins and triplets. Best Practice & Research. Clinical Obstetrics & Gynaecology 2003; 17(2): 309 –329.
630 J. F. R. Barrett 3. Evans MI, Berkowitz RL, Wapner RJ, et al. Improvement in outcomes of multifetal pregnancy reduction with increased experience. American Journal of Obstetrics and Gynecology 2001; 184(2): 97– 103. 4. Martin JA & Park M. Trends in twin and triplet births: 1980–97. CDC National Vital Statistics Report 1999; 47(24): 99–1120. 5. Zuppa AA, Maragliano G, Scapillati ME, et al. Neonatal outcome of spontaneous and assisted twin pregnancies. European Journal of Obstetrics, Gynecology, and Reproductive Biology 2001; 95(1): 68– 72. 6. Hannah ME, Hannah WJ, Hewson S, et al. for the Term Breech Trial Collaborative Group. Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial. Lancet 2000; 356: 1375–1383. 7. Joseph KS, Allen AC, Dodds L, et al. Causes and consequences of recent increases in preterm birth among twins. Obstetrics and Gynecology 2001; 98: 57–64. 8. Kiely JL. The epidemiology of perinatal mortality in multiple births. Bulletin of the New York Academy of Medicine 1990; 66: 618–637. 9. Joseph KS, Marcoux S, Ohlsson A, et al. Changes in stillbirth and infant mortality associated with increases in preterm birth among twins. Pediatrics 2001; 108: 1055–1061. 10. Cheung YB, Yip P & Karlberg J. Mortality of twins and singletons by gestational age: a varying-coefficient approach. American Journal of Epidemiology 2000; 152: 1107–1116. 11. Lie RT. Invited commentary: intersecting perinatal mortality curves by gestational age—are appearances deceiving? American Journal of Epidemiology 2000; 152: 1117– 1119. 12. Ghai V & Vidyasagar D. Morbidity and mortality factors in twins, an epidemiologic approach. Clinics in Perinatology 1988; 15: 123–140. 13. Fabre E, de Agu¨ero R, de Augustin JL, et al. Perinatal mortality in twin pregnancy: an analysis of birth weight-specific mortality rates and adjusted mortality rates for birth weight distributions. Journal of Perinatal Medicine 1988; 16: 85–91. 14. Arnold C, McLean F, Kramer M & Usher R. Respiratory distress syndrome in second-born versus firstborn twins. A matched case-control analysis. The New England Journal of Medicine 1987; 317: 1121–1125. 15. Persad V, Young D, Armson A, et al. Determinants of perinatal morbidity and death among the second of twins. American Journal of Obstetrics and Gynecology 2001; 184: S188 (Abstract 0647). 16. Joseph KS, Marcoux S, Ohlsson A, et al. Changes in stillbirth and infant mortality associated with increases in preterm birth among twins. Pediatrics 2001; 108: 1055–1061. 17. Cheung YB, Yip P & Karlberg J. Mortality of twins and singletons by gestational age: a varying-coefficient approach. American Journal of Epidemiology 2000; 152: 1107–1116. 18. Paterson-Brown S & Fisk N. Caesarean section: every woman’s right to choose? Current Opinion in Obstetrics & Gynecology 1997; 9: 351–355. 19. Hartley RS, Emanuel I & Hitti J. Perinatal mortality and neonatal morbidity rates among twin pairs at different gestational ages: optimal delivery timing at 37 to 38 weeks’ gestation. American Journal of Obstetrics and Gynecology 2001; 184: 451 –458. 20. Puissant F & Leroy F. A reappraisal of perinatal mortality factors in twins. Acta Geneticae Medicae et Gemellologiae 1982; 31: 213–219. 21. Imaizumi Y. Perinatal mortality in twins and factors influencing mortality in Japan, 1980-98. Paediatric and Perinatal Epidemiology 2001; 15: 298– 305. 22. Minakami H & Sato I. Reestimating date of delivery in multifetal pregnancies. JAMA 1996; 275: 1432. 23. Barrett J & Bocking A. The SOGC consensus statement on management of twin pregnancies (part I). JSOGC 2000; 22(7): 519–529. 24. Chasen ST, Madden A & Chervenak FA. Cesarean delivery of twins and neonatal respiratory disorders. American Journal of Obstetrics and Gynecology 1999; 181: 1052–1056. 25. Rabinovici J, Barkai G, Reichman B, et al. Randomized management of the second non-vertex twin: vaginal delivery or cesarean section. American Journal of Obstetrics and Gynecology 1987; 156: 52–56. 26. Crowther CA. Caesarean delivery for the second twin (Cochrane Review). In: The Cochrane Library, Issue 4, 2001. Oxford: Update Software. 27. Hogle K, Hutton E, McBrien KA, et al. Cesarean delivery for twins: a systematic review and meta-analysis. American Journal of Obstetrics and Gynecology 2002;. 28. Smith GCS, Pell JP & Dobbie R. Birth order, gestational age, and risk of delivery related perinatal death in twins: retrospective cohort study. BMJ 2002; 325(7371): 1004. 29. Hutton E, Barrett J & Hannah M. Survey of Canadian practitioners regarding management of breech and twin pregnancies (submitted for publication). 30. Houlihan C & Knuppel RA. Intrapartum management of multiple gestations. Complicated labor and delivery II. Clinics in Perinatology 1996; 23: 91 –116. 31. Hofmeyr GJ & Hannah ME. Planned caesarean section for term breech delivery. (Cochrane Review) In: The Cochrane Library, Issue 4, 2001. Oxford: Update Software.
8 Delivery of the term twin J. F. R. Barrett*
MBBCH, MD, FRCOG FRCSC
Specialist in Maternal Fetal Medicine and Director of Multiple Births, Sunnybrook and Women’s College Hospital; Associate Professor, University of Toronto Sunnybrook and Women’s College Hospital, University of Toronto, 60 Grosvenor Street, Toronto, Ont., Canada M5S 1B6
The ever-increasing incidence of twin pregnancies world wide, together with the increasing trend to caesarean delivery, has resulted in intense scrutiny of the most appropriate method of twin delivery. The term twin has an increased risk of twin mortality compared to term singletons and this might be a result of the increase risk of labour and delivery compared to that of singletons. There are three ways to address this from the literature. The first is to compare outcome for the second twin versus the first twin, and to compare these outcomes in those twins delivered vaginally compared to those delivered by lower section caesarean section (LSCS). The second is to compare outcomes for twins delivered vaginally and for those delivered by caesarean section (CS). These data show higher rates of adverse perinatal outcome for the twin at or near term if delivery is vaginal versus CS. The third method is to compare outcomes for twins delivered by planned vaginal birth (VB; actual VB plus emergency CS) versus planned CS. This chapter will review this data thus outline an ongoing randomized controlled trial—the Twin Birth Study. Key words: twin delivery; term twin; vaginal delivery; caesarean section; randomized controlled trial.
The incidence of twins worldwide continues to increase.1 Attempts at reducing the incidence of higher-order multiples, such as triplets or more, have met with some success in countries that have legislated against multiple embryo reimplantation during in vitro fertilization (IVF) cycles. However, even in these tightly controlled cycles twin pregnancies occur at a rate 10-fold that of normal cycle conception.2 In addition, and unfortunately, many countries (including the author’s) still do not have effective control over, or even accurate information of, practices within fertility centers, and significant numbers of higher-order multiples are still conceived. Although not without some controversy and ethical considerations, these higher-order multiples are often reduced. However, the final number reduced to is usually two3, thus further increasing the population of twins that will ultimately need delivering. Finally, in Western countries * Tel.: þ 1-416-323-7551; Fax: þ1-416-323-6319. E-mail address: [email protected] (J.F.R. Barrett). 1521-6934/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.
626 J. F. R. Barrett
the number of spontaneously conceived twins is increasing consequent on the population choosing to conceive at later maternal ages.4 The end result of all of this is reflected in the author’s hospital, a university center in Toronto, Canada, in which 6% of all deliveries were twins in 2001– 2002, compared to 2.3% just 4 years ago. There is evidence that this population of predominately IVF/assisted reproductive technology (ART) conceptions have an increased number of complications during pregnancy than the spontaneously conceived population.5 Therefore, a thorough understanding of the twin at more advanced gestation, especially delivery, is coming under intense scrutiny and demands detailed attention. The Term Breech Trial demonstrated fetal benefit for the delivery of the term singleton breech presentation by caesarian section (CS) without a major increase in maternal risk.6 These findings are being extrapolated to many other at-risk situations, especially twins, in which the second twin is often non-vertex. In Chapter 7 of Ballie`re Tindall’s Operative delivery (edited by Baskett and Arulkumaran and published in 2001), we presented the ‘how to of twin delivery’. This current chapter focuses on the specific risks of the term twin and the data that is currently guiding both research and some changes in practice in this area.
WHAT IS THE RISK OF ADVERSE PERINATAL OUTCOME IN TWINS AT 32 WEEKS OR MORE? The risk of death for twins has decreased over time in Canada but still continues to be high.7 Table 1 shows not only that the risk is appreciable but that over the past 10 years this risk has not decreased at the same rate in more mature twins (34 – 37 weeks) as in aged 32 – 34 weeks. This implies that whereas our neonatal colleagues are improving the results of more premature infants, the obstetricians are not making similar inroads in fetal loss. The next piece of sobering information for obstetricians is that there is a higher risk of death among twin fetuses that are . 2500 g at birth than among singletons of the same birth weight. Kielly reviewed the data on 16 831 multiple births from the New York City Department of Health’s computerized vital records for the period 1978 – 1984.8 The neonatal mortality rate for twins versus singletons at 2501 – 3000 and $ 3001 g was 4.3/1000 versus 3.8/1000 (RR ¼ 1.12) and 7.4/1000 versus 2.2/1000 (RR ¼ 3.32), respectively. Furthermore, even more cause for alarm is the suggestion that the intrapartum fetal death rate for twins is higher than singletons. In the same study, Kielly reported that the intrapartum death rate for twins at $ 2501 g was 1.22/1000 versus Table 1. Gestational age-specific risk of stillbirth and infant death among twin births in Canada. 1985–1987 Gestational age (weeks) 32 –33 34 –36 37 –41
1994–1996
Stillbirth rate
Infant mortality rate
Stillbirth rate
Infant mortality rate
2.1/1000 4.3/1000 7.7/1000
30.5/1000 13.1/1000 7.5/1000
2.4/1000 2.7/1000 4.5/1000
20.8/1000 9.9/1000 4.8/1000
Rates of stillbirth are expressed as per 1000 fetuses at risk; infant death rates are expressed as per 1000 live births.
Delivery of the term twin 627
0.34/1000, in singletons(RR ¼ 3.54, 95% CI ¼ 1.82, 6.88). Other studies have confirmed this higher risk of fetal and neonatal death in twins versus singletons if the pregnancy is at or near term or above 2500 g in birth weight.9 – 16 Neonatal seizures, respiratory morbidity and low Apgar scores at 1 and 5 minutes have also been shown to be higher for twins than for the singleton infant at birth weights . 1500 and . 3000 g.14 – 16 There are two possible reasons for the data discussed above. The increased risk of term twin mortality compared to singletons might be a result of: (i) an increase in the risk of stillbirth; or (ii) the increased risk of labour and delivery compared to that of singletons.
THE RISK OF STILLBIRTH IN TERM TWINS There is now overwhelming cohort and epidemiological data demonstrating the increase risk of stillbirth in twins more than 37– 38 weeks gestation compared to that of singletons.17 – 22 In the absence of a randomized controlled trial (RCT) but extrapolating from the RCT addressing the management of post-term singletons, many authorities, including the International Society for Twin Studies and the SOGC, now recommend delivery before the end of the 38th week of gestation.23 Failing that, very close fetal surveillance, such as twice weekly BPP, should be undertaken. A word of caution: One should not extrapolate this data to delivery by low section caesarean section (LSCS) too early. There is evidence that the risk of neonatal respiratory problems increases if elective CS is undertaken before 38 weeks.24 Thus the ideal time to schedule an elective CS is at 38 weeks gestation. If there is uncertainty about the gestational age, consideration can be given to confirming fetal maturity by checking the lecitin/sphingomyelin (L/S) ratio in the amniotic fluid or by managing the pregnancy expectantly using serial fetal monitoring (twice-weekly nonstress and/or biophysical profile tests) until one is confident that the fetuses are mature.
WHAT IS THE EVIDENCE THAT A POLICY OF PLANNED CS MIGHT BE BENEFICIAL FOR TWINS AT OR NEAR TERM? There are three ways of addressing this from the literature. The first is to compare the outcome for the second twin versus first twin and then to compare these outcomes in those twins delivered vaginally with those delivered by LSCS. In a recent study of 1305 twin pairs delivered between 1988 and 1999 in Nova Scotia, in which second-born twins were compared to the first-born twins at $ 1500 g birth weight, the risk of adverse perinatal outcome (intrapartum fetal death, neonatal death, moderate –severe respiratory distress syndrome, asphyxia, trauma and complications of prematurity) was significantly increased (RR ¼ 2.1, 95% CI ¼ 1.4, 3.1) for second-born twins.15 There is also evidence that the second twin is at greater risk of adverse perinatal outcome than the first twin if delivery is vaginal; the same has not been shown if delivery is by CS. Arnold and colleagues undertook a matched case-control study of preterm twin pairs.14 The risk of respiratory distress syndrome was increased for the second twin compared to the first if delivery was vaginal (OR ¼ 14.2, 95% CI ¼ 2.5, 81.1) but not if delivery was by CS (OR ¼ 0.90, 95% CI ¼ 0, 17.8). The second method is to compare outcomes for twins delivered vaginally versus those delivered by CS. These data also show higher rates of adverse perinatal outcome
628 J. F. R. Barrett
for the twin at or near term if delivery is vaginal rather than by CS. In the Kiely review8, for twins in vertex presentation weighing . 3000 g at birth, the neonatal mortality rate was 12.3/1000 versus 2.9/1000 (RR ¼ 4.22) if delivery was vaginal rather than by CS. These data are possibly strongly affected by selection bias and therefore the third method is to compare outcomes for twins delivered by planned vaginal birth (VB; actual VB plus emergency CS) versus planned CS. There has been only one RCTof planned CS versus planned VB for twins, in which 60 pairs of twins were enrolled.25 There were no perinatal deaths or cases of serious neonatal morbidity in either group. The sample size was too small to answer the question of the better approach to delivery. A Cochrane Review incorporating this one trial has recommended that a larger RCT be undertaken.26 Because of the limited information from RCTs, we undertook a systematic review of studies that compared the policies of planned VB and planned CS for the delivery of twins weighing at least 1500 g or reaching at least 32 weeks gestation.27 The metaanalysis did not find significant differences between the two approaches to delivery in terms of mortality or neonatal morbidity, although low Apgar score at 5 minutes was reduced with a policy of CS. This finding, however, was confined to the twins in which twin A presented as a breech. Since this analysis was undertaken, a further cohort study of 2890 pairs of twins $ 36 weeks found that there were no deaths of either twin in those delivered by planned CS ðn ¼ 454Þ but that in those undergoing planned VB ðn ¼ 2436Þ there were no deaths of the first twin but nine second-twin deaths.28
WHERE DO WE GO FROM HERE? It seems that many of physicians are in equipoise on this most fundamental of all aspects of twin research: What is the best way to deliver twins? In 2001, Hutton undertook a survey of Canadian practitioners to determine their views toward different delivery options for twins.29 Most respondents indicated that for twins at 32 or more weeks gestation in which twin A was vertex, they would usually recommend a planned VB, with the recommendation of planned VB being as high as 100% for the vertex/vertex combination at term to as low as 78% for the vertex/ footling breech combination at 32 – 36 weeks. However, respondents to the survey were not convinced that planned VB was the best approach to delivery, as 64% indicated they would be willing to enroll their patients with twin pregnancies in a well-designed RCT comparing planned VB with planned CS. The interest in a large twin-delivery trial was greatest for twins at term (55%) and for twins presenting vertex/ non-vertex (58%). However, 48% were willing to enroll women with twins at 32 –36 weeks gestation and 42% were willing to enroll twins presenting vertex/vertex. The high number of physicians willing to enroll vertex/vertex twins (22) probably reflects the fact that: (i) up to 20% of vertex second twins will change presentation spontaneously after twin A is delivered30; (ii) Kiely found better outcomes for the twin in vertex presentation if delivery was by cesarean versus vaginal8; (iii) it is the view of practitioners experienced in the management of labour and the delivery of twins that a substantial number of those presenting vertex/vertex will present with serious acute intrapartum problems following the delivery of twin A (e.g. conversion to transverse lie, cord prolapse, prolonged interval delivery of twin B), which might lead to emergency CS, perinatal death and neonatal morbidity; and (iv) if there are benefits to avoiding labour, both twins—regardless of presentation—should benefit.
Delivery of the term twin 629
Many policies in obstetrics have been accepted as the standard of care without adequate evidence to support them. Once a policy of clinical management has been accepted and implemented into practice, it is very difficult to undertake research that is designed to determine the effectiveness of the practice. Although the Term Breech Trial emphasized the relative safety of a policy of planned CS for the mother6, the recently updated Cochrane review has found a higher risk of serious maternal morbidity following a policy of planned CS if the fetus is a singleton breech; the longer-term impact of a policy of planned CS for the mother is not known.31 The focus among practitioners has moved away from keeping CS rates low and towards supporting maternal choice for method of delivery. When the Term Breech Trial was conducted, practice had already shifted towards planned CS. Recruitment to this study was therefore confined to a minority of practitioners who had maintained their skills and confidence in vaginal breech delivery. We believed that a large RCT of planned CS for twins should be conducted before practice changed. Thankfully, the Canadian Institutes of Health Research agreed and in Spring 2003 the Twin Birth Study was funded. This international multicentered RCTwill unroll and follow 2400 twins and their mothers, who have been randomized to deliver vaginally or by LSCS between 32 and 38 weeks gestation. At the time that this article comes to print, we will hopefully be well underway trying to answer another of the great enigmas of obstetrics.
Practice points † twin delivery is an increasing dilemma because of the large numbers of twins resulting from the assisted reproductive technologies † the term twin is subject to greater risks of antepartum stillbirth and intrapartum mortality and morbidity than the term singleton † retrospective studies have shown an advantage of delivery by caesarean section for the term twin † meta-analysis of the studies comparing planned vaginal birth to planned caesarean section do not show an advantage to delivery by caesarean section, but the number of studies are small and underpowered. A large RCT—the Twin Birth Study—is underway to solve this dilemma
Research agenda † for as many centers as possible to enroll in the ongoing RCT † to investigate the reasons behind the increased incidence of term twin stillbirth † to evaluate the effect of chorionicity on the delivery of twins
REFERENCES 1. Barrett J & Bocking A. The SOGC consensus statement: management of twin pregnancies Part 2. SOGC 2000; 22(7): 623. 2. Wimalasundera RC, Trew G & Fisk NM. Reducing the incidence of twins and triplets. Best Practice & Research. Clinical Obstetrics & Gynaecology 2003; 17(2): 309 –329.
630 J. F. R. Barrett 3. Evans MI, Berkowitz RL, Wapner RJ, et al. Improvement in outcomes of multifetal pregnancy reduction with increased experience. American Journal of Obstetrics and Gynecology 2001; 184(2): 97– 103. 4. Martin JA & Park M. Trends in twin and triplet births: 1980–97. CDC National Vital Statistics Report 1999; 47(24): 99–1120. 5. Zuppa AA, Maragliano G, Scapillati ME, et al. Neonatal outcome of spontaneous and assisted twin pregnancies. European Journal of Obstetrics, Gynecology, and Reproductive Biology 2001; 95(1): 68– 72. 6. Hannah ME, Hannah WJ, Hewson S, et al. for the Term Breech Trial Collaborative Group. Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial. Lancet 2000; 356: 1375–1383. 7. Joseph KS, Allen AC, Dodds L, et al. Causes and consequences of recent increases in preterm birth among twins. Obstetrics and Gynecology 2001; 98: 57–64. 8. Kiely JL. The epidemiology of perinatal mortality in multiple births. Bulletin of the New York Academy of Medicine 1990; 66: 618–637. 9. Joseph KS, Marcoux S, Ohlsson A, et al. Changes in stillbirth and infant mortality associated with increases in preterm birth among twins. Pediatrics 2001; 108: 1055–1061. 10. Cheung YB, Yip P & Karlberg J. Mortality of twins and singletons by gestational age: a varying-coefficient approach. American Journal of Epidemiology 2000; 152: 1107–1116. 11. Lie RT. Invited commentary: intersecting perinatal mortality curves by gestational age—are appearances deceiving? American Journal of Epidemiology 2000; 152: 1117– 1119. 12. Ghai V & Vidyasagar D. Morbidity and mortality factors in twins, an epidemiologic approach. Clinics in Perinatology 1988; 15: 123–140. 13. Fabre E, de Agu¨ero R, de Augustin JL, et al. Perinatal mortality in twin pregnancy: an analysis of birth weight-specific mortality rates and adjusted mortality rates for birth weight distributions. Journal of Perinatal Medicine 1988; 16: 85–91. 14. Arnold C, McLean F, Kramer M & Usher R. Respiratory distress syndrome in second-born versus firstborn twins. A matched case-control analysis. The New England Journal of Medicine 1987; 317: 1121–1125. 15. Persad V, Young D, Armson A, et al. Determinants of perinatal morbidity and death among the second of twins. American Journal of Obstetrics and Gynecology 2001; 184: S188 (Abstract 0647). 16. Joseph KS, Marcoux S, Ohlsson A, et al. Changes in stillbirth and infant mortality associated with increases in preterm birth among twins. Pediatrics 2001; 108: 1055–1061. 17. Cheung YB, Yip P & Karlberg J. Mortality of twins and singletons by gestational age: a varying-coefficient approach. American Journal of Epidemiology 2000; 152: 1107–1116. 18. Paterson-Brown S & Fisk N. Caesarean section: every woman’s right to choose? Current Opinion in Obstetrics & Gynecology 1997; 9: 351–355. 19. Hartley RS, Emanuel I & Hitti J. Perinatal mortality and neonatal morbidity rates among twin pairs at different gestational ages: optimal delivery timing at 37 to 38 weeks’ gestation. American Journal of Obstetrics and Gynecology 2001; 184: 451 –458. 20. Puissant F & Leroy F. A reappraisal of perinatal mortality factors in twins. Acta Geneticae Medicae et Gemellologiae 1982; 31: 213–219. 21. Imaizumi Y. Perinatal mortality in twins and factors influencing mortality in Japan, 1980-98. Paediatric and Perinatal Epidemiology 2001; 15: 298– 305. 22. Minakami H & Sato I. Reestimating date of delivery in multifetal pregnancies. JAMA 1996; 275: 1432. 23. Barrett J & Bocking A. The SOGC consensus statement on management of twin pregnancies (part I). JSOGC 2000; 22(7): 519–529. 24. Chasen ST, Madden A & Chervenak FA. Cesarean delivery of twins and neonatal respiratory disorders. American Journal of Obstetrics and Gynecology 1999; 181: 1052–1056. 25. Rabinovici J, Barkai G, Reichman B, et al. Randomized management of the second non-vertex twin: vaginal delivery or cesarean section. American Journal of Obstetrics and Gynecology 1987; 156: 52–56. 26. Crowther CA. Caesarean delivery for the second twin (Cochrane Review). In: The Cochrane Library, Issue 4, 2001. Oxford: Update Software. 27. Hogle K, Hutton E, McBrien KA, et al. Cesarean delivery for twins: a systematic review and meta-analysis. American Journal of Obstetrics and Gynecology 2002;. 28. Smith GCS, Pell JP & Dobbie R. Birth order, gestational age, and risk of delivery related perinatal death in twins: retrospective cohort study. BMJ 2002; 325(7371): 1004. 29. Hutton E, Barrett J & Hannah M. Survey of Canadian practitioners regarding management of breech and twin pregnancies (submitted for publication). 30. Houlihan C & Knuppel RA. Intrapartum management of multiple gestations. Complicated labor and delivery II. Clinics in Perinatology 1996; 23: 91 –116. 31. Hofmeyr GJ & Hannah ME. Planned caesarean section for term breech delivery. (Cochrane Review) In: The Cochrane Library, Issue 4, 2001. Oxford: Update Software.