220 Twin gestations comprise approximately 1% of all preg- nancies but account for nearly 10% of perinatal mortal- ity. 1,2 Low birth weight and prematurity are the main causes of high perinatal morbidity and mortality in twins, 1,3-6 whereas malpresentation and the hazards of de- livery are next in order of concern. 1,4 Malpresentation of one or both babies occurs in about 60% of all twin preg- nancies. 1,3,7-21 An educational bulletin released by the American Col- lege of Obstetricians and Gynecologists (ACOG) in 1999 states that vaginal birth is anticipated for vertex/vertex twin gestations unless there are specific contraindications to vaginal birth. 22 Most authors agree that vaginal birth is appropriate for vertex/vertex twins born after 33 weeks or weighing at least 1500 to 2000 g, 2,7-9,12,16,19-34 but there is less agreement about the appropriate mode of delivery for those weighing less than 1500 g or born before 33 weeks of gestation. When twin A is breech, the ACOG educational bulletin recommends cesarean section because the safety of vagi- nal birth has not been documented and because the pos- sibility of locked twins exists when twin A is nonvertex and twin B is vertex. 22 Although locked twins is a rare occurrence, it is associated with a high rate of fetal mor- tality. A recently completed multicenter randomized, controlled trial, the Term Breech Trial, showed that a pol- icy of planned cesarean section is substantially better for the singleton fetus in the breech presentation at term than a policy of planned vaginal birth. 35 It has therefore been suggested by some that a policy of planned cesarean section might also benefit breech first twins. However, opinion in the literature has been divided. Some authors have argued that cesarean section is the management of choice,* whereas others believe that cesarean section may not benefit twins weighing more than 1500 g. 15,28,33,38-42 The best approach to the delivery of vertex/nonvertex twins is very controversial. A 1999 ACOG educational bul- letin admits that the data are conflicting. It states that From the Maternal Infant and Reproductive Health Research Unit at the Centre for Research in Women’s Health, a and the Department of Obstet- rics and Gynecology, b Sunnybrook and Women’s College Health Sciences Centre, University of Toronto. Received for publication December 19, 2001; revised July 12, 2002; ac- cepted August 9, 2002. Reprint requests: Mary Hannah, MDCM, University of Toronto, Mater- nal, Infant and Reproductive Health Research Unit at the Centre for Re- search in Women’s Health, Suite 714, 790 Bay St, Toronto, Ontario M5G 1N8, Canada. E-mail: mary.hannah@utoronto.ca © 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/mob.2003.64 Cesarean delivery for twins: A systematic review and meta-analysis Karen L. Hogle, a Eileen K. Hutton, MNSc, RM, a Kerry A. McBrien, a Jon F.R. Barrett, MD, b and Mary E. Hannah, MDCM a,b Toronto, Ontario, Canada OBJECTIVE: We undertook a systematic review and meta-analysis to determine whether a policy of planned cesarean section or vaginal delivery is better for twins. STUDY DESIGN: We searched MEDLINE and EMBASE from 1980 through May 2001 using combinations of the following terms: twin, delivery, cesarean section, vaginal birth, birth weight, and gestational age. Studies that compared planned cesarean section to planned vaginal birth for babies weighing at least 1500 g or reaching at least 32 weeks’ gestation were included. We computed pooled odds ratios for perinatal or neona- tal mortality, low 5-minute Apgar score, neonatal morbidity, and maternal morbidity.The infant was the unit of statistical analysis. Results were considered statistically significant if the 95% CI did not encompass 1.0. RESULTS: We retrieved 67 articles, 63 of which were excluded. Four studies with a total of 1932 infants were included in the analysis. A low 5-minute Apgar score occurred less frequently in twins delivered by planned cesarean section (odds ratio, 0.47; 95% CI, 0.26-0.88) principally because of a reduction among twins if twin A was in breech position (odds ratio, 0.33; 95% CI, 0.17-0.65).Twins delivered by planned cesarean section spent significantly longer in the hospital (mean difference, 4.01 days; 95% CI, 0.73-7.28 days).There were no significant differences in perinatal or neonatal mortality, neonatal morbidity, or maternal morbidity. CONCLUSION: Planned cesarean section may decrease the risk of a low 5-minute Apgar score, particularly if twin A is breech. Otherwise, there is no evidence to support planned cesarean section for twins. (Am J Ob- stet Gynecol 2003;188:220-7.) Key words: Planned cesarean section, planned vaginal delivery, twins *References 2, 7-9, 12, 20, 21, 24, 26, 29, 30, 34, 36, 37.