- Email: [email protected]
Clinical indicators associated with the mode of twin delivery: an analysis of 22,712 twin pairs
European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Clinical indicators associated with the mode of twin delivery: an analysis of 22,712 twin pairs Simone M.T.A. Goossens a,*, Chantal W.P.M. Hukkelhoven b, Lotte de Vries b, Ben Willem Mol c, Jan G. Nijhuis d, Frans J.M.E. Roumen a a
Department of Obstetrics and Gynecology, Atrium Medical Center Parkstad, Heerlen, The Netherlands Netherlands Perinatal Registry, Utrecht, The Netherlands The Robinson Institute, School of Pediatrics and Reproductive Health, University of Adelaide, Australia d Department of Obstetrics and Gynecology, Maastricht University Medical Center, GROW – School for Oncology and Developmental Biology, The Netherlands b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 August 2015 Received in revised form 13 September 2015 Accepted 17 September 2015
Objective: To identify clinical indicators associated with the planned and actual mode of delivery in women with a twin pregnancy. Study design: We performed a retrospective cohort study in women with a twin pregnancy who delivered at a gestational age of 32 + 0–41 + 0 weeks and days between 2000 and 2008 in the Netherlands. Data were obtained from a nationwide database. We identified maternal, pregnancyrelated, fetal, neonatal and hospital-related indicators that were associated with planned cesarean section (CS) and, for women with planned vaginal delivery (VD), for intrapartum CS. The associations between indicators and mode of delivery were studied with uni- and multivariate logistic regression analyses. Results: We included 22,712 women with a twin pregnancy, of whom 4,310 women (19.0%) had a planned CS. Of the 18,402 women who had a planned VD, 14,034 (76.3%) delivered vaginally, 3,545 (19.3%) had an intrapartum CS, while 823 (4.5%) delivered twin A vaginally and twin B by intrapartum CS. The clinical indicators for a planned CS and an intrapartum CS were comparable: non-cephalic position of both twins (aOR 25.32; 95% CI 22.50–28.50, and aOR 21.94; 95% CI 18.67–25.78, respectively), noncephalic position of twin A only (aOR 21.67 95% CI 19.12–24.34, and aOR 13.71; 95% CI 11.75–16.00, respectively), previous CS (aOR 3.69; 95% CI 3.12–4.36, and aOR 7.00; 95% CI 5.77–8.49, respectively), nulliparity (aOR 1.51; 95% CI 1.32–1.72, and aOR 4.20; 95% CI 3.67–4.81, respectively), maternal age 41 years (aOR 3.00; 95% CI 2.14–4.22, and aOR 2.50; 95% CI 1.75–3.59, respectively), and pre-eclampsia (aOR 2.12; 95% CI 1.83–2.46, and aOR 1.34; 95% CI 1.16–1.56, respectively). Conclusion: Both planned and intrapartum CS in twins had comparable predictors: non-cephalic position of both twins or twin A only, previous CS, nulliparity, advanced maternal age, and preeclampsia. ß 2015 Published by Elsevier Ireland Ltd.
Keywords: Cesarean section Clinical indicator Twin delivery Decision
Introduction To make a balanced shared decision on the mode of delivery in women with a twin pregnancy beyond a gestational age of 32 weeks, information on clinical indicators associated with a cesarean section (CS) is relevant. Presentation of twin A is such a well-known indicator. In case of a cephalic presenting twin A, the
* Corresponding author at: Department of Obstetrics and Gynecology, Atrium Medical Center Parkstad, P.O. 4446, 6401 CX Heerlen, The Netherlands. Tel.: +31 455766513; fax: +31 455766625. E-mail address: [email protected] (Simone M.T.A. Goossens). http://dx.doi.org/10.1016/j.ejogrb.2015.09.034 0301-2115/ß 2015 Published by Elsevier Ireland Ltd.
pregnant woman is counseled that, generally, planned CS has comparable safety as planned vaginal delivery (VD) [1–3]. Although there is no firm evidence to suggest benefit of cesarean over vaginal delivery in case of non-cephalic position of the first twin, she is counseled as is usual for non-cephalic presentation of a singleton, frequently resulting in an elective CS [4,5]. However, many other clinical indicators may be involved, like maternal age, mode of conception, parity, Body Mass Index, diabetes, previous CS, chorionicity, sex concordance, fetal intrauterine growth retardation, fetal weight, and birth weight discordance [2,6–11]. Ideally, an intrapartum CS, i.e. a CS during active labor, should not be necessary. Especially a combined delivery, i.e. twin A
134
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
delivered vaginally and twin B by CS, should be prevented, as it is associated with the highest neonatal and maternal morbidity in twin births [1,12]. For women with a very high risk on a combined delivery, a planned CS may be the optimal mode of delivery [12–14]. In most national guidelines, no clear recommendations on planning the mode of delivery are expressed [15–18]. So, the planned and actual mode of delivery might differ depending on maternal, professional, hospital, fetal and environmental circumstances. The purpose of this study was to identify the most important maternal, pregnancy-related, fetal, neonatal and hospital-related indicators associated with the planned and actual mode of delivery for twin pregnancies in the Netherlands. Materials and methods We used data of our nationwide registry on pregnancy and delivery. In The Netherlands, data on pregnancy, birth and neonates are routinely collected in four separate registries: the LVR1-registry for primary midwife-led care, the LVRh-registry for care led by general practitioners, the LVR2-registry for obstetrician-led care and the LNR-registry for neonatal care. These separate registries are linked into one database, The Netherlands Perinatal Registry (PRN) [19]. Approximately 96% of all births in The Netherlands are entered into this registry. We included all women with a twin pregnancy and a gestational age of 32 + 0–41 + 0 weeks at delivery, as registered in the PRN database from January 1st, 2000 until January 1st, 2008. We excluded pregnancies with one or both twins weighing <500 g at birth or having lethal congenital anomalies, and pregnancies with a fetal demise before start of delivery. Cases with incomplete twin data sets (records with only one neonate registered) were also excluded. Because of these incomplete data sets, the total number of twin babies registered initially could be unequal. In The Netherlands, twins are delivered in a hospital by an obstetrician. As far as we know, none of the participating hospitals excluded vaginal deliveries of twins in all circumstances. In the PRN, a CS may be registered as a primary CS or a secondary CS. A primary CS is a CS in women who did not try vaginal labor. A secondary CS is defined as a CS during labor in women who were planned to try a vaginal delivery. For this study, we assumed that all primary CSs were planned, and therefore we registered them in the planned CS group, whereas women who delivered (1) both children vaginally, (2) by a secondary CS of both children, or (3) by a combined delivery, were presumed to be in the planned VD group. All CSs in the planned VD group were defined as intrapartum CSs. For analysis of both the planned CS group and the planned VD group separately, we studied maternal indicators (maternal age at delivery, parity, previous CS, socio-economic status, ethnicity, maternal diabetes mellitus, pre-existing hypertension), pregnancy-related indicators (mode of conception, preeclampsia, drug use during pregnancy, gestational age at delivery), fetal and neonatal indicators (fetal position, sex concordance, weight concordance, small for gestational age), and hospitalrelated indicators (type of hospital, mean number of twin deliveries per year), as known [1,2,5–11] or expected from a pathophysiologic perspective to be associated with the planned and actual mode of delivery. Socioeconomic status was based on the mean household income level of the neighborhood, determined by the first 4 digits of the woman’s postal code. Diabetes was defined as diabetes mellitus before a gestational age of 20 weeks or gestational diabetes. Pre-existing hypertension was defined as hypertension before a gestational age of 20 weeks. Pre-eclampsia was defined as a diastolic blood pressure 90 mmHg and proteinuria (300 mg/ day). Birth weight of twin B compared to twin A was divided into
subgroups of 80%, 80–125% and 125%. Small for gestational age was defined as a birth weight <10th percentile [20]. Statistical analysis Contingency tables were created to assess frequencies and percentages of the indicators and the outcome measures. All risk indicators mentioned above were analyzed with uni- and multivariate logistic regression analyses, calculating crude and adjusted odds ratio (OR) and 95% confidence intervals (CI). For the adjusted OR we adjusted for all risk factors mentioned in the univariate analyses. The association between mortality and the planned delivery groups was calculated with chi-square test. Reference groups were women 21–30 years old, multiparous (2), without a previous CS, with a normal socio-economic status, Caucasian, non-diabetic, and without pre-existing hypertension; no in vitro fertilization (IVF), pre-eclampsia or drug use, gestational age at delivery between 37 and 38 weeks; cephalic-cephalic position, sex concordant, weight concordant, birth weight of both children >10th percentile; women delivering in a non-teaching hospital, and women delivering in a hospital with an annual number of twin deliveries of 25–49. Analyses were performed using SAS 9.3 (SAS Institute, Cary, NC, USA) and Microsoft Excel 2011 for Mac. Ethics approval was not required under Dutch Law, as only anonymous data were used. A privacy committee checks all researches done in collaboration with the PRN to make sure procedures are as accurate as possible, the privacy of the participants is guaranteed and the data confidentiality is respected. Results The database contained 56,865 records from children born from a twin pregnancy in the study period (Fig. 1). From these, 45,424 twin babies (n = 22,712 women with a twin pair) fulfilled the inclusion criteria. In 4,310 women with a twin pair (19.0%) a planned CS was performed, while a VD was planned for 18,402 women with a twin pair (81.0%). Of these, 14,034 women (76.3%) delivered vaginally, 3,545 women (19.3%) delivered both children by an intrapartum CS, and 823 women (4.5%) by combined delivery. Fig. 2 shows the distribution of the actual mode of delivery after a planned VD over the years. The percentage of actual VD in this group remained relatively stable, with a range between 75.2% and 78.3%, and the percentage of combined deliveries ranged between 3.3% and 4.9%. Table 1 shows maternal, pregnancy related, fetal, neonatal and hospital related indicators of twins according to the planned mode of delivery, including the results of the logistic regression analyses. Most prominent indicators for a delivery by planned CS according to the univariate analysis remained prominent after multivariate analysis: non-cephalic position of both twins (adjusted (a) OR 25.32; 95% confidence interval (CI) 22.50–28.50), non-cephalic position of twin A only (aOR 21.67; 95% CI 19.12–24.34), 1 previous CS (aOR 3.69; 95% CI 3.12–4.36), maternal age 41 years (aOR 3.00; 95% CI 2.14–4.22), and pre-eclampsia (aOR 2.12; 95%; CI 1.83–2.46). Minor indicators associated with delivery by planned CS were maternal age 31–40 years, nulliparity, primiparity, cephalic-non cephalic position, weight twin B 80% compared to twin A, birth weight twin A or twin B <10th percentile, and delivery in a university hospital. Gestational age at delivery 32 + 0– 37 + 0 and 39 + 0–41 + 0 was associated with a planned VD, as was an annual number of 75–99 twin deliveries in the hospital. Table 2 shows indicators of twins planned for a vaginal delivery, according to the actual mode of delivery, including the results of the logistic regression analyses. Most prominent indicators
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
135
Total number of twin babies born from Jan 01, 2000 – Jan 01, 2008 (n=56,865)
Twin babies excluded from analysis* - Gestational age <32 + 0 or ≥ 41 + 0 (n=5,714) - Weight < 500g (n=918) - Congenital abnormalities (n=1,896) - Intrauterine fetal death (n=1,040) - Incomplete data (n=2,461)
Total number of twin babies for analysis (n=45,424) Total number of twin pairs for analysis (n=22,712 )
Planned Cesarean Section Twin pairs n=4,310 (19.0%)
Vaginal Delivery both twins Twin pairs n=14,034 (76.3%)
Planned Vaginal Delivery Twin pairs n=18,402 (81.0%)
Intrapartum Cesarean Section both twins Twin pairs n=3,545 (19.3%)
Vaginal delivery twin A / Intrapartum Cesarean Section twin B Twin pairs n=823 (4.5%)
Fig. 1. Overview of twin babies and women with a twin pair as registered in the PRN database during the eight-year study period, according to the mode of delivery. *Some have more than one exclusion criterion.
Fig. 2. Actual mode of delivery of twin pairs (%) in the planned vaginal delivery group over the years 2000–2007. VD = vaginal delivery; CS = cesarean section.
136
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
Table 1 Maternal, pregnancy related, fetal, neonatal and hospital related indicators of 22,712 twin pairs according to planned cesarean section (CS) or planned vaginal delivery (VD). Total twin pairs, na
Maternal indicators Maternal age at delivery (years) 20 21–30 31–40 41 Parity 0 1 2 Unknown Previous CS 1 Yes Socio-economic status Very high High Normal Low Very low Unknown Ethnicity Caucasian Non Caucasian Unknown Maternal diabetes mellitus Yes Pre-existent hypertension Yes Pregnancy related indicators IVF (IVF/ICSI) Yes Unknown Pre-eclampsia Yes Unknown Drug use during pregnancy Yes Gestational age at delivery (weeks + days) 32 + 0–32 + 6 33 + 0–33 + 6 34 + 0–34 + 6 35 + 0–35 + 6 36 + 0–36 + 6 37 + 0–37 + 6 38 + 0–38 + 6 39 + 0–39 + 6 40 + 0–40 + 6 Unknown Fetal and neonatal indicators Fetal position Cephalic–cephalic Cephalic–non cephalic Non cephalic–cephalic Non cephalic–non cephalic Unknown Sex concordance Concordant Discordant Unknown Weight twin B compared to twin A 80% 80–125% 125% Small for gestational ageb Both twinsp10, twin B p10 Both twins >p10 Unknown Hospital related indicators Type of hospital University Non-university teaching
Planned CS (%)
Planned VD (%)
4,310
18,402
Crude OR (95% CI)
Adjusted OR (95% CI)
44 (1.0) 1,468 (34.1) 2,712 (62.9) 86 (2.00)
207 (1.1) 6,767 (36.8) 11,221 (61.0) 207 (1.1)
0.98 (0.70–1.36) Ref. 1.11 (1.04–1.20) 1.92 (1.48–2.48)
0.79 (0.52–1.18) Ref. 1.26 (1.15–1.39) 3.00 (2.14–4.22)
2,158 (50.1) 1,465 (34.0) 659 (15.3) 28 (0.6)
8,308 (45.1) 6,808 (37.0) 3,157 (17.2) 129 (0.7)
1.24 (1.13–1.37) 1.03 (0.93–1.14) Ref.
1.51 (1.32–1.72) 1.21 (1.07–1.38) Ref.
491 (11.4)
605 (3.3)
3.78 (3.34–4.28)
3.69 (3.12–4.36)
819 (19.0) 830 (19.3) 824 (19.1) 770 (17.9) 999 (23.2) 68 (1.6)
3,662 (19.9) 3,691 (20.1) 3,375 (18.3) 3,392 (18.4) 3,990 (21.7) 292 (1.6)
0.92 0.92 Ref. 0.93 1.03
0.95 0.97 Ref. 0.94 1.03
3,747 (86.9) 548 (12.7) 15 (0.4)
16,290 (88.5) 2,060 (11.2) 52 (0.3)
Ref. 1.16 (1.05–1.28)
Ref. 1.01 (0.88–1.16)
18 (0.4)
50 (0.3)
1.54 (0.90–2.64)
1.41 (0.79–2.50)
17 (0.4)
51 (0.3)
1.43 (0.82–2.47)
0.94 (0.44–1.99)
982 (22.8) 110 (2.6)
3,564 (19.4) 540 (2.9)
1.22 (1.13–1.33)
1.09 (0.98–1.22)
465 (10.8) 127 (2.9)
1,159 (6.3) 548 (3.0)
1.80 (1.61–2.02)
2.12 (1.83–2.46)
3 (0.1)
7 (0.0)
1.84 (0.48–7.09)
3.05 (0.64–14.52)
157 (3.6) 209 (4.8) 302 (7.0) 390 (9.1) 645 (15.0) 1,416 (32.9) 989 (22.9) 162 (3.8) 38 (0.9) 2 (0.0)
618 (3.4) 915 (5.0) 1,423 (7.7) 1,925 (10.5) 2,905 (15.8) 4,703 (25.6) 3,584 (19.5) 1,715 (9.3) 596 (3.2) 18 (0.1)
0.84 0.76 0.71 0.67 0.74 Ref. 0.92 0.31 0.21
0.73 0.64 0.63 0.63 0.70 Ref. 0.94 0.38 0.20
697 (16.2) 618 (14.3) 1,344 (31.2) 1,639 (38.0) 12 (0.3)
11,013 (59.9) 5,247 (28.5) 1,031 (5.6) 1,090 (5.9) 21 (0.1)
Ref. 1.86 (1.66–2.08) 20.60 (18.42–23.03) 23.76 (21.32–26.48)
Ref. 1.85 (1.64–2.98) 21.67 (19.12–24.34) 25.32 (22.50–28.50)
2,626 (60.9) 1,684 (39.1) 0 (0.0)
11,765 (63.9) 6,633 (36.0) 4 (0.0)
Ref. 1.14 (1.06–1.22)
Ref. 0.99 (0.91–1.08)
669 (15.5) 3,340 (77.5) 301 (7.0)
1,842 (10.0) 15,479 (84.1) 1,081 (5.9)
1.68 (1.52–1.85) Ref. 1.29 (1.13–1.47)
1.74 (1.50–2.02) Ref. 1.15 (0.95–1.39)
431 (10.0) 916 (21.2) 525 (12.2) 2,438 (56.6) 0 (0.0)
2,117 (11.5) 3,086 (17.8) 2,073 (11.3) 11,120 (60.4) 6 (0.0)
0.93 (0.83–1.04) 1.35 (1.24–1.48) 1.15 (1.04–1.48) Ref.
1.15 (1.00–1.34) 1.23 (1.07–1.40) 1.20 (1.03–1.39) Ref.
602 (14.0) 1,868 (43.3)
2,271 (12.3) 8,480 (46.1)
1.10 (0.99–1.22) 0.92 (0.85–0.98)
1.38 (1.13–1.69) 0.93 (0.83–1.03)
(0.82–1.02) (0.83–1.03) (0.83–1.04) (0.93–1.14)
(0.70–1.02) (0.65–0.89) (0.61–0.81) (0.59–0.76) (0.67–0.82) (0.84–1.01) (0.26–0.37) (0.15–0.30)
(0.83–1.09) (0.84–1.11) (0.82–1.08) (0.88–1.16)
(0.57–0.93) (0.52–0.78) (0.53–0.76) (0.54–0.73) (0.61–0.80) (0.84–1.06) (0.31–0.47) (0.13–0.30)
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
137
Table 1 (Continued ) Total twin pairs, na
Planned CS (%)
Planned VD (%)
4,310
18,402
1,840 (42.7) Non-teaching Mean number of twin deliveries per year in the hospital 0–24 846 (19.6) 25–49 2,013 (46.7) 806 (18.7) 50–74 75–99 388 (9.0) 100 257 (6.0)
Crude OR (95% CI)
Adjusted OR (95% CI)
7,651 (41.6)
Ref.
Ref.
3,397 (18.5) 8,839 (48.0) 3,523 (19.1) 1,736 (9.4) 907 (4.9)
1.09 Ref. 1.01 0.98 1.24
(1.00–1.20) (0.92–1.10) (0.87–1.11) (1.07–1.44)
1.03 Ref. 0.98 0.75 1.06
(0.91–1.16) (0.86–1.10) (0.61–0.91) (0.85–1.31)
Adjusted OR >1 was associated with a higher incidence of planned CS, and adjusted OR <1 was associated with a higher incidence of planned VD. IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection. a For some subgroups, not all data were available. b p10 = 10th percentile.
associated with delivery by intrapartum CS according to the univariate analysis remained prominent after multivariate analysis: non-cephalic position of both twins (aOR 21.94; 95% CI 18.67– 25.78), non-cephalic position of twin A only (aOR 13.71; 95% CI 11.75–16.00), previous CS (aOR 7.00; 95% CI 5.77–8.49), nulliparity (aOR 4.20; 95% CI 3.67–4.81), maternal age 41 years (aOR 2.50; 95% CI 1.75–3.59), and cephalic–non-cephalic twin position (aOR 2.27; 95% CI 2.07–2.48). Minor indicators were maternal age 31–40 years, primiparity, pre-eclampsia, gestational age 38 + 0, weight twin B 125% compared to twin A, delivery in a university hospital or nonuniversity teaching hospital, and a low number (0–24) of annual twin deliveries in the hospital. Gestational age <37 + 0, birth weight twin A <10th percentile, and a higher number (50) of annual twin deliveries in the hospital, were associated with a successful VD. In the planned CS group 4 children (0.05%) died during delivery, and in the planned VD group 19 children (0.05%) (p = 0.84): 5 children (0.02%) in the VD of both twins group, 10 children (0.14%) in the intrapartum CS for both twins group, and 4 children (0.24%) in the combined delivery group. Comment During this 8-year nationwide study period in The Netherlands, a VD was planned in 80.9% of women with a twin pregnancy at 32–41 weeks. In this group, a VD of both twins was successful in 76.3% of women. These percentages are higher than mentioned in most other reports [1,9,21–23]. The percentage of combined deliveries (4.5%, ranging from 3.3 to 4.9% over the years) was in line with most other studies [1,9,12,22]. Most clinical indicators for a planned CS were identical to those for an intrapartum CS. Non-cephalic presentation of twin A was by far the most important indicator. In most national guidelines, no clear recommendation on the mode of delivery in case twin A is in non-cephalic presentation is expressed [15–18]. It is common practice in most countries to plan a CS [22–24], although there is no solid evidence to support this [1,5]. In accordance with other studies, a planned or intrapartum CS was also performed more frequently for twins with a cephalic presenting twin A in combination with a non-cephalic presenting twin B [22,24]. A previous CS was also an important indicator for a planned or intrapartum CS. A CS was planned for 44.8% of women with a previous CS. This percentage is higher than for singletons (29.3%) in The Netherlands [19] which is remarkable as the percentage of complications of a trial of labor after previous CS is not higher for twins [8,25]. Nulliparity and primiparity were minor indicators for a planned CS. Especially nulliparity was an important indicator for an intrapartum CS, which is in agreement with other studies showing that nulliparous women have twice the rate of an emergency prelabor CS compared with multiparous women, and that nulliparity is associated with adverse outcomes in twins [10,11,22,26]. Maternal age >30 years, and pre-eclampsia were also associated with more planned and more intrapartum CSs. It is
known that older women have more cesarean sections for twins compared to their younger counterparts, and that preeclampsia is associated with more unsuccessful trial of labor for twins [10,11,27]. Most planned CSs were performed at a gestational age of 37–39 weeks, which is in accordance with the optimal gestational age for a dichorionic twin delivery [15,30]. Both at lower and higher gestational ages, more VDs were planned, but only at lower gestational ages more VDs were successful, whereas at higher gestational ages more intrapartum CSs were performed. These results might be related to a higher rate of labor induction in more advanced gestational age groups. Birth weight of the twins was only a minor indicator, leading to conflicting results. As could be expected, a CS was planned more frequently in case one or both twins was small for gestational age, but also in case the weight of twin B was 80% compared to twin A. An intrapartum CS was performed more frequently in case the weight of twin B was 125% compared to twin A, whereas a vaginal delivery was more often successful when only twin A was small for gestational age. Planned CSs were performed slightly more frequent in university hospitals, which could be due to more complex pregnancies in these hospitals, although we corrected for some important case-mix factors. Possibly for the same reason, intrapartum CSs were more frequently performed in university and non-university teaching hospitals. Hospitals with a low number of twin deliveries per year had more intrapartum CSs, whereas hospitals with a higher number of twin deliveries per year had more successful VDs. This might be related to the experience of the obstetrical staff in these hospitals. The strength of this study is the large cohort of 22,712 women delivering of twins at 32–41 weeks gestation. The data are derived from a reliable and validated system, including almost all deliveries in one country and therefore representing a good reflection of current clinical decision making. We showed that most prominent indicators for planning a CS were identical to those for an intrapartum CS, suggesting that indicators like twin presentation, previous CS, nulliparity, advanced maternal age, and pre-eclampsia are important items for counseling women in planning the mode of delivery. Most indicators have already been defined in literature, confirming that current practice of twin delivery in the Netherlands is not quite different from practice elsewhere. The study has also some limitations. We could not differentiate between CS for fetal distress or failure to progress or other obstetrical reasons and we did not present any information regarding newborn outcomes. These data are important when weighing the relative risks and benefits of route of delivery and will be subject of continued study. Some women who delivered vaginally might have been planned to deliver by CS and will wrongfully be placed in the planned vaginal delivery group. Also, we presented data on the position of the twins at delivery, which
138
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
Table 2 Maternal, pregnancy related, fetal, neonatal and hospital related indicators of 18,402 twin pairs planned for a vaginal delivery (VD), according to the actual mode of delivery. Total twin pairs, na
Maternal indicators Maternal age at delivery (years) 20 21–30 31–40 41 Parity 0 1 2 Unknown Previous CS (1) Yes Socio-economic status Very high High Normal Low Very low Unknown Ethnicity Caucasian Non Caucasian Unknown Maternal diabetes mellitus Yes Pre-existent hypertension Yes
VD/VD (%)
Intrapartum CS (%)
VD/CS (%)
14,034
3,545
823
Adjusted OR (95% CI)
157 5,161 8,574 142
(1.1) (36.8) (61.1) (1.0)
42 1,327 2,127 49
(1.2) (37.4) (60.0) (1.4)
8 279 520 16
(1.0) (33.9) (63.2) (1.9)
1.02 (0.74–1.41) Ref. 0.99 (0.92–1.07) 1.47 (1.09–1.98)
0.73 (0.50–1.08) Ref. 1.29 (1.18–1.41) 2.50 (1.75–3.59)
5,662 5,622 2,645 105
(40.3) (40.1) (18.8) (0.7)
2,318 869 340 18
(65.4) (24.5) (9.6) (0.5)
328 317 172 6
(39.9) (38.5) (20.9) (0.7)
2.41 (2.17–2.68) 1.09 (0.97–1.22) Ref.
4.20 (3.67–4.81) 1.36 (1.19–1.56) Ref.
5.02 (4.25–5.93)
7.00 (5.77–8.49)
1.13 1.04 Ref. 1.07 1.13
1.13 1.08 Ref. 1.10 1.13
246 (1.8) 2,766 2,840 2,619 2,589 3,011 209
(19.7) (20.2) (18.7) (18.4) (21.5) (1.5)
12,408 (88.4) 1,586 (11.3) 40 (0.3)
332 (9.4) 719 688 617 640 812 69
(20.3) (19.4) (17.4) (18.1) (22.9) (1.9)
3,133 (88.4) 403 (11.4) 9 (25.4)
27 (3.3) 177 163 139 163 167 14
(21.5) (19.8) (16.9) (19.8) (20.3) (1.7)
749 (91.0) 71 (8.6) 3 (0.4)
(1.01–1.25) (0.96–1.20) (0.96–1.20) (1.01–1.26)
(0.99–1.28) (0.95–1.23) (0.97–1.26) (0.99–1.29)
Ref. 0.96 (0.86–1.07)
Ref. 0.92 (0.80–1.05)
47 (0.3)
21 (0.6)
1 (0.1)
1.51 (0.91–2.50)
1.47 (0.82–2.65)
35 (0.2)
10 (0.3)
6 (0.7)
1.47 (0.81–2.66)
1.46 (0.71–3.00)
178 (21.6) 23 (2.8)
1.37 (1.26–1.49)
1.10 (1.00–1.22)
Pregnancy related indicators IVF (IVF/ICSI) Yes 2,544 Unknown 428 Pre-eclampsia Yes 812 Unknown 414 Drug use during pregnancy Yes 4 Gestational age at delivery (weeks + days) 32 + 0–32 + 6 438 33 + 0–33 + 6 657 34 + 0–34 + 6 1,056 35 + 0–35 + 6 1,470 36 + 0–36 + 6 2,260 37 + 0–37 + 6 3,704 38 + 0–38 + 6 2,767 39 + 0–39 + 6 1,264 40 + 0–40 + 6 403 Unknown 15 Fetal and neonatal indicators Fetal position Cephalic–cephalic Cephalic–non cephalic Non cephalic–cephalic Non cephalic–non cephalic Unknown Sex concordance Concordant Discordant Unknown Weight twin B compared to twin A 80% 80–125% 125% Small for gestational ageb Both twinsp10 Twin A >p10, twin B p10 Unknown
Crude OR (95% CI)
9,433 3,874 385 332 10
(18.1) (3.0)
842 (23.8) 89 (2.5)
(5.8) (2.9)
299 (8.4) 112 (3.2)
48 (5.8) 22 (2.7)
1.41 (1.24–1.61)
1.34 (1.16–1.56)
(0.0)
0 (0.0)
3 (0.4)
2.42 (0.54–10.80)
2.01 (0.40–10.11) 0.76 0.73 0.73 0.74 0.85 Ref. 1.16 1.43 1.75
(3.1) (4.7) (7.5) (10.5) (16.1) (26.4) (19.7) (9.0) (2.9) (0.1)
149 211 326 383 538 764 646 368 159 1
(4.2) (6.0) (9.2) (10.8) (15.2) (21.6) (18.2) (10.4) (4.5) (0.0)
31 47 41 72 109 235 171 83 34 2
(3.8) (5.7) (5.0) (8.7) (13.0) (28.9) (20.8) (10.1) (4.1) (0.2)
1.52 1.46 1.29 1.15 1.06 Ref. 1.10 1.32 1.78
(1.26–1.83) (1.24–1.71) (1.12–1.48) (1.01–1.30) (0.95–1.18)
(67.2) (27.6) (2.7) (2.4) (0.1)
1,326 842 633 735 9
(37.4) (23.8) (17.9) (20.7) (0.3)
254 531 13 23 2
(30.9) (64.5) (1.6) (2.8) (0.2)
Ref. 2.12 (1.95–2.29) 10.02 (8.74–11.49) 13.63 (11.86–15.67)
Ref. 2.27 (2.07–2.48) 13.71 (11.75–16.00) 21.94 (18.67–25.78)
(0.99–1.22) (1.16–1.50) (1.48–2.14)
(0.60–0.95) (0.60–0.89) (0.61–0.86) (0.63–0.86) (0.75–0.97) (1.03–1.31) (1.23–1.66) (1.41–2.18)
9,071 (64.6) 4,961 (35.4) 2 (0.0)
2,201 (62.1) 1,342 (37.9) 2 (0.1)
493 (59.9) 330 (40.1) 0 (0.0)
Ref. 1.14 (1.06–1.22)
Ref. 1.03 (0.95–1.12)
1,370 (9.8) 11,891 (76.8) 773 (71.5)
411 (11.6) 2,891 (81.6) 243 (6.9)
61 (7.4) 697 (84.7) 65 (7.9)
1.14 (1.02–1.28) Ref. 1.32 (1.15–1.52)
1.03 (0.88–1.20) Ref. 1.43 (1.19–1.72)
363 370 620 2,190 2
108 95 122 498 0
0.90 (0.80–1.00) 0.91 (0.81–1.02) 0.99 (0.90–1.09) Ref.
0.91 (0.79–1.04) 0.78 (0.67–0.90) 0.90 (0.79–1.02) Ref.
1,646 1,608 2,344 8,432 4
(11.7) (11.5) (16.7) (60.1) (0.0)
(10.2) (10.4) (17.5) (61.8) (0.0)
(13.1) (11.5) (14.8) (60.5) (0.0)
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
139
Table 2 (Continued ) Total twin pairs, na
VD/VD (%)
Intrapartum CS (%)
VD/CS (%)
14,034
3,545
823
462 (13.0) 1,648 (46.5) 1,435 (40.5) 708 1,689 663 334 151
Hospital related indicators Type of hospital University 1,693 (21.1) Non-university teaching 6,471 (46.1) 5,870 (41.8) Non-teaching Mean number of twin deliveries per year in the hospital 0–24 2,553 (18.2) 25–49 6,718 (47.9) 50–74 2,727 (19.4) 75–99 1,325 (9.4) 100 711 (5.1)
(20.0) (47.6) (18.7) (9.4) (4.3)
Crude OR (95% CI)
Adjusted OR (95% CI)
116 (14.1) 361 (43.9) 346 (42.0)
1.13 (1.01–1.25) 1.02 (0.95–1.10) Ref.
1.46 (1.20–1.77) 1.10 (1.00–1.22) Ref.
136 432 133 77 45
1.05 Ref. 0.93 0.98 0.87
1.17 Ref. 0.81 0.70 0.63
(16.5) (52.5) (16.2) (9.4) (5.5)
(0.96–1.15) (0.84–1.01) (0.87–1.11) (0.74–1.03)
(1.04–1.31) (0.72–0.91) (0.58–0.84) (0.51–0.79)
Adjusted OR >1 was associated with a higher incidence of intrapartum cesarean section (CS) of one or both twins, and adjusted OR <1 was associated with a higher incidence of successful vaginal delivery of both twins. VD/VD, vaginal delivery both twins; intrapartum CS, cesarean section for both twins; VD/CS, vaginal delivery twin A followed by cesarean section twin B; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection. a For some subgroups, not all data were available. b p10 = 10th percentile.
is not necessarily the same as the position at the start of labor [21,28]. We do not have any information about chorionicity, as chorionicity is not scored in the PRN database. Chorionicity might be an important indicator, as some gynecologists deliver all monochorionic twins by CS [29]. Some indicators such as diabetes, hypertension, and drug use during pregnancy may be underreported because caregivers are not obliged to register them. Since the proportion of underreporting is expected to be similar in both the outcome groups, the observed odds ratios may approximate the ‘true’ strength of the association. Unfortunately, the PRN database does not provide information on other potential confounding factors, like fetal distress, possible indications for planned CS or VD, antenatal surveillance, and ultrasound examination. Finally, we did not have any information on the pregnant woman’s preference on mode of delivery. In conclusion, we found a high percentage of planned and successful VD of twins 32–41 weeks gestation. Fetal position, previous CS, nulliparity, advanced maternal age, and pre-eclampsia were prominent indicators for both planned and intrapartum CS. Further studies need to focus on neonatal and maternal outcomes. Based on these findings, a prediction model for the optimal mode of delivery for twins should be made. Acknowledgments None. References [1] Rossi AC, Mullin PM, Chmait RH. Neonatal outcomes of twins according to birth order, presentation and mode of delivery: a systematic review and metaanalysis. BJOG 2011;118:523–32. [2] Barrett JF, Hannah ME, Hutton EK, et al. A randomized trial of planned cesarean or vaginal delivery for twin pregnancy. N Engl J Med 2013;369:1295–305. [3] van der Garde M, Winkens B, Roumen FJ. Increased elective caesarean section rate is not associated with a decreased serious morbidity rate for twins 32 weeks’ gestation. J Obstet Gynaecol 2012;32:453–7. [4] NVOG. Richtlijn Meerlingzwangerschap. NVOG; 2011, Available at: http:// nvog-documenten.nl/index.php?pagina=/richtlijn/pagina. php&fSelectTG_62=75&fSelectedSub=62&fSelectedParent=75. [5] Steins Bisschop CN, Vogelvang TE, May AM, Schuitemaker NW. Mode of delivery in non-cephalic presenting twins: a systematic review. Arch Gynecol Obstet 2012;286:237–47. [6] Smith GC, Shah I, White IR, Pell JP, Dobbie R. Mode of delivery and the risk of delivery-related perinatal death among twins at term: a retrospective cohort study of 8073 births. BJOG 2005;112:1139–44. [7] Smith GC, Fleming KM, White IR. Birth order of twins and risk of perinatal death related to delivery in England, Northern Ireland, and Wales, 1994–2003: retrospective cohort study. BMJ 2007;334:576.
[8] Christopher D, Robinson BK, Peaceman AM. An evidence-based approach to determining route of delivery for twin gestations. Rev Obstet Gynecol 2011;4:109–16. [9] Wenckus DJ, Gao W, Kominiarek MA, Wilkins I. The effects of labor and delivery on maternal and neonatal outcomes in term twins: a retrospective cohort study. BJOG 2014;121:1137–44. [10] Ko HJ, Jun JK. Clinical factors associated with failed trials of labor in late preterm and term twin pregnancies. J Perinat Med 2014;42:449–55. [11] Fox NS, Gupta S, Melka S, et al. Risk factors for cesarean delivery in twin pregnancies attempting vaginal delivery. Am J Obstet Gynecol 2015;212: 106.e1–.e5. [12] Persad VL, Baskett TF, O’Connell CM, Scott HM. Combined vaginal–cesarean delivery of twin pregnancies. Obstet Gynecol 2001;98:1032–7. [13] Haest KM, Roumen FJ, Nijhuis JG. Neonatal and maternal outcomes in twin gestations 32 weeks according to the planned mode of delivery. Eur J Obstet Gynecol Reprod Biol 2005;123:17–21. [14] Zipori Y, Smolkin T, Makhoul IR, Weissman A, Blazer S, Drugan A. Optimizing outcome of twins by routine cesarean section beyond 37 weeks. Am J Perinatol 2011;28:51–6. [15] National Collaborating Centre for Women’s and Children’s Health (UK). Multiple Pregnancy: The Management of Twin and Triplet Pregnancies in the Antenatal Period. London: RCOG Press; 2011, September (NICE Clinical Guidelines, No. 129.) Available from: http://www.ncbi.nlm.nih.gov/books/NBK83105/. [16] American College of Obstetricians and Gynecologists, Society for MaternalFetal Medicine. ACOG Practice Bulletin No. 144: multifetal gestations: twin, triplet, and higher-order multifetal pregnancies. Obstet Gynecol 2014;123: 1118–32. [17] Vayssiere C, Benoist G, Blondel B, et al. Twin pregnancies: guidelines for clinical practice from the French College of Gynaecologists and Obstetricians (CNGOF). Eur J Obstet Gynecol Reprod Biol 2011;156:12–7. [18] Institute of Obstetricians and Gynaecologists, Royal College of Physicians of Ireland and Directorate of strategy and Clinical Care, Health Service Executive. Clinical practice guideline management of multiple pregnancy. 1st ed. 2012, Available at: http://www.hse.ie/eng/about/Who/vlinical/natclinprog/ obsandgynaeprogramme/pregmulti.pdf. [19] Stichting Perinatale Registratie Nederland. Grote Lijnen 10 jaar Perinatale Registratie Nederland. Utrecht: Stichting Perinatale Registratie Nederland; 2011, Available at: http://www.perinatreg.nl/uploads/173/123/10_jaar_ Perinatale_Zorg_in_Nederland_de_grote_lijnen.PDF. [20] Visser GH, Eilers PH, Elferink-Stinkens PM, Merkus HM, Wit JM. New Dutch reference curves for birthweight by gestational age. Early Hum Dev 2009;85: 737–44. [21] Engelbrechtsen L, Nielsen EH, Perin T, Oldenburg A, Tabor A, Skibsted L. Cesarean section for the second twin: a population-based study of occurrence and outcome. Birth 2013;40:10–6. [22] Breathnach F, McAuliffe F, Geary M, et al. Prediction of safe and successful vaginal twin birth. Am J Obstet Gynecol 2011;205:237.e1–.e7. [23] Lee HC, Gould JB, Boscardin WJ, El-Sayed YY, Blumenfeld YJ. Trends in cesarean delivery for twin births in the United States: 1995–2008. Obstet Gynecol 2011;118:1095–101. [24] Schmitz T, Carnavalet Cde C, Azria E, Lopez E, Cabrol D, Goffinet F. Neonatal outcomes of twin pregnancy according to the planned mode of delivery. Obstet Gynecol 2008;111:695–703. [25] Ford AA, Bateman BT, Simpson LL. Vaginal birth after cesarean delivery in twin gestations: a large, nationwide sample of deliveries. Am J Obstet Gynecol 2006;195:1138–42. [26] Schutte JM, Steegers EA, Schuitemaker NW, et al. Rise in maternal mortality in the Netherlands. BJOG 2010;117:399–406.
140
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
[27] Lisonkova S, Joseph KS, Bell R, Glinianaia SV. Effect of advanced maternal age on perinatal outcomes in twins: the impact of chorionicity. Ann Epidemiol 2013;23:428–34. [28] Dodd JM, Deussen AR, Grivell RM, Crowther CA. Elective birth at 37 weeks’ gestation for women with an uncomplicated twin pregnancy. Cochrane Database Syst Rev 2014;2. CD003582.
[29] Houlihan C, Knuppel RA. Intrapartum management of multiple gestations. Clin Perinatol 1996;23:91–116. [30] Hoffmann E, Oldenburg A, Rode L, Tabor A, Rasmussen S, Skibsted L. Twin births: cesarean section or vaginal delivery? Acta Obstet Gynecol Scand 2012;91:463–9.
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Clinical indicators associated with the mode of twin delivery: an analysis of 22,712 twin pairs Simone M.T.A. Goossens a,*, Chantal W.P.M. Hukkelhoven b, Lotte de Vries b, Ben Willem Mol c, Jan G. Nijhuis d, Frans J.M.E. Roumen a a
Department of Obstetrics and Gynecology, Atrium Medical Center Parkstad, Heerlen, The Netherlands Netherlands Perinatal Registry, Utrecht, The Netherlands The Robinson Institute, School of Pediatrics and Reproductive Health, University of Adelaide, Australia d Department of Obstetrics and Gynecology, Maastricht University Medical Center, GROW – School for Oncology and Developmental Biology, The Netherlands b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 August 2015 Received in revised form 13 September 2015 Accepted 17 September 2015
Objective: To identify clinical indicators associated with the planned and actual mode of delivery in women with a twin pregnancy. Study design: We performed a retrospective cohort study in women with a twin pregnancy who delivered at a gestational age of 32 + 0–41 + 0 weeks and days between 2000 and 2008 in the Netherlands. Data were obtained from a nationwide database. We identified maternal, pregnancyrelated, fetal, neonatal and hospital-related indicators that were associated with planned cesarean section (CS) and, for women with planned vaginal delivery (VD), for intrapartum CS. The associations between indicators and mode of delivery were studied with uni- and multivariate logistic regression analyses. Results: We included 22,712 women with a twin pregnancy, of whom 4,310 women (19.0%) had a planned CS. Of the 18,402 women who had a planned VD, 14,034 (76.3%) delivered vaginally, 3,545 (19.3%) had an intrapartum CS, while 823 (4.5%) delivered twin A vaginally and twin B by intrapartum CS. The clinical indicators for a planned CS and an intrapartum CS were comparable: non-cephalic position of both twins (aOR 25.32; 95% CI 22.50–28.50, and aOR 21.94; 95% CI 18.67–25.78, respectively), noncephalic position of twin A only (aOR 21.67 95% CI 19.12–24.34, and aOR 13.71; 95% CI 11.75–16.00, respectively), previous CS (aOR 3.69; 95% CI 3.12–4.36, and aOR 7.00; 95% CI 5.77–8.49, respectively), nulliparity (aOR 1.51; 95% CI 1.32–1.72, and aOR 4.20; 95% CI 3.67–4.81, respectively), maternal age 41 years (aOR 3.00; 95% CI 2.14–4.22, and aOR 2.50; 95% CI 1.75–3.59, respectively), and pre-eclampsia (aOR 2.12; 95% CI 1.83–2.46, and aOR 1.34; 95% CI 1.16–1.56, respectively). Conclusion: Both planned and intrapartum CS in twins had comparable predictors: non-cephalic position of both twins or twin A only, previous CS, nulliparity, advanced maternal age, and preeclampsia. ß 2015 Published by Elsevier Ireland Ltd.
Keywords: Cesarean section Clinical indicator Twin delivery Decision
Introduction To make a balanced shared decision on the mode of delivery in women with a twin pregnancy beyond a gestational age of 32 weeks, information on clinical indicators associated with a cesarean section (CS) is relevant. Presentation of twin A is such a well-known indicator. In case of a cephalic presenting twin A, the
* Corresponding author at: Department of Obstetrics and Gynecology, Atrium Medical Center Parkstad, P.O. 4446, 6401 CX Heerlen, The Netherlands. Tel.: +31 455766513; fax: +31 455766625. E-mail address: [email protected] (Simone M.T.A. Goossens). http://dx.doi.org/10.1016/j.ejogrb.2015.09.034 0301-2115/ß 2015 Published by Elsevier Ireland Ltd.
pregnant woman is counseled that, generally, planned CS has comparable safety as planned vaginal delivery (VD) [1–3]. Although there is no firm evidence to suggest benefit of cesarean over vaginal delivery in case of non-cephalic position of the first twin, she is counseled as is usual for non-cephalic presentation of a singleton, frequently resulting in an elective CS [4,5]. However, many other clinical indicators may be involved, like maternal age, mode of conception, parity, Body Mass Index, diabetes, previous CS, chorionicity, sex concordance, fetal intrauterine growth retardation, fetal weight, and birth weight discordance [2,6–11]. Ideally, an intrapartum CS, i.e. a CS during active labor, should not be necessary. Especially a combined delivery, i.e. twin A
134
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
delivered vaginally and twin B by CS, should be prevented, as it is associated with the highest neonatal and maternal morbidity in twin births [1,12]. For women with a very high risk on a combined delivery, a planned CS may be the optimal mode of delivery [12–14]. In most national guidelines, no clear recommendations on planning the mode of delivery are expressed [15–18]. So, the planned and actual mode of delivery might differ depending on maternal, professional, hospital, fetal and environmental circumstances. The purpose of this study was to identify the most important maternal, pregnancy-related, fetal, neonatal and hospital-related indicators associated with the planned and actual mode of delivery for twin pregnancies in the Netherlands. Materials and methods We used data of our nationwide registry on pregnancy and delivery. In The Netherlands, data on pregnancy, birth and neonates are routinely collected in four separate registries: the LVR1-registry for primary midwife-led care, the LVRh-registry for care led by general practitioners, the LVR2-registry for obstetrician-led care and the LNR-registry for neonatal care. These separate registries are linked into one database, The Netherlands Perinatal Registry (PRN) [19]. Approximately 96% of all births in The Netherlands are entered into this registry. We included all women with a twin pregnancy and a gestational age of 32 + 0–41 + 0 weeks at delivery, as registered in the PRN database from January 1st, 2000 until January 1st, 2008. We excluded pregnancies with one or both twins weighing <500 g at birth or having lethal congenital anomalies, and pregnancies with a fetal demise before start of delivery. Cases with incomplete twin data sets (records with only one neonate registered) were also excluded. Because of these incomplete data sets, the total number of twin babies registered initially could be unequal. In The Netherlands, twins are delivered in a hospital by an obstetrician. As far as we know, none of the participating hospitals excluded vaginal deliveries of twins in all circumstances. In the PRN, a CS may be registered as a primary CS or a secondary CS. A primary CS is a CS in women who did not try vaginal labor. A secondary CS is defined as a CS during labor in women who were planned to try a vaginal delivery. For this study, we assumed that all primary CSs were planned, and therefore we registered them in the planned CS group, whereas women who delivered (1) both children vaginally, (2) by a secondary CS of both children, or (3) by a combined delivery, were presumed to be in the planned VD group. All CSs in the planned VD group were defined as intrapartum CSs. For analysis of both the planned CS group and the planned VD group separately, we studied maternal indicators (maternal age at delivery, parity, previous CS, socio-economic status, ethnicity, maternal diabetes mellitus, pre-existing hypertension), pregnancy-related indicators (mode of conception, preeclampsia, drug use during pregnancy, gestational age at delivery), fetal and neonatal indicators (fetal position, sex concordance, weight concordance, small for gestational age), and hospitalrelated indicators (type of hospital, mean number of twin deliveries per year), as known [1,2,5–11] or expected from a pathophysiologic perspective to be associated with the planned and actual mode of delivery. Socioeconomic status was based on the mean household income level of the neighborhood, determined by the first 4 digits of the woman’s postal code. Diabetes was defined as diabetes mellitus before a gestational age of 20 weeks or gestational diabetes. Pre-existing hypertension was defined as hypertension before a gestational age of 20 weeks. Pre-eclampsia was defined as a diastolic blood pressure 90 mmHg and proteinuria (300 mg/ day). Birth weight of twin B compared to twin A was divided into
subgroups of 80%, 80–125% and 125%. Small for gestational age was defined as a birth weight <10th percentile [20]. Statistical analysis Contingency tables were created to assess frequencies and percentages of the indicators and the outcome measures. All risk indicators mentioned above were analyzed with uni- and multivariate logistic regression analyses, calculating crude and adjusted odds ratio (OR) and 95% confidence intervals (CI). For the adjusted OR we adjusted for all risk factors mentioned in the univariate analyses. The association between mortality and the planned delivery groups was calculated with chi-square test. Reference groups were women 21–30 years old, multiparous (2), without a previous CS, with a normal socio-economic status, Caucasian, non-diabetic, and without pre-existing hypertension; no in vitro fertilization (IVF), pre-eclampsia or drug use, gestational age at delivery between 37 and 38 weeks; cephalic-cephalic position, sex concordant, weight concordant, birth weight of both children >10th percentile; women delivering in a non-teaching hospital, and women delivering in a hospital with an annual number of twin deliveries of 25–49. Analyses were performed using SAS 9.3 (SAS Institute, Cary, NC, USA) and Microsoft Excel 2011 for Mac. Ethics approval was not required under Dutch Law, as only anonymous data were used. A privacy committee checks all researches done in collaboration with the PRN to make sure procedures are as accurate as possible, the privacy of the participants is guaranteed and the data confidentiality is respected. Results The database contained 56,865 records from children born from a twin pregnancy in the study period (Fig. 1). From these, 45,424 twin babies (n = 22,712 women with a twin pair) fulfilled the inclusion criteria. In 4,310 women with a twin pair (19.0%) a planned CS was performed, while a VD was planned for 18,402 women with a twin pair (81.0%). Of these, 14,034 women (76.3%) delivered vaginally, 3,545 women (19.3%) delivered both children by an intrapartum CS, and 823 women (4.5%) by combined delivery. Fig. 2 shows the distribution of the actual mode of delivery after a planned VD over the years. The percentage of actual VD in this group remained relatively stable, with a range between 75.2% and 78.3%, and the percentage of combined deliveries ranged between 3.3% and 4.9%. Table 1 shows maternal, pregnancy related, fetal, neonatal and hospital related indicators of twins according to the planned mode of delivery, including the results of the logistic regression analyses. Most prominent indicators for a delivery by planned CS according to the univariate analysis remained prominent after multivariate analysis: non-cephalic position of both twins (adjusted (a) OR 25.32; 95% confidence interval (CI) 22.50–28.50), non-cephalic position of twin A only (aOR 21.67; 95% CI 19.12–24.34), 1 previous CS (aOR 3.69; 95% CI 3.12–4.36), maternal age 41 years (aOR 3.00; 95% CI 2.14–4.22), and pre-eclampsia (aOR 2.12; 95%; CI 1.83–2.46). Minor indicators associated with delivery by planned CS were maternal age 31–40 years, nulliparity, primiparity, cephalic-non cephalic position, weight twin B 80% compared to twin A, birth weight twin A or twin B <10th percentile, and delivery in a university hospital. Gestational age at delivery 32 + 0– 37 + 0 and 39 + 0–41 + 0 was associated with a planned VD, as was an annual number of 75–99 twin deliveries in the hospital. Table 2 shows indicators of twins planned for a vaginal delivery, according to the actual mode of delivery, including the results of the logistic regression analyses. Most prominent indicators
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
135
Total number of twin babies born from Jan 01, 2000 – Jan 01, 2008 (n=56,865)
Twin babies excluded from analysis* - Gestational age <32 + 0 or ≥ 41 + 0 (n=5,714) - Weight < 500g (n=918) - Congenital abnormalities (n=1,896) - Intrauterine fetal death (n=1,040) - Incomplete data (n=2,461)
Total number of twin babies for analysis (n=45,424) Total number of twin pairs for analysis (n=22,712 )
Planned Cesarean Section Twin pairs n=4,310 (19.0%)
Vaginal Delivery both twins Twin pairs n=14,034 (76.3%)
Planned Vaginal Delivery Twin pairs n=18,402 (81.0%)
Intrapartum Cesarean Section both twins Twin pairs n=3,545 (19.3%)
Vaginal delivery twin A / Intrapartum Cesarean Section twin B Twin pairs n=823 (4.5%)
Fig. 1. Overview of twin babies and women with a twin pair as registered in the PRN database during the eight-year study period, according to the mode of delivery. *Some have more than one exclusion criterion.
Fig. 2. Actual mode of delivery of twin pairs (%) in the planned vaginal delivery group over the years 2000–2007. VD = vaginal delivery; CS = cesarean section.
136
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
Table 1 Maternal, pregnancy related, fetal, neonatal and hospital related indicators of 22,712 twin pairs according to planned cesarean section (CS) or planned vaginal delivery (VD). Total twin pairs, na
Maternal indicators Maternal age at delivery (years) 20 21–30 31–40 41 Parity 0 1 2 Unknown Previous CS 1 Yes Socio-economic status Very high High Normal Low Very low Unknown Ethnicity Caucasian Non Caucasian Unknown Maternal diabetes mellitus Yes Pre-existent hypertension Yes Pregnancy related indicators IVF (IVF/ICSI) Yes Unknown Pre-eclampsia Yes Unknown Drug use during pregnancy Yes Gestational age at delivery (weeks + days) 32 + 0–32 + 6 33 + 0–33 + 6 34 + 0–34 + 6 35 + 0–35 + 6 36 + 0–36 + 6 37 + 0–37 + 6 38 + 0–38 + 6 39 + 0–39 + 6 40 + 0–40 + 6 Unknown Fetal and neonatal indicators Fetal position Cephalic–cephalic Cephalic–non cephalic Non cephalic–cephalic Non cephalic–non cephalic Unknown Sex concordance Concordant Discordant Unknown Weight twin B compared to twin A 80% 80–125% 125% Small for gestational ageb Both twins
Planned CS (%)
Planned VD (%)
4,310
18,402
Crude OR (95% CI)
Adjusted OR (95% CI)
44 (1.0) 1,468 (34.1) 2,712 (62.9) 86 (2.00)
207 (1.1) 6,767 (36.8) 11,221 (61.0) 207 (1.1)
0.98 (0.70–1.36) Ref. 1.11 (1.04–1.20) 1.92 (1.48–2.48)
0.79 (0.52–1.18) Ref. 1.26 (1.15–1.39) 3.00 (2.14–4.22)
2,158 (50.1) 1,465 (34.0) 659 (15.3) 28 (0.6)
8,308 (45.1) 6,808 (37.0) 3,157 (17.2) 129 (0.7)
1.24 (1.13–1.37) 1.03 (0.93–1.14) Ref.
1.51 (1.32–1.72) 1.21 (1.07–1.38) Ref.
491 (11.4)
605 (3.3)
3.78 (3.34–4.28)
3.69 (3.12–4.36)
819 (19.0) 830 (19.3) 824 (19.1) 770 (17.9) 999 (23.2) 68 (1.6)
3,662 (19.9) 3,691 (20.1) 3,375 (18.3) 3,392 (18.4) 3,990 (21.7) 292 (1.6)
0.92 0.92 Ref. 0.93 1.03
0.95 0.97 Ref. 0.94 1.03
3,747 (86.9) 548 (12.7) 15 (0.4)
16,290 (88.5) 2,060 (11.2) 52 (0.3)
Ref. 1.16 (1.05–1.28)
Ref. 1.01 (0.88–1.16)
18 (0.4)
50 (0.3)
1.54 (0.90–2.64)
1.41 (0.79–2.50)
17 (0.4)
51 (0.3)
1.43 (0.82–2.47)
0.94 (0.44–1.99)
982 (22.8) 110 (2.6)
3,564 (19.4) 540 (2.9)
1.22 (1.13–1.33)
1.09 (0.98–1.22)
465 (10.8) 127 (2.9)
1,159 (6.3) 548 (3.0)
1.80 (1.61–2.02)
2.12 (1.83–2.46)
3 (0.1)
7 (0.0)
1.84 (0.48–7.09)
3.05 (0.64–14.52)
157 (3.6) 209 (4.8) 302 (7.0) 390 (9.1) 645 (15.0) 1,416 (32.9) 989 (22.9) 162 (3.8) 38 (0.9) 2 (0.0)
618 (3.4) 915 (5.0) 1,423 (7.7) 1,925 (10.5) 2,905 (15.8) 4,703 (25.6) 3,584 (19.5) 1,715 (9.3) 596 (3.2) 18 (0.1)
0.84 0.76 0.71 0.67 0.74 Ref. 0.92 0.31 0.21
0.73 0.64 0.63 0.63 0.70 Ref. 0.94 0.38 0.20
697 (16.2) 618 (14.3) 1,344 (31.2) 1,639 (38.0) 12 (0.3)
11,013 (59.9) 5,247 (28.5) 1,031 (5.6) 1,090 (5.9) 21 (0.1)
Ref. 1.86 (1.66–2.08) 20.60 (18.42–23.03) 23.76 (21.32–26.48)
Ref. 1.85 (1.64–2.98) 21.67 (19.12–24.34) 25.32 (22.50–28.50)
2,626 (60.9) 1,684 (39.1) 0 (0.0)
11,765 (63.9) 6,633 (36.0) 4 (0.0)
Ref. 1.14 (1.06–1.22)
Ref. 0.99 (0.91–1.08)
669 (15.5) 3,340 (77.5) 301 (7.0)
1,842 (10.0) 15,479 (84.1) 1,081 (5.9)
1.68 (1.52–1.85) Ref. 1.29 (1.13–1.47)
1.74 (1.50–2.02) Ref. 1.15 (0.95–1.39)
431 (10.0) 916 (21.2) 525 (12.2) 2,438 (56.6) 0 (0.0)
2,117 (11.5) 3,086 (17.8) 2,073 (11.3) 11,120 (60.4) 6 (0.0)
0.93 (0.83–1.04) 1.35 (1.24–1.48) 1.15 (1.04–1.48) Ref.
1.15 (1.00–1.34) 1.23 (1.07–1.40) 1.20 (1.03–1.39) Ref.
602 (14.0) 1,868 (43.3)
2,271 (12.3) 8,480 (46.1)
1.10 (0.99–1.22) 0.92 (0.85–0.98)
1.38 (1.13–1.69) 0.93 (0.83–1.03)
(0.82–1.02) (0.83–1.03) (0.83–1.04) (0.93–1.14)
(0.70–1.02) (0.65–0.89) (0.61–0.81) (0.59–0.76) (0.67–0.82) (0.84–1.01) (0.26–0.37) (0.15–0.30)
(0.83–1.09) (0.84–1.11) (0.82–1.08) (0.88–1.16)
(0.57–0.93) (0.52–0.78) (0.53–0.76) (0.54–0.73) (0.61–0.80) (0.84–1.06) (0.31–0.47) (0.13–0.30)
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
137
Table 1 (Continued ) Total twin pairs, na
Planned CS (%)
Planned VD (%)
4,310
18,402
1,840 (42.7) Non-teaching Mean number of twin deliveries per year in the hospital 0–24 846 (19.6) 25–49 2,013 (46.7) 806 (18.7) 50–74 75–99 388 (9.0) 100 257 (6.0)
Crude OR (95% CI)
Adjusted OR (95% CI)
7,651 (41.6)
Ref.
Ref.
3,397 (18.5) 8,839 (48.0) 3,523 (19.1) 1,736 (9.4) 907 (4.9)
1.09 Ref. 1.01 0.98 1.24
(1.00–1.20) (0.92–1.10) (0.87–1.11) (1.07–1.44)
1.03 Ref. 0.98 0.75 1.06
(0.91–1.16) (0.86–1.10) (0.61–0.91) (0.85–1.31)
Adjusted OR >1 was associated with a higher incidence of planned CS, and adjusted OR <1 was associated with a higher incidence of planned VD. IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection. a For some subgroups, not all data were available. b p10 = 10th percentile.
associated with delivery by intrapartum CS according to the univariate analysis remained prominent after multivariate analysis: non-cephalic position of both twins (aOR 21.94; 95% CI 18.67– 25.78), non-cephalic position of twin A only (aOR 13.71; 95% CI 11.75–16.00), previous CS (aOR 7.00; 95% CI 5.77–8.49), nulliparity (aOR 4.20; 95% CI 3.67–4.81), maternal age 41 years (aOR 2.50; 95% CI 1.75–3.59), and cephalic–non-cephalic twin position (aOR 2.27; 95% CI 2.07–2.48). Minor indicators were maternal age 31–40 years, primiparity, pre-eclampsia, gestational age 38 + 0, weight twin B 125% compared to twin A, delivery in a university hospital or nonuniversity teaching hospital, and a low number (0–24) of annual twin deliveries in the hospital. Gestational age <37 + 0, birth weight twin A <10th percentile, and a higher number (50) of annual twin deliveries in the hospital, were associated with a successful VD. In the planned CS group 4 children (0.05%) died during delivery, and in the planned VD group 19 children (0.05%) (p = 0.84): 5 children (0.02%) in the VD of both twins group, 10 children (0.14%) in the intrapartum CS for both twins group, and 4 children (0.24%) in the combined delivery group. Comment During this 8-year nationwide study period in The Netherlands, a VD was planned in 80.9% of women with a twin pregnancy at 32–41 weeks. In this group, a VD of both twins was successful in 76.3% of women. These percentages are higher than mentioned in most other reports [1,9,21–23]. The percentage of combined deliveries (4.5%, ranging from 3.3 to 4.9% over the years) was in line with most other studies [1,9,12,22]. Most clinical indicators for a planned CS were identical to those for an intrapartum CS. Non-cephalic presentation of twin A was by far the most important indicator. In most national guidelines, no clear recommendation on the mode of delivery in case twin A is in non-cephalic presentation is expressed [15–18]. It is common practice in most countries to plan a CS [22–24], although there is no solid evidence to support this [1,5]. In accordance with other studies, a planned or intrapartum CS was also performed more frequently for twins with a cephalic presenting twin A in combination with a non-cephalic presenting twin B [22,24]. A previous CS was also an important indicator for a planned or intrapartum CS. A CS was planned for 44.8% of women with a previous CS. This percentage is higher than for singletons (29.3%) in The Netherlands [19] which is remarkable as the percentage of complications of a trial of labor after previous CS is not higher for twins [8,25]. Nulliparity and primiparity were minor indicators for a planned CS. Especially nulliparity was an important indicator for an intrapartum CS, which is in agreement with other studies showing that nulliparous women have twice the rate of an emergency prelabor CS compared with multiparous women, and that nulliparity is associated with adverse outcomes in twins [10,11,22,26]. Maternal age >30 years, and pre-eclampsia were also associated with more planned and more intrapartum CSs. It is
known that older women have more cesarean sections for twins compared to their younger counterparts, and that preeclampsia is associated with more unsuccessful trial of labor for twins [10,11,27]. Most planned CSs were performed at a gestational age of 37–39 weeks, which is in accordance with the optimal gestational age for a dichorionic twin delivery [15,30]. Both at lower and higher gestational ages, more VDs were planned, but only at lower gestational ages more VDs were successful, whereas at higher gestational ages more intrapartum CSs were performed. These results might be related to a higher rate of labor induction in more advanced gestational age groups. Birth weight of the twins was only a minor indicator, leading to conflicting results. As could be expected, a CS was planned more frequently in case one or both twins was small for gestational age, but also in case the weight of twin B was 80% compared to twin A. An intrapartum CS was performed more frequently in case the weight of twin B was 125% compared to twin A, whereas a vaginal delivery was more often successful when only twin A was small for gestational age. Planned CSs were performed slightly more frequent in university hospitals, which could be due to more complex pregnancies in these hospitals, although we corrected for some important case-mix factors. Possibly for the same reason, intrapartum CSs were more frequently performed in university and non-university teaching hospitals. Hospitals with a low number of twin deliveries per year had more intrapartum CSs, whereas hospitals with a higher number of twin deliveries per year had more successful VDs. This might be related to the experience of the obstetrical staff in these hospitals. The strength of this study is the large cohort of 22,712 women delivering of twins at 32–41 weeks gestation. The data are derived from a reliable and validated system, including almost all deliveries in one country and therefore representing a good reflection of current clinical decision making. We showed that most prominent indicators for planning a CS were identical to those for an intrapartum CS, suggesting that indicators like twin presentation, previous CS, nulliparity, advanced maternal age, and pre-eclampsia are important items for counseling women in planning the mode of delivery. Most indicators have already been defined in literature, confirming that current practice of twin delivery in the Netherlands is not quite different from practice elsewhere. The study has also some limitations. We could not differentiate between CS for fetal distress or failure to progress or other obstetrical reasons and we did not present any information regarding newborn outcomes. These data are important when weighing the relative risks and benefits of route of delivery and will be subject of continued study. Some women who delivered vaginally might have been planned to deliver by CS and will wrongfully be placed in the planned vaginal delivery group. Also, we presented data on the position of the twins at delivery, which
138
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
Table 2 Maternal, pregnancy related, fetal, neonatal and hospital related indicators of 18,402 twin pairs planned for a vaginal delivery (VD), according to the actual mode of delivery. Total twin pairs, na
Maternal indicators Maternal age at delivery (years) 20 21–30 31–40 41 Parity 0 1 2 Unknown Previous CS (1) Yes Socio-economic status Very high High Normal Low Very low Unknown Ethnicity Caucasian Non Caucasian Unknown Maternal diabetes mellitus Yes Pre-existent hypertension Yes
VD/VD (%)
Intrapartum CS (%)
VD/CS (%)
14,034
3,545
823
Adjusted OR (95% CI)
157 5,161 8,574 142
(1.1) (36.8) (61.1) (1.0)
42 1,327 2,127 49
(1.2) (37.4) (60.0) (1.4)
8 279 520 16
(1.0) (33.9) (63.2) (1.9)
1.02 (0.74–1.41) Ref. 0.99 (0.92–1.07) 1.47 (1.09–1.98)
0.73 (0.50–1.08) Ref. 1.29 (1.18–1.41) 2.50 (1.75–3.59)
5,662 5,622 2,645 105
(40.3) (40.1) (18.8) (0.7)
2,318 869 340 18
(65.4) (24.5) (9.6) (0.5)
328 317 172 6
(39.9) (38.5) (20.9) (0.7)
2.41 (2.17–2.68) 1.09 (0.97–1.22) Ref.
4.20 (3.67–4.81) 1.36 (1.19–1.56) Ref.
5.02 (4.25–5.93)
7.00 (5.77–8.49)
1.13 1.04 Ref. 1.07 1.13
1.13 1.08 Ref. 1.10 1.13
246 (1.8) 2,766 2,840 2,619 2,589 3,011 209
(19.7) (20.2) (18.7) (18.4) (21.5) (1.5)
12,408 (88.4) 1,586 (11.3) 40 (0.3)
332 (9.4) 719 688 617 640 812 69
(20.3) (19.4) (17.4) (18.1) (22.9) (1.9)
3,133 (88.4) 403 (11.4) 9 (25.4)
27 (3.3) 177 163 139 163 167 14
(21.5) (19.8) (16.9) (19.8) (20.3) (1.7)
749 (91.0) 71 (8.6) 3 (0.4)
(1.01–1.25) (0.96–1.20) (0.96–1.20) (1.01–1.26)
(0.99–1.28) (0.95–1.23) (0.97–1.26) (0.99–1.29)
Ref. 0.96 (0.86–1.07)
Ref. 0.92 (0.80–1.05)
47 (0.3)
21 (0.6)
1 (0.1)
1.51 (0.91–2.50)
1.47 (0.82–2.65)
35 (0.2)
10 (0.3)
6 (0.7)
1.47 (0.81–2.66)
1.46 (0.71–3.00)
178 (21.6) 23 (2.8)
1.37 (1.26–1.49)
1.10 (1.00–1.22)
Pregnancy related indicators IVF (IVF/ICSI) Yes 2,544 Unknown 428 Pre-eclampsia Yes 812 Unknown 414 Drug use during pregnancy Yes 4 Gestational age at delivery (weeks + days) 32 + 0–32 + 6 438 33 + 0–33 + 6 657 34 + 0–34 + 6 1,056 35 + 0–35 + 6 1,470 36 + 0–36 + 6 2,260 37 + 0–37 + 6 3,704 38 + 0–38 + 6 2,767 39 + 0–39 + 6 1,264 40 + 0–40 + 6 403 Unknown 15 Fetal and neonatal indicators Fetal position Cephalic–cephalic Cephalic–non cephalic Non cephalic–cephalic Non cephalic–non cephalic Unknown Sex concordance Concordant Discordant Unknown Weight twin B compared to twin A 80% 80–125% 125% Small for gestational ageb Both twins
Crude OR (95% CI)
9,433 3,874 385 332 10
(18.1) (3.0)
842 (23.8) 89 (2.5)
(5.8) (2.9)
299 (8.4) 112 (3.2)
48 (5.8) 22 (2.7)
1.41 (1.24–1.61)
1.34 (1.16–1.56)
(0.0)
0 (0.0)
3 (0.4)
2.42 (0.54–10.80)
2.01 (0.40–10.11) 0.76 0.73 0.73 0.74 0.85 Ref. 1.16 1.43 1.75
(3.1) (4.7) (7.5) (10.5) (16.1) (26.4) (19.7) (9.0) (2.9) (0.1)
149 211 326 383 538 764 646 368 159 1
(4.2) (6.0) (9.2) (10.8) (15.2) (21.6) (18.2) (10.4) (4.5) (0.0)
31 47 41 72 109 235 171 83 34 2
(3.8) (5.7) (5.0) (8.7) (13.0) (28.9) (20.8) (10.1) (4.1) (0.2)
1.52 1.46 1.29 1.15 1.06 Ref. 1.10 1.32 1.78
(1.26–1.83) (1.24–1.71) (1.12–1.48) (1.01–1.30) (0.95–1.18)
(67.2) (27.6) (2.7) (2.4) (0.1)
1,326 842 633 735 9
(37.4) (23.8) (17.9) (20.7) (0.3)
254 531 13 23 2
(30.9) (64.5) (1.6) (2.8) (0.2)
Ref. 2.12 (1.95–2.29) 10.02 (8.74–11.49) 13.63 (11.86–15.67)
Ref. 2.27 (2.07–2.48) 13.71 (11.75–16.00) 21.94 (18.67–25.78)
(0.99–1.22) (1.16–1.50) (1.48–2.14)
(0.60–0.95) (0.60–0.89) (0.61–0.86) (0.63–0.86) (0.75–0.97) (1.03–1.31) (1.23–1.66) (1.41–2.18)
9,071 (64.6) 4,961 (35.4) 2 (0.0)
2,201 (62.1) 1,342 (37.9) 2 (0.1)
493 (59.9) 330 (40.1) 0 (0.0)
Ref. 1.14 (1.06–1.22)
Ref. 1.03 (0.95–1.12)
1,370 (9.8) 11,891 (76.8) 773 (71.5)
411 (11.6) 2,891 (81.6) 243 (6.9)
61 (7.4) 697 (84.7) 65 (7.9)
1.14 (1.02–1.28) Ref. 1.32 (1.15–1.52)
1.03 (0.88–1.20) Ref. 1.43 (1.19–1.72)
363 370 620 2,190 2
108 95 122 498 0
0.90 (0.80–1.00) 0.91 (0.81–1.02) 0.99 (0.90–1.09) Ref.
0.91 (0.79–1.04) 0.78 (0.67–0.90) 0.90 (0.79–1.02) Ref.
1,646 1,608 2,344 8,432 4
(11.7) (11.5) (16.7) (60.1) (0.0)
(10.2) (10.4) (17.5) (61.8) (0.0)
(13.1) (11.5) (14.8) (60.5) (0.0)
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
139
Table 2 (Continued ) Total twin pairs, na
VD/VD (%)
Intrapartum CS (%)
VD/CS (%)
14,034
3,545
823
462 (13.0) 1,648 (46.5) 1,435 (40.5) 708 1,689 663 334 151
Hospital related indicators Type of hospital University 1,693 (21.1) Non-university teaching 6,471 (46.1) 5,870 (41.8) Non-teaching Mean number of twin deliveries per year in the hospital 0–24 2,553 (18.2) 25–49 6,718 (47.9) 50–74 2,727 (19.4) 75–99 1,325 (9.4) 100 711 (5.1)
(20.0) (47.6) (18.7) (9.4) (4.3)
Crude OR (95% CI)
Adjusted OR (95% CI)
116 (14.1) 361 (43.9) 346 (42.0)
1.13 (1.01–1.25) 1.02 (0.95–1.10) Ref.
1.46 (1.20–1.77) 1.10 (1.00–1.22) Ref.
136 432 133 77 45
1.05 Ref. 0.93 0.98 0.87
1.17 Ref. 0.81 0.70 0.63
(16.5) (52.5) (16.2) (9.4) (5.5)
(0.96–1.15) (0.84–1.01) (0.87–1.11) (0.74–1.03)
(1.04–1.31) (0.72–0.91) (0.58–0.84) (0.51–0.79)
Adjusted OR >1 was associated with a higher incidence of intrapartum cesarean section (CS) of one or both twins, and adjusted OR <1 was associated with a higher incidence of successful vaginal delivery of both twins. VD/VD, vaginal delivery both twins; intrapartum CS, cesarean section for both twins; VD/CS, vaginal delivery twin A followed by cesarean section twin B; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection. a For some subgroups, not all data were available. b p10 = 10th percentile.
is not necessarily the same as the position at the start of labor [21,28]. We do not have any information about chorionicity, as chorionicity is not scored in the PRN database. Chorionicity might be an important indicator, as some gynecologists deliver all monochorionic twins by CS [29]. Some indicators such as diabetes, hypertension, and drug use during pregnancy may be underreported because caregivers are not obliged to register them. Since the proportion of underreporting is expected to be similar in both the outcome groups, the observed odds ratios may approximate the ‘true’ strength of the association. Unfortunately, the PRN database does not provide information on other potential confounding factors, like fetal distress, possible indications for planned CS or VD, antenatal surveillance, and ultrasound examination. Finally, we did not have any information on the pregnant woman’s preference on mode of delivery. In conclusion, we found a high percentage of planned and successful VD of twins 32–41 weeks gestation. Fetal position, previous CS, nulliparity, advanced maternal age, and pre-eclampsia were prominent indicators for both planned and intrapartum CS. Further studies need to focus on neonatal and maternal outcomes. Based on these findings, a prediction model for the optimal mode of delivery for twins should be made. Acknowledgments None. References [1] Rossi AC, Mullin PM, Chmait RH. Neonatal outcomes of twins according to birth order, presentation and mode of delivery: a systematic review and metaanalysis. BJOG 2011;118:523–32. [2] Barrett JF, Hannah ME, Hutton EK, et al. A randomized trial of planned cesarean or vaginal delivery for twin pregnancy. N Engl J Med 2013;369:1295–305. [3] van der Garde M, Winkens B, Roumen FJ. Increased elective caesarean section rate is not associated with a decreased serious morbidity rate for twins 32 weeks’ gestation. J Obstet Gynaecol 2012;32:453–7. [4] NVOG. Richtlijn Meerlingzwangerschap. NVOG; 2011, Available at: http:// nvog-documenten.nl/index.php?pagina=/richtlijn/pagina. php&fSelectTG_62=75&fSelectedSub=62&fSelectedParent=75. [5] Steins Bisschop CN, Vogelvang TE, May AM, Schuitemaker NW. Mode of delivery in non-cephalic presenting twins: a systematic review. Arch Gynecol Obstet 2012;286:237–47. [6] Smith GC, Shah I, White IR, Pell JP, Dobbie R. Mode of delivery and the risk of delivery-related perinatal death among twins at term: a retrospective cohort study of 8073 births. BJOG 2005;112:1139–44. [7] Smith GC, Fleming KM, White IR. Birth order of twins and risk of perinatal death related to delivery in England, Northern Ireland, and Wales, 1994–2003: retrospective cohort study. BMJ 2007;334:576.
[8] Christopher D, Robinson BK, Peaceman AM. An evidence-based approach to determining route of delivery for twin gestations. Rev Obstet Gynecol 2011;4:109–16. [9] Wenckus DJ, Gao W, Kominiarek MA, Wilkins I. The effects of labor and delivery on maternal and neonatal outcomes in term twins: a retrospective cohort study. BJOG 2014;121:1137–44. [10] Ko HJ, Jun JK. Clinical factors associated with failed trials of labor in late preterm and term twin pregnancies. J Perinat Med 2014;42:449–55. [11] Fox NS, Gupta S, Melka S, et al. Risk factors for cesarean delivery in twin pregnancies attempting vaginal delivery. Am J Obstet Gynecol 2015;212: 106.e1–.e5. [12] Persad VL, Baskett TF, O’Connell CM, Scott HM. Combined vaginal–cesarean delivery of twin pregnancies. Obstet Gynecol 2001;98:1032–7. [13] Haest KM, Roumen FJ, Nijhuis JG. Neonatal and maternal outcomes in twin gestations 32 weeks according to the planned mode of delivery. Eur J Obstet Gynecol Reprod Biol 2005;123:17–21. [14] Zipori Y, Smolkin T, Makhoul IR, Weissman A, Blazer S, Drugan A. Optimizing outcome of twins by routine cesarean section beyond 37 weeks. Am J Perinatol 2011;28:51–6. [15] National Collaborating Centre for Women’s and Children’s Health (UK). Multiple Pregnancy: The Management of Twin and Triplet Pregnancies in the Antenatal Period. London: RCOG Press; 2011, September (NICE Clinical Guidelines, No. 129.) Available from: http://www.ncbi.nlm.nih.gov/books/NBK83105/. [16] American College of Obstetricians and Gynecologists, Society for MaternalFetal Medicine. ACOG Practice Bulletin No. 144: multifetal gestations: twin, triplet, and higher-order multifetal pregnancies. Obstet Gynecol 2014;123: 1118–32. [17] Vayssiere C, Benoist G, Blondel B, et al. Twin pregnancies: guidelines for clinical practice from the French College of Gynaecologists and Obstetricians (CNGOF). Eur J Obstet Gynecol Reprod Biol 2011;156:12–7. [18] Institute of Obstetricians and Gynaecologists, Royal College of Physicians of Ireland and Directorate of strategy and Clinical Care, Health Service Executive. Clinical practice guideline management of multiple pregnancy. 1st ed. 2012, Available at: http://www.hse.ie/eng/about/Who/vlinical/natclinprog/ obsandgynaeprogramme/pregmulti.pdf. [19] Stichting Perinatale Registratie Nederland. Grote Lijnen 10 jaar Perinatale Registratie Nederland. Utrecht: Stichting Perinatale Registratie Nederland; 2011, Available at: http://www.perinatreg.nl/uploads/173/123/10_jaar_ Perinatale_Zorg_in_Nederland_de_grote_lijnen.PDF. [20] Visser GH, Eilers PH, Elferink-Stinkens PM, Merkus HM, Wit JM. New Dutch reference curves for birthweight by gestational age. Early Hum Dev 2009;85: 737–44. [21] Engelbrechtsen L, Nielsen EH, Perin T, Oldenburg A, Tabor A, Skibsted L. Cesarean section for the second twin: a population-based study of occurrence and outcome. Birth 2013;40:10–6. [22] Breathnach F, McAuliffe F, Geary M, et al. Prediction of safe and successful vaginal twin birth. Am J Obstet Gynecol 2011;205:237.e1–.e7. [23] Lee HC, Gould JB, Boscardin WJ, El-Sayed YY, Blumenfeld YJ. Trends in cesarean delivery for twin births in the United States: 1995–2008. Obstet Gynecol 2011;118:1095–101. [24] Schmitz T, Carnavalet Cde C, Azria E, Lopez E, Cabrol D, Goffinet F. Neonatal outcomes of twin pregnancy according to the planned mode of delivery. Obstet Gynecol 2008;111:695–703. [25] Ford AA, Bateman BT, Simpson LL. Vaginal birth after cesarean delivery in twin gestations: a large, nationwide sample of deliveries. Am J Obstet Gynecol 2006;195:1138–42. [26] Schutte JM, Steegers EA, Schuitemaker NW, et al. Rise in maternal mortality in the Netherlands. BJOG 2010;117:399–406.
140
S.M.T.A. Goossens et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 195 (2015) 133–140
[27] Lisonkova S, Joseph KS, Bell R, Glinianaia SV. Effect of advanced maternal age on perinatal outcomes in twins: the impact of chorionicity. Ann Epidemiol 2013;23:428–34. [28] Dodd JM, Deussen AR, Grivell RM, Crowther CA. Elective birth at 37 weeks’ gestation for women with an uncomplicated twin pregnancy. Cochrane Database Syst Rev 2014;2. CD003582.
[29] Houlihan C, Knuppel RA. Intrapartum management of multiple gestations. Clin Perinatol 1996;23:91–116. [30] Hoffmann E, Oldenburg A, Rode L, Tabor A, Rasmussen S, Skibsted L. Twin births: cesarean section or vaginal delivery? Acta Obstet Gynecol Scand 2012;91:463–9.