CLINICAL ARTICLE A population-based cohort study of stillbirth among twins in Lusaka, Zambia Elizabeth M. Stringer a, ⁎, Carla Chibwesha b , Marie Stoner c , Bellington Vwalika d , Jessica Joseph e , Benjamin H. Chi a,b , Eugene Kaunda d , William Goodnight a , Jeffrey S.A. Stringer a a Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA b Centre for Infectious Disease Research in Zambia, Lusaka, Zambia c Department of Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA d Department of Obstetrics and Gynaecology, University Teaching Hospital, Lusaka, Zambia e Clinton Health Access Initiative, Boston, MA, USA abstract article info Article history: Received 7 August 2014 Received in revised form 17 December 2014 Accepted 13 March 2015 Keywords: Africa Stillbirth Twins Urban Zambia Objective: To determine rates of stillbirth and the associated risk factors for stillbirth among twins delivered in Lusaka, Zambia. Methods: A retrospective cohort analysis was conducted of singletons and twins delivered at 26 public sector facilities between February 1, 2006, and May 31, 2013. Data were obtained from the Zambian Electronic Perinatal Record System. Risk of stillbirth was estimated using logistic regression. Results: Overall, 260 657 singletons and 4021 twin pairs were included. There were 5105 stillbirths; 317 twins were stillborn. The crude stillbirth rate for twins was 39.4 per 1000 births (95% confidence interval [CI] 35.2–43.7) whereas the rate for singletons was 18.4 per 1000 births (95% CI 17.9–18.9; P b 0.001). Factors associated with stillbirth among twins were increased interval between delivery (N 60 minutes), low birth weight (b 2500 g), birth order (being the second-born), and difference in birth weights (N 30% discordance). Conclusion: Twins were at an increased risk of stillbirth. Improved understanding of factors associated with stillbirth in this population could help to improve perinatal outcomes globally. © 2015 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. 1. Introduction The incidence of twins in high-income countries such as the USA has increased from 18.9 per 1000 live births to 33.9 per 1000 live births dur- ing the past 20 years, largely owing to a marked rise in the use of assisted reproductive technologies [1]. Globally, twin birth outcomes are consistently worse than are those of singletons [2,3]. Twins repre- sent 2% of all births in the USA [4]; however, this population comprises 12% of preterm births, 15% of neonatal deaths, and 10% of stillbirths [5]. Factors known to be associated with increased stillbirth rates among twins born in high-income countries include monochorionicity [6],a difference in fetal weights of more than 20% [7], and continuation of pregnancies beyond certain gestational ages according to chorionicity [6,8]. For example, a national US consensus panel on timing of twin births recommended delivery at 38 weeks for dichorionic twins and 34–37 weeks for monochorionic/diamnionic twins [9]. However, still- birth rates among twins in high-income countries have progressively decreased over the past 30 years, possibly owing to improved detection of chorionicity with increased fetal surveillance and uptake of scheduled deliveries rather than spontaneous labor [10–13]. Numerous studies from the USA and Canada [10–14] suggest that elective early delivery of twin gestations correlates with overall reductions in stillbirths, but the trade-off is increased rates of prematurity. An increasing amount of information is available on twin gestation and stillbirth in low-resource countries [3,15,16]. However, reported risk factors associated with twin pregnancy and stillbirths in this setting do not always correlate with those identified in the USA. Enhanced understanding of risk factors associated with stillbirth among twin pregnancies could help low-income countries to develop specific sur- veillance strategies and interventions to improve outcomes. The aim of the present study was, therefore, to investigate the risk of stillbirth among twin gestations in Zambia and to identify specific risk factors for stillbirth in this group. 2. Materials and methods A retrospective cohort analysis was conducted of twin and singleton births at public sector facilities in Lusaka, Zambia. The routine clinical data used in this analysis had been collected by the Zambia Electronic Perinatal Record System (ZEPRS) between February 1, 2006, and May International Journal of Gynecology and Obstetrics 130 (2015) 74–78 ⁎ Corresponding author at: Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill, 3010 Old Clinic Building, Campus Box 7516, Chapel Hill, NC 27599, USA. Tel.: +1 919 260 1601; fax: +1 919 966 6377. E-mail address: elizabeth_stringer@med.unc.edu (E.M. Stringer). http://dx.doi.org/10.1016/j.ijgo.2014.12.015 0020-7292/© 2015 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. Contents lists available at ScienceDirect International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo