A retrospective cohort study was carried out in the Obstetrics Department of Government medical college, Chandigarh from December, 2011 to November 2015 after approval from the Institute’s Ethical Committee. The study population included all twin foetuses who delivered after 20 weeks gestation in this institute. Foetuses of patients with chronic disease like chronic hypertenstion, diabetes mellitus etc., were excluded. Foetal data were obtained from the institute’s labour room delivery data entry register along with analysis of individual mother’s records. Analysis of foetal data was done according to chorionicity of the placenta and data was also compared among foetuses of patients who were booked at our institute (atleast 4 antenatal visits at our hospital) and with those referred to our institute.

Chorionicity was determined by ultrasound in the antenatal period or in postnatal period by analysis of placenta and sex of the baby was considered whenever there was doubt about the chorionicity. The chorionicity was classified as: Dichorionic (DC) and Monochorionic (MC). Three groups were defined based on chorionicity and amniotic sac: Diamniotic Dichorionic (DADC), Diamniotic Monochorionic (DAMC), Monoamniotic Monochorionic (MAMC).

The foetal parameters that were included: gestational age at birth, mode of delivery, intrauterine death, discordant twin, live birth, and weight of the twins at birth. Other outcomes which were included in the study were Apgar scores and gross congenital malformations which were diagnosed by ultrasound examination during the antenatal period or by clinical examination of the neonate.

The gestational age at birth was defined as extremely preterm <28 weeks, very preterm (28-<32 weeks) and moderate or late preterm (32-<37 completed weeks of gestation). Discordant twin was diagnosed when estimated twin weights differences were greater than 25 percent, based on the weight of the larger twin. Newborns weighing less than 2500 g were classified as low birth weight, those weighing less than 1500 g as very low birth weight and extremely low birth weight when it was less than 1000g. Apgar score less than 7 at 5 minute of birth was the criterion for immediate neonatal morbidity.

Discrete categorical data were presented as n (%); continuous data were given as mean±SD and range. Normality of quantitative data was checked by measures of Kolmogorov-Smirnov tests of normality. As our data was normally distributed data, means of different parameters for three groups (DADC, DAMC, MAMC) were compared using One-Way ANOVA followed by Post-Hoc Multiple Comparisons test. Proportions were compared between the referred and booked groups using the chi-square test. All statistical tests were two-sided and performed at a significance level of α=0.05. Analysis was conducted using IBM SPSS STATISTICS (version 22.0).

During this study period, 862 twin foetuses were born. Out of total, 564/862 (65.2%) twins born to patients who were referred from other hospital and 298/862 (34.5%) were born to patients who were booked at our hospital. Commonest presentation of twins was both vertex in 205/431 (47.5%), 78/431 (18%) were vertex breech, 67/431 (15.5%) were both breech, 61/431 (14%) were breech vertex and 21/431 (4.8%) had transverse lie of first twin. Cesarean section was done in 224/431 (51.9%) and cesarean section for delivery of second twin was done in five cases.

With respect to chorionicity, 258 (59.8%) pairs were DADC, 162 (37.5%) were DAMC and 11 (2.5%) were Monoamniotic Monochorionic (MAMC) twins. Out of 862 pairs, 608/862 (70.5%) were premature born. There were 62/422 (14.6%) vs 02/186 (1.07%), p<0.001 indicated premature born twins in referred and booked patients.

Foetal parameters according to chorionicity are presented in [Table/Fig-1]. Foetal data among referred and booked patients are given in [Table/Fig-2].

Intrauterine foetal deaths were 35 (both twins in 6 pregnancies and one twin in 23 pregnancies). Congenital malformations were hydrocephalus in 1, one case of ambigous genitalia 1 and one case of occipital encephalocele in dichorionic pregnancy. Monochorionic group have 2 cases of bilateral club foot, anancephaly in 1, 1 case of exomphalos, absent limbs and absent umbillical cord and one case was of acardiac twin.

The prevalence of twins in this study was 22/1000 of birth that is on higher side when compared to other studies by Deepthi HR et al., and Tomer SP et al., [7,8]. This is explained by referral of patient’s with multiple pregnancies from near by health care centre to our Institute. In this study 86.6% foetuses were born through spontaneous conception that co-relates with Assuncao RA et al., who had reported only 3.8% rate of twins conception by assisted reproduction and Behávková K et al., reported 80% rate of spontaneous conception in twin pregnancy [15,16].

As per a WHO fact sheet (2012), India is amongst the top most country contributing to world’s premature deliveries and rising incidence of multiple pregnancies is one of the contributing factor [17]. In present study, out of 70% prematurely delivered twins, 10% of twins baby were born at very preterm gestation (<before 32 weeks of gestation) and 60% were delivered at late preterm gestation (>32 to <37 weeks). Martin JA et al., has reported 60.4% rate of preterm birth in twins and 12.2% of them were born at very preterm (<32 weeks) gestation in United States population [6].

The mean gestational age at birth was 34.9±2.6 weeks in this study whereas, it was reported as 35.2 weeks, 34.6 weeks (SD: 3.9, range: 21.3 to 40.3), and 33.48 weeks±3.32 weeks in a study by Martin JA et al., Assuncao RA et al., and Singh and Trivedi respectively [6,15,18]. According to chorionicity there were no differences in mean gestational age in present study but Assuncao RA et al., did find significant difference in gestational age among three group and it was lesser in MCMA and MCDA twins as compared to Dichorionic group [15], 32.9 weeks, 33.6 weeks and 35.4 weeks respectively. Although, mean gestational age at birth was significantly different in referred and booked patients in this study that may be due to higher percentage of induced preterm delivery in referred patients.

Overall 88% of babies were low birth weight (<2.5 kg) in this study and this incidence was higher than reported by Martin JA et al., Behavkova K et al., and Bangal VB et al., in 57.5%, 78.67%, and 82% of twins respectively that may be explained by poor nutritional status and low BMI in our population [6,16,19]. Mean birth weight of twins was 1.89±0.48 kg in this study whereas it was 2.33 kg by Martin JA et al., and 1.95 kg in a study by Singh and Trivedi [6,18]. Birth weights were significantly different among DADC, DAMC, MAMC group and also in referred and booked patients due to higher percentage of discordant twin and intrauterine foetal death in monochorionic group and in referred patients.

Growth discordancy was observed in 8.9% of twins in this study lower than reported by Assuncao RA et al., (12.8%) and Puccio G et al., (19.35%). This may be due to difference in definition of discordant twins that was taken as twin weight difference of 20%, whereas, it was taken as 25% twin weight difference in this study [15,20]. Although, no definitive causes of discordant twin have been determined but it has been co-related with conception with ovulation induction, multiple pregnancy reduction and foetal malformation by Audibert F et al., [21]. In this study, discordant twins were more commonly associated with monochorionic pregnancy that co-relates with increased perinatal morbidity and mortality in monochorionic pregnancy.

In this study, 35 (4.5%) of twins had intrauterine death and maximum rate of intrauterine death were in monochorionic monoamniotic twins 06 (27.3%), followed by monochorionic diamniotic 20 (6.2%) and was lowest in dichorionic diamniotic twins 09 (1.7%). Assuncao RA et al., reported 38 (6.6%) rate of intrauterine foetal deaths in twin pregnancy and 26/216 (12%) were from monochorionic group and 9/350 (2.5%) from dichorionic group, and three were from unknown group [15]. Hillman SC et al., has reported intrauterine foetal demise in monochorionic and dichorionic twins in 15 and 3% of twins whereas in this study it was 7.5% and 1.7% in monochorionic and dichorionic foetuses respectively [22].

Growth discordancy was observed in 43% of intrauterine death whereas pregnancy induced hypertension and antepartum haemorrhage were seen in 31% and 20% of intrauterine death respectively. Mahony R et al., reported twin -twin transfusion syndrome as the most common cause of intrauterine death before 30 weeks gestation whereas only 1 case of intrauterine death with twin-twin transfusion syndrome was noticed in this study as most of the intrauterine demise were after 31 weeks in this study except three were at 21, 25 and 28 weeks of gestation, all were from monochorionic group with two presented with antapartum haemorrhage and 1 with TTTS [23].

Most important is the role of antenatal care, out of 35 (4.0 %) IUFD, 30 (5.3%) were in unbooked patients and 05 (1.7%) intrauterine death were in booked patients which were observed in monochorionic group only. There were intrauterine death of both foetuses in 6 twins and foetal distress was observed in another twin in 10/23 (43%) twins who had one intrauterine death in utero that suggest increase risk of neonatal morbidity in second twin also.

This study revealed intrauterine death in all normally grown foetus at around 38±1.7 weeks in dichorionic group and in monochorionic group at 31.9±6.29 weeks. This observation was conflicting to that reported by Mahony R et al., who had reported low risk of intrauterine death in normally grown foetus after 34 weeks [23].

Twin pregnancy always presumed to be a risk factor of increased foetal maformations but in this study only 8/862 (0.9%) had some foetal malformations. Other studies had reported foetal maformations in 12.8% and 6.6% of twin foetuses by Assuncao RA et al., and Puccio G et al., respectively [15,20]. None of the congenital malformations were associated with growth discordancy in this study except acardiac twin. Low incidence of foetal malformations may be due to earlier termination of pregnancy by better availability of sonography nowadays.

There was no significant difference in 5-minute Apgar score of ≤ 7 among all the three groups but it was significantly different when compared between referred and booked patient 24 (4.2%) vs 05 (1.7%), (p, 0.04) respectively that may be explained by more number of referred patients with complications.

Major drawback of this study is being a retrospective study and non-availability of neonatal data regarding survival rate after birth. Reason being limited availability of intensive care unit at our hospital as compared to number of premature delivery that may have given variable results in those neonates who were not able to get intensive care.

Monochorionic pregnancy needs more frequent antenatal sonography from early gestation because of association with discordant twins and subsequent risk of intrauterine deaths. Normally grown foetus in dichorionic pregnancy seeks frequent follow-ups in later gestation and delivery should be conducted at optimum time to prevent intrauterine deaths at advanced gestation.