Introduction

Over the last four decades, globally, the rate of twin pregnancy has been increased by a third (from 9.1 per 1000 births in 1980 to 12.0 per 1000 births in 2021), constituting one in 42 births or 1.6 million twin births each year [1]. The increment has mainly been attributed to the advancement of assisted reproductive technology [2–4] and postponement of parenthood to the later age [5–7], particularly in high countries. In Africa, although the practice of assisted reproductive technology and delaying childbearing age is not common, naturally conceived twin pregnancy is becoming more prevalent [8]. A current estimate suggested that about 80% of twin pregnancies occurred either in Africa or Asia [1].

Twin pregnancy results from a complex interaction of different factors such as maternal age, parity, family history of multiple pregnancies, and others [9, 10]. Due to inherent biological factors, a twin pregnancy is associated with various adverse perinatal outcomes and maternal obstetrics complications [11]. Preterm birth, low birth weight, twin to twin transfusion syndrome, perinatal deaths, and admission to neonatal intensive care unit (NICU) are among many adverse perinatal outcomes associated with twin pregnancy [12, 13]. For example, the rate of perinatal mortality among twin pregnancies could be up to six times higher compared to singleton pregnancies [11].

Maternal obstetrics complications such as hemorrhage, preeclampsia, and maternal deaths are also common among twin pregnancies [14, 15]. Further, twin pregnancies increase both medical and surgical interventions including caesarean section [16, 17]. Mothers with twin pregnancies are also about six times more likely to be hospitalized due to complications during pregnancy compared to those with singleton pregnancies, posing additional healthcare costs [18].

Despite this, in most developing countries including Ethiopia, little is known about the current estimate of twin pregnancy and related adverse perinatal outcomes. Thus, for the proper resource allocation and management of mothers with twin pregnancies, there is a need to estimate the incidence of twin pregnancies and related adverse perinatal outcomes and understand the associated factors. The purpose of this study, therefore, to estimate the incidence of adverse perinatal outcomes and identify factors associated among twin pregnancies in Ethiopia.

Methods and materials

Results

Socio-demographic and obstetrics characteristics

A total of 10,850 births were recorded during the study period at Adama Hospital Medical College. Of these, twin-births accounted for 354 (3.26%). The mean maternal age at birth was 26.6 years (SD± 4.7years). About 44% of twin births were recorded within the age range of 25–29 years. Regarding obstetrics characteristics, the mean parity was 2.6 (SD± 1.7) and the majority of mothers (98.1%) had 2–4 para. Three hundred sixteen mothers had ANC follow up and ultrasound examinations were performed for 54% and 97.2% of mothers during ANC follow up and during labour, respectively (See Table 1 for the details).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic factors & maternal health service utilization characteristics among mothers who gave twin birth at AHMC, Ethiopia.

https://doi.org/10.1371/journal.pone.0275307.t001

Obstetric characteristics

Regarding obstetrics characteristics, the mean parity was 2.6 (SD± 1.7) and the majority of mothers (98.1%) had 2–4 para. Approximately half of the mothers gave births at term and nearly the same number of mothers before term (less than 37 completed weeks). While half of the mothers gave birth through spontaneous vaginal birth, 46.6% of them gave birth by caesarean section. Moreover, the commonest indication for caesarean birth was found to be fetal mal-presentation (47.7%) followed by non-reassuring fetal heart rate patterns (NRFHRP) [10.5%] (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Obstetrics characteristics of the twin-births recorded at AHMC, Ethiopia.

https://doi.org/10.1371/journal.pone.0275307.t002

Adverse perinatal outcome

Of the total 322 paired (twin) birth records (644 individual birth records), 199 (61.8%) paired births had at least one adverse perinatal outcome on one or both twins. The commonest adverse perinatal outcomes observed in this study were low birth weight (50.8%) followed by preterm birth (46.6) and NICU admission (24.7%). As expected although, 24.5% of newborns had a low one-minute Apgar score, only 5.0% of them had a low five-minute Apgar score. Table 3 presents these adverse perinatal outcomes and others.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Percentage distribution of adverse perinatal outcomes.

https://doi.org/10.1371/journal.pone.0275307.t003

Factors associated with perinatal outcome

In a multivariable logistic regression model, factors such as younger maternal age and not having an ultrasound scan during ANC follow-up were significantly associated with Adverse perinatal outcomes after controlling for confounding factors. The odds of having adverse perinatal outcomes among twin births recorded to mothers within the age range of 15–24 years were four (AOR = 4.1, 95% CI; 1.3, 12.5) times higher than those births recorded to mothers within the age range of 35 years and above. In addition, not having ultrasound scan during ANC follow-up increased the odds of adverse perinatal outcomes by two folds (AOR = 2.0, 95% CI: 1.2, 3.1) (Table 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Multivariable logistic regression analysis for factors associated with adverse perinatal outcomes.

https://doi.org/10.1371/journal.pone.0275307.t004

Discussion

The purpose of this study was to estimate the incidence of twin pregnancies and related-adverse perinatal outcomes and identify factors associated with adverse perinatal outcomes in twin pregnancies. The findings indicated that the incidence of twin pregnancies was 32.6 per 1000 births and close to 62% of twin birth records had at least one adverse perinatal outcome on one or both twins. Moreover, the findings revealed that younger maternal age (15–24 years) and not having ultrasound scan during antenatal care were associated with adverse perinatal outcomes.

The incidence of twin pregnancies observed in the current study (32.6 per 1000 births) was higher than the one reported in the studies conducted in Addis Ababa (24 per 1000) [19], Mekelle (14 per 1000) [20], and Gondar (14.9 per 1000) [21]. The high incidence of twin pregnancy in the current study could be partly attributed to the study setting. That is, since Adama Hospital Medical College is a referral hospital, it receives and manages mothers with a high-risk pregnancy from the local primary health facilities which probably might increase the number of twin pregnancies in the hospital. The findings of this study, however, in agreement with a findings reported from south west Ethiopia [22] and slightly lower than those reported in other African countries such as Niger, Congo, and Nigeria [8, 23].

In this study, the majority of twin births (44.4%) were recorded among mothers within the age range of 24–29 years. However, in a study conducted in southwest Ethiopia, it was found that more than half of twin births occurred among mothers who were under the age of 24 years [22]. Such variations in the sociodemographic characteristics related to twin pregnancies might be partly explained by the difference in socio-cultural issues such as early marriage.

Regarding the mode of births among twin pregnancies, it was found that 50% of mothers gave birth through spontaneous vaginal deliveries while 47% required caesarean section, which is comparable with the findings reported in the previous studies [24]. Twin pregnancy has been identified, alongside factors, such as advanced maternal age and maternal request without clinical indication, as one of the contributing factors for increasing rates of caesarean section [17].

In this study, it was found that the incidence of adverse perinatal outcomes was 61.8%, with low birth weight being the commonest adverse perinatal outcomes followed by preterm birth and NICU admission. The findings of these adverse perinatal outcomes, particularly low birth weight, were comparable with the one reported in a study conducted in Nigeria [25], but lower than the one reported in some Asian countries such as Thailand, India, and Korea [26, 27]. The observed discrepancies between the current findings and those reported from Asian countries might be due to contextual factors, such as the availability of advanced medical technologies that help for the survival of high-risk newborns (if born too early), increasing the rate of preterm, low birth weight, and NICU admission.

In this study, it was found that the odds of adverse perinatal outcomes were higher among the younger maternal age group. This finding is somewhat surprising and in contrast with findings from several studies [6]. However, similar findings were reported in a few studies [23, 28], warranting further study that explicitly addresses the effect of maternal age on adverse perinatal outcomes in twin pregnancies, preferably with a population-based sample. Not having an ultrasound scan during antenatal care also was found to be significantly associated with an increased risk of adverse perinatal outcomes in twin pregnancies. This might be due to the fact that pregnant women with twin pregnancies may not early identified and managed to avert adverse perinatal outcomes.

This study was not without some limitations. As it was based on retrospective hospital data, information on some sociodemographic and maternal-related variables that may influence the perinatal outcomes in twin pregnancies was not available, warranting a future study that addresses all the relevant variables in relation to the perinatal outcomes among twin pregnancies. Although antenatal ultrasound had been performed for most of the pregnant women, there were some limitations in the registry, meaning some variables including chorionicity were not clearly documented and were not assessed in this study. As the focus of this study was on fetus/newborn outcomes in twin pregnancies and some maternal-related variables were not available, the maternal outcomes in twin pregnancies were not included in this study, warranting further studies. Although assisted reproduction is less common in the study setting, we were not able to differentiate twins occurred after assisted reproduction, particularly in vitro fertilization (IVF). Furthermore, although the adverse perinatal outcomes in twin pregnancies were comprehensively assessed, the extent by which identified factors contribute for each adverse perinatal outcome might be slightly varies, indicating the need to address each perinatal outcome separately, particularly in settings where targeted-interventions are more feasible. Similarly, the adverse perinatal outcomes in twin pregnancies better compared with that of singleton pregnancies to indicate differential risks, whenever feasible.

Conclusion

The incidence of adverse perinatal outcomes in twin pregnancies was high in this study. Factors such as younger maternal age and not having an ultrasound scan during antenatal care were found to be the strong predictors of adverse perinatal outcomes, indicating the for early screening and continuous monitoring of women with twin pregnancies to reduce the adverse perinatal outcomes.

Acknowledgments

The authors would like to acknowledge Adama Hospital Medical College for their support for this research. The authors are very much grateful to record room workers and data collectors.