A retrospective study design and a data collection protocol that adhered to the tenets of the Declaration of Helsinki were used to collect information on pregnant women with singleton foetuses, who delivered at gestation ages of ≥ 24 weeks at the study institute in the year 2012. Women with multiple pregnancies and those who delivered earlier than 24 gestation weeks were excluded. Details of pregnancy, including demographic details, prior and current obstetric, medical and surgical histories, details of personal risk behaviours, family histories, types of conception, details of clinical exams and laboratory investigations as part of antenatal care, trimester specific ultrasound examinations and foetal growth assessments, details of delivery and post-partum period were entered in medical records, that were transcribed to an electronic database. Women were assessed for height and weight at baseline and a baseline body mass index was derived. Thyroid disorders were screened for at baseline through a universal screening protocol. We defined gestational diabetes, based on the International Association of Diabetes and Pregnancy Study Group (IADPSG) criteria [9]. Pregnancy induced hypertension was defined by using the criteria which was developed by the International Society for the Study of Hypertension in Pregnancy and it included proteinuria as a criteria [10].

The details of labour and delivery, including gestational age at delivery, type of labour-induced or spontaneous, mode of delivery, and obstetric complications were recorded. The details of the baby, including neonatal morbidity and mortality, NICU admission, APGAR scores, still born status and intrauterine foetal deaths were recorded.

Assisted conception was considered to include all invasive and non-invasive methods like OI, IUI, IVF, ICSI. Primary infertility is defined, for clinical purposes, as the inability to conceive within two years of exposure to pregnancy (sexually active, not using contraceptives, non-lactating) among women who are aged 15 to 49 years.

The data was initially entered into a Microsoft Excel spreadsheet and it was then exported to SPSS statistical software (version 14.0) for statistical analysis. The primary outcome of interest was preterm births that were defined as a spontaneous onset of labour and those with preterm rupture of the membranes, which occurred before 37 completed gestation weeks. The data was initially explored for normality of distribution by using measures like skewness and kurtosis, as well as the Shapiro Wilk test for Normality. The frequency distribution, point estimates and 95% CI around point estimates were determined. Bivariate analysis was done by using One Way Analysis of Variance (ANOVA) test for continuous variables and Chi-square or Fisher’s Exact test for categorical variables. A univariate logistic regression model was used to determine odds ratios and the 95% confidence intervals for outcomes which were associated with exposure. Factors that were significantly different between the two groups were further assessed by using a multivariate logistic regression model. We considered a p value of <0.05 as statistically significant for this study.

This study included 6,712 women who were pregnant with singleton foetuses, who delivered at the study institute in the year 2012. Four hundred and sixty (6.85%, 95% CI: 6.27, 7.48) women in the study population conceived with assistance. Methods of assistance included OI (n=283), IVF (n=99), IUI (n=72) and ICSI (n=10). The characteristics of the study population, which were stratified by conception, have been presented in [Table/Fig-1].

On bivariate analysis, assisted conceptions were significantly associated with preterm gestations of < 37 weeks (p=0.009). On further exploration of preterm births by using bivariate analysis, assisted conceptions were significantly associated with preterm births at 32-36 gestation weeks (p=0.02) but not with preterm births at < 32 gestation weeks (p=0.43). Birth weights of < 1,500 grams (p=0.99) were not significantly associated with assisted conceptions; however, birth weights which were between 1,500 and 2,499 grams (p=0.006) were significantly associated with assisted conceptions on bivariate analysis. Assisted conceptions were not significantly associated with small for gestational age babies (p=0.09), still births (p=0.56), or neonatal deaths (p=0.89). Birth weight (1,500-2,499 grams) and caesarean sections were significantly associated with assisted conceptions after adjusting for maternal age, body mass index, parity, pre-gestational and gestational diabetes, and pre-pregnancy hypothyroidism [Table/Fig-2]. Birth weights should be lower in babies who are delivered at earlier gestation weeks (preterm deliveries). The association of birth weight (1,500-2,499 grams) with assisted conceptions, which was explored further by adding gestational age at delivery to the multivariate logistic regression model, showed that birth weights (1,500-2,499 grams) were not significantly associated with assisted conceptions (adjusted OR: 1.12, 95% CI: 0.81, 1.55).

The results were further explored and stratified, based on type of assisted conception. OI was not associated with gestational age at delivery of <32 weeks (p=0.55), gestational age at delivery of 32-36 weeks (p=0.52) or birth weights of < 1500 grams (p=0.56), small for gestational age babies (p=0.06), caesarean sections (p=0.71) or still born babies (p=0.19) in the regression model. IUI was not associated with gestational age at delivery of <32 weeks (p=0.34), gestational age at delivery of 32-36 weeks (p=0.85) or birth weights of < 1500 grams (p=0.60), small for gestational age babies (p=0.51), or caesarean sections (p=0.38) in the regression model. IVF was associated with gestational age at delivery of <32 weeks (p=0.02), gestational age at delivery of 32-36 weeks (p=0.001) and caesarean sections (p<0.001), but it was not associated with birth weights of < 1500 grams (p=0.38), small for gestational age babies (p=0.53) or still born babies (p=0.92) in the regression model.

Assisted conceptions were associated with preterm gestations, mild low birth weights and an increased incidence of caesarean sections, but they were not associated with small for gestational age babies and perinatal mortality on bivariate analysis. The results of this study indicated that women who conceived with assistance were older or nulliparous, with a higher prevalence of morbidity, like obesity (body mass index >30), pre-gestational diabetes mellitus, gestational diabetes and pre-pregnancy hypothyroidism as compared to women who conceived naturally in this population. The difference in maternal characteristics between women who conceived naturally and with assistance is important, as these characteristics can act as confounders that affect perinatal outcomes. After adjusting for these differences in maternal characteristics, assisted conceptions were associated only with a higher incidence of caesarean sections.

Several studies worldwide have reported differing results for associations of adverse perinatal events with assisted conceptions, although they are limited by a lack of adequate power to determine associations with reasonable accuracy [11–18]. Several of these studies were limited by a lack of adequate power to determine associations with reasonable accuracy. A systematic review of controlled studies on assisted conceptions reported an increased incidence of adverse perinatal events- preterm deliveries (pooled relative risk 2.04, 95% CI: 1.80-2.32), low birth weight babies (pooled relative risk 1.70, 95% CI: 1.50-1.92), caesarean sections (pooled relative risk 1.27, 95% CI: 1.16-1.40), small for gestational age babies (pooled relative risk 1.54, 95% CI: 1.44-1.66) and perinatal mortality (pooled relative risk 1.68, 95% CI: 1.11-2.55)- with assisted conceptions [19].

Several of these reported associations are inter-linked, which include caesarean sections for preterm deliveries or planned interventions for small for gestational age babies, and consequent lower birth weights of the babies, with an increased risk for morbidity and mortality. We found that the association between preterm delivery and low birth weight were no longer significant, after adjusting for these confounding factors. Preterm delivery of a baby may be spontaneous or it may be a planned decision which is based on the health of the mother and/or the foetus, including suspicion of foetal compromise. Obstetricians may be more cautious with assisted conceptions, considering the difficulties that women have in conception, the relatively older ages, nulliparous status, associated medical co-morbidities and the relative importance of such pregnancies and they may tend to deliver the babies by using planned elective procedures if there is any suspicion of foetal or maternal distress. The availability of neonatal services (neonatal intensive care units and nurseries) has improved outcomes of preterm deliveries, including very preterm deliveries, thus making the decisions for planned preterm deliveries more feasible. Unfortunately, we could not distinguish between planned preterm and spontaneous preterm deliveries with reasonable accuracy in this study; a prospective study design will help us in making that distinction.

The documentation and the retrospective cohort study design may be considered as strengths of this study. However, it does have several limitations. It was done at a single tertiary care centre. While the single centre nature of this study offers advantages in terms of standardized procedures and documentations, it does have a limitation in terms of representativeness of the population and hence, in terms of generalizability of the results to a larger population. In conclusion, the results of this study indicate that women with assisted conceptions do not have significantly different perinatal outcomes as compared to natural conceptions in singleton pregnancies, if we adjust for the difference in maternal characteristics. Women with assisted conceptions tend to have higher rates of caesarean sections, probably as part of a planned obstetric approach that minimizes risk to the foetus and mother and intervenes on the side of caution. Further studies are required to determine as to whether the rates of caesarean sections can be reduced with more term vaginal deliveries in women with singleton foetuses, after undergoing assisted conceptions.