Study selection

From a systematic search of the literature to identify the maternal and fetal adverse effects of weight management interventions in pregnancy, 14,832 potentially relevant records were obtained (up to 31 March 2011). A search of the reference lists of the relevant articles led to the identification of 26 further citations. After reviewing the abstracts, the full texts of 180 papers were obtained for detailed assessment. After exclusion of 154 publications, 26 papers were included in the review. Figure 35 provides details of the process of study selection.

FIGURE 35

Flow chart of study identification and selection for the evaluation of adverse maternal and fetal outcomes.

Of the included studies, two were RCTs (involving 277 women)^129,132 and 24 were observational studies (19 cohort studies and five case–control studies, involving 468,581 women).^{63,64,67,68,70,73–77, 80,85,89,133–143} The studies evaluated the effect of dietary, physical activity and other lifestyle interventions in pregnancy on maternal and fetal outcomes. Appendices 7 and 10 provide details of the included RCTs and observational studies, respectively, that assessed the adverse effects of outcomes.

Quality of the included studies

Randomised controlled trials

The quality of the two included RCTs^129,132 is shown in Figure 36. The details regarding sequence generation, allocation concealment and blinding for subjective outcomes were unclear in both studies. A detailed quality assessment of the included RCTs is provided in Appendix 8.

FIGURE 36

Quality of the included RCTs for the adverse effects review.

Results

The adverse outcomes included in the review were defined as those that occurred unintentionally with potential harm to the mother or baby. We also included those outcomes that may have been the direct result of the intervention itself, for example risk of preterm delivery due to strenuous physical exercise.

Observational studies

A total of 18 studies^{68,73–76,80,85,89,133–139,141–143} observed the effect of diet on maternal and fetal outcomes. The majority of the included studies produced data on the effects of a severe reduction in caloric intake in extreme conditions such as war or famine (Table 7). The studies on physical activity included women undergoing exercises of various intensities or other recreational physical activity in pregnancy. The rates of congenital abnormalities such as neural tube defects (NTDs) were observed in those following dietary interventions that aimed to significantly reduce weight¹³³ or in those intaking food with a very high- or a very low-glycaemic index.¹⁴³ The risks of coronary artery disease, metabolic syndrome, breast cancer and diabetes were studied in infants born to mothers who were severely diet restricted owing to famine.^68,135,139

TABLE 7

Adverse outcomes associated with any diet or physical activity in pregnancy.

The observational studies on physical activity in pregnancy did not show any significant adverse maternal or fetal outcomes. This was consistently observed for different activities of varying severity.

The detailed clinical characteristics of the included studies for the evaluation of adverse effects are provided in Appendix 10.

Summary

The review of adverse effects identified two RCTs and a relatively large number of observational studies. The data from the observational studies showed a possible association between extremes of diet (exposure to famine) and adverse outcomes; however, there was no evidence to suggest that dietary interventions evaluated in the review or currently offered in clinical practice could be associated with adverse maternal or fetal outcomes. Physical activity in pregnancy and maternal and fetal outcomes were studied in the randomised trials and observational studies. Various forms of physical activity such as structured exercises, running and recreational activities of differing intensities were not associated with adverse maternal and fetal outcomes.

The strength of the review is the systematic search for evidence using a broad search strategy. The inclusion of both randomised and non-randomised data including case series has ensured that the review identifies the evidence for all potential adverse effects of interventions. The review was limited by the RCTs being of poor quality. A large proportion of the evidence from the observational studies was devoted to extremes of diet rather than the components of a balanced healthy diet. There was insufficient evidence on popular diets such as the Atkins diet, the Slimming World diet and ‘high-protein’ diets. The studies on physical activity in pregnancy were mainly concerned with cord abnormalities and abnormal fetal heart rate patterns. The data from RCTs on women undergoing physical activity in pregnancy show no effect on gestational age at delivery or preterm delivery provide reassuring evidence on the safety of these interventions for these outcomes.