How should women be advised on weight management in pregnancy?BMJ 2012; 344 doi: http://dx.doi.org/10.1136/bmj.e2774 (Published 17 May 2012) Cite this as: BMJ 2012;344:e2774
- Lucilla Poston, head of division of women’s health, King’s College London,
- Lucy C Chappell, clinical senior lecturer in maternal and fetal medicine
Obesity in pregnant women has considerable resource implications, with increased odds of caesarean or instrumental delivery, haemorrhage, infection, longer duration of hospital stay, and need for neonatal intensive care.1 At a time when more than half the women of reproductive age in the United Kingdom are overweight or obese,2 any analysis of weight management interventions in pregnancy is timely and welcome. In the linked paper (doi:10.1136/bmj.e2088), Thangaratinam and colleagues present a comprehensive and well conducted meta-analysis of studies that have tried to improve maternal weight and obstetric outcomes through dietary and lifestyle interventions across the body mass index (BMI) range.3 However, many of the included studies are of small size and limited quality. The authors conclude that interventions can improve some outcomes for the mother and baby and that dietary advice, rather than advice on physical activity, is most effective. Importantly, there was no evidence of harm.
Internationally, the guidelines for weight management in pregnancy vary. Because the US Institute of Medicine (IOM) recommends limits for gestational weight gain (table⇓),4 most published intervention studies focus mainly on this parameter. In the UK, however, the National Institute for Health and Clinical Excellence (NICE) guidelines for weight management in pregnancy do not advise regular weighing of pregnant women beyond their first visit because evidence for an effective intervention to improve clinical outcomes in a UK population has been lacking.5
Does the meta-analysis by Thangaratinam and colleagues provide the evidence needed for NICE to reassess the guidelines? We think that this would be premature. Although the findings suggest that interventions in pregnancy are safe and can produce modest reductions in gestational weight gain, control and intervention groups did not differ in the proportion of women who achieved IOM gestational weight gain limits. It is therefore not surprising that there was no effect on clinically relevant outcomes, including birth weight or macrosomia, and no reduction in the caesarean section rate. Although a positive benefit was reported for pre-eclampsia and gestational hypertension, the quality of evidence for these measures was rated as low or very low. The authors did not include evidence for an effect on postpartum weight retention (presumably through lack of adequate data), which has been shown to be robustly associated with excessive gestational weight gain.4
The authors undertook a subanalysis of overweight and obese pregnant women and found a reduction in gestational weight gain for this group, but with no change in pre-eclampsia or birthweight indices, an important determinant of adverse obstetric outcomes. This has been confirmed by another recent systematic review of randomised controlled trials in obese and overweight women, which reported a significant reduction in gestational weight gain of 2.21 kg, but again with no concurrent improvement in clinical outcomes, except for a lower incidence of gestational diabetes, for which the strength of evidence was low.6 Most overweight and obese women gain weight above the IOM limits, as was shown in the recent meta-analysis by presentation of the absolute weight gain in the experimental and control arms of each trial; both groups gained weight even though the increase may have been less marked in the experimental arm.6 (Thangaratinam and colleagues presented the mean difference only.) Thus a lack of any convincing improvement in relevant clinical outcomes was not unexpected.
Gestational weight gain, which incorporates fetal and placental weight, amniotic fluid volume, and maternal plasma volume expansion in addition to maternal fat accretion, varies substantially from woman to woman. For obese women, pre-pregnancy BMI may contribute as much to the determination of adverse pregnancy outcomes as excessive gestational weight gain. This may explain the lack of clinical benefit in obese and overweight women reported in the linked meta-analysis despite the larger reduction in gestational weight gain compared with women of any pre-pregnancy weight. Moreover, recent studies have challenged the IOM weight gain limits for obese pregnant women.7
The systematic review shows that dietary interventions achieved a greater reduction in gestational weight gain than physical activity interventions or those with a mixed approach but the significant effect on birth weight was seen only in trials of physical activity. A better understanding of the barriers to changes to dietary and physical activity behaviour in women of differing pre-pregnancy BMI are needed before we abandon interventions that include physical activity. Women with a lower pre-pregnancy BMI may have lower barriers to changing physical activity than those with a higher BMI,8 so more emphasis of effective interventions to promote physical activity in overweight and obese women is needed. Pregnant women who already consume a diet with excess fat and sugar content are also more likely to intend to lower the intake of these foods.9 A recent review of 10 studies that tried to define the most effective components of successful interventions for limiting gestational weight gain found no evidence to favour one behavioural technique over another, but there was some suggestion that interventions were more likely to succeed in non-smokers and if started early in pregnancy.10 The lack of detail and standardised reporting of the psychological theory underpinning the behavioural change or the specific techniques used in the interventions were notable. The authors of the current meta-analysis plan an analysis of individual patient data, which may shed further light on the elements of a successful intervention, but lack of detail in the source data is likely to make this approach challenging.
The focus on gestational weight gain at the expense of adequate power for clinical outcomes and the heterogeneity between the many different study populations and protocols limit the interpretation of the current meta-analysis, as the authors acknowledge. We agree with the authors that it is crucially important that future effectiveness studies focus on clinically relevant outcomes and provide information that will enable the role of individual components of the intervention to be assessed.
Ongoing randomised controlled trials that are adequately powered for clinical outcomes and rigorously assess the different elements of the intervention include the Australian LIMIT trial in overweight and obese pregnant women11 and the UK UPBEAT trial in obese pregnant women.12 Because of increasing evidence that maternal obesity and excessive gestational weight gain are associated, through intrauterine exposure, to increased risk of childhood obesity, both trials are also assessing body composition in the child in control and intervention arms. Work on developing a core set of outcomes for studies in pregnancy has recently started under the COMET initiative at the University of Liverpool and will inform future work (www.comet-initiative.org). This should also enable data on safety to be obtained from the trials themselves rather than relying on other, less appropriate, assessments of adverse outcomes obtained from observational studies in very different populations (such as women who are pregnant during war or famine).
Cite this as: BMJ 2012;344:e2774
Competing interests: Both authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years. LP is chief investigator on the UPBEAT trial; LCC has no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.