select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
Researchers publishing in The Lancet journal have examined whether change in a woman’s BMI between two pregnancies impacts on mortality of the second child, and report that an increase in BMI may be a risk factor for stillbirth and infant mortality.
Prof. Basky Thilaganathan, Director of the Fetal Medicine Unit, St George’s University Hospital, said:
“This study used very sound, population-based data, but I do have some concerns about how the results might be interpreted. The magnitude of the effect (RR of 1.55 for stillbirth) is comparable or less than for other factors such as primiparity (giving birth for the first time), ethnic minorities, high blood pressure and poor fetal growth.
“The increase in risk conferred by modest weight gain between pregnancy is likely to be negated by the protective effect of multiparity (having had a baby before) which is known to reduce the relative risk of stillbirth. Although the relative risk appears substantial – 55% increase- the absolute risk remains low for stillbirth at around 1 in 200 in the UK (a slightly higher rate than for this dataset from Sweden). The distinction between these two sorts of risks should be made to prevent unnecessary alarm.”
“The authors correctly discuss that the biological mechanism of this finding causing stillbirth is unknown. Despite this, I am surprised that the authors go on to infer that a statistical association is causation. Afro-Caribbean women have higher risks for stillbirth – do we assume that this is the cause for fetal death or accept that skin colour is a proxy for an elusive biological mechanism?”
“Whilst weight loss or normalisation is an important health objective, well conducted clinical trials have failed to demonstrate effective interventions to reduce weight in relation to pregnancy.”
Dr Patrick McSharry, Head of Catastrophe Risk Financing, University of Oxford, said:
“With obesity on the rise, this robust statistical study of over 450 thousand women conveys a striking message about the dangers of weight gain in childbearing women. The statistical analysis is rigorous and clearly presented. The data supports the broad conclusions that weight gain increases the risk of stillbirth and infant mortality.
“Evidence for a rise in the risk of stillbirth and infant mortality due to BMI increases already exist in the literature. It would be useful to have more details about the numbers of women in these previous studies. The authors claim that the main contribution of this new paper is the dose–response association between weight gain and stillbirth risk.
“I think it is extremely important that the increase in risk is communicated in a meaningful manner. For women with a stable BMI (change between –1 kg/m2 and <1 kg/m2) between pregnancies, the stillbirth rate is 2 per 1,000 births. Women who gained at least 4 BMI units between pregnancies are at greater risk, as indicated by the adjusted relative risk of 1.55, corresponding to 3.1 per 1000 births.
“For women with a stable BMI (change between –1 kg/m2 and <1 kg/m2) between pregnancies, the infant mortality rate is 1.9 per 1,000 births. Women who gained at least 4 BMI units between pregnancies are at greater risk, as indicated by the adjusted relative risk of 1.29, corresponding to 2.5 per 1000 births.”
Prof. Andrew Whitelaw, Emeritus Professor of Neonatal Medicine, University of Bristol, said:
“This is a high-quality, robust study carried out by two experienced researchers who have previously published several important papers on maternal obesity, inter-pregnancy weight gain and pregnancy complications. This is the first study to investigate a link between inter-pregnancy weight gain and stillbirth or infant mortality.
“The investigators used the Swedish Birth Registry, analyzing 98% of all births in Sweden over 11 years. Stillbirth and infant mortality rates in Sweden are among the lowest in the world but the increase in BMI equivalent to about 2.8 kg between pregnancies was associated with increases of more than 50% increase for stillbirth and nearly a third for infant mortality. These substantial increases in mortality are very well supported by the data, as the researchers have taken great pains to eliminate bias from maternal smoking, education, country of birth, height and age.
“Importantly, the researchers had sufficient numbers to show that BMI gain in women not obese to start with also increased infant mortality. They were also able to show that obese women who reduced BMI between pregnancies had a reduction in neonatal mortality. Although this is the first study to show this, the size of the study group and thorough methodology make it very likely that a similar effect would be found in other western developed countries.
“The obvious question is how can fat deposition between pregnancies harm the foetus and newborn? Sadly the researchers wrote that even their 450,000 women was too small a group to investigate individual cause of death. Raised blood pressure and gestational diabetes are two complications of pregnancy possibly leading to more stillbirths that may have been increased by the increased fat tissue. Gestational diabetes increases congenital malformations such as heart defects. Excessive fat tissue increases the risk of critical oxygen deprivation during labour and also increases inflammation and that could trigger preterm birth which certainly increases mortality. Treatment programs to reduce weight gain between pregnancies have proved difficult in the past but are clearly needed. It would help if we knew about causes of death. To investigate that would require an even larger study. The British NHS would be ideal as we have a universal health system and a population about 7 times greater than Sweden. ”
Additional information from Prof Andrew Whitelaw:
“All residents in Scandinavia have a unique personal number from birth which is used for all interactions with official and public resources including health, tax, social welfare, education, passport, employment, bank etc. This number sticks throughout life and means that every individual can be traced over many decades regardless of move or name change. Each new baby’s personal number is linked to the mother’s personal number. This routine collection of data enabled the researchers to acquire complete data on over 450,000 women who had first and second singleton babies during the 11 year period.
“However, there are two barriers to such a study in the UK: cost and confidentiality. The use of the personal number in Sweden without any name or address meant the data could be handed over to the researchers with the data protection authority in Sweden being confident that no identification of individuals was possible. Because of this confidentiality no individual consent needed to be obtained and therefore nobody could opt out. Because the information is collected routinely anyway in order to run services, no information is collected specifically for research and it costs very little to package it into a file suitable for research analysis. This is why so much good perinatal epidemiology can be done in Scandinavia.
“In the UK, most large high quality perinatal research studies such as EPICURE have had to be separately funded and were therefore very costly because of the inadequate routinely collected clinical information. The British have a deep suspicion of government surveillance, identity cards, and personal registers. They value privacy and confidentiality, demanding individual informed consent for even the most harmless of surveys. Think what important health questions could be answered if the UK with 64 million people had routinely and universally collected health data of Swedish quality.”
Prof. Roland Devlieger, Professor of Obstetrics, University of Leuven, said:
“The paper is robust and the data supports the conclusions of the paper. It confirms the importance of going back to your initial weight between pregnancies. This was also found in a recent study of ours (Bogaerts et al 2014)
“The main message for the public is that the increase in risks appear spectacular when represented as a relative risk, while the absolute numbers are small. The risk for stillbirth for example increases with 50% in women with an increase in BMI with 4 points between pregnancies, increasing this risk from about 2 in 1000 to about 3 in 1000.
“But globally, more attention is needed for weight management before pregnancy and in between pregnancies, as the interpregnancy period is the preconception period of the next pregnancy. Effective strategies to help women to regain a heathy BMI after pregnancy are urgently needed.”
‘Weight change between successive pregnancies and risks of stillbirth and infant mortality: a nationwide cohort study’ by Cnattingius et al. published in The Lancet on Wednesday 2nd December.
Declared interests
Prof Basky Thilaganathan: I have an academic/research interest in placental dysfunction and stillbirth – but not with population based work as with this group. I am also a member on the RCOG Stillbirth Clinical Study Group – devised to help prioritise research and advise on funding.
Dr Patrick McSharry: No conflicts of interest
Prof. Andrew Whitelaw: I have no conflicts of interest
Prof. Roland Devlieger: No conflicts of interest