The Mummy's Club

This research compared the chemical profiles of urine samples from a small group of autistic children to those from their non-autistic brothers and sisters, and another group of unrelated non-autistic children. It did find that autistic children had differences in the levels of some chemicals, but those that were most different are not chemicals produced by gut bacteria.

This is encouraging research, but it is too early to say whether it will develop into another diagnostic tool for autism. Importantly, it did not directly investigate the role of gut bacteria in autism, but looked at levels of chemicals in urine. Also, it is not possible to determine whether these differences indicate a cause or consequence of autism. These children were already diagnosed with autism, and the study used urine samples from only one point in time. Further research in a larger group of children over time is needed.

Where did the story come from?

The study was carried out by researchers from Imperial College London and the University of South Australia. It was funded by Cure Autism Now and a grant from the International Study of Macro-Micronutrients and Blood Pressure. The study was published in the peer-reviewed Journal of Proteome Research.

The Daily Express reported that this study had distinguished children with autism by “looking at bugs from their intestinal tracts and the body’s metabolic process in their urine”. The Daily Mail reported that the test “builds on research showing that people with autism have different bacteria in their guts from others”. Additionally, The Daily Telegraph focused on the gut bacteria being a potential therapeutic target based on this research.

However, the research did not actually make any direct measurements of gut bacteria. The extent to which the chemical composition of urine reflects the bacterial population in the gut was not demonstrated in this study.

What kind of research was this?

This was a cross-sectional analysis, in which the researchers conducted chemical analysis of urine samples from children who had been diagnosed with autism, their non-autistic siblings and children without autism. The researchers aimed to assess whether there was any difference in the chemical composition of the urine samples from these children.

What did the research involve?

Urine samples from autistic children, their siblings and unrelated non-autistic children were obtained from the University of South Australia. Further urine samples from non-autistic children were also collected from Switzerland. The children were between three and nine years old. Children with autism met psychiatric criteria for autism disorder or Asperger syndrome.

There were 35 samples from autistic boys and four from autistic girls. The ‘sibling group’ was made up of 17 brothers of these children and 17 of their sisters. The control group consisted of 17 samples from non-autistic boys and 17 from non-autistic girls.

To identify the different types of chemicals in the children’s urine the researchers used a technique called nuclear magnetic resonance spectroscopy.

What were the basic results?

Initial analysis of the chemicals in the urine samples showed that the major source of variation between them was not related to autism, but variation between individuals. However, further statistical analysis indicated that the chemical pattern was different in the autistic children compared with non-autistic children and was partially different between the siblings and the non-autistic children.

The researchers then assessed whether the types of chemical found in the urine samples differed between each group. This analysis involved a statistical test that compared all of the chemicals in the children’s samples at the same time rather than assessing each chemical separately. This showed there were higher levels of some chemicals in autistic than non-autistic children, and vice versa.

For example, autistic children had higher levels of the breakdown products of nicotinic acid, N-methyl nicotinic acid (NMNA), and N-methyl nicotinamide (NMND), but lower levels of chemicals that are suggested to be associated with gut bacteria such as hippurate and phenylacetylglutamine (PAG).

The urine samples of the siblings of the autistic children were not significantly different to those of either of the other groups.

Statistical analysis of individual chemicals showed that autistic children had higher levels of NMNA and NMND compared with non-autistic children. NMNA and NMND, and succinate were also higher in the autistic children’s samples compared with those of their non-autistic siblings. Chemicals typically associated with gut bacteria, such as hippurate and phenylacetylglutamine (PAG), were no longer significantly different between autistic and non-autistic children.

How did the researchers interpret the results?

The researchers suggest that there are significant differences in the metabolic composition of urine between children with autism and unrelated non-autistic children. They say that the breakdown products of nicotinic acid (NMNA and NMND) were the group of chemicals that had the greatest power to indicate differences between autistic and non-autistic children.

However, they say that larger scale, longitudinal studies are needed to determine “whether the metabolic differences are related to the cause or the progression of the disease”.

Conclusion

This small study showed that there were differences between the chemicals in urine samples from autistic children compared with non-autistic children, however there was a degree of variation between all of the children’s samples regardless of whether they had autism or not. Further work is required to see whether the observed differences are consistent in a larger sample. Although the researchers suggest that some chemicals found in urine may be associated with bacteria in the gut, measuring chemicals in urine samples is an indirect way of assessing whether gut bacteria are associated with autism.

The research has several limitations:

• The researchers point out that, as it is not possible to tell whether these differences indicate a cause or consequence of the disease, further research is needed in a larger group of children over time.
• Different statistical analyses had different results, some showing differences in certain chemical levels in autistic children, while others did not.
• The researchers did not assess the medications the children with autism were taking for their condition or the diet they were following. Both would affect the chemicals they found in the children’s urine samples.

Finally, these children had already been diagnosed with autism, and the study design was cross-sectional, looking at their urine samples from only one point in time. It is not possible to say whether there would be any differences in the chemicals found in the urine in younger children prior to standard diagnosis, and whether it could be used as a diagnostic tool.

This is encouraging research, but it is too early to say whether this research would be of benefit in terms of providing an additional diagnostic tool for autism in children.

Links To The Headlines

Revolutionary urine test for autism could soon diagnose children with the condition. Daily Mail, June 4 2010

Autism test could make the condition ‘preventable’. The Daily Telegraph, June 4 2010

New test may spot autism in children. Daily Mail, June 4 2010

Links To Science

Yap IKS, Angley M, Veselkov KA, et al. Urinary Metabolic Phenotyping Differentiates Children with Autism from Their Unaffected Siblings and Age-Matched Controls. Journal of Proteome Research, 2010

The news story is based on a seven-year study of hospital births in the Netherlands. It found an increased risk of adverse outcomes in newborn infants with evening and night births in hospitals, and with night births in specialist centres. These adverse outcomes were uncommon: in 655,961 hospital deliveries, 1.7% had an adverse outcome and in 0.19% of the births, the newborn died.

The study’s strengths are its large size and the completeness of the data that was collected. However, several factors may have contributed to this association, and it is not possible to say conclusively that poor staff cover during the evening and night shifts were responsible, though it may have contributed.

Furthermore, as this study was in the Netherlands, it is not known if these results apply to the UK or other countries, which may have very different obstetric care protocols and organisation of staff and medical training posts.

Where did the story come from?

The study was carried out by researchers from the University Medical Centre Rotterdam and the University of Amsterdam. The research received no funding. The study was published in the peer-reviewed British Journal of Obstetrics and Gynaecology.

The news reports have generally reflected the main findings of this research, but have not been able to analyse these results in their full context or recognise some of the difficulties in interpretation. It is not correct to conclude from this study that the increased risk of adverse outcomes with night-time deliveries is due to poor staffing.

What kind of research was this?

This aim of this Dutch cohort study in hospitals in the Netherlands was to investigate whether time of birth, and certain features of the maternity unit affected the risk of newborn deaths or other adverse outcomes. The researchers used national registry data to examine births in all 98 hospitals across the Netherlands between 2000 and 2006.

What did the research involve?

The researchers used the Netherlands Perinatal Registry to identify all pregnancies continuing beyond 20 weeks between 2000 and 2006. The Registry contains data on maternal characteristics, pregnancy and labour, and outcomes in the newborn.

After excluding multiple births, home births and deliveries at hospitals that had not participated in the registry for two years, they were left with 792,954 births in 98 hospitals. Further refinement of the hospital birth dataset excluded foetal deaths during pregnancy, very premature babies and babies born with congenital malformations, as all of these would normally be cared for in specialist centres.

Births in specialist perinatal centres (109,858 births) were analysed separately from those in hospitals. From this data, foetal deaths during pregnancy and births prior to 22 weeks of pregnancy were excluded.

These further exclusions resulted in a final data set of 655,961 births in hospitals and 108,445 births in specialist perinatal centres.

Evening deliveries were defined as happening between 6pm and 11.59pm, and night deliveries between 12am and 7.59am. Survey data was collected on the staffing and organisation within the hospitals (such as whether it was a teaching hospital, and the numbers of obstetricians, midwives and doctors in training posts).

The main outcomes considered were perinatal mortality (death during delivery or within the first seven days following birth), and adverse outcomes in the newborn (including perinatal mortality,
poor score on a standard measure of neonatal health immediately after birth [5-minute Apgar score below 7] or transfer of the newborn to a neonatal intensive care unit).

Statistical analyses were conducted to examine the relationship between organisational factors and pregnancy and delivery characteristics with adverse outcomes in the newborn. The analyses took into account factors that can affect the risk of adverse outcomes, including the mother’s age, number of previous children, type of delivery, pregnancy length, ethnicity and calendar year of the birth.

What were the basic results?

Of the 655,961 hospital deliveries, 1.7% (11,118) were associated with an adverse outcome and 0.19% (1,206) with death in the newborn. Of the 108,445 births in specialist perinatal centres, 11.7% (12,705) were associated with an adverse outcome and 1.8% (1,915) with death in the newborn. Of the births analysed, 53% were to first-time mums and 20% of mums were over 35 years old. Approximately half of the births occurred during the evening and night-time hours.

In hospitals, evening births had a 32% increased risk of death in the newborn compared to daytime births (odds ratio [OR] 1.32, 95% confidence interval [CI] 1.15 to 1.52). Night births had a 47% increased odds of newborn death compared to day births (OR 1.47, 95% CI 1.28 to 1.69). In specialist centres, only night-time births had an increased risk(20%) of newborn death (OR 1.20, 95% CI 1.06 to 1.37). Compared to day births, evening births had a higher risk for an adverse outcome (OR 1.30 for hospitals and 1.21 for specialist centres) as did night-time births (OR 1.28 for hospitals and 1.25 for specialist centres).

Further analysis found links with various other factors. There was a clear increased risk of adverse outcomes for premature births (prior to 37 weeks) compared to term births (at 40 weeks), both in hospitals and specialist centres. Emergency caesarean sections were also associated with an increased risk of adverse outcomes in hospitals and specialist centres compared with spontaneous delivery, but with a decreased risk of death in the newborn in specialist centres. There was no clear relationship between the annual number of deliveries and the risk of adverse outcomes. More senior staff were associated with a significantly reduced risk of adverse outcomes, but did not affect risk of perinatal mortality in hospital deliveries.

How did the researchers interpret the results?

The researchers conclude that hospital deliveries at night are associated with increased perinatal mortality and adverse perinatal outcomes. They say that the time of delivery and other organisational features such as staff experience may explain the variation between hospitals.

Conclusion

This is a good quality study, but the data needs to be interpreted in the correct context. These adverse outcomes were uncommon, and of the 655,961 hospital deliveries analysed, 1.7% had an adverse outcome and 0.19% were associated with the newborn’s death. The rates in specialist centres were higher, but this is likely to be because these centres deal with more complicated pregnancies and births.

It is difficult to identify specific reasons why evening and night-time births in hospital were associated with an increased risk of adverse outcomes. It cannot be assumed that it is due to reduced staffing levels or facilities at night, although these and other factors may contribute. As the researchers say, they had limited information on actual staffing levels just before and during each individual delivery, and therefore they could not look at the effects of these factors. The relationship may even be due to an unknown biological factor.

In examining the links between adverse outcomes and the numerous possible contributing variables, it is difficult to unpick the exact cause and see how the variables interact with each other. There is also the increased risk of chance findings when examining associations with many variables.

The study was in births in hospitals in the Netherlands and therefore the results may not apply to the UK. The UK may have very different obstetric care protocols, including staff numbers and facilities during the evening and night.

Links To The Headlines

‘Higher death risk’ for births at night. Metro, June 18 2010

Babies born at night in hospital ‘have higher risk of dying’. Daily Mail, June 18 2010

de Graaf J, Ravelli A, Visser G, Hukkelhoven C, Tong W, Bonsel G, Steegers E. Increased adverse perinatal outcome of hospital delivery at night. BJOG 2010

This large cohort study looked at data from 8,552 seven-year-old children in 1965, and compared it to 1,889 of their four to nine-year-old children in 1991. The clearest finding was an increase in childhood obesity/overweight in both girls and boys, and the most consistent association in both generations was between the mother’s BMI and child BMI.

As reported, higher maternal employment seemed to have some association with higher child BMI in the 1991 cohort. However, there are likely to be several different factors involved in the increase in children’s BMI, including diet and physical activity levels, which were not measured here. This is a good-quality study, but translating its findings to mean that more working mothers equates to more obese children is an oversimplification of the facts.

Where did the story come from?

The study was carried out by researchers from the University College London Institute of Child Health. It was funded by several sources, including the Great Ormond Street Hospital NHS Trust and a grant from the Institute of Child Health Special Project Initiative. The study was published in the peer-reviewed medical journal American Journal of Epidemiology.

The news has over-simplified the findings of this study. Although a positive link was seen between higher child BMI and maternal employment in the later generation, the association is not completely clear, and there are many other unmeasured factors that could be having an effect too. Therefore it is not possible to conclude that maternal employment is the only cause of higher childhood BMI.

What kind of research was this?

The purpose of this large British cohort study was to look at how risk factors for obesity have changed over time. The researchers compared the data from a population born in 1958 to that from any offspring they had by they time they were 33 years of age (in 1991). The parents’ data was collected when they were about seven years old, and the offspring data was collected when they were between four and nine.

The researchers used statistical models to look at the relationships between mothers and their offspring and their respective BMIs, and also looked at the factors that were associated at the time of the assessments.

Although the study can tell us how the prevalence of obesity has changed over time, it cannot tell us the reasons for this. The higher prevalence of obesity in the offspring cohort may be associated with a higher prevalence of working mums compared to the previous generation. However, it cannot be concluded that greater maternal employment is the only cause of the higher child BMI.

What did the research involve?

The first cohort included all people born during a particular week in 1958. About 17,000 of these people were then followed up on eight occasions until they reached the age of 45. In 1991, when the cohort was aged 33 years, 11,407 members were interviewed and of those who were parents, a random sample of one-third was selected. This gave about 4,300 children who formed the second, offspring cohort.

As all of the offspring had been born when their parent was aged 33 or younger, they formed two comparable cohort groups:

• Children born in 1958 who had been assessed in 1965 when they were seven years old and who had been born when their mother was under the age of 30 (8,552 members).
• The offspring cohort who were aged between four and nine years in 1991, and who had been born when their mother was under the age of 30 (1,889 members).

Height and weight measurements of the children were collected in 1991. Their parents were asked to complete a questionnaire providing extra information on their children, such as socio-demographic details, including whether they were married/cohabitating, employment status and housing, and information on various pre- and postnatal medical and lifestyle factors, e.g. whether the parents smoked, whether the baby was breastfed, etc. Similar information had been collected for the 1958 cohort, which also had their height and weight measured when they were seven. Their own parents BMI had also been measured (grandparents of the offspring cohort).

What were the basic results?

The authors found that the prevalence of overweight/obesity had increased by more than 50% between generations. Between the first cohort in 1965 and their offspring cohort in 1991, there had been a BMI increase of about 0.64 units among seven year old girls. For seven year old boys there was a BMI increase of about 0.23 units.

The researchers looked at numerous social factors to see whether they were associated with BMI in both cohorts, including maternal employment, housing and family size. A significant positive relationship was found between offspring BMI and their mothers’ BMI, i.e. there was a higher chance of the child being overweight/obese if their mother was. The association between child and maternal BMI had become more significant across generations. There was also a positive trend between increased BMI in the offspring cohort if their mother was in full-time employment; a relationship that was not seen in the 1958 cohort.

There was no clear relationship with any other factors, and they showed different associations in the two cohorts. For example, in the 1958 cohort, lower parental socioeconomic status was linked to lower child BMI; in the offspring cohort lower parental socioeconomic status was linked to higher child BMI.

Maternal employment had increased over the generations, and socioeconomic factors had improved.

How did the researchers interpret the results?

The researchers conclude that parental obesity, maternal employment and socioeconomic factors may play an increasing role in the childhood obesity epidemic.

Conclusion

This large cohort study looked at the BMI and socio-demographic status of 8,552 seven-year-old children in 1965, and then looked at 1,889 of their four to nine-year-old children in 1991. The researchers were interested in the change in obesity prevalence between generations, and to see whether it was related to any other factors.

The clearest finding from this research is that the prevalence of childhood obesity/overweight has increased, with a 0.64 BMI unit increase for seven-year-old girls and 0.23 units for boys. The researchers carried out statistical models to look at associations with other factors. They found a variety of trends and associations in both generations. Maternal BMI and child BMI had the most established association, which was consistent in both generations. Others were less consistent and some associations reversed. For example, in 1965 lower socioeconomic status was associated with lower childhood BMI, whereas in 1991 it was linked with higher BMI.

Unsurprisingly, maternal employment was found to increase from the first to the second generation. As reported, this higher maternal employment was also associated with a higher child BMI – a relationship which had not been seen in the first cohort. However,  there are likely to be several different factors involved in this increase in BMI, including diet and physical activity levels – factors which have not been assessed for either of the cohorts in this study. This is a good-quality study, but translating it to mean that more working mothers equates to more obese children is an oversimplification of the facts.

Links To The Headlines

Children of working mothers are telly tubbies: Obesity concerns for latchkey kids. Daily Mail, May 26 2010

Links To Science

Pinot de Moira A, Power C and Li L. Changing Influences on Childhood Obesity: A Study of 2 Generations of the 1958 British Birth Cohort. American Journal of Epidemiology 2010; Advance Access published online on May 20

These reports are based on research examining evidence on the quality of the diets of UK women throughout their lives. It found that, among most age groups, intakes of key micronutrients, particularly iron, vitamin D, calcium and folate, remain below recommended levels. Many women’s diets are also too high in saturated fat and salt, and low in fibre, oily fish and fruit and vegetables.

This narrative review was not a systematic analysis of research in this field, but its finding are in line with a body of quality evidence that suggests that many women (as well as men and children) are not consuming recommended levels of nutrients and, generally, do not have healthy diets. This is a serious, recognised issue that clearly needs to be addressed. However, while this study concludes that fortified foods or vitamin and mineral supplements may have a role to play in improving women’s nutrition, it is important to have a systematic approach to the evidence on supplementation and nutritional health.

Where did the story come from?

The review was carried out by Carrie Ruxton, a dietitian who runs a nutrition consultancy, and Emma Derbyshire, a researcher from Manchester Metropolitan University. The study was funded by the Health Supplements Information Service (HSIS), an online information programme funded by the Proprietary Association of Great Britain, which is the trade association for manufacturers of food supplements and over-the-counter medicines.

The study was published in Nutrition Bulletin, a publication of the British Nutrition Foundation, a registered charity.

The review’s overall finding that many women eat poorly was generally reported accurately, if somewhat sensationally, in the media. The Daily Telegraph’s headline about millions on “danger diets” and claims in the Daily Express about “food fads” are not borne out by the evidence presented in this study. The newspapers also failed to mention the fact that there is evidence for some improvements in dietary quality over the last few years. Stories in some papers used phrasing and quotes that were identical to those found in the press release on the review. The Daily Mail correctly pointed out (near the end of its report) that the study was industry-funded, but the Express incorrectly called HSIS an “independent body”.

What kind of research was this?

This was a non-systematic narrative review discussing women’s diets. It used supporting evidence from various studies of nutrition. It also referenced studies on the relationship between diet and health. It relied heavily on data from the main source of dietary information in the UK, the National Diet and Nutrition Survey (NDNS), as collated by the Scientific Advisory Committee on Nutrition (SACN). This evidence was supplemented by other studies of women’s diet. The government surveys of nutrition are reputable and generally thought to be accurate.

The study also describes evidence that a poor diet and high alcohol intake are both related to diseases such as cancer, cardiovascular disease, diabetes, osteoporosis and mental health problems. It does not systematically examine the quality of all this evidence, but uses some of it as part of an argument for the importance of improving women’s diets. Some research experts might describe this method for selecting evidence as ‘cherry picking’.

What did the research involve?

The review summarised over 100 studies addressing women’s diets and dietary deficiencies, and also how diet relates to other aspects of health. The review did not state how it identified and selected the studies to be included and, therefore, may not identify or include all relevant studies. The studies that were not included may not agree with the findings of studies that were included. Those studies that were included were summarised in a narrative way.

What were the basic results?

The review of UK dietary surveys broadly found that, while some improvements have occurred, intakes of key micronutrients, particularly iron, vitamin D, calcium and folate, remain below recommended levels across most age groups. Women’s diets are also too high in saturated fat and salt, and low in fibre, oily fish and fruit and vegetables.

The authors also make numerous specific observations on nutrition, such as:

• Among girls of school age, 52% have low intakes of magnesium, 25% have low intakes of zinc and about a quarter have poor iron status.
• Among women in their childbearing years 20% fail to meet the Lower Reference Nutrient Intake (LRNI) for iron and 83% eat more than the maximum recommended salt intake of 6g daily.
• The authors also think vitamin D intake may be inadequate, although there is no consensus about recommended intake (the body can obtain vitamin D through either dietary intake of certain foods or by producing it when the skin is exposed to sunlight).
• On average, pregnant women do not meet recommendations for vitamin D or folate.
• Middle-aged women (aged 50-64) tend to eat better quality diets and meet recommended levels of most nutrients.
• Older women (over 65 years) have low intakes of calcium, magnesium and zinc, and vitamin D status is poor.

The researchers also conclude that the evidence suggests that certain chronic conditions are influenced by dietary components. For example, they say that:

• inadequate calcium and vitamin D intakes reduce bone density
• salt and saturated fat increase cardiovascular disease risk
• excessive alcohol intakes increase cancer risk
• low intakes of long-chain n-3 fatty acids may adversely affect foetal development and mental health
• adequate folic acid reduces the risk of birth defects

How did the researchers interpret the results?

The authors conclude that, although there have been positive changes in UK women’s diets, “there is still room for improvement”. High intakes of salt, saturated fat and alcohol remain a problem, as do low intakes of fibre, long-chain n-3 fatty acids, iron, calcium, folate and vitamin D. They say that women need to make better dietary choices in order to ensure adequate levels of micronutrients and, thus, safeguard health.

However, they suggest that attaining optimal levels of some vitamins and minerals from food sources may be “challenging” for some groups of women and, therefore, fortified foods and supplements should continue to play a role in public health campaigns. They argue that, because of insufficient year-round exposure to sunlight and because few foods are naturally rich in vitamin D, vitamin supplements play an important role in helping vulnerable groups achieve acceptable vitamin D status.

Conclusion

This review does not tell us anything new, but it does highlight the worrying fact that many women in the UK of all ages have poor diets. Diet and food choices are complex issues and are also influenced by many environmental factors such as income, lifestyle and access to cheap, nutritious food, some of which are not entirely within an individual’s control.

Sticking to a healthy diet is easier said than done, but it’s important to try to make healthy choices. The current advice is to:

• Eat a variety of food, including plenty of fruit, vegetables and starchy foods, such as wholemeal bread and wholegrain cereals; some protein-rich foods, such as meat, fish, eggs and lentils; and some milk and dairy foods.
• Eat more fish, including one portion of oily fish a week.
• Cut down on sugar and saturated fats (found in butter, cream, cheese and many processed foods).
• Eat less salt (no more than 6g a day).

As a review of nutrition, the authors do not pay much attention to sunlight as a source of vitamin D.

While some groups may benefit from vitamin supplements (women who are pregnant or planning to be for example, are advised to take supplements of folic acid and vitamin D), it is not clear from this report which subgroups of women can get nutrients from eating a healthy diet and who may need supplementation.

It is worth considering the commercially-affiliated funding source of this review and the choice of evidence presented before deciding what role vitamins, mineral supplements or fortified foods may have in improving the health of women.

Links To The Headlines

Millions of British women on danger diets. The Daily Telegraph, June 1 2010

Too much alcohol and not enough vitamins: How poor diet is putting women of all ages at risk. Daily Mail, June 1 2010

Food fads that are putting women’s health in danger. Daily Express, June 1 2010

Links To Science

Ruxton CHS and Derbyshire E et al. Women’s diet quality in the UK. Nutrition Bulletin, Volume 35 Issue 2, Pages 126-137 [Published Online May 20 2010]

The current study looked at more than 5,000 rare genetic differences (called copy number variations, or CNVs) in the DNA of 996 people with autistic spectrum disorders (ASD) and 1,287 people without the condition. It found that people with ASD had a greater number of genes affected by these differences than those without ASD. Some of the CNVs affected genes that were already thought to play a role in ASD, while others affected genes not previously known to play a role.

This important study provides a lot of information about genes that may play a role in the development of ASD. More research will be needed to examine these genes. While this may eventually suggest new targets for drug treatments, their development would still be some way off. The study also showed that different people with ASD had different sets of variations, suggesting that a single, universal genetic test for ASD is unlikely to be developed in the near future.

Where did the story come from?

The study was carried out by a large international consortium of researchers. The main funders were Autism Genome Project Consortium, Autism Speaks (US), the Health Research Board (Ireland), The Medical Research Council (UK), Genome Canada/Ontario Genomics Institute, and the Hilibrand Foundation (US). The study was published in the peer-reviewed journal Nature.

The Independent, Daily Mirror, The Guardian, Daily Mail and The Daily Telegraph have covered this research. The Mirror and Daily Mail suggest in their headlines that a new test for autism could be developed based on these results. Most of the other papers discussed this possibility in their articles. However, not all individuals with ASD carry the same genetic variations, meaning a single diagnostic test based on these results seems unlikely to be developed in the near future. At best, such a test might indicate the chance of developing ASD, but would not be able to predict definitively whether a person would or would not develop ASD. Much more research into how well such a test might perform would be needed before it could become a reality.

The Independent importantly highlights that caution should be applied when interpreting the implications of this study, which the researchers themselves describe as preliminary. They say that “it will take many more years of intensive investigation to understand and treat the genetic alterations that increase an individual’s susceptibility to the disorder”.

What kind of research was this?

This was a case-control study looking for genetic differences that could contribute to the development of autistic spectrum disorders. In particular, they were looking at sections of DNA that can exist in different numbers of copies in different individuals. These are called “copy number variations” or CNVs, and they can arise when pieces of DNA are duplicated or deleted. They wanted to determine whether individuals with autistic spectrum disorders have different numbers of copies at rare CNV sites than people without the condition.

This type of study design is commonly used to investigate the potential genetic causes of disease. This type of study has been made easier by advances in genetic technology, which means that researchers can now look for a large number of differences across multiple DNA samples in a relatively short space of time.

What did the research involve?

The researchers analysed DNA that was collected from 996 people with autistic spectrum disorders (ASD cases) and their parents, plus DNA from 1,287 matched individuals without ASD (the control group). To ensure that people’s ethnic backgrounds did not contribute to any genetic differences seen, the researchers only looked at people with European ancestry.

The researchers looked at CNV sites throughout the DNA and compared these between the cases and controls. They then specifically focused on 5,478 rare CNVs, which occurred in less than 1% of the sample population. They also looked for differences between cases and controls in the number of rare CNVs per individual, how long the CNV DNA regions were, or the number of genes affected by CNVs (that is, where the CNV was within or near to the gene).

What were the basic results?

The researchers found no differences between people with autistic spectrum disorders and controls in the number of CNVs per individual. Cases and controls each had 2.4 CNVs on average. There was also no difference between cases and controls in the length of their CNVs.

However, compared with controls, people with autistic spectrum disorders were found to have 19% more genes affected by CNVs (i.e. a CNV near to or within a gene). In some cases, these variations had been inherited from their parents. In other cases, the variations had arisen spontaneously in the affected individual.

Of the rare CNVs, 226 were found to affect a single gene and were found only in people with ASD, not the controls. Several of these CNVs were in genes or areas of the DNA previously thought to be involved in ASD or intellectual development. Some of the CNVs were in areas of the DNA containing genes not previously suspected to be involved in ASD. Other CNVs were in and around genes with roles in cell division, movement and signalling, and some played roles in nerve cells.

How did the researchers interpret the results?

The researchers concluded that their results “provide strong support for the involvement of multiple rare genic CNVs” in ASD. They say that this and subsequent research “may broaden the targets amenable to genetic testing and therapeutic intervention”.

Conclusion

There is much ongoing research into the genetic risk factors for complex conditions such as ASD, which is helping us to understand their role. One school of thought suggests that the genetic basis of these conditions is due to the cumulative influence of numerous common genetic factors contributing to the overall risk of developing the disease. The current study also suggests a role for rare copy number variations.

This particular study advances researchers’ knowledge of genes that may play a role in ASD, and they will use these results to target specific genetic areas for further study. Ideally, these results should now be verified by replicating the study in other groups of cases and controls.

Some news sources have suggested that these findings could lead to a test for ASD. However, not all individuals with ASD carry the same genetic variations, so a single diagnostic test based on these results seems unlikely in the near future. At best, such a test might indicate the chance of developing ASD, but it would not be able to predict definitively whether a person would or would not develop ASD. Much more research would be needed to look at the practical feasibility of a test, and how accurate and predictive such a test might be before it could become a reality.

Links To The Headlines

Autism and genetics: A breakthrough that sheds light on a medical mystery. The Independent, June 10 2010

Autism find could lead to new test. Daily Mirror, June 10 2010

The genetics of autism. The Guardian, June 10 2010

Autism study means scientists have genetic causes in their sights. The Guardian, June 10 2010

Autism blood test could be available in three years. Daily Mail, June 10 2010

Dozens of genetic mutations linked to autism in children discovered. The Daily Telegraph, June 10 2010

Links To Science

Pinto D, Pagnamenta AT, Klei L et al. Functional impact of global rare copy number variation in autism spectrum disorders. Nature, June 9 2010 [published online]

This news story is from a small laboratory study testing a new non-invasive method for extracting genetic information from the gut cells found in infant stool samples. The researchers also wanted to compare the intestinal genetic profiles of babies that were breastfed to those of formula-fed babies.

This study did find a difference in gene activity in the cells from the guts of breastfed compared with bottle-fed babies. However, it is uncertain whether this is due to any difference between the immune systems of breastfed and bottle-fed babies. Also, the babies were not randomly assigned to receive either breast or formula milk, so it is possible that other factors could have caused these differences. Further larger studies are needed.

Breast milk is known to help protect babies against various disorders and boost their immune defences. While this study may be of scientific interest and merit, these results do not offer any new information about the relative benefits of breastfeeding.

Where did the story come from?

The study was carried out by researchers from Texas A & M University, the University of Illinois and Mead Johnson Nutrition (a manufacturer of infant formula) in Evansville, Indiana. It was funded by grants from Mead Johnson and the National Institutes of Health. It was published in the peer-reviewed American Journal of Physiology, Gastrointestinal and Liver Physiology.

The study was given some prominence in the Telegraph, but these early findings by themselves do not warrant the Telegraph’s conclusion that breast milk is best because it “kick starts the immune system”.

What kind of research was this?

This was a small, proof-of-concept laboratory study. The researchers aimed to test a new non-invasive method for extracting genetic information about gut cells from infant stool samples. They intended to use this method, patented by one of them, to ‘fingerprint’ and compare the intestinal genetic profiles of babies that were either exclusively breast or formula fed.

Although breastfeeding is known to help protect the developing infant against a range of infections and other disorders, less is known about how it does this. The researchers report that there is some evidence from animal studies and human infants that the intestinal tract (or ‘gut’) undergoes marked structural and functional changes in response to feeding. They also say that there is evidence that components in breast milk may contribute to the way the gut matures. However, exploring whether this is the case has been difficult because of the ethical constraints around obtaining gut tissue from healthy babies for research purposes.

The infants in this study were not randomly assigned to receive either breast or formula milk, but were categorised according to what they were already being fed. The study may, therefore, be affected by confounding. This means that factors other than the type of feeding may be contributing to any differences in gene activity seen between formula-fed and breastfed babies. The researchers tried to reduce these ‘confounding effects’ by ensuring that the babies were similar in many other ways. However, it was not possible to avoid all potential confounding factors.

What did the research involve?

Researchers recruited the mothers of 22 healthy full-term infants: 12 exclusively breastfed and 10 fed only on formula. The mothers in both groups had similar average ages and number of previous children, and there was a similar number of male and female babies of similar birth lengths and weights. Most of the infants were Caucasian.

The study excluded infants who were intolerant to cows’ milk or who were receiving a mix of breast and formula milk. Infants consuming juice or solid foods were also excluded, as were babies who became clinically ill or who had antibiotic treatment. Mothers of formula-fed babies were all provided with a certain type of formula, Enfamil LIPIL, for the duration of the study.

The parents collected stool samples from their babies at three months of age, with instructions on how to do this. Samples were kept in a sterile solution, and frozen and transported to the laboratory by research staff. The parents also weighed the babies before and after each feeding for 24 hours before the stool sample was taken and the change in body weight was used as an estimate of breast-milk or formula intake. Maternal age and infant birth weight and length were also recorded.

The researchers used a technique to isolate any intestinal cells in the stool samples and analysed the genetic material from these in the laboratory. The activity of certain genes was then compared between the groups.

What were the basic results?

The researchers were able to identify several gene sets or combinations showing different levels of activity in the formula- or breastfed infants. They say that their non-invasive technique for isolating genetic material from the developing infant gut was successful.

There was also some variability in gene activity among breastfed infants. The researchers say this could be explained by variations in breast-milk composition.

How did the researchers interpret the results?

The researchers say these findings support the theory that genetic material isolated from babies’ stools can be used to help understand the development of the intestine and the effect of different types of nutrition. Nutrition is believed to play a major role in the way the infant gut develops and is particularly crucial during the early postnatal period, having an impact not only on immunity, but also digestion, absorption and colonisation by different types of bacteria.

They say their study indicates for the first time that gene expression in infants fed breast milk is different from gene expression in formula-fed infants.

Conclusion

This study found that genetic material from the infant intestine can be isolated from stool samples. The researchers say that analysis of this genetic material suggests that the genetic pathways of breastfed infants differ from those found in infants fed formula milk.

The results of this small study could be of great interest to other scientists in the field. Its main purpose appears to have been to test the practicality and usefulness of a new method for examining cells from babies’ guts. However, it does not, by itself, tell us anything new about the practical health benefits of breastfeeding compared to formula feeding.

In terms of investigating the effects of breastfeeding compared to bottle-feeding, the infants were not randomly assigned to either type of feeding, so the study may be affected by confounding. This means that differences between the babies other than the type of feeding may be contributing to any differences in gene activity seen. The researchers did try to ensure that the babies in the groups were similar in some areas, but other factors could still be having an effect. For example, the researchers did not test exactly which cells the genetic material was coming from, and this could have differed between infants.

It has been established that breast milk helps protect babies against various disorders and boosts their immune defences at a time when their immune system is still developing.

The technique is likely to need further development, and the results of this study will need replication in larger, more comprehensive studies.

Links To The Headlines

Breast milk best because it ‘kick-starts babies’ immune system’. The Daily Telegraph, June 9 2010

Links To Science

Chapkin RS, Zhao C, Ivanov I, et al. Noninvasive stool-based detection of infant gastrointestinal development using gene expression profiles from exfoliated epithelial cells. Am J Physiol Gastrointest Liver Physiol 2010: 298: G582-G589