Date: February 5th, 2009

Seminar Title: "Divergence in Tracking of Development of BMI and Sleep Duration during Puberty: A Role of Leptin"

Speaker's Name and Affiliation:

• Margriet Westerterp-Plantenga
• Mastricht University, Netherlands

Presiding Chair: Harry R. Kissileff, Ph.D.

Rapporteur: Kathleen L. Keller, Ph.D.

Attendees and their Affiliation:

Kathleen Keller	Columbia/Obesity Research Center
Harry Kissileff	Obesity Research Center
Eva Kovacs	Unilever
Susan Carnell	Columbia
Allan Geliebter	Columbia
Michael Lewis	Hunter

Summary:

Dr. Westerterp-Plantenga presented prospective data from a representative sample of 105 Dutch children, ages 0 ¨C 12 years. The aims of the study were to identify predictive variables for the development of childhood obesity and to confirm hypotheses related to BMI tracking, catch-up growth, and parental influences. Anthropometric and other key study measures were taken at birth, 0.5 y, 1 y, 2 y, 3 y, 4 y, and 7 y.

Results suggested that overweight children exhibited more dietary restraint and disinhibition and reported lower levels of physical activity than normal weight children. Parental overweight, particularly in the father, was positively associated with children's overweight. Maternal level of restraint and disinhibition were also positively associated with child overweight. While overweight children reported less physical activity (by both the Baecke questionnaire and the TRACMOR accelerometer), their perception of physical activity was greater than in normal weight children. But, total TRACMOR counts per activity were equivalent in overweight and normal weight children, suggesting the level of intensity per activity were the same in both groups. With respect to tracking of body mass index (BMI), after 1 year of age, tracking of BMI occurred at each timepoint. Body composition at each time was related to birth weight. The tracking of both body composition and BMI were greater as children were measured at later timepoints in the study. For example, the relationship between BMI at ages 5 y. and 12 y. was greater than the relationship between BMI at ages 2 y. and 12 y, and this has also been found in other studies.

The second part of the talk reviewed genetic and neuroendocrine findings from a cohort of 94 Dutch children, age birth ¨C 13 y. Results indicated that leptin played a key role in the onset of puberty, showing peaks for both girls and boys at around age 12 y. Leptin peaks for overweight girls were the highest, followed by normal weight girls, and all boys. Leptin likely plays a role in determining the onset of puberty. As a follow-up objective, the associations between leptin concentrations, gonadotropic hormones, and body composition were investigated during puberty. In both boys and girls, factors that are independent of fat mass become important regulators of plasma leptin concentration. For girls only, leptin acts as a permissive factor in the onset of puberty, and peaks in leptin concentration preceded the peaks in LH and FSH, estradiol, and testosterone concentrations. In boys, leptin appears to have a different timing and function with respect to the onset of puberty.

An additional objective of this study was to investigate the effect of FTO (rs9939609) on BMI, body composition, leptin concentrations, and physical activity in a longitudinal study design. Children with a copy of "A" at this site had higher BMIs and fat mass indices than children without this sequence. Children with "A" at this site also reported lower levels of physical activity, higher leptin concentrations, and a higher leptin peak at age 12 y.

A final objective was to investigate the relationship between anthropometric indices, body composition, leptin, physical activity, and sleep duration. Previous studies have found inverse associations between sleep duration and BMI, for both children and adults. Results of the present study showed that short sleep duration becomes more prevalent during puberty and was associated with earlier onset of puberty, increased body fat, BMI, and waist circumference, even after adjusting for Tanner stage.

Discussion:

Q. Is there a difference between boys and girls in these relationships (between physical activity and obesity)?
A. We didn't find any.

Q. Do father show lower levels of dietary restraint?
A. Yes, the men typically are less restrained. That's what most studies show.

Q. How are you assessing growth in the first year? Is that a % increase?
A. It doesn't matter how you measure it. For us, it was a very clear phenomenon.

Q. There were no differences in male and female children?
A. Not yet, those usually come out during/after puberty.

Q. Why would dietary restraint lead to obesity? It seems counter-intuitive.
A. Because of disinhibition and emotional eating (which often go along with dietary restraint). We distinguish successful dietary restraint from unsuccessful dietary restraint.

Q. How do you measure dietary restraint in kids?
A. We use the TFEQ and it's actually quite simple. You have to change a few of the questions (ones related to work, etc), but it's not difficult.

Q. What is the interpretation of a decrease in leptin in boys?
A. We are not sure.

Q. There are data to suggest that leptin is more related to subqutaneous fat than visceral fat. Maybe what you are seeing is a shift in fat depots?
A. Yes, we could have a look at that in our data.

Q. Did you find the "u" shaped relationship between sleep time and children's BMI?
A. This relationship is mostly found in adults where those who sleep more than 9 hours / night may be very sedentary, depressed, or have other health problems.

Q. What is the cause and effect of the relationship between sleep and obesity?
A. There are multiple possibilities. It could be related to diurnal fluctuations in leptin and ghrelin that are thrown off by lack of sleep. It could be due to reduced fat oxidation. It could be due to fluctuations in melatonin. There are other possibilities as well.

Q. Is sleep apnea a factor?
A. It could be, but it seems to be more of an effect of obesity, not a cause.

Q. Was this self-reported sleep duration?
A. Yes.

Q. Did it take into account the quality of sleep?
A. Yes, the quality of sleep was assessed.

Q. In adults, why do you think that high sleep is associated with obesity?
A. It could be the extreme of lazy or depressed people, or it could be a result of depression.

Q. Have you looked at orexin?
A. It is on our list of things to look at. We are looking for a software at the moment that is capable of detecting slow wave sleep, as this will improve the accuracy of our sleep reporting.

Q. Do people sleep longer when they lose weight?
A. I'm not sure, but I think it would be a good therapy.
Comment: I think weight loss definitely improves sleep quality.

Q. What age ranges have you looked at?
A. We studied 0-12 years and after that, 13 and up.