New directions in childhood obesity research: how a comprehensive biorepository will allow better prediction of outcomes

Weight change and all-cause and cause-specific mortality: A 25-year follow-up study

BACKGROUND

Whether the dynamic weight change is an independent risk factor for mortality remains controversial. This study aimed to examine the association between weight change and risk of all-cause and cause-specific mortality based on the Linxian Nutrition Intervention Trial (NIT) cohort.

METHODS

Body weight of 21,028 healthy residents of Linxian, Henan province, aged 40-69 years was measured two times from 1986 to 1991. Outcome events were prospectively collected up to 2016. Weight maintenance group (weight change <2 kg) or stable normal weight group was treated as the reference. Cox proportional hazard model was performed to calculate hazard ratios (HRs) and 95% confidence intervals (95% CIs) to estimate the risk of mortality.

RESULTS

A total of 21,028 subjects were included in the final analysis. Compared with the weight maintenance group, subjects with weight loss ≥2 kg had an increased risk of death from all-cause (HR All-cause  = 1.14, 95% CI: 1.09-1.19, P  <0.001), cancer (HR Cancer  = 1.12, 95% CI: 1.03-1.21, P  = 0.009), and heart disease (HR Heart diseases  = 1.21, 95% CI: 1.11-1.31, P  <0.001), whereas subjects with weight gain ≥5 kg had 11% (HR Cancer  = 0.89, 95% CI: 0.79-0.99, P  = 0.033) lower risk of cancer mortality and 23% higher risk of stroke mortality (HR Stroke  = 1.23,95% CI: 1.12-1.34, P  <0.001). For the change of weight status, both going from overweight to normal weight and becoming underweight within 5 years could increase the risk of total death (HR Overweight to normal = 1.18, 95% CI: 1.09-1.27; HR Becoming underweight = 1.35, 95% CI: 1.25-1.46) and cancer death (HR Overweight to normal = 1.20, 95% CI: 1.04-1.39; HR Becoming underweight = 1.44, 95% CI: 1.24-1.67), while stable overweight could increase the risk of total death (HR Stable overweight = 1.11, 95% CI: 1.05-1.17) and death from stroke (HR Stable overweight = 1.44, 95% CI: 1.33-1.56). Interaction effects were observed between age and weight change on cancer mortality, as well as between baseline BMI and weight change on all-cause, heart disease, and stroke mortality (all Pinteraction  <0.01).

CONCLUSIONS

Weight loss was associated with an increased risk of all-cause, cancer, and heart disease mortality, whereas excessive weight gain and stable overweight were associated with a higher risk of stroke mortality. Efforts of weight management should be taken to improve health status.

TRIAL REGISTRATION

https://classic.clinicaltrials.gov/ , NCT00342654.

Increased innate immune responses in adolescents with obesity and its relation to subclinical cardiovascular measures: An exploratory study

Cardiometabolic risk accrues across the life course and childhood and adolescence are key periods for effective prevention. Obesity is associated with inflammation in adults, but pediatric data are scarce. In a cross-sectional and longitudinal study, we investigated immune cell composition and activation in 31 adolescents with obesity (41.9% male, BMIz>2.5, 14.4 years) and 22 controls with healthy weight (45.1% male, -1.5 Collapse

Key Words

• Disease
• Health sciences
• Immunology
• Pathophysiology
• Risk factor

Collapse

MESH Headings Collapse

Grants Collapse

Affiliation(s)

Siroon Bekkering Department of Internal Medicine, Radboud University Medical Centre, Nijmegen 6525 GA, the Netherlands Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia
Christoph Saner Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Division of Pediatric Endocrinology, Diabetology and Metabolism, Department of Pediatrics, University Children`s Hospital Bern, Inselspital, Bern 3010, Switzerland Department of Biomedical Research, University of Bern, 3010 Bern, Switzerland
Boris Novakovic Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Department of Paediatrics, The University of Melbourne, Parkville, VIC 3052, Australia
Toby Mansell Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Department of Paediatrics, The University of Melbourne, Parkville, VIC 3052, Australia
Danielle K. Longmore Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Department of Endocrinology, The Royal Children`s Hospital, Parkville, VIC 3052, Australia
Zoe McCallum Department of Paediatrics, The University of Melbourne, Parkville, VIC 3052, Australia Department of Neurodevelopment and Disability, Royal Children’s Hospital, Parkville, VIC 3052, Australia
Anne-Louise Ponsonby Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, VIC 3052, Australia
Markus Juonala Department of Medicine, University of Turku, 20500 Turku, Finland Division of Medicine, Turku University Hospital, 20500 Turku, Finland
Mihai G. Netea Department of Internal Medicine, Radboud University Medical Centre, Nijmegen 6525 GA, the Netherlands Department of Immunology and Metabolism, Life and Medical Sciences Institute, University of Bonn, 53115 Bonn, Germany
Matthew A. Sabin Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Department of Endocrinology, The Royal Children`s Hospital, Parkville, VIC 3052, Australia Department of Paediatrics, The University of Melbourne, Parkville, VIC 3052, Australia
Richard Saffery Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Department of Paediatrics, The University of Melbourne, Parkville, VIC 3052, Australia
Niels P. Riksen Department of Internal Medicine, Radboud University Medical Centre, Nijmegen 6525 GA, the Netherlands
David P. Burgner Murdoch Children’s Research Institute, The Royal Children’s Hospital, Parkville, VIC 3052, Australia Department of Paediatrics, The University of Melbourne, Parkville, VIC 3052, Australia Department of Pediatrics, Monash University, Clayton, VIC 3168, Australia

Collapse

Change in adiposity is associated with change in glycoprotein acetyls but not hsCRP in adolescents with severe obesity

BACKGROUND

Obesity-associated chronic inflammation mediates the development of adverse cardiometabolic outcomes. There are sparse data on associations between severe obesity and inflammatory biomarkers in adolescence; most are cross-sectional and limited to acute phase reactants. Here, we investigate associations between adiposity measures and inflammatory biomarkers in children and adolescents with severe obesity both cross-sectionally and longitudinally.

METHODS

From the Childhood Overweight Biorepository of Australia (COBRA) study, a total of n = 262 participants, mean age 11.5 years (SD 3.5) with obesity had measures of adiposity (body mass index, BMI; % above the 95th BMI-centile, %>95th BMI-centile; waist circumference, WC; waist/height ratio, WtH; % total body fat, %BF; % truncal body fat, %TF) and inflammation biomarkers (glycoprotein acetyls, GlycA; high-sensitivity C-Reactive Protein, hsCRP; white blood cell count, WBC; and neutrophil/lymphocyte ratio, NLR) assessed at baseline. Ninety-eight individuals at mean age of 15.9 years (3.7) participated in a follow-up study 5.6 (2.1) years later. Sixty-two individuals had longitudinal data. Linear regression models, adjusted for age and sex for cross-sectional analyses were applied. To estimate longitudinal associations between change in adiposity measures with inflammation biomarkers, models were adjusted for baseline measures of adiposity and inflammation.

RESULTS

All adiposity measures were cross-sectionally associated with GlycA, hsCRP and WBC at both time points. Change in BMI, %>95th BMI-centile, WC, WtH and %TF were associated with concomitant change in GlycA and WBC, but not in hsCRP and NLR.

CONCLUSION

GlycA and WBC but not hsCRP and NLR may be useful in assessing adiposity-related severity of chronic inflammation over time.

Decreasing severity of obesity from early to late adolescence and young adulthood associates with longitudinal metabolomic changes implicated in lower cardiometabolic disease risk

BACKGROUND

Obesity in childhood is associated with metabolic dysfunction, adverse subclinical cardiovascular phenotypes and adult cardiovascular disease. Longitudinal studies of youth with obesity investigating changes in severity of obesity with metabolomic profiles are sparse. We investigated associations between (i) baseline body mass index (BMI) and follow-up metabolomic profiles; (ii) change in BMI with follow-up metabolomic profiles; and (iii) change in BMI with change in metabolomic profiles (mean interval 5.5 years).

METHODS

Participants (n = 98, 52% males) were recruited from the Childhood Overweight Biorepository of Australia study. At baseline and follow-up, BMI and the % >95th BMI-centile (percentage above the age-, and sex-specific 95th BMI-centile) indicate severity of obesity, and nuclear magnetic resonance spectroscopy profiling of 72 metabolites/ratios, log-transformed and scaled to standard deviations (SD), was performed in fasting serum. Fully adjusted linear regression analyses were performed.

RESULTS

Mean (SD) age and % >95th BMI-centile were 10.3 (SD 3.5) years and 134.6% (19.0) at baseline, 15.8 (3.7) years and 130.7% (26.2) at follow-up. Change in BMI over time, but not baseline BMI, was associated with metabolites at follow-up. Each unit (kg/m²) decrease in sex- and age-adjusted BMI was associated with change (SD; 95% CI; p value) in metabolites of: alanine (-0.07; -0.11 to -0.04; p < 0.001), phenylalanine (-0.07; -0.10 to -0.04; p < 0.001), tyrosine (-0.07; -0.10 to -0.04; p < 0.001), glycoprotein acetyls (-0.06; -0.09 to -0.04; p < 0.001), degree of fatty acid unsaturation (0.06; 0.02 to 0.10; p = 0.003), monounsaturated fatty acids (-0.04; -0.07 to -0.01; p = 0.004), ratio of ApoB/ApoA1 (-0.05; -0.07 to -0.02; p = 0.001), VLDL-cholesterol (-0.04; -0.06 to -0.01; p = 0.01), HDL cholesterol (0.05; 0.08 to 0.1; p = 0.01), pyruvate (-0.08; -0.11 to -0.04; p < 0.001), acetoacetate (0.07; 0.02 to 0.11; p = 0.005) and 3-hydroxybuturate (0.07; 0.02 to 0.11; p = 0.01). Results using the % >95th BMI-centile were largely consistent with age- and sex-adjusted BMI measures.

CONCLUSIONS

In children and young adults with obesity, decreasing the severity of obesity was associated with changes in metabolomic profiles consistent with lower cardiovascular and metabolic disease risk in adults.

Association of the Brain-derived Neurotrophic Factor Val66Met Polymorphism with Body Mass Index, Fasting Glucose Levels and Lipid Status in Adolescents

Brain-derived neurotrophic factor (BDNF) has an important role in energy balance. It suppresses food intake, reduces hepatic glucose production and converts white fat into brown fat in adipose tissue, leading to energy dissipation, lowered blood glucose and a lean phenotype. Studies have shown that the single nucleotide polymorphism (SNP) Val66Met within BDNF may be associated with obesity, insulin sensitivity, type 2 diabetes mellitus (T2DM) and dyslipidemia. The objective of the study was to investigate the association of the Val66Met polymorphism with body mass index (BMI), fasting glucose levels and lipid profile in Serbian adolescents. The study included 308 randomly selected healthy adolescents, 153 (49.68%) boys and 155 girls (50.32%), 15 years of age. Data including age, gender, height, weight, lipid profile and fasting glucose were recorded. Genotyping was performed by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. No association of this polymorphism was found with BMI and lipid profile. However, significant association was observed between this polymorphism and fasting blood glucose (FBG). Carriers of a Val/Val genotype had significantly higher mean values of fasting glucose level compared to carriers of Val/ Met and Met/Met genotypes (p = 0.01). To confirm these results multiple linear regression analysis was performed. Body mass index and gender were taken as covariates. Carriers of the Val/Val genotype had significantly higher levels of FBG (β = -0.152, p = 0.02). A statistically significant association between BMI and glucose level was also observed (β = 0.124,p = 0.033). This polymorphism could be associated with fasting glucose level in Serbian adolescents, thus further research would be of great interest to validate these results.

Lower grip strength in youth with obesity identifies those with increased cardiometabolic risk

BACKGROUND

We examined whether grip strength differentiates youth with obesity with increased cardiometabolic risk.

METHODS

The sample comprised 43 youth with severe obesity (mean age 14.8, standard deviation 3.0 years) enrolled in the Childhood Overweight BioRepository of Australia. Grip strength was normalized to body mass and categorized as low and moderate/high.

RESULTS

Youth with low grip strength had higher systolic blood pressure (mean difference 13mmHg), low-density lipoprotein cholesterol (0.26mmol/l), continuous metabolic syndrome score (0.36), and carotid intima-media thickness (0.05mm) compared with those with moderate/high grip strength.

CONCLUSIONS

Low grip strength may differentiate youth with obesity with increased cardiometabolic risk.

Time spent watching television impacts on body mass index in youth with obesity, but only in those with shortest sleep duration

AIM

To determine the interplay between sleep and sedentary behaviours on body mass index (BMI) in children with obesity.

METHODS

Cross-sectional study of 343 children with obesity aged 4-17 years, from a tertiary care weight management clinic in Melbourne, Victoria, Australia. Multifaceted data relating to activity and sleep from child and parent questionnaires analysed with anthropometric data collected during routine clinical care. Associations between sleep duration and activity measures were examined via regression models with adjustment for potential confounders.

RESULTS

Higher BMI was associated with more hours spent watching television (P = 0.04), as well as less reported enjoyment of physical activity (P = 0.005) and less time spent in organised sport activity (P = 0.005). Higher BMI was also associated with higher levels of obstructive sleep apnoea (P = 0.002). Less time in bed was associated with higher levels of BMI (P = 0.03) but analysis by sex revealed this association to only hold for males. In the whole group, a significant television and sleep interaction was seen, such that increasing television watching was associated with higher BMI, but only in those with shortest sleep duration.

CONCLUSIONS

Both poor sleep and increasing screen time (including television viewing, smart-phone use, internet use or video-gaming) appear to impact BMI in children with obesity, with a particular detrimental effect of television viewing in those who sleep less. Efforts to improve sleep time and quality in children may minimise negative effects of screen time on increasing BMI and should be included in public health strategies to combat obesity in childhood.

Evidence for Protein Leverage in Children and Adolescents with Obesity

OBJECTIVE

The aim of this study was to test the protein leverage hypothesis in a cohort of youth with obesity.

METHODS

A retrospective study was conducted in a cohort of youth with obesity attending a tertiary weight management service. Validated food questionnaires revealed total energy intake (TEI) and percentage of energy intake from carbohydrates (%EC), fats (%EF), and proteins (%EP). Individuals with a Goldberg cutoff ≥ 1.2 of the ratio of reported TEI to basal metabolic rate from fat-free mass were included. A subgroup had accelerometer data. Statistics included modeling of percentage of energy from macronutrients and TEI, compositional data analysis to predict TEI from macronutrient ratios, and mixture models for sensitivity testing.

RESULTS

A total of 137 of 203 participants were included (mean [SD] age 11.3 [2.7] years, 68 females, BMI z score 2.47 [0.27]). Mean TEI was 10,330 (2,728) kJ, mean %EC was 50.6% (6.1%), mean %EF was 31.6% (4.9%), and mean %EP was 18.4% (3.1%). The relationship between %EP and TEI followed a power function (L coefficient -0.48; P < 0.001). TEI was inversely associated with increasing %EP. In the subgroup with < 60 min/d of moderate to vigorous physical activity (n = 48), lower BMI z scores were associated with higher %EP and moderate %EC.

CONCLUSIONS

In youth with obesity, protein dilution by either carbohydrates or fats increases TEI. Assessment of dietary protein may be useful to assist in reducing TEI and BMI in youth with obesity.

Sex and puberty-related differences in metabolomic profiles associated with adiposity measures in youth with obesity

BACKGROUND

Specific patterns of metabolomic profiles relating to cardiometabolic disease are associated with increased weight in adults. In youth with obesity, metabolomic data are sparse and associations with adiposity measures unknown.

OBJECTIVES

Primary, to determine associations between adiposity measures and metabolomic profiles with increased cardiometabolic risks in youth with obesity. Secondary, to stratify associations by sex and puberty.

METHODS

Participants were from COBRA (Childhood Overweight BioRepository of Australia; a paediatric cohort with obesity). Adiposity measures (BMI, BMI z-score, %truncal and %whole body fat, waist circumference and waist/height ratio), puberty staging and NMR metabolomic profiles from serum were assessed. Statistics included multivariate analysis (principal component analysis, PCA) and multiple linear regression models with false discovery rate adjustment.

RESULTS

214 participants had metabolomic profiles analyzed, mean age 11.9 years (SD ± 3.1), mean BMI z-score 2.49 (SD ± 0.24), 53% females. Unsupervised PCA identified no separable clusters of individuals. Positive associations included BMI z-score and phenylalanine, total body fat % and lipids in medium HDL, and waist circumference and tyrosine; negative associations included total body fat % and the ratio of docosahexaenoic acid/total fatty acids and histidine. Stratifying by sex and puberty, patterns of associations with BMI z-score in post-pubertal males included positive associations with lipid-, cholesterol- and triglyceride-content in VLDL lipoproteins; total fatty acids; total triglycerides; isoleucine, leucine and glycoprotein acetyls.

CONCLUSION

In a paediatric cohort with obesity, increased adiposity measures, especially in post-pubertal males, were associated with distinct patterns in metabolomic profiles.

Psychosocial measures and weight change in a clinical paediatric population with obesity

PURPOSE

Poor quality of life has been shown to occur in youth with obesity. This study aimed to assess associations between health-related quality of life, general mental health and general psychological distress measures, collectively termed psychosocial health questionnaires (PSH), with weight outcomes in a busy paediatric weight management service.

METHODS

A cross-sectional longitudinal clinical cohort, 'Childhood Overweight BioRepository of Australia (COBRA)', was used (n = 250, median age 11, range 2-18 year, mean BMI z-score 2.5 ± 0.2). Clinical data were collected and HRQOL questionnaires; Pediatric Quality of Life 4.0 (PedsQL), 'Sizing Me/Them Up' (SMU/STU), and psychological well-being questionnaires; strengths and difficulties questionnaire (SDQ) and Kessler 10 (K10) were completed by the child and primary caregiver. PSH results were compared to age- and sex-adjusted BMI z-score at baseline and follow-up. Direct logistic regression modelling was performed to assess the impact of PSH factors on the likelihood of successful weight reduction over a period of ≥ 12 months.

RESULTS

Mean self-report PSH scores were: 68.0 ± 15.28 (PedsQL, range 0-100), 64.8 ± 15.8, (SMU, range 0-100), 17.3 ± 4.4 (SDQ, range 0-40) and 20.0 ± 7.7 (K10, range 0-50). A significant negative correlation was observed between PSH scores and childhood obesity (baseline BMI z-scores (p < 0.01)). No correlations were observed between psychological well-being measures and BMI z-scores. Higher subscale scores of the PedsQL and SDQ, which measure impaired psychosocial health and more difficulties with hyperactivity and inattention, significantly predict weight loss in children with obesity after 12 months.

CONCLUSION

PSH questionnaires may be useful in identifying individuals who require additional support to achieve weight loss goals in a tertiary weight management service.

Serum IGFBP-2 levels are associated with reduced insulin sensitivity in obese children

Insulin-like growth factor binding protein 2 (IGFBP-2) may represent a critical link between body composition and insulin sensitivity. We investigated the relationship between circulating IGFBP-2 levels, body composition, insulin sensitivity, energy intake and physical activity in children with obesity. Children were recruited via the Weight Management Service at the Royal Children's Hospital, Melbourne, as part of the Childhood Overweight BioRepository of Australia (COBRA). Comprehensive anthropometric, biochemical and environmental data were collected and compared to serum IGFBP-2 levels (measured by enzyme-linked immunosorbent assay). Multiple regression modelling was used to assess the influence of circulating IGFBP-2 levels on anthropometric and biochemical measures. One hundred and ninety-four children were included in this study (46% male). Circulating IGFBP-2 negatively correlated with age, anthropometric measures, blood pressure and insulin concentration. Positive associations were observed between insulin sensitivity index-homeostasis model assessment (ISI-HOMA) and serum IGFBP-2. In multiple regression modelling, IGFBP-2 significantly contributes to variance in systolic blood pressure (-19%, P < 0.05), circulating triglycerides (-16%, P < 0.05) and ISI-HOMA (18%, P < 0.05). No associations were observed between dietary energy intake or physical activity and IGFBP-2 levels. Circulating IGFBP-2 levels in children with obesity correlate inversely with body mass and markers of metabolic dysfunction, and positively with insulin sensitivity. These findings suggest that reduced levels of IGFBP-2 may play an important role in the pathogenesis of obesity complications in early life.

Viewpoint article: Childhood obesity--looking back over 50 years to begin to look forward

The last 50 years have seen the emergence of childhood obesity as a major public health concern and a condition now regularly encountered in routine general paediatric practice. Causes are extremely complex, bringing together multifactorial environmental factors and individual genetics, and we still do not have a clear understanding of why some children appear predisposed to exaggerated and sometimes extreme weight gain. Overweight and obese children of today face an uncertain future. They are likely to experience higher rates of type 2 diabetes and heart disease, as well as many other health problems. However, while the prevalence of childhood obesity has progressively increased over the last few decades, so has research into its underlying causes. This has led to large-scale trials aimed at improving prevention or treatment. As data have emerged from such studies, we have begun to accept that the heterogeneity of obesity means that broad 'common sense' strategies to address diet and activity will not lead to success on their own. Now is the time to begin to build on this information, dispelling myths and beliefs, in order to focus research efforts and take first steps towards more sophisticated strategies that go beyond the surface behaviours that simply potentiate obesity. Through carefully designed studies, aimed at tackling fundamental questions missed in the hasty development of 'common sense' approaches, will come answers that can lead to the development of more effective community- and health-care-orientated prevention and treatment programmes.

Childhood obesity in secondary care: national prospective audit of Australian pediatric practice

OBJECTIVE

In many countries, pediatricians offer skilled secondary care for children with conditions more challenging than can readily be managed in the primary care sector, but the extent to which this sector engages with the detection and management of obesity remains largely unexplored. This study aimed to audit the prevalence, diagnosis, patient, and consultation characteristics of obesity in Australian pediatric practices.

METHODS

This was a national prospective patient audit in Australia. During the course of 2 weeks, members of the Australian Paediatric Research Network prospectively recorded consecutive outpatient consultations by using a brief standardized data collection form. Measures included height, weight, demographics, child and parent health ratings, diagnoses, referrals, investigations, and consultation characteristics. We compared the prevalence of pediatrician-diagnosed and measured obesity (body mass index ≥95th percentile) and top-ranked diagnoses, patient, and consultation characteristics in (a) obese and nonobese children, and (b) obese children with and without a diagnosis.

RESULTS

A total of 198 pediatricians recorded 5466 consultations with 2-17 year olds, with body mass index z-scores calculated for 3436 (62.9%). Of the 12.6% obese children, only one-third received an "overweight/obese" diagnosis. Obese children diagnosed as overweight/obese were heavier, older, and in poorer health than those not diagnosed and incurred more Medicare (government-funded health system) cost and referrals.

CONCLUSIONS

Obesity is infrequently clinically diagnosed by Australian pediatricians and measurement practices vary widely. Further research could focus on supporting and normalizing clinical obesity activities from which pediatricians and parents could see clear benefits.

The LIFE child study: a life course approach to disease and health

BACKGROUND

Profound knowledge about child growth, development, health, and disease in contemporary children and adolescents is still rare. Epidemiological studies together with new powerful research technologies present exciting opportunities to the elucidation of risk factor-outcome associations with potentially major consequences for prevention, diagnosis and treatment.

AIM

To conduct a unique prospective longitudinal cohort study in order to assess how environmental, metabolic and genetic factors affect growth, development and health from fetal life to adulthood.

METHODS

The 'Leipzig Research Centre for Civilization Diseases (LIFE) Child Study' focuses on two main research objectives: (1) monitoring of normal growth, development and health; (2) non-communicable diseases such as childhood obesity and its co-morbidities, atopy and mental health problems. Detailed assessments will be conducted alongside long-term storage of biological samples in 2,000 pregnant women and more than 10,000 children and their families.

RESULTS

Close coordination and engagement of a multidisciplinary team in the LIFE Child study successfully established procedures and systems for balancing many competing study and ethical needs. Full participant recruitment and complete data collection started in July 2011. Early data indicate a high acceptance rate of the study program, successful recruitment strategies and the establishment of a representative cohort for the population of Leipzig. A series of subprojects are ongoing, and analyses and publications are on their way.

DISCUSSION

This paper addresses key elements in the design and implementation of the new prospective longitudinal cohort study LIFE Child. Given the recognized need for long-term data on adverse effects on health and protective factors, our study data collection should provide magnificent opportunities to examine complex interactions that govern the emergence of non-communicable diseases.

Design and phenomenology of the FEBSTAT study

PURPOSE

Febrile status epilepticus (FSE) has been associated with hippocampal injury and subsequent hippocampal sclerosis (HS) and temporal lobe epilepsy. The FEBSTAT study was designed to prospectively examine the association between prolonged febrile seizures and development of HS and associated temporal lobe epilepsy, one of the most controversial issues in epilepsy. We report on the baseline phenomenology of the final cohorts as well as detailed aims and methodology.

METHODS

The "Consequences of Prolonged Febrile Seizures in Childhood" (FEBSTAT) study is a prospective, multicenter study. Enrolled are children, aged 1 month to 6 years of age, presenting with a febrile seizure lasting 30 min or longer based on ambulance, emergency department, and hospital records, and parental interview. At baseline, procedures included a magnetic resonance imaging (MRI) study and electroencephalography (EEG) recording done within 72 h of FSE, and a detailed history and neurologic examination. Baseline development and behavior are assessed at 1 month. The baseline assessment is repeated, with age-appropriate developmental testing at 1 and 5 years after enrollment as well as at the development of epilepsy and 1 year after that. Telephone calls every 3 months document additional seizures. Two other groups of children are included: a "control" group consisting of children with a first febrile seizure ascertained at Columbia University and with almost identical baseline and 1-year follow-up examinations and a pilot cohort of FSE from Duke University.

KEY FINDINGS

The FEBSTAT cohort consists of 199 children with a median age at baseline of 16.0 months (interquartile range [IQR] 12.0-24.0) and a median duration of FSE of 70.0 min (IQR 47.0-110.0). Seizures were continuous in 57.3% and behaviorally intermittent (without recovery in between) in 31.2%; most were partial (2.0%) or secondary generalized (65.8%), and almost all (98.0%) culminated in a generalized tonic-clonic seizure. Of the 199 children, 86.4% had normal development and 20% had prior febrile seizures. In one third of cases, FSE was unrecognized in the emergency department. The Duke existing cohort consists of 23 children with a median age of FSE onset of 18.0 months (IQR 14.0-28.0) and median duration of FSE of 90.0 min (IQR 50.0-170.0). The Columbia control cohort consists of 159 children with a first febrile seizure who received almost the same workup as the FEBSTAT cohort at baseline and at 1 year. They were followed by telephone every 4 months for a median of 42 months. Among the control cohort, 64.2% had a first simple FS, 26.4% had a first complex FS that was not FSE, and 9.4% had FSE. Among the 15 with FSE, the median age at onset was 14.0 months (IQR 12.0-20.0) and the median duration of FSE was 43.0 min (IQR 35.0-75.0).

SIGNIFICANCE

The FEBSTAT study presents an opportunity to prospectively study the relationship between FSE and acute hippocampal damage, the development of mesial temporal sclerosis, epilepsy (particularly temporal lobe epilepsy), and impaired hippocampal function in a large cohort. It is hoped that this study may illuminate a major mystery in clinical epilepsy today, and permit the development of interventions designed to prevent the sequelae of FSE.

APOE genotype and cardio-respiratory fitness interact to determine adiposity in 8-year-old children from the Tasmanian Infant Health Survey

APOE plays a well established role in lipid metabolism. Animal model evidence suggests APOE may also be associated with adiposity, but this has not been thoroughly investigated in humans. We measured adiposity (BMI, truncal fat mass, waist circumference), physical activity (PA), cardiorespiratory fitness and APOE genotype (E2, E3, E4) in 292 8-year-old children from the Tasmanian Infant Health Survey (TIHS), an Australian population-based prospective birth cohort. Our aims were to examine the association of APOE with child adiposity, and to examine the interplay between this association and other measured factors. We found that APOE was associated with child lipid profiles. APOE was also associated with child adiposity measures. The association was E4 allele-specific, with adiposity lower in the E4-containing group (BMI: Mean difference -0.90 kg/m²; 95% confidence intervals (CI) -1.51, -0.28; p = 0.004). The association of APOE4 with lower BMI differed by fitness status (difference in effect p = 0.002), and was more evident among the less fit (mean difference -1.78 kg/m²; 95% CI -2.74, -0.83; p<0.001). Additionally, associations between BMI and lipids were only apparent in those of lower fitness who did not carry APOE4. Similar overall findings were observed when truncal fat mass and waist circumference were used as alternative adiposity measures. APOE4 and cardiorespitatory fitness could interact to influence child adiposity. In studies addressing the genetic determinants of childhood obesity, the context of child fitness should also be taken into account.