ASSOCIATION BETWEEN TELOMERE LENGTH AND SYMPTOMS OF ATTENTION DEFICIT HYPERACTIVITY DISORDER (ADHD) IN CHILDREN AGED 6 TO 12 YEARS

Abstract

S74 whether the pMO is a good proxy of puberty onset and, therefore, a good candidate to be used in epidemiological consortia where TS evaluation is unavailable. Methods: The sample includes 760 children from the Spanish clinic cohort GENOBOX and 2,750 children from the Portuguese birth cohort Generation XXI (G21), with complete information on TS, child sex, height, and age at the evaluation moment, collected by trained professionals. For this study, the sex-specific MO prediction modified equation previously proposed by Koziel & Malina (Sports Med. 2018; 48(1):221-236) has been considered, including stature and age. The TS data was defined by breast development in girls and genitalia in boys, obtained by a nurse/paediatrician using visual inspection/palpation. Normal puberty is described as a TS ≥ 2 in girls with 9 years or more and boys with 10 years or more. Results: The mean age in the normal puberty group was 11.8 ± 1.6 years and 12.9 ± 2.0 for females, and 12.6 ± 1.4 and 13.5 ± 1.7 for males, in G21 and GENOBOX, respectively. Most girls (50.0% in GENOBOX and 55.3% in G21) had achieved PHV (pMO > 0) by TS 3 and most boys by TS 4 in GENOBOX (75.0%) and by TS 5 in G21 (49.4%). The pMO mean values at each TS differed between early and normal developers (e.g. GENOBOX: Tanner II-Early puberty: pMO Female mean =-2.97 ± 0.32 vs. Normal puberty: pMO Female mean =-0.54 ± 1.51). Conclusions/Recommendations: This study emphasizes that pMO is related to the pubertal stages defined by TS, being a good predictor for puberty onset in normal developers. However, pMO does not seem to be a good predictor for puberty onset in early developers since pMO mean levels differed between early and normal developers at each TS. Despite being a more invasive method, TS is still the best approach to define puberty onset in epidemiological studies. Antecedentes/Objetivos: La longitud del telómero (LT) se conoce como un biomarcador de envejecimiento celular. La evidencia previa ha mostrado que el acortamiento de la LT está asociado a patologías neurodegenerativas y psiquiátricas en adultos. Sin embargo, poco se conoce sobre esta asociación en población infantil. Por ello, el objetivo del presente estudio fue explorar la asociación entre la LT y los sínto-mas de déficit de atención e hiperactividad (TDAH) en niños de 6 a 12 años. Métodos: Se analizaron los datos de 1922 niños/as pertenecientes a las cohortes del proyecto HELIX (http://www.projecthelix.eu/es) y las cohortes de Asturias, Gipuzkoa y Valencia de INMA (http://www. proyectoinma.org/). La LT se determinó a los 6-12 años mediante una muestra sanguínea usando un protocolo PCR. Los síntomas de TDAH se midieron usando la escala Conner's (compuesta por 4 subescalas) administrada a los padres. Para estimar las asociaciones se utilizaron técnicas meta-analíticas de modelos de regresión binomial ajustados por potenciales confusores. Resultados: La LT se asoció significativamente con una ratio de in-cidencia de 0,65 (IC95%: 0,46-0,92) en la subescala de síntomas de hiperactividad en la cohorte HELIX mientras que en INMA fue de 1,14 (IC95%: 0,74-1,75). Al combinar los datos, la asociación protectora en-contrada en los/as niños/as de HELIX no se mantuvo. Conclusiones/Recomendaciones: No se observa una asociación clara entre la LT y los síntomas de TDAH al combinar ambas cohortes de nacimiento. Son necesarios más estudios que exploren esta asocia-ción. Financiación: Instituto de Salud Carlos III (PI18/00825): Proyecto TeloDiPA; Generalitat Valenciana (GVA/2021/191): Proyecto TeloNeu-ro; Ayuda a la Formación de Profesorado Universitario (FPU21/01323). Background/Objectives: Obesity is a multifactorial disease that has become one of the most alarming public health diseases. Overweight and obesity in childhood have been associated with a higher risk of developing type 2 diabetes, cardiovascular diseases, and adverse mental health outcomes in childhood and later in life, so early prevention is needed. We aim to develop a predictive model for over-weight/obesity at 13 years, considering variables regarding pregnancy and the birth. Methods: The study was conducted among 4,232 mother-child pairs of Generation XXI, a population-based prospective birth cohort study in Porto, Portugal. We assessed 18 lifestyles, sociodemographic and clinical variables regarding pregnancy and birth. At 13 years, body mass index z-scores adjusted for age and sex were obtained and categorized (overweight/obesity ≥ 1 SD). Logistic regression model with cross-validation was fitted to the imputed dataset. Backward elimination was performed to exclude the subsequent least significant variable to obtain the final model. Performance analysis of the model included computing the AUC, sensitivity, specificity, and prediction error. Results: The final model showed that the most important predic-tors of overweight/obesity at 13 years were the following: low maternal education (Odds Ratio 5.3; 95% Confidence Interval 1.6-17.5), mothers with no partner (1.6; 1.1-2.4), smoking during pregnancy (1.31; 1.1-1.6), higher maternal body mass index (1.2; 1.1-1.2), excessive weight gain during pregnancy (1.2; 1.0-1.5), higher birth weight (1.03; 1.01-1.06) and being male (3.1, 1.1-8.9). Regarding the performance , we obtained: AUC = 0.67, specificity = 0.92, sensitivity = 0.24, and prediction error = 33%. These results show that the model is only predicting well the newborns that will not be obese. Conclusions/Recommendations: This study suggested that maternal education, marital status, smoking habits, body mass index, and weight gain during pregnancy are the most important modifi-able pregnancy and birth predictors of overweight/obesity at 13 years, so early prevention strategies should target these factors. We also showed that predicting the risk of overweight/obesity in adolescence might be suboptimal if only considering pregnancy and birth data.