Fragestellung: Gegenstand dieser Untersuchung ist der diagnostische Nutzen der Neuro-Biomarker S100B-Protein und Neuronenspezifische Enolase (NSE) bei leichten Schädel-Hirn-Verletzungen im Kindesalter. Es wird untersucht, ob anhand der posttraumatischen S100B- und NSE-Serum-Konzentrationen Kinder mit einer Schädelprellung und einem leichten Schädel-Hirn-Trauma (SHT) differenziert werden können. Material und Methode: In einer prospektiven, klinischen Studie werden die posttraumatischen S100B- und NSE-Serum-Konzentrationen von Kindern im Alter zwischen 6 Monaten und 15 Jahren mit einer Schädelprellung oder einem leichten SHT untersucht. Kinder mit extrakraniellen Begleitverletzungen und Begleiterkrankungen sind ausgeschlossen. Die Blutentnahme erfolgt innerhalb von 6 Stunden nach dem Trauma. Es werden 2 diagnostische Gruppen definiert: Gruppe 1: asymptomatische Schädelprellungen (Glasgow-Coma-Scale [GCS] 15 Punkte), Gruppe 2: leichte SHT (GCS 13-15 Punkte) mit klinischen Zeichen einer Gehirnerschütterung (kurze Bewusstlosigkeit, Amnesie, Übelkeit, Erbrechen, Somnolenz, Kopfschmerzen, Schwindel, Sehstörungen, Kreislaufreaktion). Die S100B- und NSE- Konzentrationen werden zwischen beiden Diagnosegruppen verglichen. Die Korrelation zwischen S100B und NSE sowie zwischen den Markern und dem Alter der Kinder, dem Zeitraum zwischen Trauma und Blutentnahme, dem GCS-Wert und dem Vorhandensein von Kopfplatzwunden wird analysiert. Ergebnisse: 148 Kinder sind in die Studie eingeschlossen (53 Kinder mit einer Schädelprellung und 95 mit einem leichten SHT). Nach Adjustierung der gemessenen Marker-Konzentrationen auf Unterschiede im Alter und Zeitraum zwischen Trauma und Blutentnahme unterscheiden sich die S100B- und NSE-Konzentrationen nicht signifikant zwischen Kindern mit einer Schädelprellung und einem leichten SHT. Zwischen den S100B- und NSE-Konzentrationen besteht eine signifikant positive Korrelation. Beide Marker korrelieren signifikant negativ mit dem Alter und dem Entnahmezeitraum. Der GCS-Wert und das Vorhandensein von Kopfplatzwunden zeigen keinen Effekt auf die Marker-Konzentrationen. Schlussfolgerung: Die posttraumatischen S100B- und NSE-Serum-Konzentrationen zeigen keinen diagnostischen Nutzen bei der Differenzierung zwischen Kindern mit einer Schädelprellung und Kindern mit einem leichten SHT. S100B und NSE sind altersabhängige Marker.

Le pronostic de l’arrêt cardiaque (AC) reste dramatique. Diagnostiquer sa cause rapidement et prédire précocement son pronostic ("pronostication") de manière fiable permettrait de mieux guider les traitements initiaux, en évitant de traiter futilement les patients avec faible probabilité d’évolution favorable ou à l’inverse de permettre d’intensifier le traitement de patients avec forte probabilité d’évolution favorable. Les biomarqueurs, dont l’utilité diagnostique et pronostique reste débattue, semblent actuellement insuffisamment sensibles et précis, surtout dans les 1ères heures après la reprise de l’activité circulatoire spontanée (RACS). Dans l’algorithme pronostique, seule la Neuron Specific Enolase (NSE) est validée après le 3ème jour post-AC et en 2ème intention. Notre première étude a montré que la valeur diagnostique des biomarqueurs "spécifiques" des lésions cérébrales en post-AC (protéine S100B : S100 et surtout NSE) était insuffisante, à l’admission en réanimation, pour étayer précisément le diagnostic de cause neurologique d’AC. Si la coronarographie précoce est l’outil diagnostique de référence de l’AC de probable cause cardiaque, les biomarqueurs ne peuvent remplacer le scanner cérébral pour diagnostiquer une cause neurologique d’AC. La deuxième étude a évalué, au 1er jour post-AC, S100 et NSE avec 2 témoins d’œdème cérébral proposés comme outils pronostiques : le diamètre de l’enveloppe du nerf optique (DENO) par échographie et le rapport de dédifférenciation substance grise / substance blanche (DSG/B) par scanner cérébral. Même si une relation directe ne peut être affirmée formellement entre ces paramètres, l’élargissement du DENO à J1 post-AC était corrélé aux lésions cérébrales, surtout l’œdème cérébral et les lésions neuronales suspectés sur l’élévation de la NSE (à l’admission et à J1) et la baisse de DSG/B. Si NSE, DSG/B et DENO à J1 étaient liés, S100, plus spécifique de la glie, n’était pas corrélée au DENO ni au DSG/B. NSE et S100 à l’admission, à J1 et J2 post-RACS et DENO à J1 étaient associées à la mortalité hospitalière. La troisième étude évaluait la valeur pronostique des biomarqueurs à la phase précoce de l’AC (NSE et S100 étant prélevées en médiane 220 min après la RASC). S100, réalisée en aveugle des cliniciens, était le biomarqueur le plus précis à l’admission en réanimation pour prédire correctement le pronostic défavorable à la sortie de l’hôpital et à 3 mois après AC, par rapport au lactate, pH et créatininémie, et surtout à la NSE. Les variations de S100 dans le temps permettaient d’affiner cette prédiction. S100 à l’admission était un facteur indépendant du pronostic défavorable à la sortie de l’hôpital, avec la durée sans massage cardiaque, le rythme initial non-choquable, le lactate initial et la présence de convulsion clinique. Selon les recommandations, la pronostication nécessite théoriquement d’être différée et multimodale, les biomarqueurs seuls n’étant pas recommandés, surtout précocement. Les biomarqueurs ne peuvent constituer une alternative, en comparaison à l’imagerie, pour l’aide diagnostique de la cause d’AC. A l’inverse, certains biomarqueurs comme la S100 après admission pourraient facilement et spécifiquement discriminer les AC ayant une certitude de pronostic défavorable. Associée à d’autres outils prédictifs clinico-radiologiques, la S100 pourrait être incorporée dans des algorithmes permettant de guider les thérapeutiques initiales. Une pronostication correcte précoce pourrait éviter des traitements invasifs inutiles, ou au contraire optimiser certaines thérapeutiques agressives. Le choix de méthodes recommandées et automatisées de contrôle ciblé de la température, très efficaces mais invasives et onéreuses, ou l’indication d’utiliser -ou pas- une assistance cardio-circulatoire extra-corporelle pourrait bénéficier d’une telle stratégie précoce de sélection des patients. / Outcome of cardiac arrest (CA) remains dramatic. To quickly diagnose the cause of CA and establish a reliable outcome prediction (prognostication) as early as possible could help to guide initial treatments. It could avoid futile treatments in patients with low chance of survival or of good neurological recovery, or conversely allow treatment optimization in patients expected to have a high likelihood of good neurological outcome. Usefulness of biomarkers to guide clinicians in finding the CA diagnosis and helping prognostication is debated. Biomarkers are considered as not sensitive and accurate enough, especially within the first hours after return of spontaneous circulation (ROSC). Their use is only recommended in prognostication for Neuron Specific Enolase (NSE) as a second line tool and after the third day from CA. Our first study confirmed that biomarkers “specific” of brain injury (S100B protein: S100 and moreover NSE) cannot sufficiently discriminate the neurological cause of CA on ICU admission. If early coronary angiogram is the standard for diagnosing a probable cardiac cause of CA, biomarkers cannot replace brain computed-tomography (CT) in CA from a neurological cause. The second study evaluated, during the 1st day after ROSC, the link between biomarkers (S100 and NSE) and 2 surrogates of brain oedema recently proposed as outcome predictors: echography of the optic nerve sheath diameter (ONSD), and grey to white matter attenuation ratio (GWR) on brain CT-scan. Even though we cannot conclude on a definitive relationship between these parameters, ONSD enlargement at day 1 was associated with specific brain damage after CA, such as brain oedema and mostly axonal injuries, as reflected by increases in NSE (on admission and at day 1) and low GWR measurements. Whereas NSE, GWR and ONSD at day 1 were correlated, S100, which is more specific of glial injuries, did not reach significance. NSE and S100 on admission, at days 1 and 2 after ROSC, as well as ONSD at day 1, were associated with survival at hospital discharge. The third study evaluated the prognostic value of several biomarkers in the early phase after CA (NSE and S100 being sampled at median 220 min after ROSC). S100, blinded to physicians, was the biomarker with the best accuracy after ICU admission to correctly predict unfavourable outcome at hospital discharge and at 3 months after CA, compared with all other biomarkers such as lactate, pH, creatinine, and especially NSE. S100 variations during the first day after admission refined prognostication. Initial S100 was an early independent predictive factor associated with unfavourable outcome at hospital discharge, with the no-flow duration, initial lactate value, initial non-shockable rhythm, and the presence of clinical seizure. According to guidelines, prognostication theoretically needs to be delayed and multimodal, biomarkers alone not being recommended especially in the early phase after CA. Biomarkers cannot seem to be an alternative option compared to imaging to precisely diagnose the CA cause. By contrast, some biomarkers, such as S100 after admission, could easily and specifically discriminate CA patients with certainty of unfavourable outcome. Associated with other predictive tools (clinical or using imaging), biomarkers could interestingly be incorporated in early decisional algorithms to optimally guide initial therapies. This correct patient classification could help to avoid unuseful treatments versus to maximize aggressive therapies. The choice of recommended servo-controlled targeted temperature management devices, very efficient but invasive and expensive, or the indication -or not- of a cardio-circulatory assist device implementation should be guided in the early stage after ROSC using this simple strategy of patient selection.

Abstract There is a lack of reliable methods of evaluation of brain ischemic injury in patients undergoing cardiac surgery. The present study was, therefore, planned to evaluate whether serum S100β protein (I), brain cortical microdialysis (II), intracranial pressure (III) and electroencephalography (EEG) (IV) are predictive of postoperative death and brain ischemic injury in an experimental surviving porcine model of hypothermic circulatory arrest (HCA). One hundred and twenty eight (128) female, juvenile (8 to 10 weeks of age) pigs of native stock, weighing 21.0 to 38.2 kg, underwent cardio-pulmonary bypass prior to, and following, a 75-minute period of HCA at a brain temperature of 18°C. During the operation, hemodynamic, electrocardiograph and temperature monitoring was performed continuously. Furthermore, metabolic parameters were monitored at baseline, end of cooling, at intervals of two, four and eight hours after HCA and before extubation. Electroencephalographic recording was performed in all animals, serum S100β protein measurement in 18 animals, cortical microdialysis in 109 animals, and intracranial pressure monitoring in 58 animals. After the operation, assessment of behavior was made on a daily basis until death or elective sacrifice on the seventh postoperative day. All four studies showed that these parameters were predictive of postoperative outcome. Animals with severe histopathological injury had higher serum S100β protein levels at every time interval after HCA. Analysis of cortical brain microdialysis showed that the lactate/glucose ratio was significantly lower and the brain glucose concentration significantly higher among survivors during the early postoperative hours. Intracranial pressure increased significantly after 75 minutes of HCA, and this was associated with a significantly increased risk of postoperative death and brain infarction. A slower recovery of EEG burst percentage after HCA was significantly associated with the development of severe cerebral cortex, brain stem and cerebellum ischemic injury. In conclusion, serum S100β protein proved to be a reliable marker of brain ischemic injury as assessed on histopathological examination. Cerebral microdialysis is a useful method of cerebral monitoring during experimental HCA. Low brain glucose concentrations and high brain lactate/ glucose ratios after HCA are strong predictors of postoperative death. Increased intracranial pressure severely affected the postoperative outcome and may be a potential target for treatment. EEG burst percentage as a sum effect of anesthetic agent and ischemic brain damage is a useful tool for early prediction of severe brain damage after HCA. Among these monitoring methods, brain cortical microdialysis seems to be the most powerful one in predicting brain injury after experimental hypothermic circulatory arrest.

Fragestellung: Gegenstand dieser Untersuchung ist der diagnostische Nutzen der Neuro-Biomarker S100B-Protein und Neuronenspezifische Enolase (NSE) bei leichten Schädel-Hirn-Verletzungen im Kindesalter. Es wird untersucht, ob anhand der posttraumatischen S100B- und NSE-Serum-Konzentrationen Kinder mit einer Schädelprellung und einem leichten Schädel-Hirn-Trauma (SHT) differenziert werden können. Material und Methode: In einer prospektiven, klinischen Studie werden die posttraumatischen S100B- und NSE-Serum-Konzentrationen von Kindern im Alter zwischen 6 Monaten und 15 Jahren mit einer Schädelprellung oder einem leichten SHT untersucht. Kinder mit extrakraniellen Begleitverletzungen und Begleiterkrankungen sind ausgeschlossen. Die Blutentnahme erfolgt innerhalb von 6 Stunden nach dem Trauma. Es werden 2 diagnostische Gruppen definiert: Gruppe 1: asymptomatische Schädelprellungen (Glasgow-Coma-Scale [GCS] 15 Punkte), Gruppe 2: leichte SHT (GCS 13-15 Punkte) mit klinischen Zeichen einer Gehirnerschütterung (kurze Bewusstlosigkeit, Amnesie, Übelkeit, Erbrechen, Somnolenz, Kopfschmerzen, Schwindel, Sehstörungen, Kreislaufreaktion). Die S100B- und NSE- Konzentrationen werden zwischen beiden Diagnosegruppen verglichen. Die Korrelation zwischen S100B und NSE sowie zwischen den Markern und dem Alter der Kinder, dem Zeitraum zwischen Trauma und Blutentnahme, dem GCS-Wert und dem Vorhandensein von Kopfplatzwunden wird analysiert. Ergebnisse: 148 Kinder sind in die Studie eingeschlossen (53 Kinder mit einer Schädelprellung und 95 mit einem leichten SHT). Nach Adjustierung der gemessenen Marker-Konzentrationen auf Unterschiede im Alter und Zeitraum zwischen Trauma und Blutentnahme unterscheiden sich die S100B- und NSE-Konzentrationen nicht signifikant zwischen Kindern mit einer Schädelprellung und einem leichten SHT. Zwischen den S100B- und NSE-Konzentrationen besteht eine signifikant positive Korrelation. Beide Marker korrelieren signifikant negativ mit dem Alter und dem Entnahmezeitraum. Der GCS-Wert und das Vorhandensein von Kopfplatzwunden zeigen keinen Effekt auf die Marker-Konzentrationen. Schlussfolgerung: Die posttraumatischen S100B- und NSE-Serum-Konzentrationen zeigen keinen diagnostischen Nutzen bei der Differenzierung zwischen Kindern mit einer Schädelprellung und Kindern mit einem leichten SHT. S100B und NSE sind altersabhängige Marker.

Estudos prévios demonstram que intervenções precoces levam a alterações comportamentais e neuroendócrinas na vida adulta. Nossos achados anteriores demonstram que animais manipulados no período neonatal consomem mais alimento palatável na vida adulta e apresentam um menor aumento do depósito de gordura abdominal, após exposição crônica a dieta palatável (chocolate) em relação aos animais intactos. Neste trabalho de tese, nosso objetivo foi avaliar os efeitos da manipulação neonatal sobre a preferência alimentar bem como sobre a regulação metabólica de ratas adultas. Para isso, investigamos parâmetros metabólicos e neuroquímicos em resposta à exposição crônica a uma dieta hipercalórica e palatável, assim como a um período de abstinência desse tipo de alimento em ratas fêmeas adultas expostas ou não à manipulação neonatal (10 min/dia, 10 primeiros dias de vida). A manipulação neonatal induziu maior ingestão de alimento palatável após um curto período de privação. Entretanto, o consumo durante a exposição crônica a essa dieta não diferiu entre os grupos experimentais. Também observamos que ratas fêmeas manipuladas, quando cronicamente expostas à dieta palatável na vida adulta, têm menor aumento da gordura abdominal e esse efeito persiste após a privação da dieta. Ratas não-manipuladas apresentaram níveis mais elevados de colinesterases no soro após exposição crônica a dieta palatável, entretanto, sem alterações na atividade de colinesterases no córtex cerebral. Foi identificado, após exposição crônica a dieta palatável, uma redução na atividade da enzima Na+,K+-ATPse no hipocampo e na amígdala e um aumento nos níveis plasmáticos de S100B em ratas não-manipuladas no período neonatal. Após as primeiras 24h de privação do alimento palatável, ratas fêmeas não-manipuladas demonstraram maior frequência de sinais de abstinência (tremores de cabeça) em comparação com ratas manipuladas no período neonatal. Assim, esses achados sugerem que a manipulação neonatal determina alterações persistentes no comportamento alimentar e previne algumas alterações periféricas e centrais induzidas pela exposição crônica a uma dieta hiperpalatável, modulando a resposta metabólica de modo a reduzir a vulnerabilidade de dano metabólico e neural. / Previous studies have demonstrated that an intervention early in life leads to behavior and neuroendocrine alterations in adulthood. According to our previous findings neonatallyhandled animals have an increased consumption of palatable food, as well as a lower increase in abdominal fat accumulation after being chronically exposed to a highly palatable diet (chocolate) as compared with intact rats. In the present study, our aim was to evaluate the effects of neonatal handling on food preference and metabolic regulation in adult female rats. Therefore, we investigated metabolic and neurochemical parameters in response to a chronic exposure to a highly palatable diet, and to its withdrawal in adult female rats exposed or not to neonatal handling procedure (10 min/day, 10 first days of life). We observed an effect of neonatal handling inducing an increased palatable food intake after one week of chocolate withdrawal. However, chocolate consumption during long-term exposure to this type of diet did not differ between experimental groups. After a 30-days-period of chocolate exposure, non-handled female rats exhibited an increased abdominal fat deposition in comparison to neonatally-handled rats, and this effect persisted even after chocolate withdrawal. Nonhandled rats had increased serum cholinesterase levels after chronic exposure to palatable diet, without alterations in cerebral cortex cholinesterase activity. We also observed that chocolate consumption lead to a reduced Na+,K+-ATPse activity in hippocampus and amygdala, as well as an increased plasma S100B levels in non-handled females rats. After the first 24h of chocolate withdrawal, non-handled female rats exhibited an increased frequency of head shakes, during the Open Field task, in comparison to handled rats. Therefore, these findings suggest that neonatal handling leads to persistent alterations in feeding behavior, and also prevents some peripheral and central alterations induced by chronic exposure to a highly palatable diet; modulating the metabolic response in order to reduce the vulnerability to metabolic and neuronal damage in adulthood.

Estudos prévios demonstram que intervenções precoces levam a alterações comportamentais e neuroendócrinas na vida adulta. Nossos achados anteriores demonstram que animais manipulados no período neonatal consomem mais alimento palatável na vida adulta e apresentam um menor aumento do depósito de gordura abdominal, após exposição crônica a dieta palatável (chocolate) em relação aos animais intactos. Neste trabalho de tese, nosso objetivo foi avaliar os efeitos da manipulação neonatal sobre a preferência alimentar bem como sobre a regulação metabólica de ratas adultas. Para isso, investigamos parâmetros metabólicos e neuroquímicos em resposta à exposição crônica a uma dieta hipercalórica e palatável, assim como a um período de abstinência desse tipo de alimento em ratas fêmeas adultas expostas ou não à manipulação neonatal (10 min/dia, 10 primeiros dias de vida). A manipulação neonatal induziu maior ingestão de alimento palatável após um curto período de privação. Entretanto, o consumo durante a exposição crônica a essa dieta não diferiu entre os grupos experimentais. Também observamos que ratas fêmeas manipuladas, quando cronicamente expostas à dieta palatável na vida adulta, têm menor aumento da gordura abdominal e esse efeito persiste após a privação da dieta. Ratas não-manipuladas apresentaram níveis mais elevados de colinesterases no soro após exposição crônica a dieta palatável, entretanto, sem alterações na atividade de colinesterases no córtex cerebral. Foi identificado, após exposição crônica a dieta palatável, uma redução na atividade da enzima Na+,K+-ATPse no hipocampo e na amígdala e um aumento nos níveis plasmáticos de S100B em ratas não-manipuladas no período neonatal. Após as primeiras 24h de privação do alimento palatável, ratas fêmeas não-manipuladas demonstraram maior frequência de sinais de abstinência (tremores de cabeça) em comparação com ratas manipuladas no período neonatal. Assim, esses achados sugerem que a manipulação neonatal determina alterações persistentes no comportamento alimentar e previne algumas alterações periféricas e centrais induzidas pela exposição crônica a uma dieta hiperpalatável, modulando a resposta metabólica de modo a reduzir a vulnerabilidade de dano metabólico e neural. / Previous studies have demonstrated that an intervention early in life leads to behavior and neuroendocrine alterations in adulthood. According to our previous findings neonatallyhandled animals have an increased consumption of palatable food, as well as a lower increase in abdominal fat accumulation after being chronically exposed to a highly palatable diet (chocolate) as compared with intact rats. In the present study, our aim was to evaluate the effects of neonatal handling on food preference and metabolic regulation in adult female rats. Therefore, we investigated metabolic and neurochemical parameters in response to a chronic exposure to a highly palatable diet, and to its withdrawal in adult female rats exposed or not to neonatal handling procedure (10 min/day, 10 first days of life). We observed an effect of neonatal handling inducing an increased palatable food intake after one week of chocolate withdrawal. However, chocolate consumption during long-term exposure to this type of diet did not differ between experimental groups. After a 30-days-period of chocolate exposure, non-handled female rats exhibited an increased abdominal fat deposition in comparison to neonatally-handled rats, and this effect persisted even after chocolate withdrawal. Nonhandled rats had increased serum cholinesterase levels after chronic exposure to palatable diet, without alterations in cerebral cortex cholinesterase activity. We also observed that chocolate consumption lead to a reduced Na+,K+-ATPse activity in hippocampus and amygdala, as well as an increased plasma S100B levels in non-handled females rats. After the first 24h of chocolate withdrawal, non-handled female rats exhibited an increased frequency of head shakes, during the Open Field task, in comparison to handled rats. Therefore, these findings suggest that neonatal handling leads to persistent alterations in feeding behavior, and also prevents some peripheral and central alterations induced by chronic exposure to a highly palatable diet; modulating the metabolic response in order to reduce the vulnerability to metabolic and neuronal damage in adulthood.

Estudos prévios demonstram que intervenções precoces levam a alterações comportamentais e neuroendócrinas na vida adulta. Nossos achados anteriores demonstram que animais manipulados no período neonatal consomem mais alimento palatável na vida adulta e apresentam um menor aumento do depósito de gordura abdominal, após exposição crônica a dieta palatável (chocolate) em relação aos animais intactos. Neste trabalho de tese, nosso objetivo foi avaliar os efeitos da manipulação neonatal sobre a preferência alimentar bem como sobre a regulação metabólica de ratas adultas. Para isso, investigamos parâmetros metabólicos e neuroquímicos em resposta à exposição crônica a uma dieta hipercalórica e palatável, assim como a um período de abstinência desse tipo de alimento em ratas fêmeas adultas expostas ou não à manipulação neonatal (10 min/dia, 10 primeiros dias de vida). A manipulação neonatal induziu maior ingestão de alimento palatável após um curto período de privação. Entretanto, o consumo durante a exposição crônica a essa dieta não diferiu entre os grupos experimentais. Também observamos que ratas fêmeas manipuladas, quando cronicamente expostas à dieta palatável na vida adulta, têm menor aumento da gordura abdominal e esse efeito persiste após a privação da dieta. Ratas não-manipuladas apresentaram níveis mais elevados de colinesterases no soro após exposição crônica a dieta palatável, entretanto, sem alterações na atividade de colinesterases no córtex cerebral. Foi identificado, após exposição crônica a dieta palatável, uma redução na atividade da enzima Na+,K+-ATPse no hipocampo e na amígdala e um aumento nos níveis plasmáticos de S100B em ratas não-manipuladas no período neonatal. Após as primeiras 24h de privação do alimento palatável, ratas fêmeas não-manipuladas demonstraram maior frequência de sinais de abstinência (tremores de cabeça) em comparação com ratas manipuladas no período neonatal. Assim, esses achados sugerem que a manipulação neonatal determina alterações persistentes no comportamento alimentar e previne algumas alterações periféricas e centrais induzidas pela exposição crônica a uma dieta hiperpalatável, modulando a resposta metabólica de modo a reduzir a vulnerabilidade de dano metabólico e neural. / Previous studies have demonstrated that an intervention early in life leads to behavior and neuroendocrine alterations in adulthood. According to our previous findings neonatallyhandled animals have an increased consumption of palatable food, as well as a lower increase in abdominal fat accumulation after being chronically exposed to a highly palatable diet (chocolate) as compared with intact rats. In the present study, our aim was to evaluate the effects of neonatal handling on food preference and metabolic regulation in adult female rats. Therefore, we investigated metabolic and neurochemical parameters in response to a chronic exposure to a highly palatable diet, and to its withdrawal in adult female rats exposed or not to neonatal handling procedure (10 min/day, 10 first days of life). We observed an effect of neonatal handling inducing an increased palatable food intake after one week of chocolate withdrawal. However, chocolate consumption during long-term exposure to this type of diet did not differ between experimental groups. After a 30-days-period of chocolate exposure, non-handled female rats exhibited an increased abdominal fat deposition in comparison to neonatally-handled rats, and this effect persisted even after chocolate withdrawal. Nonhandled rats had increased serum cholinesterase levels after chronic exposure to palatable diet, without alterations in cerebral cortex cholinesterase activity. We also observed that chocolate consumption lead to a reduced Na+,K+-ATPse activity in hippocampus and amygdala, as well as an increased plasma S100B levels in non-handled females rats. After the first 24h of chocolate withdrawal, non-handled female rats exhibited an increased frequency of head shakes, during the Open Field task, in comparison to handled rats. Therefore, these findings suggest that neonatal handling leads to persistent alterations in feeding behavior, and also prevents some peripheral and central alterations induced by chronic exposure to a highly palatable diet; modulating the metabolic response in order to reduce the vulnerability to metabolic and neuronal damage in adulthood.