OBJECTIVE:To identify prenatal predictors of poor perinatal outcome in fetuses with isolated sacrococcygeal teratoma (SCT). METHODS:This was a retrospective study of fetuses with isolated (non-syndromic) SCT managed at one of five pediatric surgery and/or fetal medicine centers between January 2007 and December 2017. The primary outcome was the occurrence of poor perinatal outcome, defined as prenatal death (including termination), or neonatal death or severe compromise (hemorrhagic shock). Data regarding prenatal diagnosis (sonographic features both at referral and at the last ultrasound examination before pregnancy outcome, assessment of SCT growth velocity), perinatal complications and outcome, and neonatal course were analyzed to determine prenatal SCT characteristics associated with adverse perinatal outcome. RESULTS:Fifty-five fetuses were included, diagnosed with isolated SCT at a median gestational age of 22 (interquartile range, 18-23) weeks. There was a poor perinatal outcome in 31% (n = 17) of these cases, including intrauterine fetal demise (4%, n = 2), pregnancy termination (13%, n = 7) and neonatal severe compromise (15%, n = 8), leading to neonatal death in five cases. The overall survival rate after prenatal diagnosis of isolated SCT was 75% (n = 41 of 55). Earlier gestational age at diagnosis (P = 0.02), large tumor volume at referral (P < 0.001), presence of one or more hemodynamic complications (P = 0.02), fast tumor growth velocity (P < 0.001) and high tumor grade (highest tumor grade ≥ 3) (P = 0.049) were associated with poor perinatal outcome on univariate analysis. On stepwise logistic regression analysis, tumor growth velocity was the only remaining independent factor associated with poor perinatal outcome (odds ratio (OR) (per 1-mm/week increase), 1.48 (95% CI, 1.22-1.97), P = 0.001). The best predictive cut-off of tumor growth velocity for poor perinatal outcome was 7 mm/week (OR, 25.7 (95% CI, 5.6-191.3), P < 0.001), yielding a sensitivity of 88% and a specificity of 77%. CONCLUSIONS:Approximately 30% of fetuses with a diagnosis of isolated SCT have poor perinatal outcome. Tumor growth velocity ≥ 7 mm/week appears to be an appropriate discriminative cut-off for poor perinatal outcome. These results could help to inform prenatal management and counseling of parents with an affected pregnancy. © 2024 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Abstract Objective  The aim of this retrospective study was to describe the risk of postoperative recurrence (POR) after ileocecal resection, the occurrence of surgical complications, and identify predictors of these adverse postoperative outcomes in pediatric Crohn's disease (CD). Patients and methods  All the children less than 18 years of age with a diagnosis of CD, who underwent primary ileocecal resection for CD between January 2006 and December 2016 in our tertiary center, were considered for inclusion. Factors related to POR were investigated. Results  A total of 377 children were followed for CD between 2006 and 2016. During this period, 45 (12%) children needed an ileocecal resection. POR was diagnosed in 16% ( n  = 7) at 1 year and 35% ( n  = 15) at the end of the follow-up, with a median follow-up of 2.3 years (Q1–Q3 1.8–3.3). Median duration of the postoperative clinical remission was 1.5 years (range 0.5–2). Multivariate Cox regression analysis identified only young age at diagnosis as a risk factor for POR. In total, 7 of the 43 patients (16%) developed severe postoperative complications, defined as requiring surgical, endoscopic, or radiological intervention. The only risk factor was intraoperative abscess. Conclusion  Only young age at diagnosis was associated with POR. This information could be useful to develop targeted therapeutic strategies for young CD children. At the end of follow-up with a median follow-up of 2.3 years (Q1–Q3 1.8–3.3), there was no surgical POR: endoscopic dilatation for POR should be considered in order to delay or prevent surgery.

Contexte et objectifs Les tumeurs pulmonaires primaires (TPP) chez les enfants sont rares, et la chirurgie reste le traitement principal. L’objectif est de décrire leurs caractéristiques cliniques et thérapeutiques, ainsi que l’impact respiratoire au décours du traitement. Méthodes Nous avons réalisé une cohorte nationale française de TPP pédiatriques de 2013 à 2023. Résultats et discussion Soixante-deux patients ont été inclus. L’âge médian au diagnostic était de 3,6 ans, avec des symptômes chez 54 patients. Soixante patients ont eu une résection chirurgicale : 32 lobectomies, 15 wedges, 5 segmentectomies et 5 pneumectomies. Un abord par thoracoscopie n’a été réalisé que dans 14 % des cas. Le pleuropneumoblastome représente 50 % de la cohorte. Les autres histologies sont principalement des tumeurs neuroendocrines, des carcinomes mucoépidermoïdes, des tumeurs inflammatoires myofibroblastiques. Six mois après la chirurgie et au dernier suivi, respectivement 11 puis 8 patients avaient des symptômes respiratoires ; et respectivement 10 puis 3 patients ont eu des complications chirurgicales. Au dernier suivi, seulement 20 enfants ont eu des EFR. Conclusions La chirurgie est essentielle pour assurer la rémission complète dans les TPP et semble bien tolérée. Cependant, un suivi pneumologique accru après la chirurgie semble nécessaire pour optimiser la rééducation pulmonaire post-chirurgie et donc la qualité de vie de l’enfant.

Background: In children with sickle cell anemia (SCA), the spleen is altered early in life and may be the site of serious complications such as acute splenic sequestration (ASS) or hypersplenism. Surgical splenectomy may be necessary in those who experience such complications, but both the consequent infectious and thrombo-embolic risks, and the age at which splenectomy may be safely performed, remain unclear. Aims: The objective of our study was to determine the post-splenectomy incidence of invasive bacterial infections and thrombo-embolic complications in a large cohort of children with SCA. The secondary objective was to analyze the effect of age at splenectomy on these risks. Methods: Retrospective review of children with SCA (SS or SB0-thalassemia) splenectomized during the 2000-2018 period at Robert Debré University Hospital (Paris, France) was undertaken. The study received ethical approval. The main clinical characteristics of patients, indications for splenectomy and outcomes were analyzed. Results: A total of 188 children were included, with 164 (87.2%) SS children. Median (Q1-Q3) age at splenectomy was 4.1 yrs (2.5-7.3). Median duration of post splenectomy follow-up (FU) was 5.9 yrs (2.7-9.2) yielding 1192.6 patients-years (PY) of observation. One death (0.5%) occurred during FU, unrelated to splenectomy. Indications for splenectomy were ASS in 101 (53.7%) cases, hypersplenism in 75 (39.9%) and diverse reasons in the remaining children (splenomegaly n=5; delayed hemolytic post transfusion reaction n=5, other n=2). Pneumococcal immunization coverage was good with 166 (95.9%) eligible children immunized with polysaccharidic pneumococcal vaccination at the time of splenectomy and 7 (4.0%) after the procedure. Regarding the conjugated pneumococcal vaccination, 158 (91.9%) of eligible patients had received at least one injection at the time of splenectomy. All patients received penicillin prophylaxis. Overall incidence of invasive bacterial infection was 0.005 /PY (n=6 in 5 patients, no pneumococcal infections). Overall incidence of thrombo-embolic events was 0.003 /PY (n= 4 in 4 patients). In this cohort, 123 (65.4%) and 65 (34.6%) children were splenectomized > or < 3 years of age, respectively. Overall incidence of invasive bacterial infection and thrombo-embolic events was not different according to the age at splenectomy (0.005/ PY in both groups) and 0.004/ PY versus 0.003/ PY (p= 0.7), respectively. Likewise, there was no difference between these groups in the incidence of acute chest syndrome or vaso-occlusive events. Conversely, a significant increase in the overall incidence of cerebral vasculopathy (including abnormal TCD, stroke and cerebral stenosis) was found in children splenectomized before 3 (0.037/PY) versus after 3 (0.011/ PY), p<0.01). Patients who presented with ASS (n=101) were splenectomized at a median age of 3.3 yrs (2.4-5.4), following a median number of 3 (2-4) episodes of ASS and therefore contributed to a large proportion of patients splenectomized under 3 (n=44/65, 67.7 %). Summary/Conclusion: To date this is the largest pediatric cohort study of splenectomized patients with SCA. We show that surgical splenectomy does not result in a significant increased risk of complications, notably invasive bacterial infections or thrombo-embolic complication. Splenectomy should not be delayed in patients once there is an indication for surgical removal, and the children have received recommended vaccines and prophylactic penicillin therapy. The relationship between ASS, splenectomy and cerebral vasculopathy needs to be further assessed.

Abstract Introduction Long-gap esophageal atresia (LGEA) leads to multifactorial morbidity. The aim of this study is to analyze midterm gastrointestinal and respiratory morbidities with special assessment of feeding difficulties. Secondary goals were to determine risk factors of malnutrition and the impact of surgical techniques on morbidity. Methods We conducted a prospective observational national study. Using French national database, we reviewed medical charts of every patient treated for LGEA between 2008 and 2010 in France. Phone contact was proposed to assess orality disorders using Functional Oral Intake Scale (FOIS). Patients with complete data were included. We compared conservative management with esophageal replacement. Results We included 31 cases with a median age of follow-up of 9 years [7–10]. Median z-score for weight was −0.97 [−3.52–2.50], Conclusion Midterm morbidity in LGEA concerns 80% of our population. Gastrointestinal morbidity includes mainly dysphagia, GER, and orality disorders. This study suggests that conservative management provides more orality disorders. Other studies are mandatory to confirm it.

Abstract Objectives and study Anastomotic stricture (AS) is a frequent complication of the surgery for œsophageal atresia (OA) during the first year of life. The primary objective of this study was to evaluate the prevalence of AS before 1 year old in infants with type A and C OA who were operated on. Secondary objectives were to determine risk factors for AS in OA, for recurrent and refractory AS, and to establish if AS is associated with antireflux surgery. Methods A prospective national multicentric study was conducted including all infants born with OA between 2008 and 2015. Patients deceased before one year old, OA types B and E, and patients for whom data about AS were missing were excluded from the study. Data were collected at birth and at 12 months of age. Anastomosis under tension was defined by the surgeon and a delayed anastomosis was defined by an anastomosis after 15 days of life. Recurrent stricture was defined by the need of ≥3 dilations and refractory stricture was defined by the need of ≥5 dilations. Univariate and multivariate statistical analyses were conducted. Results Of the 1258 eligible patients (84%), 1054 were included in the study from 38 centers. The prevalence of AS in the first year of life was 23.3% [20.7–28.9]. Anastomosis under tension (AUT) and delayed anastomosis (DA) were found to be independent risk factors for AS (respectively 2.5 [1.73–3.45] and 3.7 [1.95–7.2] (OR [CL 95%])) in the total population. Neither sex, birth weight, prematurity, intrauterine growth retardation, associated malformations, type A OA, nor the type of surgical approach was a risk factor for AS. In type C OA, DA was the only risk factor for AS (OR: 3.1 [1.65–5.86]). The group with AS had 2.5-fold more fundoplication compared to the patients without AS (P = 0.0005) in the total population and in type C OA. AUT and DA were found to be independent risk factors for recurrent stricture (OR: 2.4 [1.47–3.9] and 4.7 [2.2–10.4], respectively) and DA was the only risk factor for refractory stricture (OR: 6.23 [2.4–16.2]). Conclusion Surgical factors at the time of first repair of OA are the only risk factors for AS.

With advances in surgical and neonatal care, survival of patients with esophageal atresia (EA) has improved over time. While a number of conditions associated with EA may have an impact on feeding development (delayed primary anastomosis, anastomotic leaks, recurrent tracheoesophageal fistula, anastomotic stricture, gastroesophageal reflux, esophageal dysmotility, etc.) and although children with EA experience a number of oral aversive events in their first year of life, feeding disorders (FD) are poorly described and frequently unrecognized. The primary aim of this study was to describe FD in children born with EA, with a standardized scale. The secondary aim was to describe conditions associated with FD. FEED-EASY is a multicentric French study. Parents of children born with EA between 2013 and 2016 in one of the 22 participating centers were asked to participate and received the French version of the standardized and reproductive ‘Montreal Children's Hospital Feeding Scale (MCH-FS)’. One hundred and forty-five children were included; 61 (42%) had FD according to the MCH-FS. These children were characterized by disinterest in food, oral hypersensitivity, difficulty in touching some textures and food avoidance, with an influence in quality of life. Nineteen (13%) were tube-fed between 1 and 4 years of age. Birth weight and chronic respiratory difficulties were associated with FD in children with EA. Anastomotic stricture (present in 31% of the included children) was not associated with FD. FD is frequent and unrecognized in children with EA, and can influence growth and quality of life. MCH-FS allows pediatricians to identify FD in children with EA within a couple of minutes.

Aim of the study: Congenital Central Hypoventilation Syndrome (CCHS) is a rare affection associated to Hirschsprung disease (HD) in 20% of the cases. Using the French CCHS registry, we described the population of patients suffering from both CCHS and HD reporting the outcome on these patients. Methods: Medical records were reviewed. Epidemiological, clinical, histological and genetic data were analyzed and extracted from the national French registry data. Results: 33 patients had CCHS and HD. Thirty percent had a severe form of CCHS (Death owing to CCHS or 24/24 ventilation beyond 1 year old). Fifty four percent required tracheotomy. HD's pathologic segment was classic (Rectosigmoid and left colic form) in 20% and long (Above the splenic flexure) in 80%. Twenty four percent were treated with daily irrigation, 21% had colostomy without undergoing pullthrough, and 55% underwent optimal treatment (pull through). We failed to demonstrate a correlation between severity of CCHS and HD's length. The rate of mortality was 57% and was higher in the long HD group (p = 0.0005). Fourteen patients were still alive, aged 1 to 31 years old. Ninety two percent were weaned off the 24/24 ventilation. Regarding the intestinal function, 38% presented with soiling and 30% with chronic diarrhea. Hundred percent had CCHS follow-up while only 35% had no surgical follow-up in regard to the HD. Conclusions: This is the largest study regarding the CCHS / HD association and its long-term followup. Mortality is high demonstrating that a multidisciplinary follow-up on respiratory and intestinal function is necessary to improve outcome. (C) 2019 Elsevier Inc. All rights reserved.

Summary Respiratory diseases are common in children with esophageal atresia (EA), leading to an increased morbidity and mortality in the first months of life. Objective Assess the prevalence of hospitalizations linked to a respiratory disease and of maintenance inhaled therapy at the age of 1 year in French children. Methods Population based-study using data from the French national EA register. We included all children born between 2010 and 2015 with data available at birth and at follow-up at one year of age. Results A total of 981 patients born with EA were included in the register, 75 of them (8%) being deceased at the age of 1 year. Data were missing for 60 patients, thus 846 children (86%) were retained for analysis. EAs were type III of Ladd classification in 89% and type I in 7%. Rate of prematurity was 37% while 51% presented associated malformations. At 1 year of age, 1297 hospitalizations were reported for 508 patients (60%), at least one hospitalization for a respiratory disease for 251 children (51%). Factors significantly associated with respiratory hospitalizations were longer median length of oxygen supplementation (P < 10−2) and noninvasive ventilation in the neonatal period (P = 0.02), gastrostomy tube (P < 10−2), esophageal anastomosis dilation (P < 10−2). At 1 year of age, 29% had an inhaled maintenance treatment, and 85% inhaled corticosteroids. Factors significantly associated with inhaled maintenance treatment were male gender (P = 0.04), advanced older median age at gastrostomy tube insertion (P < 10−3), enteral feeding at one year of age (P < 10−3). Conclusion This study shows a high rate of respiratory problems responsive of frequent hospitalizations and inhaled maintenance treatment in children born with EA in the first year of life.

Abstract Objectives and Study The aim of our study is to assess the risk factor of mortality and morbidity in a large population-based registry in a population of type III/C esophageal atresia (EA). Methods Under the umbrella of the national plan for rare disease, a population based register was set up in 2008 recording the data of all the live newborns with EA in France. Based on the registry data, survival and morbidity at 1 year were studied. Morbidity was approached by calculating the rate of full oral autonomy and the hospital length of stay during the first year. Multivariate analysis was performed via multinomial logistic regression to evaluate independent predictors of overall survival and morbidity items among all significant category variables identified on univariate analysis. Results A total of 1008 patients with type III EA were extracted from the national database born from January first 2008 till 31 December 2014. Mean birth weight was 2610 g (2060 to 3075). Right lateral thoracotomy was used in 93% of cases and primary anastomosis was possible in 95% of cases. Associated abnormalities were present in 53% of patients, VACTERL in 19%, and CHARGE syndrome in 3%. Mortality at 3 months was at 5% and at one year at 6% and was correlated to prenatal diagnosis (odd ratio (OR):2.96 (1.08 to 8.08)), low birth weight (OR: 0.52 (0.38 to 0.72)), and heart defect (OR: 6.09 (1.96 to 18.89)). Length of hospitalization was correlated to birth weight (OR: 0.83 (0.61–1.13)), the difficulty of the anastomosis (OR: 1.59 (0.71–3.55)) and associated abnormalities (OR: 1.93(0.65–5.68)), the prenatal diagnosis is correlated only to the rate of full oral autonomy and not with the length of hospitalization. Conclusions Surgical procedure or difficulties did not seem to affect survival in this group of patients. In addition to the continuous need to improve care of low weight neonates, our results clearly identify that efforts should be focused on the small group of EA with prenatal diagnosis and/or severe cardiac malformation to reduce even more mortality and morbidity in EA patients.

Abstract Aim of the Study Long-gap esophageal atresia (LGEA) remains a surgical challenge. This study aimed to report the results of thoracoscopic esophageal axial internal traction in LGEA. Methods This multicenter observational study included retrospectively neonates who underwent primary thoracoscopic esophageal axial internal traction for LGEA between June 2017 and July 2018. LGEA was defined as the technical impossibility to perform a primary esophageal anastomosis. The Ethical Review Board of our institution approved the study. Main Results Eight neonates were included with a median gestational age at birth of 35 weeks [25; 37] and a median birth weight of 2266 g [890; 3800], 6 types I and 2 types II according to Ladd's classification. Initial median gap between 2 esophageal ends was 5 vertebral bodies [4.5; 7]. Internal traction was performed at a median age of 5 weeks of life [1; 17] with a median operative time of 87 minutes. Four patients required at least 2 internal traction procedures. After a mean traction time of 1.5 weeks [1; 13.5], esophageal anastomosis was successfully performed in 7 patients (5 thoracoscopies, 2 thoracotomies) with a median operative time of 165 minutes. One patient needed a colonic interposition. Five of these 7 patients required an esophageal endoscopic dilatation (median number: 4 [2; 6]). Median follow-up was 9.75 months [3; 16]. Conclusions Thoracoscopic esophageal axial internal traction for LGEA was a safe and feasible procedure that allowed an esophageal anastomosis in 7 of the 8 patients. Improvement of the procedure requires setting a common protocol concerning the timing of the first internal axial traction and the duration of traction before considering esophageal anastomosis.

Abstract Introduction With advances in surgical and neonatal care, survival of patients with esophageal atresia (EA) has improved over time. While a number of conditions associated with EA may have an impact on feeding development (delayed primary anastomosis, anastomotic leaks, recurrent tracheoesophageal fistula, anastomotic stricture, gastroesophageal reflux, esophageal dysmotility, etc.) and although children with EA experience a number of oral aversive events in their first year of life, feeding disorders (FD) are poorly described and frequently unrecognized. The primary aim of this study was to describe FD in children born with EA, with a standardized scale. The secondary aim was to describe conditions associated with FD. Methods FEED-EASY is a multicentric French study. Parents of children born with EA between 2013 and 2016 in one of the 22 participating centers were asked to participate and received the French version of the standardized and reproductive ‘Montreal Children's Hospital Feeding Scale (MCH-FS)’. Results One hundred and forty-five children were included; 61 (42%) had FD according to the MCH-FS. These children were characterized by disinterest in food, oral hypersensitivity, difficulty in touching some textures and food avoidance, with an influence in quality of life. Nineteen (13%) were tube-fed between 1 and 4 years of age. Birth weight and chronic respiratory difficulties were associated with FD in children with EA. Anastomotic stricture (present in 31% of the included children) was not associated with FD. Conclusions FD is frequent and unrecognized in children with EA, and can influence growth and quality of life. MCH-FS allows pediatricians to identify FD in children with EA within a couple of minutes.

Evaluate the neonatal management and outcomes of neonates with prenatal diagnosis of esophageal atresia (EA) type A.

Desmoid tumors (DT) are rare and nonmetastasizing fibroblastic neoplasms, characterized by local invasiveness. They occur sporadically or arise in the context of familial adenomatous polyposis (FAP; 5-10% of cases). Most cases develop sporadically in young adults, but some cases also occur in children. We report the case of an adolescent girl with FAP and DT, and we discuss the therapeutic strategies. An adolescent girl with FAP underwent surgery at the age of 14 years with total proctocolectomy. She had a neo-mutation in the APC gene at codon 1068, which is not usually associated with DT. Three years later, she had painful defecations Imaging showed two abdominal DT. After a multidisciplinary team meeting, the patient was refused for surgery, and medical treatment with antihormonal agents and nonsteroidal anti-inflammatory drugs was started. Imaging 18 months later showed DT stabilization, but the patient had difficulties to control chronic pains, which required morphine treatment, hypnotic sessions, and transcutaneous electric nerve stimulation. This case highlights the importance of DT screening in patients with FAP, mainly after surgery, regardless of their age and genetic mutation. Progress remains to be made in determining DT risk factors and in developing treatment. DT are still difficult to cure because of their potential for local invasion and local recurrence, and need to be managed by a multidisciplinary team. (C) 2016 Elsevier Masson SAS. All rights reserved.

Evaluer la valeur pronostic fonctionnelle d'une variation du marquage à la calrétinine après ATA selon Swenson pour MH recto sigmoïdienne. Etude rétrospective de 2008 à 2012 de tous les patients opérés d'un ATA selon Swenson pour MH recto sigmoidienne. L'analyse du marquage par la calrétinine distinguait 2 groupes : un groupe (P-) avec une absence totale de marquage et un groupe variant (P+) avec une positivité de la sous-muqueuse. Les 2 groupes étaient comparés sur les suites post opératoires à court terme. 33 patients étaient inclus. Le recul moyen était de 4,1 ans (1–6 ans). Le groupe variant P+ incluait 17 patients (51.5%). Les deux groupes étaient comparables en terme d'âge gestationnel ou poids à la naissance, de comorbidités, de présentation clinique initiale ou de type de chirurgie. Il n'était pas retrouvé de différence statistiquement significative en terme de pronostic fonctionnel digestif entre ces 2 groupes Un variant anatomopathologique de marquage à la calrétinine était identifié dans 51.5% des formes recto sigmoidiennes de MH mais ne semble pas être un facteur pronostic. Ceci est une étape supplémentaire dans la recherche de corrélation clinico-pathologique dans la MH.

Les complications postopératoires sont réputées être plus fréquentes durant les premiers âges de la vie. Ceci est dû à l'extrême fragilité de ces patients (tout particulièrement en période néonatale), à la plus grande vulnérabilité de ces patients aux événements hémodynamiques et respiratoire périopératoire et aux difficultés chirurgicales liées aux pathologies malformatives ou au petit poids [1]. Toutefois, il serait intéressant de pouvoir prédire les patients à risques de complications afin d'adapter la prise en charge périopératoire. Après accord du comité d'éthique, nous avons recueilli prospectivement les données périopératoires de patients âgés de moins de 6 mois et opérés dans notre établissement sur une période de 6 mois. Les données étaient l'âge des patients, l'âge gestationnel, l'âge corrigé, les données démographiques, les données biologiques préopératoires (taux d'hémoglobine et de fibrinogène), le type de chirurgie et les complications postopératoires (complications nécessitant la prise en charge en réanimation ou une reprise chirurgicale). Ces dernières étaient classées en hémodynamiques, respiratoires, infectieuses, neurologiques, hémorragiques ou échec précoce de la chirurgie. Plusieurs modèles était testé par le logiciel SPSS modeler : régression logistique, arbre de segmentation et réseau de neurones. Quatre-vingt-quinze patients ont été inclus dans cette analyse. L'âge était de 42 ± 44 jours, l'âge corrigé de 42 ± 7 semaines, le pourcentage de prématurés était de 29,4 % et celui des patients ASA III de 18 %. Les chirurgies pratiquées étaient : digestives (60 %), thoraciques (13,7 %), urologiques (9,5 %), ORL (9,5 %), orthopédiques (4,2 %) et ophtalmologiques (3,2 %). Des complications postopératoires sont survenues dans 20 % des cas et se répartissaient comme indiqué en Tableau 1. Le modèle le plus prédictif de la survenue de ces complications était le réseau de neurones avec les facteurs suivants : le type d'intervention, le score ASA III, les taux d'hémoglobines et de fibrinogènes préopératoires et l'âge corrigé. Ce modèle expliquait 100 % de la variabilité (aire soue la courbe ROC 1 [IC95 % : 1–1]). La cross-validation a donné les même résultats. Les deux autres modèles testés donnaient des variabilités expliquées de respectivement 24 et 36 % pour la régression logistique et l'arbre de segmentation. Le modèle statistique par réseau de neurones, peu utilisé dans le domaine médical, permet de traiter des informations complexes. Dans le cas présent il a pu donner un modèle prédictif à 100 % et reproductible (cross-validation) de la survenue de complications postopératoires. Bien qu'il reste à valider à plus grande échelle, ce modèle pourrait permettre d'anticiper des optimisations de prise en charge de patients et tout particulièrement concernant le taux d'hémoglobine préopératoire ou l'âge ou les patients doivent être opérés dans le cas de chirurgies non urgentes.

Rapporter le suivi à moyen terme des enfants porteurs de Maladie de Hirschsprung (MH) colique étendue ou pancolique étude rétrospective des patients pris en charge de 1991 à 2013. Tous avaient un abaissement laparoscopique selon Duhamel. Age et poids, sexe, longueur d'aganglionose, comorbidités, suites post opératoires, fonction digestive et croissance à moyen terme, données de l'endoscopie réalisée pour entérocolite répétée étaient notés. 26 patients étaient opérés : 10 coliques étendues et 16 iléales (moyenne = 14 cm). Le suivi médian était de 10 ans (1-19 ans). La mortalité était nulle et la morbidité de 20%. 53% avaient une fonction digestive satisfaisante. La courbe de croissance staturo-pondérale était <1 DS chez 50% des patients. Des entérocolites tardives (délai moyen 3.7 ans (1-8 ans)) survenaient chez 27%. Des lésions Crohn-like étaient retrouvées dans 100% des cas à l'endoscopie. Ces lésions étaient traitées par corticoïdes, immunosuppresseurs ou anti-TNF. Une iléostomie était nécessaire pour 3. L'évolution à moyen terme des formes longues de MH peut être marquée par la fréquence d'épisodes d'entérocolite, décrites en endoscopie comme des lésions Crohn-like.