Search

Background: Weight loss trajectories after bariatric surgery vary widely between individuals, and predicting weight loss before the operation remains challenging. We aimed to develop a model using machine learning to provide individual preoperative prediction of 5-year weight loss trajectories after surgery., Methods: In this multinational retrospective observational study we enrolled adult participants (aged ≥18 years) from ten prospective cohorts (including ABOS [NCT01129297], BAREVAL [NCT02310178], the Swedish Obese Subjects study, and a large cohort from the Dutch Obesity Clinic [Nederlandse Obesitas Kliniek]) and two randomised trials (SleevePass [NCT00793143] and SM-BOSS [NCT00356213]) in Europe, the Americas, and Asia, with a 5 year follow-up after Roux-en-Y gastric bypass, sleeve gastrectomy, or gastric band. Patients with a previous history of bariatric surgery or large delays between scheduled and actual visits were excluded. The training cohort comprised patients from two centres in France (ABOS and BAREVAL). The primary outcome was BMI at 5 years. A model was developed using least absolute shrinkage and selection operator to select variables and the classification and regression trees algorithm to build interpretable regression trees. The performances of the model were assessed through the median absolute deviation (MAD) and root mean squared error (RMSE) of BMI., Findings: 10 231 patients from 12 centres in ten countries were included in the analysis, corresponding to 30 602 patient-years. Among participants in all 12 cohorts, 7701 (75·3%) were female, 2530 (24·7%) were male. Among 434 baseline attributes available in the training cohort, seven variables were selected: height, weight, intervention type, age, diabetes status, diabetes duration, and smoking status. At 5 years, across external testing cohorts the overall mean MAD BMI was 2·8 kg/m 2 (95% CI 2·6-3·0) and mean RMSE BMI was 4·7 kg/m 2 (4·4-5·0), and the mean difference between predicted and observed BMI was -0·3 kg/m 2 (SD 4·7). This model is incorporated in an easy to use and interpretable web-based prediction tool to help inform clinical decision before surgery., Interpretation: We developed a machine learning-based model, which is internationally validated, for predicting individual 5-year weight loss trajectories after three common bariatric interventions., Funding: SOPHIA Innovative Medicines Initiative 2 Joint Undertaking, supported by the EU's Horizon 2020 research and innovation programme, the European Federation of Pharmaceutical Industries and Associations, Type 1 Diabetes Exchange, and the Juvenile Diabetes Research Foundation and Obesity Action Coalition; Métropole Européenne de Lille; Agence Nationale de la Recherche; Institut national de recherche en sciences et technologies du numérique through the Artificial Intelligence chair Apprenf; Université de Lille Nord Europe's I-SITE EXPAND as part of the Bandits For Health project; Laboratoire d'excellence European Genomic Institute for Diabetes; Soutien aux Travaux Interdisciplinaires, Multi-établissements et Exploratoires programme by Conseil Régional Hauts-de-France (volet partenarial phase 2, project PERSO-SURG)., Competing Interests: Declaration of interests PP reports a grant from I-Site Université Lille Nord Europe, University of Lille, Métropole Européenne de Lille, Inria, Région Hauts-de-France. RP reports a grant from Swiss National Science; and foundation fees from Johnson & Johnson and the Falik Foundation. RVC reports grants from Johnson & Johnson Medical Brazil, Medtronic Brazil, Jansen Pharmaceuticals, NovoNordisk, and Abbott; and being a member of a Scientific Advisory Board for Baritek and GI Dynamics. CWLR reports grants from the Irish Research Council, Health Research Board, Science Foundation Ireland, and Anabio; being a member of the Global Advisory Board for NovoNordisk, Eli Lilly, Johnson & Johnson, Boehringer Ingelheim, GI Dynamics, Herbalife, and Irish Life Health; and has stock or stock options in Keyron and Beyond BMI. FP reports consulting fees from Novo Nordisk, Eli lilly, Medtronic, and Johnson & Johnson. All other authors declare no competing interests., (Copyright © 2023 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.)

Background: Pre-pregnancy obesity and excessive gestational weight gain (GWG) are established risk factors for adverse pregnancy, delivery and birth outcomes. Pregnancy is an ideal moment for nutritional interventions in order to establish healthier lifestyle behaviors in women at high risk of obstetric and neonatal complications., Methods: Electronic-Personalized Program for Obesity during Pregnancy to Improve Delivery (ePPOP-ID) is an open multicenter randomized controlled trial which will assess the efficacy of an e-health web-based platform offering a personalized lifestyle program to obese pregnant women in order to reduce the rate of labor procedures and delivery interventions in comparison to standard care. A total of 860 eligible pregnant women will be recruited in 18 centers in France between 12 and 22 weeks of gestation, randomized into the intervention or the control arm and followed until 10 weeks of postpartum. The intervention is based on nutrition, eating behavior, physical activity, motivation and well-being advices in which personalization is central, as well as the use of a mobile/tablet application. Inputs includes data from the medical record of participants (medical history, anthropometric data), from the web platform (questionnaires on dietary habits, eating behavior, physical activity and motivation in both groups), and adherence to the program (time of connection for the intervention group only). Data are collected at inclusion, 32 weeks, delivery and 10 weeks postpartum. As primary outcome, we will use a composite endpoint score of obstetrical interventions during labor and delivery, defined as caesarean section and instrumental delivery (forceps and vacuum extractor). Secondary outcomes will consist of data routinely collected as part of usual antenatal and perinatal care, such as GWG, hypertension, preeclampsia, as well as fetal and neonatal outcomes including premature birth, gestational age at birth, birth weight, macrosomia, Apgar score, arterial umbilical cord pH, neonatal traumatism, hyperbilirubinemia, respiratory distress syndrome, transfer in neonatal intensive care unit, and neonatal adiposity. Post-natal outcomes will be duration of breastfeeding, maternal weight retention and child weight at postnatal visit., Discussion: The findings of the ePPOP-ID trial will help design e-health intervention program for obese women in pregnancy., Trial Registration: ClinicalTrials.gov Identifier: NCT02924636 / October 5th 2016.