Effect of bariatric surgery on cancer risk: results from an emulated target trial using population-based data.

BACKGROUND: The impact of weight loss induced by bariatric surgery on cancer occurrence is controversial. To study the causal effect of bariatric surgery on cancer risk from an observational database, a target-trial emulation technique was used to mimic an RCT. METHODS: Data on patients admitted between 2010 and 2019 with a diagnosis of obesity were extracted from a national hospital discharge database. Criteria for inclusion included eligibility criteria for bariatric surgery and the absence of cancer in the 2 years following inclusion. The intervention arms were bariatric surgery versus no surgery. Outcomes were the occurrence of any cancer and obesity-related cancer; cancers not related to obesity were used as negative controls. RESULTS: A total of 1 140 347 patients eligible for bariatric surgery were included in the study. Some 288 604 patients (25.3 per cent) underwent bariatric surgery. A total of 48 411 cancers were identified, including 4483 in surgical patients and 43 928 among patients who did not receive bariatric surgery. Bariatric surgery was associated with a decrease in the risk of obesity-related cancer (hazard ratio (HR) 0.89, 95 per cent c.i. 0.83 to 0.95), whereas no significant effect of surgery was identified with regard to cancers not related to obesity (HR 0.96, 0.91 to 1.01). CONCLUSION: When emulating a target trial from observational data, a reduction of 11 per cent in obesity-related cancer was found after bariatric surgery.

Comparison of predictive models for cumulative live birth rate after treatment with ART.

RESEARCH QUESTION: Can a machine learning model better predict the cumulative live birth rate for a couple after intrauterine insemination or embryo transfer than Cox regression based on their personal characteristics? STUDY DESIGN: Retrospective cohort study conducted in two French infertility centres (Créteil and Tenon Hospitals) between 2012 and 2019, including 1819 and 1226 couples at Créteil and Tenon, respectively. Two models were applied: a Cox regression, which is almost exclusively used in assisted reproductive technology (ART) predictive modelling, and a tree ensemble-based model using XGBoost implementation. Internal validations were performed on each hospital dataset separately; an external validation was then carried out on the Tenon Hospital's population. RESULTS: The two populations were significantly different, with Tenon having more severe cases than Créteil, although internal validations show comparable results (C-index of 60% for both populations). As for the external validation, the XGBoost model stands out as being more stable than Cox regression, with the latter having a higher performance loss (C-index of 60% and 58%, respectively). The explicability method indicates that the XGBoost model relies strongly on features such as the ages of a couple, causes of infertility, and the woman's body mass index or infertility duration, which is consistent with the ART literature about risk factors. CONCLUSIONS: Overall performances are still relatively modest, which is coherent with all reported ART predictive models. Explicability-based methods would allow access to new knowledge, to gain a greater comprehension of which characteristics and interactions really influence a couple's journey. These models can be used by practitioners and patients to make better informed decisions about performing ART.

Local Explanation-Based Method for Healthcare Risk Stratification.

Decision support tools in healthcare require a strong confidence in the developed Machine Learning (ML) models both in terms of performances and in their ability to provide users a deeper understanding of the underlying situation. This study presents a novel method to construct a risk stratification based on ML and local explanations. An open-source dataset was used to demonstrate the efficiency of this method that well identified the main subgroups of patients. Therefore, this method could help practitioners adjust and build protocols to improve care deliveries that would better reflect patient's risk level and profile.

Evolution of hospitalized patient characteristics through the first three COVID-19 waves in Paris area using machine learning analysis.

Characteristics of patients at risk of developing severe forms of COVID-19 disease have been widely described, but very few studies describe their evolution through the following waves. Data was collected retrospectively from a prospectively maintained database from a University Hospital in Paris area, over a year corresponding to the first three waves of COVID-19 in France. Evolution of patient characteristics between non-severe and severe cases through the waves was analyzed with a classical multivariate logistic regression along with a complementary Machine-Learning-based analysis using explainability methods. On 1076 hospitalized patients, severe forms concerned 29% (123/429), 31% (66/214) and 18% (79/433) of each wave. Risk factors of the first wave included old age (≥ 70 years), male gender, diabetes and obesity while cardiovascular issues appeared to be a protective factor. Influence of age, gender and comorbidities on the occurrence of severe COVID-19 was less marked in the 3rd wave compared to the first 2, and the interactions between age and comorbidities less important. Typology of hospitalized patients with severe forms evolved rapidly through the waves. This evolution may be due to the changes of hospital practices and the early vaccination campaign targeting the people at high risk such as elderly and patients with comorbidities.

Analysis of COVID-19 inpatients in France during first lockdown of 2020 using explainability methods.

The COVID-19 pandemic rapidly puts a heavy pressure on hospital centers, especially on intensive care units. There was an urgent need for tools to understand typology of COVID-19 patients and identify those most at risk of aggravation during their hospital stay. Data included more than 400 patients hospitalized due to COVID-19 during the first wave in France (spring of 2020) with clinical and biological features. Machine learning and explainability methods were used to construct an aggravation risk score and analyzed feature effects. The model had a robust AUC ROC Score of 81%. Most important features were age, chest CT Severity and biological variables such as CRP, O2 Saturation and Eosinophils. Several features showed strong non-linear effects, especially for CT Severity. Interaction effects were also detected between age and gender as well as age and Eosinophils. Clustering techniques stratified inpatients in three main subgroups (low aggravation risk with no risk factor, medium risk due to their high age, and high risk mainly due to high CT Severity and abnormal biological values). This in-depth analysis determined significantly distinct typologies of inpatients, which facilitated definition of medical protocols to deliver the most appropriate cares for each profile. Graphical Abstract Graphical abstract represents main methods used and results found with a focus on feature impact on aggravation risk and identified groups of patients.

Deep learning-based classification of retinal vascular diseases using ultra-widefield colour fundus photographs.

OBJECTIVE: To assess the ability of a deep learning model to distinguish between diabetic retinopathy (DR), sickle cell retinopathy (SCR), retinal vein occlusions (RVOs) and healthy eyes using ultra-widefield colour fundus photography (UWF-CFP). METHODS AND ANALYSIS: In this retrospective study, UWF-CFP images of patients with retinal vascular disease (DR, RVO, and SCR) and healthy controls were included. The images were used to train a multilayer deep convolutional neural network to differentiate on UWF-CFP between different vascular diseases and healthy controls. A total of 224 UWF-CFP images were included, of which 169 images were of retinal vascular diseases and 55 were healthy controls. A cross-validation technique was used to ensure that every image from the dataset was tested once. Established augmentation techniques were applied to enhance performances, along with an Adam optimiser for training. The visualisation method was integrated gradient visualisation. RESULTS: The best performance of the model was obtained using 10 epochs, with an overall accuracy of 88.4%. For DR, the area under the receiver operating characteristics (ROC) curve (AUC) was 90.5% and the accuracy was 85.2%. For RVO, the AUC was 91.2% and the accuracy 88.4%. For SCR, the AUC was 96.7% and the accuracy 93.8%. For healthy controls, the ROC was 88.5% with an accuracy that reached 86.2%. CONCLUSION: Deep learning algorithms can classify several retinal vascular diseases on UWF-CPF with good accuracy. This technology may be a useful tool for telemedicine and areas with a shortage of ophthalmic care.

Frequencies, Modalities, Doses and Duration of Computerized Prescriptions for Sedative, Analgesic, Anesthetic and Paralytic Drugs in Neonates Requiring Intensive Care: A Prospective Pharmacoepidemiologic Cohort Study in 30 French NICUs From 2014 to 2020.

Objectives: No consensus exists about the doses of analgesics, sedatives, anesthetics, and paralytics used in critically ill neonates. Large-scale, detailed pharmacoepidemiologic studies of prescription practices are a prerequisite to future research. This study aimed to describe the detailed prescriptions of these drug classes in neonates hospitalized in neonatal intensive care units (NICU) from computerized prescription records and to compare prescriptions by gestational age. Materials and Methods: We included all neonates requiring intensive care in 30 French level III units from 2014 through 2020 with a computerized prescription for an analgesic, sedative, anesthetic, or paralytic agent. We described frequencies of prescription, methods of administration, concomitant drug prescriptions, and dosing regimen, and compared them across gestational ages. Results: Among 65,555 neonates, 29,340 (44.8%) were prescribed at least one analgesic (acetaminophen in 37.2% and opioids in 17.8%), sedative (9.8%), anesthetic (8.5%), and/or paralytic agent (1%). Among preterm infants born before 28 weeks, 3,771/4,283 (88.0%) were prescribed at least one of these agents: 69.7% opioids, 41.2% sedatives, 32.5% anesthetics, and 5.8% paralytics. The most frequently prescribed agents were sufentanil (in 10.3% of neonates) and morphine (in 8.0% of neonates) for opioids, midazolam (9.3%) for sedatives, ketamine (5.7%) and propofol (3.3%) for anesthetics. In most neonates, opioids and sedatives were prescribed as continuous infusion, whereas anesthetics were prescribed as single doses. Opioids, sedatives and paralytics were mostly prescribed in association with another agent. Doses varied significantly by gestational age but within a limited range. Gestational age was inversely related to the frequency, cumulative dose and duration of prescriptions. For example, morphine prescriptions showed median (IQR) cumulative doses of 2601 (848-6750) vs. 934 (434-2679) µg/kg and median (IQR) durations of 7 (3-15) vs. 3 (2-5) days in infants born <28 vs. ≥ 37 weeks of gestation, respectively (p-value<0.001). Conclusion: The prescriptions of analgesic, sedative, anesthetic, or paralytic agent were frequent and often combined in the NICU. Lower gestational age was associated with higher frequencies, longer durations and higher cumulative doses of these prescriptions. Dose-finding studies to determine individualized dosing regimens and studies on long-term neurodevelopmental outcome according to received cumulative doses are required.

Trends in metabolic bariatric surgery in adolescents in France: a nationwide analysis on an 11- year period.

BACKGROUND: As the prevalence of obesity in adolescents has reached an alarming level of 16%, the rate of metabolic bariatric surgery (MBS) in this population is also rising in several countries. OBJECTIVES: This study aimed to compare the trends in types of MBS, short-term safety, and revisional rates, in younger adolescents aged < 18 years, compared with older adolescents (aged 18-19 yr) and adults aged >20 years. SETTING: Clinical research center, general hospital in France. METHODS: Using a national administrative database (Programme de Médicalisation des Systèmes d'Information [PMSI]), data regarding all patients undergoing MBS between 2008 and 2018 in France were examined. Demographic parameters, body mass index (BMI), co-morbidities, types of surgery, early complications, and long-term revisional rates were analyzed, comparing younger adolescents (<18 yr), older adolescents (18-19 yr), and adults (≥20 yr). RESULTS: The number of bariatric procedures in adolescents initially increased from 59 in 2008 to 135 in 2014, and then progressively declined to 56 procedures in 2018. Adjustable gastric banding (AGB) decreased from 83.1% (n = 49) of procedures to 32.1% (n = 18) of procedures during the study period, while sleeve gastrectomy (SG) increased from 6.8% (n = 4) to 46.4% (n = 26). In the early postoperative period, younger adolescents undergoing MBS experienced fewer episodes of reoperation (1.0% versus 1.3% in older adolescents and 2.6% in adults, P < .001) and intensive care unit (ICU) stays (.2% versus .2% in older adolescents and .6% in adults, P < .001), and no deaths were observed in younger adolescents (.02% in older adolescents and .1% in adults, P = .18). At 10 years, the AGB removal rate was lower in younger adolescents (24.8%) compared with that in older adolescents (29.6%) and adults (50.3%, P < .001). Similarly, rates of revisional surgery after SG were different in the 3 groups: 2.9%, 4.6% and 12.2% in younger adolescents, older adolescents, and adults, respectively. CONCLUSION: Despite significantly lower early complication rates and long-term revisional rates in young adolescents (<18 yr), we observed a progressive decrease in the utilization of MBS in this population in France, compared with adults (≥20 yr) and older adolescents (18-19 yr).