RESUMO
BACKGROUND: The pilot study addresses the challenge of predicting postoperative outcomes, particularly body mass index (BMI) trajectories, following bariatric surgery. The complexity of this task makes preoperative personalized obesity treatment challenging. OBJECTIVES: To develop and validate sophisticated machine learning (ML) algorithms capable of accurately forecasting BMI reductions up to 5 years following bariatric surgery aiming to enhance planning and postoperative care. The secondary goal involves the creation of an accessible web-based calculator for healthcare professionals. This is the first article that compares these methods in BMI prediction. SETTING: The study was carried out from January 2012 to December 2021 at GZOAdipositas Surgery Center, Switzerland. Preoperatively, data for 1004 patients were available. Six months postoperatively, data for 1098 patients were available. For the time points 12 months, 18 months, 2 years, 3 years, 4 years, and 5 years the following number of follow-ups were available: 971, 898, 829, 693, 589, and 453. METHODS: We conducted a comprehensive retrospective review of adult patients who underwent bariatric surgery (Roux-en-Y gastric bypass or sleeve gastrectomy), focusing on individuals with preoperative and postoperative data. Patients with certain preoperative conditions and those lacking complete data sets were excluded. Additional exclusion criteria were patients with incomplete data or follow-up, pregnancy during the follow-up period, or preoperative BMI ≤30 kg/m2. RESULTS: This study analyzed 1104 patients, with 883 used for model training and 221 for final evaluation, the study achieved reliable predictive capabilities, as measured by root mean square error (RMSE). The RMSE values for three tasks were 2.17 (predicting next BMI value), 1.71 (predicting BMI at any future time point), and 3.49 (predicting the 5-year postoperative BMI curve). These results were showcased through a web application, enhancing clinical accessibility and decision-making. CONCLUSION: This study highlights the potential of ML to significantly improve bariatric surgical outcomes and overall healthcare efficiency through precise BMI predictions and personalized intervention strategies.
RESUMO
Hospitals are facing difficulties in predicting, evaluating, and managing cost-affecting parameters in patient treatments. Inaccurate cost prediction leads to a deficit in operational revenue. This study aims to determine the ability of Machine Learning (ML) algorithms to predict the cost of care in bariatric and metabolic surgery and develop a predictive tool for improved cost analysis. 602 patients who underwent bariatric and metabolic surgery at Wetzikon hospital from 2013 to 2019 were included in the study. Multiple variables including patient factors, surgical factors, and post-operative complications were tested using a number of predictive modeling strategies. The study was registered under Req 2022-00659 and approved by an institutional review board. The cost was defined as the sum of all costs incurred during the hospital stay, expressed in CHF (Swiss Francs). The data was preprocessed and split into a training set (80%) and a test set (20%) to build and validate models. The final model was selected based on the mean absolute percentage error (MAPE). The Random Forest model was found to be the most accurate in predicting the overall cost of bariatric surgery with a mean absolute percentage error of 12.7. The study provides evidence that the Random Forest model could be used by hospitals to help with financial calculations and cost-efficient operation. However, further research is needed to improve its accuracy. This study serves as a proof of principle for an efficient ML-based prediction tool to be tested on multi-center data in future phases of the study.
Assuntos
Cirurgia Bariátrica , Custos Hospitalares , Humanos , Aprendizado de Máquina , Tempo de Internação , Estudos RetrospectivosRESUMO
BACKGROUND: This is the first randomized trial to evaluate the efficacy of intraoperative cholangiography (IOC) and magnetic resonance cholangiopancreatography (MRCP) in patients with suspected CBDS. METHODS: This unblinded, multicenter RCT was conducted at five swiss hospitals. Eligibility criteria were suspected CBDS. Patients were randomized to IOC and laparoscopic cholecystectomy (LC), followed by endoscopic retrograde cholangiopancreatography (ERCP) if needed, or MRCP followed by ERCP if needed, and LC. Primary outcome was length of stay (LOS), secondary outcomes were cost, stone detection, and complication rates. RESULTS: 122 Patients were randomised to the IOC Group (63) or the MRCP group (59). Median LOS for the IOC and the MRCP groups were 4 days IQR [3, 6] and [4, 6], with an estimated increase of LOS of 1.2 days in the MRCP group (p = 0.0799) in the linear model. Median cost in the IOC and MRCP groups were 10 473 Swiss Francs (CHF) and 10 801 CHF, respectively (p = 0.694). CBDS were found in 24 and 12 patients in the IOC and the MRCP groups, respectively (p = 0.0387). The complication rate did not differ between both groups. CONCLUSION: There is equipoise between both pathways. IOC has a significantly higher diagnostic yield than MRCP. TRIAL REGISTRATION: Clinicaltrials.gov identifier NCT02351492: Radiological Investigation of Bile Duct Obstruction (RIBO).
Assuntos
Colecistectomia Laparoscópica , Cálculos Biliares , Humanos , Estudos Retrospectivos , Colangiografia , Cálculos Biliares/diagnóstico por imagem , Cálculos Biliares/cirurgia , Cálculos Biliares/complicações , Colecistectomia Laparoscópica/efeitos adversos , Colangiopancreatografia Retrógrada Endoscópica/efeitos adversos , Espectroscopia de Ressonância Magnética , Ducto ColédocoRESUMO
BACKGROUND: Paraesophageal hernias (PEH) tend to occur in elderly patients and the assumed higher morbidity of PEH repair may dissuade clinicians from seeking a surgical solution. On the other hand, the mortality rate for emergency repairs shows a sevenfold increase compared to elective repairs. This analysis evaluates the complication rates after elective PEH repair in patients 80 years and older in comparison with younger patients. METHODS: In total, 3209 patients with PEH were recorded in the Herniamed Registry between September 1, 2009 and January 5, 2018. Using propensity score matching, 360 matched pairs were formed for comparative analysis of general, intraoperative, and postoperative complication rates in both groups. RESULTS: Our analysis revealed a disadvantage in general complications (6.7% vs. 14.2%; p = 0.002) for patients ≥ 80 years old. No significant differences were found between the two groups for intraoperative (4.7% vs. 5.8%, p = 0.627) and postoperative complications (2.2% vs. 2.8%, p = 0.815) or for complication-related reoperations (1.7% vs. 2.2%, p = 0.791). CONCLUSIONS: Despite a higher risk of general complications, PEH repair in octogenarians is not in itself associated with increased rates of intraoperative and postoperative complications or associated reoperations. Therefore, PEH repair can be safely offered to elderly patients with symptomatic PEH, if general medical risk factors are controlled.