BJS Open最新文献

Background: A real-time deep learning system was developed to identify the extrahepatic bile ducts during indocyanine green fluorescence-guided laparoscopic cholecystectomy.

Methods: Two expert surgeons annotated surgical videos from 113 patients and six class structures. YOLOv7, a real-time object detection model that enhances speed and accuracy in identifying and localizing objects within images, was trained for structures identification. To evaluate the model's performance, single-frame and short video clip validations were used. The primary outcomes were average precision and mean average precision in single-frame validation. Secondary outcomes were accuracy and other metrics in short video clip validations. An intraoperative prototype was developed for the verification experiments.

Results: A total of 3993 images were extracted to train the YOLOv7 model. In single-frame validation, all classes' mean average precision was 0.846, and average precision for the common bile duct and cystic duct was 0.864 and 0.698 respectively. The model was trained to detect six different classes of objects and exhibited the best overall performance, with an accuracy of 94.39% for the common bile duct and 84.97% for the cystic duct in video clip validation.

Conclusion: This model could potentially assist surgeons in identifying the critical landmarks during laparoscopic cholecystectomy, thereby minimizing the risk of bile duct injuries.

Background: Evidence-based medicine relies on randomized clinical trials, which should represent the patients encountered in clinical practice. Characteristics of patients recruited to randomized clinical trials involving vascular index operations (carotid endarterectomy, abdominal aortic aneurysm repair, infrainguinal bypass and major lower limb amputations) were compared with those recorded in the National Vascular Registry across England and Wales.

Methods: MEDLINE, Embase, Web of Science, CENTRAL, clinicaltrials.gov and World Health Organization International Trials Registry Platform (CRD42021247905) were searched for randomized clinical trials involving the index operations. Demographic (age, sex, ethnicity) and clinical (co-morbidities, medications, body mass index, smoking, alcohol, cognition) data were extracted, by operation. Characteristics of operated on patients were extracted from publicly available National Vascular Registry reports (2014-2020). All findings are reported according to PRISMA guidelines. Rayyan.AI, Excel and GraphPad Prism were used for screening and analysis.

Results: A total of 307 randomized clinical trials (66 449 patients) were included and compared with National Vascular Registry data for 119 019 patients. Randomized clinical trial patients were younger across all operations; for carotid endarterectomy, bypass and major lower limb amputation randomized clinical trials, there were differences in female patient representation. Further comparisons were limited by the insufficient baseline data reporting across randomized clinical trials, though reporting improved over decades. National Vascular Registry reports lacked information on patient factors such as patient ethnicity or body mass index.

Conclusions: There are significant differences in demographic and clinical factors between patients recruited to vascular surgery randomized clinical trials and the real-world National Vascular Registry vascular surgery patient population. Minimum reporting standards for baseline data should be defined to allow future randomized clinical trials to represent real-world patient populations and ensure the external validity of their results.

Background: Acute Stanford type A aortic dissection is a severe emergency condition that, if left untreated, is associated with a high mortality rate. The extent of surgical repair may impact the outcomes of these patients.

Method: Patients operated for acute type A aortic dissection from a multicentre European registry were included. Patients were categorized based on the following types of surgical intervention: isolated ascending aortic replacement, ascending aortic replacement with concomitant aortic valve replacement, aortic root replacement, partial or total arch replacement, and partial or total arch replacement with concomitant aortic root replacement. The primary outcome was mortality rate, both in-hospital and at 10 years. Secondary outcomes were acute kidney injury requiring dialysis, neurological complications, a composite endpoint including in-hospital death, neurological complications and/or dialysis, and proximal endovascular or surgical aortic re-operations at 10 years.

Results: 3702 patients were included. The adjusted risk of in-hospital mortality was higher in all subsets of patients compared to those who underwent isolated ascending aortic replacement. The adjusted rates of in-hospital mortality ranged from 16.4% (95% c.i. 15.3 to 17.4) among patients who underwent isolated ascending aortic replacement to 27.7% (95% c.i. 23.3 to 31.2) among those who underwent aortic arch and concomitant aortic root replacement. The adjusted risks of neurological complications, renal replacement therapy and of the composite endpoint were significantly higher in patients who underwent partial/total aortic arch replacement. The adjusted risk estimates of 10-year mortality rate were markedly higher in patients who underwent partial/total aortic arch replacement with or without concomitant aortic root replacement. Extensive aortic repair did not significantly reduce the risk of distal or proximal aortic reoperations.

Conclusion: These findings suggest that, when feasible, limiting the extent of aortic replacement for acute type A aortic dissection may be beneficial in reducing mortality rate and major complications both in the short and long term.

Trial registration: ClinicalTrials.gov identifier: NCT04831073.

Background: The incidence of early-onset colorectal cancer is increasing, but in Europe this growth shows a heterogeneous pattern in different countries and regions.

Methods: Patients from six countries who participated in the Global Early-Onset COlorectal Cancer DatabasE (GEOCODE)-Europe group were included. The inclusion criteria were patients with colorectal adenocarcinoma diagnosed between 18 and 49 years of age, between January 2010 and December 2017, with at least 3 years of follow-up. Patients with inherited colorectal cancer syndromes were excluded.

Results: A total of 851 patients were included with almost equal sex distribution, most were diagnosed at age 39 years or older and 42% of patients were overweight or obese. Diagnoses were predominantly at later stages (62.5% stage III-IV) and tumours were predominantly located in the distal colon (76.9% left colon and rectum). Comparative analysis between countries demonstrated that the UK had a younger age at diagnosis and the Italian cohort had a higher prevalence of being overweight or obese. Patients from Luxembourg had more advanced stage diagnoses and those from The Netherlands had more polyps. Patients from the UK had a greater family history of colorectal cancer. Comparison of Mediterranean versus non-Mediterranean countries showed significant differences in the age at diagnosis and body mass index. The prevalence of early-onset colorectal cancer over the age of 40 years in Mediterranean versus non-Mediterranean countries was 71.4% versus 62.1% (P = 0.002), and early-onset colorectal cancer was diagnosed at a more advanced stage in Mediterranean countries versus non-Mediterranean countries (65.3% versus 54.7%; P = 0.033). Family history of colorectal cancer in a first-degree relative was more common in non-Mediterranean versus Mediterranean countries (19.1% versus 11.4%; P < 0.001).

Conclusion: This study highlights significant geographical disparities in the clinical, pathological and familial features of early-onset colorectal cancer across European countries.

Background: Large bowel obstruction is a possible complication in patients undergoing neoadjuvant treatment for rectal cancer; however, it may be prevented by placing a pretreatment defunctioning stoma. The aim of this retrospective study was to investigate complication rates in patients with rectal cancer undergoing long-course neoadjuvant therapy, comparing those with and without a prophylactic stoma.

Methods: All patients with rectal cancer undergoing neoadjuvant therapy between 2007 and 2022 in Region Västerbotten, Sweden, were identified using the Swedish Colorectal Cancer Registry. Patients not planned for curative long-course neoadjuvant therapy and those requiring a stoma due to urgent bowel-related issues before treatment were excluded. The primary outcome was the incidence of complications between diagnosis and resection surgery or end of follow-up. The secondary outcomes were 30-day complications following resection, time to treatment (neoadjuvant therapy and surgery), and overall survival. Multivariable regression analysis was used, with adjustment for age, sex, American Society of Anesthesiologists fitness grade, and clinical tumour stage.

Results: Of 482 identified patients, 105 were analysed after exclusion. Among these, 22.9% (24 of 105) received a pretreatment stoma, whereas 77.1% (81 of 105) received upfront neoadjuvant therapy. The complication incidence before resection in the group with a defunctioning stoma and in the group without a defunctioning stoma was 75.0% (18 of 24) and 29.6% (24 of 81) respectively. A considerable number of complications were directly caused by the stoma surgery. Patients in the stoma group had an adjusted OR of 6.71 (95% c.i. 2.17 to 20.76) for any complication. However, for 30-day complications following resection, an adjusted non-significant OR of 2.05 (95% c.i. 0.62 to 6.81) was documented for the stoma group, in comparison with the control group. Neoadjuvant treatment was also delayed for the stoma group (adjusted mean time difference: 21 (95% c.i. 14 to 27) days), whereas the difference was not significant for the time to resection surgery. The median survival after diagnosis was 4.7 years in the stoma group and 12.2 years in the control group (P = 0.015); however, adjustment in the multivariable analysis rendered the estimate non-significant (HR 1.71 (95% c.i. 0.93 to 3.14)).

Conclusion: Patients with rectal cancer who receive a stoma before long-course neoadjuvant therapy, in the absence of urgent symptoms, experience more complications than those without a stoma and a delay with regard to the start of neoadjuvant treatment.