BACKGROUND: Emergency conversion from minimally invasive liver resection (MILR) to open surgery, typically due to unexpected bleeding, is associated with poorer outcomes than elective conversion or completed MILR. However, prior studies rarely distinguished emergency from elective conversion and lacked integrated preoperative prediction tools. This study aimed to develop and validate an artificial intelligence (AI)-based preoperative model to predict emergency conversion. METHODS: We retrospectively analyzed patients undergoing MILR for hepatocellular carcinoma between 2010 and 2023, including only those with emergency conversion due to unexpected bleeding. Thirteen preoperative variables were used to develop a Random Forest model with five-fold cross-validation. Model performance was assessed by the area under the ROC curve (AUC) and calibration. Patients were stratified into four risk groups, and operative and postoperative outcomes were compared. RESULTS: Among 994 patients, 128 (12.9%) underwent emergency conversion. The model demonstrated good discrimination (AUC 0.79; 95% CI, 0.72-0.86). Observed conversion rates increased stepwise across risk groups (7.3%, 9.5%, 16.0%, and 37.7%). Higher risk groups were associated with longer operative times, greater blood loss, and prolonged hospital stays (all p < 0.001). CONCLUSION: This AI-based model enables preoperative risk stratification for emergency conversion during MILR and may support optimized patient selection and surgical planning.
ABSTRACT Background/Purpose This study evaluated associations of hospital surgical volume and perioperative transfusion with 30‐day mortality in patients undergoing biliary tract surgery. Methods This nationwide cohort study utilized data from the Korean National Health Insurance Service to identify 641 915 patients who underwent biliary tract surgery between 2012 and 2020. The primary outcome was 30‐day mortality. Perioperative transfusion was classified into four levels, and hospitals were grouped into quartiles by surgical volume. Mortality trends were evaluated using linear regression and Cochran–Armitage tests. Results The number of biliary tract surgeries increased over time, while the 30‐day mortality rate declined from 0.5% to 0.3%. For all types of surgery, crude mortality increased across the transfusion‐volume categories (p‐trend < 0.0001). Compared with the lowest‐volume hospitals, the highest‐volume hospitals had higher transfusion rates (4.8% vs. 3.9%), but lower crude mortality rates among transfused patients (3.3% vs. 4.4%). Among transfused patients, volume‐mortality patterns were more prominent in tertiary referral hospitals. Conclusions Hospital surgical volume and perioperative transfusion were associated with 30‐day mortality following biliary tract surgery. The higher mortality observed among transfused patients may partly reflect differences in operative complexity and clinical severity. These findings support further evaluation of selective centralization.
ABSTRACT Objectives A sufficient amount of cancer tissue for next‐generation sequencing (NGS) is difficult to obtain from patients with unresectable cholangiocarcinoma. We evaluated whether bile cell block (CB) samples provide a DNA source of sufficient quantity and quality for NGS compared with surgical specimens. Methods NGS results of paired bile CB and surgical specimens from 14 patients were compared. Results Median number of malignant cells on one glass slide of CB specimen was 50. Although median DNA concentration and amount of extracted DNA were significantly lower in the CB than those in the surgical specimens, NGS success rate did not differ significantly (CB, 64%; surgical specimen, 86%). In a pair‐wise comparison of eight patients with successful NGS of both bile CB and surgical specimens, median gene matching rate was 99%, and median coverage rate of surgical genes based on CB genes was 100%. The presence of ≥ 30 malignant cells on a single slide was identified as a practical indicator of successful NGS of CB using receiver operating characteristic curve, and this threshold increased the NGS success rate to 80%. Conclusions As bile CB is easy to prepare in general hospitals, bile cell‐based NGS may help address the difficulty of obtaining surgical NGS samples for cholangiocarcinoma.
ABSTRACT Background/Purpose Optimal perioperative antithrombotic therapy (ATT) management during emergency cholecystectomy for acute cholecystitis remains controversial. We examined whether continuing single‐agent ATT is associated with hemorrhagic risk compared with no ATT. Methods This retrospective cohort study included 1018 patients (826 without ATT, 192 continued on single‐agent ATT) at seven institutions (2020–2023). Propensity score overlap weighting balanced 12 baseline covariates. The primary outcome was intraoperative blood loss > 100 mL; risk ratios and 95% confidence intervals were estimated via modified Poisson regression, with upper confidence bound compared against a pre‐specified risk ratio (1.5). Results The primary outcome occurred in 27.3% and 25.4% of the non‐ATT and continued‐ATT groups, respectively (risk ratio: 0.93, 95% confidence interval: 0.69–1.26). The upper bound of the confidence interval was within the pre‐specified margin. No significant differences were observed in perioperative bleeding complications or hospital stay. Secondary outcomes, sensitivity analyses, and drug‐specific subgroup analyses (including direct oral anticoagulants) yielded consistent results. Conclusions Continuing single‐agent ATT was comparable to no ATT regarding bleeding during cholecystectomy for acute cholecystitis. These findings suggest that emergency cholecystectomy can be performed without meaningful increases in hemorrhagic risk in patients for whom adequate preoperative ATT washout is not feasible.
