Annals of hepato-biliary-pancreatic surgery最新文献

Backgrounds/aims: This study evaluates whether positive resection margins after colorectal liver metastasis (CRLM) surgery are linked to tumor recurrence and impact long-term survival.

Methods: The oncological outcomes of patients undergoing curative resection for CRLM at a single institution were analyzed concerning clinicopathological factors using both univariate and multivariate statistical methods.

Results: Among 138 patients who underwent liver resection for CRLM, hepatic tumor recurrence was noted in 70 patients (50.7%), with no significant difference between those with R0 and R1 resections (p = 0.33). Positive resection margins were associated with tumor proximity to major liver vascular structures, while negative margins corresponded more frequently with T4 stage colorectal cancer. Multivariate analysis indicated that R1 margins in CRLM resections do not affect overall or disease-free survival. Nonetheless, the proximity of tumors to major liver vascular structures and R1 margins from initial colorectal cancer resections were significant independent predictors of poorer survival outcomes.

Conclusions: With the advent of modern perioperative systemic therapies, both hepatic recurrence and survival outcomes following hepatectomy for colorectal liver metastases seem unaffected by the presence of R1 resection margins.

Non-cirrhotic portal fibrosis (NCPF), a leading cause of non-cirrhotic portal hypertension (NCPH), commonly presents with splenomegaly, esophageal varices, and preserved liver function. While minimally invasive splenectomy (MIS) offers advantages over open splenectomy, concerns persist due to the risks associated with portal hypertension. This study evaluates the feasibility, safety, and long-term outcomes of MIS in non-bleeder NCPF patients, highlighting perioperative challenges and techniques. Thirteen consecutive non-bleeder NCPF patients undergoing MIS between November 2017 and December 2023 were analyzed. Procedures included eight laparoscopic and four robotic splenectomies, with one conversion to open surgery. Additionally, two patients underwent laparoscopic and one robotic gastric devascularization. Perioperative parameters such as operative time, blood loss, hospital stay, and complications were recorded. The median operative time was 240 minutes, and median blood loss was 150 mL. One patient required transfusion, and one developed splanchnic venous thrombosis, managed conservatively. Median hospital stay was three days, with no 90-day mortality. Follow-up assessments included clinical evaluation, blood tests, Doppler ultrasound of the splenoportal axis, and upper gastrointestinal endoscopy. At a median follow-up of 16 months (interquartile range 12-43), significant improvements were observed in hemoglobin, leukocyte, and platelet counts (p < 0.01). Esophageal variceal grades decreased from 2 to 1, while portal vein peak systolic velocity improved from 18 to 27.7 cm/sec (p < 0.01), indicating reduced portal hypertension. No postoperative infections or variceal bleeding recurrences were noted. MIS is a safe and effective treatment option for non-bleeder NCPF with favorable long-term outcomes when performed by skilled surgeons.

Backgrounds/aims: Postoperative pain recurrence is a challenging issue in chronic pancreatitis (CP). This study explores the incidence and factors contributing to recurrent pain after surgery.

Methods: An ambispective observational study evaluated patients with painful CP undergoing surgery from 2011 to 2022. The intensity frequency consequence (IFC) pain score and the painDETECT questionnaire were utilized to assess pain before and after surgery. Patients were categorized into 2 groups based on their IFC pain scores: a pain group and a pain-free group. Clinical, radiological, surgical, and post-surgical parameters were compared between these groups using the student t-test and logistic regression for continuous and categorical variables, respectively. A p-value of < 0.05 was deemed significant. Multivariate analysis was conducted.

Results: A total of 125 patients were enrolled (pain group, 71 [56.8%]; pain-free group, 54 [43.2%]). In the pain group, 65 experienced mild and 6 experienced moderate pain. The average post-surgery pain score was significantly lower than the pre-surgery score (13.7 vs 2.5, p < 0.001). Multivariate analyses revealed that intravenous (IV) analgesics and preoperative endoscopic interventions were independent predictors of recurrent pain.

Conclusions: The incidence of recurrent pain in CP patients post-surgery was 56.8%, with a significant reduction in pain intensity postoperatively. Patients who required preoperative IV analgesics and underwent endoscopic interventions demonstrated a higher risk of recurrent pain. Neuropathic pain was not identified as a cause of pain recurrence in this study.

Backgrounds/aims: Surgical data science (SDS) is an emerging discipline that aims to enhance the quality of interventional healthcare by capturing and analyzing intraoperative data. Our study focused on identifying human errors (HEs) and adverse events (AEs) during elective liver surgery using an SDS-based approach.

Methods: Intraoperative data from 15 patients undergoing elective open liver resection were collected using an operating room data system (audio, room, and operative field videos) over a 6-month period in a tertiary hepatobiliary surgical center. Two independent researchers analyzed the data to identify HEs and AEs according to two distinct classifications.

Results: A total of 154 HEs (median number per intervention: 7) and 42 AEs (33 minor, 9 major) were identified. All except one major AE were associated with HEs, while 15 minor AEs had no identifiable underlying HEs. The type of HEs significantly varied depending on the presence or absence of AEs. The majority of HEs (n = 128, 83.1%), which did not result in any AEs, primarily involved lapses in attention, whereas approximately half of the AEs were linked to failures in recognition.

Conclusions: This preliminary study indicates that failures in recognition were frequently associated with major AEs during elective liver resection, as per the SDS approach. Larger multicenter studies are necessary to confirm these findings and develop preventive strategies.

Backgrounds/aims: Post-hepatectomy liver failure (PHLF) is a significant complication with an incidence rate between 8% and 12%. Machine learning (ML) can analyze large datasets to uncover patterns not apparent through traditional methods, enhancing PHLF prediction and potentially mitigate complications.

Methods: Using the National Surgical Quality Improvement Program (NSQIP) database, patients who underwent hepatectomy were randomized into training and testing sets. ML algorithms, including LightGBM, Random Forest, XGBoost, and Deep Neural Networks, were evaluated against logistic regression. Performance metrics included receiver operating characteristic area under the curve (ROC AUC) and Brier score loss. Shapley Additive exPlanations was used to identify individual variable relevance.

Results: 28,192 patients from 2013 to 2021 who underwent hepatectomy were included; PHLF occurred in 1,305 patients (4.6%). Preoperative and intraoperative factors most contributed to PHLF. Preoperative factors were international normalized ratio > 1.0, sodium < 139 mEq/L, albumin < 3.9 g/dL, American Society of Anesthesiologists score > 2, total bilirubin > 0.65 mg/dL. Intraoperative risks include transfusion requirements, trisectionectomy, operative time > 266.5 minutes, open surgical approach. The LightGBM model performed best with an ROC AUC of 0.8349 and a Brier Score loss of 0.0834.

Conclusions: While topical, the role of ML models in surgical risk stratification is evolving. This paper shows the potential of ML algorithms in identifying important subclinical changes that could affect surgical outcomes. Thresholds explored should not be taken as clinical cutoffs but as a proof of concept of how ML models could provide clinicians more information. Such integration could lead to improved clinical outcomes and efficiency in patient care.

Backgrounds/aims: Diabetes is a recognized risk factor for pancreatic cancer; however, precise molecular mechanisms remain unclear. This study aimed to assess the influence of inflammation on the progression of pancreatic cancer in a diabetic murine model utilizing circulating tumor cells (CTC).

Methods: Fifty mice were randomly allocated into five groups. The P group were injected Panc02 cells only. In the streptozotocin (STZ), STZ/P, and P/STZ groups, mice were administered intraperitoneal STZ solution (50 mg/kg) alone, prior to Panc02 cell injection, and following Panc02 cell injection, respectively. Tumor development was assessed by gross inspection. Immunohistochemistry was performed to evaluate inflammatory cytokine expression, and CTCs were detected using quantum dot-conjugated aptamers.

Results: All mice exposed to STZ developed marked hyperglycemia. Tumor volume to body weight ratio was significantly higher in both P/STZ and STZ/P groups (p < 0.001). Liver metastasis rate was highest in the P/STZ group (p = 0.05). Malondialdehyde (p < 0.001), interleukin-1β (p < 0.05), tumor necrosis factor-α (p < 0.001), and interleukin-6 (p < 0.05) levels were significantly elevated in the STZ/P group. Expression of Signal Transducer and Activator of Transcription 3 and Snail1 was increased in both STZ/P and P/STZ groups. In addition, seven mice in the STZ/P group (70%) and nine mice in the P/STZ group (90%) exhibited larger CTC-like cells (p < 0.001).

Conclusions: In STZ-induced murine models, both hyperglycemia and elevated inflammatory markers were observed. Within this diabetes-associated inflammatory microenvironment, pancreatic cancer cells demonstrated increased proliferation and metastasis, as verified by aptasensor-based CTC detection.

Backgrounds/aims: Groove pancreatitis (GP) is a seldom encountered form of chronic pancreatitis characterized by inflammation of the pancreatoduodenal groove. Our study presents our experience with pancreatoduodenectomy (PD) for GP at a tertiary referral center.

Methods: We conducted a retrospective review of patients who underwent PD for a preoperative diagnosis of GP at a tertiary referral center from 2010 to 2024. The primary outcomes were long-term pain relief and risks of recurrent pancreatitis, exocrine, and endocrine insufficiency. Secondary outcomes included perioperative complications.

Results: During the study period, 19 patients underwent PD for GP. The median age was 45.5 years, and all patients were male. Eighty percent of the patients had a history of alcohol consumption and smoking. The median duration of symptoms was 24 months, with pain being the most prevalent symptom (94.73%). The overall complication rate (Clavien-Dindo grades 1-5) was 52.63% (10/19), and the major complication rate (Clavien-Dindo grades 3-5) was 21.05%. The median follow-up period was 67.25 months. Complete pain relief was achieved in 73.33% (11/15) of the patients, with the remaining 26.66% (4/15) experiencing partial resolution of pain. Among these, all had recurrent pancreatitis in the remnant pancreas, with ongoing alcohol consumption (n = 3) or smoking (n = 4). New-onset diabetes mellitus and steatorrhea were observed in 42.85% (6/14) and 21.42% (3/14) of patients, respectively. Furthermore, 71.42% (10/14) reported weight gain, with a median increase of 13.5 kg (range 5.00-22.75 kg).

Conclusions: PD for GP offers substantial long-term pain relief with acceptable levels of perioperative morbidity and mortality.

Backgrounds/aims: Laparoscopic cholecystectomy (LC) is the standard therapy for acute calculous cholecystitis (ACC). However, high-risk patients often require percutaneous gallbladder drainage (PGBD) as a bridge to surgery. The optimal interval between PGBD and LC remains uncertain.

Methods: We retrospectively reviewed 177 patients who underwent LC after PGBD for ACC at Nhan dan Gia Dinh Hospital (2018-2024). Patients were stratified by drainage-to-surgery interval: Q1, ≤ 9 days; Q2, 10-17 days; Q3, 18-32 days; Q4, ≥ 32 days. Primary outcomes were operative time, conversion, intraoperative bleeding, postoperative stay, and Clavien-Dindo complications. Multivariable logistic regression was performed after adjusting for age, Charlson Comorbidity Index (CCI), TG18 severity, sex, prior endoscopic retrograde cholangiopancreatography, and Concurrent PGBD and LC during the same admission.

Results: Median operative time, conversion rate, and postoperative stay were similar across intervals. Crude bleeding differed significantly (p = 0.019), being the highest in Q4 (68.2%) and the lowest in Q2 (36.4%). Q3 showed the shortest operative time (median: 90 min) and the lowest complication rate (11.4%). In adjusted analysis, only CCI independently predicted bleeding (adjusted odds ratio: 1.42; 95% confidence interval: 1.02-2.03), while timing lost its statistical significance.

Conclusions: Scheduling LC 18-32 days after PGBD offers the most balanced surgical profile, whereas delaying beyond 32 days increases bleeding without added benefit. Comorbidity burden rather than calendar interval per se appears to increase bleeding risk. Prospective studies are warranted to confirm the intermediate 2- to 4-week window.