Published papers

Purpose: This study aimed to compare perioperative outcomes of robotic and laparoscopic transabdominal peritoneal repair (TAPP) for unilateral inguinal hernia. Methods: This single institutional retrospective cohort study used de-identified data of patients who underwent robotic TAPP (R-TAPP) or laparoscopic TAPP (L-TAPP) for unilateral inguinal hernia between January 1, 2016 and October 31, 2021. Two cohorts were propensity matched, and data were analyzed. The learning curve was evaluated in the R-TAPP group. Results: Among 938 patients analyzed, 704 were included. After propensity-score matching, 80 patients were included in each group. The difference in operative time between R-TAPP and L-TAPP groups was 10 min (99.5 and 89.5 min, p = 0.087); however, console/laparoscopic time was similar (67 and 66 min, p = 0.71). The dissection time for medial-type hernia in the R-TAPP group was marginally shorter than that in the L-TAPP group (17 and 27 min, p = 0.056); however, there was no difference for lateral-type hernia (38.5 and 40 min p = 0.37). Perioperative variables, including estimated blood loss, postoperative hospital stay, and postoperative pain, had no significant difference, and chronic pain, which needed medication or intervention, was not observed in each group. The number of cases needed to achieve plateau performance was 7–10 in the R-TAPP group. Conclusion: This study suggests that R-TAPP was safely introduced, and its perioperative outcomes were not inferior to those of L-TAPP. A shorter dissection time for medial-type hernia might be due to the robot’s advantages, and a fast-learning curve could help with the early standardization of the procedure.

Objective: To establish global benchmark outcomes indicators after laparoscopic liver resections (L-LR). Background: There is limited published data to date on the best achievable outcomes after L-LR. Methods: This is a post hoc analysis of a multicenter database of 11,983 patients undergoing L-LR in 45 international centers in 4 continents between 2015 and 2020. Three specific procedures: left lateral sectionectomy (LLS), left hepatectomy (LH), and right hepatectomy (RH) were selected to represent the 3 difficulty levels of L-LR. Fifteen outcome indicators were selected to establish benchmark cutoffs. Results: There were 3519 L-LR (LLS, LH, RH) of which 1258 L-LR (40.6%) cases performed in 34 benchmark expert centers qualified as low-risk benchmark cases. These included 659 LLS (52.4%), 306 LH (24.3%), and 293 RH (23.3%). The benchmark outcomes established for operation time, open conversion rate, blood loss ≥500 mL, blood transfusion rate, postoperative morbidity, major morbidity, and 90-day mortality after LLS, LH, and RH were 209.5, 302, and 426 minutes; 2.1%, 13.4%, and 13.0%; 3.2%, 20%, and 47.1%; 0%, 7.1%, and 10.5%; 11.1%, 20%, and 50%; 0%, 7.1%, and 20%; and 0%, 0%, and 0%, respectively. Conclusions: This study established the first global benchmark outcomes for L-LR in a large-scale international patient cohort. It provides an up-to-date reference regarding the "best achievable" results for L-LR for which centers adopting L-LR can use as a comparison to enable an objective assessment of performance gaps and learning curves.

Objective: This study aimed to establish global benchmark outcomes indicators for robotic liver resections (R-LR). Background: In recent years, minimally invasive liver resections and in particular R-LR has seen an increase in uptake in recent years. Although, benchmark outcomes have been recently established for laparoscopic (L) -LR, this has not been established for R-LR. Methods: This is a post hoc analysis of a multicenter database of 5,213 patients undergoing R-LR in 51 international centers between 2016 and 2022. Benchmark cutoffs for 16 outcome indicators of low-risk R-LR were established. The 75th percentile of individual center medians for a given outcome indicator was set as the benchmark cutoff. Four procedures were selected for benchmarking: left lateral sectionectomy(LLS)/H23, left hepatectomy(LH)/H234±1, right hepatectomy(RH)/H5678±1 and right posterior sectionectomy/H67 (RPS). Results: There were 1,654 R-LR cases (528 LLS/H23, 432 LH/H234±1, 408 RH/H5678±1, 286 RPS/H67) performed in 24 expert centers, of which 518 (31.3%) R-LR cases qualified as low risk benchmark cases. Benchmark outcomes were established for R-LLS/H23, R-LH/ H234±1, R-RH/ H5678±1 and R-RPS/H67 for operation time (190, 323, 474, 413) min, open conversion rate (0.0, 0.0, 1.3, 0.0)%, estimated blood loss (100, 250, 600, 550) mls, blood transfusion rate (0.0,0.0, 20.0, 29.2)%, postoperative major morbidity (0.0,0.0, 20.9, 16.7)%, 90-day mortality (0.0, 0.0,0.0, 0.0)% and textbook outcome (12.5,24.3,0,0)%. Conclusions: The present study established the first global benchmark values for R-LR. It provided an up-to-date reference of best achievable outcomes for auditing and benchmarking.

Hepatocellular carcinoma (HCC) is the fifth most common primary tumor and the third leading cause of cancer-related deaths worldwide. Rodent models of HCC have contributed to the advancement of studies investigating liver carcinogenesis, tumor-host interactions, and drug screening. However, their small size renders them unsuitable for surgical or clinical imaging studies, necessitating the development of larger-size HCC models. Here, we developed a xenograft model of human HCC in X-linked interleukin-2 receptor gamma chain gene (Il2rg)-targeted severe combined immunodeficient (SCID) pigs. HepG2 cell suspension in serum-free medium containing 50% membrane matrix was directly injected into the liver parenchyma of eight X-linked Il2rg-targeted SCID pigs (6.6-15.6 kg) via ultrasonography- guided percutaneous puncture. Tumor engraftment was evaluated weekly using ultrasonography, and cone-beam computed tomography was performed during arterial portography (CTAP) and hepatic arteriography (CTHA) to evaluate the hemodynamics of engrafted tumors. The engrafted tumors were histologically analyzed following necropsy and assessed for pathological similarities to human HCCs. Macroscopic tumor formation was observed in seven of the eight pigs (simple nodular tumors in three and multinodular tumors in four). Engrafted tumors were identified as low-echoic upon ultrasonography and as perfusion-defect nodules on the CTAP images. Meanwhile, CTHA showed that the tumors were hyperattenuating. Further, histopathological findings of the engrafted tumors were consistent with those of human HCC. In conclusion, the porcine model of human HCC, successfully generated herein, might help develop more effective therapeutic strategies for HCC.

Background: Although robotic pancreaticoduodenectomy (RPD) offers several technical advantages, prolonged operative time remains a major clinical challenge. Mesenteric Kocherization (MK) has recently been introduced as a technical modification for mobilizing the duodenum and pancreatic head during RPD. The present study evaluated the early clinical impact of MK on operative efficiency and perioperative outcomes. Methods: We retrospectively reviewed 56 consecutive patients who underwent totally robotic pancreaticoduodenectomy between July 2021 and June 2025. Conventional Kocherization (CK) was performed in 36 patients and MK in 20 patients. The primary outcome was total operative time. Secondary outcomes included console time, Kocherization time, estimated blood loss, transfusion, harvested lymph nodes, R0 resection rate, major postoperative complications (Clavien–Dindo ≥ IIIa), time to first oral intake, and 90-day unplanned readmission rate. To reduce selection bias, propensity score matching was performed using a 1:2 allocation (MK:CK). Results: In the full cohort, operative time tended to be shorter in the MK group than in the CK group (560 [516–598] vs. 592 [556–667] min; p = 0.057). Kocherization time was significantly shorter with MK (28 [27–36] vs. 50 [40–58] min; p < 0.001), while estimated blood loss was comparable (p = 0.875). After propensity score matching (MK n = 11; CK n = 22), MK was associated with significantly shorter total operative time (533 [511–575] vs. 580 [557–611] min; p = 0.031) and Kocherization time (28 [24–30] vs. 49 [43–55] min; p < 0.001). Blood loss, R0 resection rate, and major postoperative complications were comparable between groups. Conclusions: Mesenteric Kocherization was associated with improved operative efficiency in RPD, primarily through shortening the resection phase of the procedure while maintaining perioperative safety.

Background: Early laparoscopic cholecystectomy (ELC) for acute cholecystitis (AC) poses multiple challenges. The Tokyo Guidelines 2018 (TG18) eliminated the time limit (< 72 h) and expanded the surgical indication to severe AC. This study aimed to evaluate the clinical outcomes of ELC for AC following the TG18 in a single high-volume center. Methods: From 2019 to 2021, we managed all AC patients with a TG18 flowchart and prospectively enrolled those who underwent ELC within 7 days of symptom onset. The primary outcome was overall morbidity, with a comparison between mild (Grade I) and moderate/severe (Grade II/III) AC. Results: During the study period, 201 patients underwent ELC was for Grade I (56.2%), II (40.3%), and III (3.5%) ACs. Mean age was 69 ± 15.2 years and time to surgery from symptom onset was 0 (12.9%), 1–3 (66.7%), and 4–7 days (20.4%). Mean operative time and blood loss were 118.9 ± 42.7 min and 57.8 ± 99.4 mL, respectively. The critical view of safety (CVS) was achieved in 76.1% of patients, and bailout procedures were performed in 21.4%. There were no open conversions or bile duct injuries. Major morbidities (Clavien–Dindo classification ≥ IIIa) were observed in 5.5% of cases and mortality in 0.5%. Comparing Grades II/III to Grade I, operative time was longer (112.3 vs. 127.3 min, p = 0.014), blood loss was higher (40.3 vs. 80.1 mL, p = 0.005), the CVS rate was lower (83.2 vs. 67.0%, p = 0.012), and the major morbidity rate was higher (1.8 vs. 10.2%, p = 0.012). In the subgroup analysis of Grade II/III, there were no significant differences in major morbidities (p = 0.288) between the two groups (0–3 vs. 4–7 days). Conclusion: ELC for AC following TG18 is feasible with low morbidity rates. However, ELC for Grade II/III ACs remains challenging, and surgeons must carefully assess intraoperative difficulties and surgical risks before proceeding.

Background and Objectives: Laparoscopic cholecystectomy (LC) is one of the most performed surgeries worldwide. Procedure difficulty and patient outcomes depend on several factors which are not considered in the current literature, including the learning curve, generating confusing and subjective results. This study aims to create a scoring system to calculate the learning curve of LC based on hepatobiliopancreatic (HPB) experts’ opinions during an educational course. Materials and Methods: A questionnaire was submitted to the panel of experts attending the HPB course at Research Institute against Digestive Cancer-IRCAD (Strasbourg, France) from 27–29 October 2022. Experts scored the proposed variables according to their degree of importance in the learning curve using a Likert scale from 1 (not useful) to 5 (very useful). Variables were included in the composite scoring system only if more than 75% of experts ranked its relevance in the learning curve assessment ≥4. A positive or negative value was assigned to each variable based on its effect on the learning curve. Results: Fifteen experts from six different countries attended the IRCAD HPB course and filled out the questionnaire. Ten variables were finally included in the learning curve scoring system (i.e., patient body weight/BMI, patient previous open surgery, emergency setting, increased inflammatory levels, presence of anatomical bile duct variation(s), and appropriate critical view of safety (CVS) identification), which were all assigned positive values. Minor or major intraoperative injuries to the biliary tract, development of postoperative complications related to biliary injuries, and mortality were assigned negative values. Conclusions: This is the first scoring system on the learning curve of LC based on variables selected through the experts’ opinions. Although the score needs to be validated through future studies, it could be a useful tool to assess its efficacy within educational programs and surgical courses.

Background: The use of fluorescence agents and imaging systems is a promising adjunct in the surgical management of colorectal cancer. This systematic review and meta-analysis aimed to assess the safety and efficacy of fluorescence-guided surgery in the management of colorectal cancer, with a comparison to conventional (non-fluorescence-guided) surgery. Methods: A literature search of MEDLINE, Embase, Emcare, and CINAHL databases was performed for studies that reported data on the outcomes of fluorescence-guided surgery, with or without a comparison group undergoing conventional surgery, for colorectal cancer between January 2000 and January 2024. A meta-analysis was performed using random-effect models, and between-study heterogeneity was assessed. Results: 35 studies of 3217 patients with colorectal cancer were included: 26 studies (964 patients) reported on fluorescence-guided surgery and 9 studies (2253 patients) reported on fluorescence versus conventional surgery. The weighted mean of the cancer detection rate of fluorescence-guided surgery was 71% (95% CI 0.55–0.85), with no significant difference in lymph node yield ratio (WMD −0.04; 95% CI −0.10–0.02; p = 0.201) between fluorescence and conventional surgery groups. There was a significantly lower blood loss (WMD −4.38; 95% CI −7.05–−1.70; p = 0.001) and complication rate (WMD −0.04; 95% CI −0.07–0.00; p = 0.027) in the fluorescence-guided surgery group, with a potentially lower anastomotic leak rate (WMD −0.05; 95% CI −0.10–0.01; p = 0.092). Conclusions: Fluorescence-guided surgery is a safe and effective approach in the management of colorectal cancer, potentially reducing blood loss and complications. Further randomised controlled trials are required comparing fluorescence-guided surgery with conventional surgery to determine its prognostic benefit and where it should precisely fit within the management pathway of colorectal cancer.

Background/Purpose: Laparoscopic liver mono-segmentectomy (LLMS) may improve patient outcomes, but it is difficult and its accuracy and safety are unknown. We evaluated the accuracy of LLMS using Glissonian approach with indocyanine green fluorescence (ICG) negative staining. Methods: Seventy-four patients eligible for LLMS except for segment 1 were enrolled. Preoperative three-dimensional CT-based surgical simulation was used to determine estimated liver resection volume (ELRV), which was compared with modified actual liver resection volume (ALRV) obtained from actual liver resection mass. The LLMS accuracy was also evaluated based on operator's experience (attending surgeon [AS] or trainee surgeon [TS]). Results: Estimated liver resection volumes significantly correlated with ALRVs (r =.82) in all cases. Moreover, TS-conducted LLMS also showed acceptable difference between ELRV and ALRV compared with AS-conducted LLMS. There were no intergroup differences in estimated blood loss, operation time, time of Pringle maneuver, postoperative complications, and length of postoperative hospitalization (P <.05). Moreover, R0 resection was comparable between the AS and TS groups. Conclusions: Laparoscopic liver mono-segmentectomy with Glissonian approach using ICG negative imaging ensured safe and accurate procedure owing to facilitated visualization of the resection line. Our approach was effective in avoiding postoperative liver dysfunction and securing radical resection. In addition, it might be helpful in TS education of LLMS.

Background: Artificial intelligence (AI) is becoming more useful as a decision-making and outcomes predictor tool. We have developed AI models to predict surgical complexity and the postoperative course in laparoscopic liver surgery for segments 7 and 8. Methods: We included patients with lesions located in segments 7 and 8 operated by minimally invasive liver surgery from an international multi-institutional database. We have employed AI models to predict surgical complexity and postoperative outcomes. Furthermore, we have applied SHapley Additive exPlanations (SHAP) to make the AI models interpretable. Finally, we analyzed the surgeries not converted to open versus those converted to open. Results: Overall, 585 patients and 22 variables were included. Multi-layer Perceptron (MLP) showed the highest performance for predicting surgery complexity and Random Forest (RF) for predicting postoperative outcomes. SHAP detected that MLP and RF gave the highest relevance to the variables “resection type” and “largest tumor size” for predicting surgery complexity and postoperative outcomes. In addition, we explored between surgeries converted to open and non-converted, finding statistically significant differences in the variables “tumor location,” “blood loss,” “complications,” and “operation time.” Conclusion: We have observed how the application of SHAP allows us to understand the predictions of AI models in surgical complexity and the postoperative outcomes of laparoscopic liver surgery in segments 7 and 8.

Introduction: Despite the increasing widespread adoption and experience in minimally invasive liver resections (MILR), open conversion occurs not uncommonly even with minor resections and as been reported to be associated with inferior outcomes. We aimed to identify risk factors for and outcomes of open conversion in patients undergoing minor hepatectomies. We also studied the impact of approach (laparoscopic or robotic) on outcomes. Methods: This is a post-hoc analysis of 20,019 patients who underwent RLR and LLR across 50 international centers between 2004–2020. Risk factors for and perioperative outcomes of open conversion were analysed. Multivariate and propensity score-matched analysis were performed to control for confounding factors. Results: Finally, 10,541 patients undergoing either laparoscopic (LLR; 89.1%) or robotic (RLR; 10.9%) minor liver resections (wedge resections, segmentectomies) were included. Multivariate analysis identified LLR, earlier period of MILR, malignant pathology, cirrhosis, portal hypertension, previous abdominal surgery, larger tumor size, and posterosuperior location as significant independent predictors of open conversion. The most common reason for conversion was technical issues (44.7%), followed by bleeding (27.2%), and oncological reasons (22.3%). After propensity score matching (PSM) of baseline characteristics, patients requiring open conversion had poorer outcomes compared with successful MILR cases as evidenced by longer operative times, more blood loss, higher requirement for perioperative transfusion, longer duration of hospitalization and higher morbidity, reoperation, and 90-day mortality rates. Conclusions: Multiple risk factors were associated with conversion of MILR even for minor hepatectomies, and open conversion was associated with significantly poorer perioperative outcomes.

Background: Despite the rapid advances that minimally invasive liver resection has gained in recent decades, open conversion is still inevitable in some circumstances. In this study, we aimed to determine the risk factors for open conversion after minimally invasive left lateral sectionectomy, and its impact on perioperative outcomes. Methods: This is a post hoc analysis of 2,445 of 2,678 patients who underwent minimally invasive left lateral sectionectomy at 45 international centers between 2004 and 2020. Factors related to open conversion were analyzed via univariate and multivariate analyses. One-to-one propensity score matching was used to analyze outcomes after open conversion versus non-converted cases. Results: The open conversion rate was 69/2,445 (2.8%). On multivariate analyses, male gender (3.6% vs 1.8%, P =.011), presence of clinically significant portal hypertension (6.1% vs 2.6%, P =.009), and larger tumor size (50 mm vs 32 mm, P <.001) were identified as independent factors associated with open conversion. The most common reason for conversion was bleeding in 27/69 (39.1%) of cases. After propensity score matching (65 open conversion vs 65 completed via minimally invasive liver resection), the open conversion group was associated with increased operation time, blood transfusion rate, blood loss, and postoperative stay compared with cases completed via the minimally invasive approach. Conclusion: Male sex, portal hypertension, and larger tumor size were predictive factors of open conversion after minimally invasive left lateral sectionectomy. Open conversion was associated with inferior perioperative outcomes compared with non-converted cases.

Liver resection is the first curative option for most hepatic primary and secondary malignancies. However, post-hepatectomy liver failure (PHLF) still represents a non-negligible postoperative complication, embodying the most frequent cause of hepatic-related mortality. In the absence of a specific treatment, the most effective way to deal with PHLF is its prevention through a careful preoperative assessment of future liver remnant (FLR) volume and function. Apart from the clinical score and classical criteria to define the safe limit of resectability, new imaging modalities have shown their ability to assist surgeons in planning the best operative strategy with a precise estimation of the FLR amount. New technologies leading to liver and tumor 3D reconstruction may guide the surgeon along the best resection planes combining the least liver parenchymal sacrifice with oncological appropriateness. Integration with imaging modalities, such as hepatobiliary scintigraphy, capable of estimating total and regional liver function, may bring about a decrease in postoperative complications. Magnetic resonance imaging with hepatobiliary contrast seems to be predominant since it simultaneously integrates hepatic function and volume information along with a precise characterization of the target malignancy.