Analysis of technical failure after 1,613 surgical microwave ablations: A propensity score-matched analysis.

BACKGROUND: Microwave ablation is becoming increasingly common for the treatment of liver tumors. Despite numerous studies aimed at identifying risk factors for local recurrence after microwave ablation, a consensus on modifiable risk factors for failure remains elusive, partly because of the limited statistical power of these studies. This study investigated the incidence of technical failure after microwave ablation, encompassing both incomplete ablation and local recurrence, and aimed to identify modifiable factors that reduce technical failure. METHODS: This retrospective review included patients who underwent surgical microwave ablation at a high-volume institution between October 2006 and March 2023. Univariate analysis, multivariate analysis, and propensity score matching were performed to identify risk factors for technical failure. RESULTS: A total of 1,613 surgical microwave ablations were performed on 3,035 tumors, with 226 instances (14% per procedure, 7.4% per tumor) of technical failure. Incomplete ablation occurred at a rate of 1.7% per tumor, whereas local recurrence was identified in 6.5% of ablations in per-tumor analysis. Body mass index >25 was significant for failure (odds ratio, 1.50; 95% confidence interval, 1.07-2.11; P < .05), suggesting that more difficult targeting may lead to increased technical failure rates. African American race (odds ratio, 1.62; 95% confidence interval, 1.16-2.27; P < .05), pre-microwave ablation transarterial chemoembolization (odds ratio, 1.54; 95% confidence interval, 1.08-2.21; P < .05), and previous ablation (odds ratio, 1.58; 95% confidence interval, 1.09-2.29; P < .05) were found to be statistically significant. CONCLUSION: On the basis of the largest microwave ablation database available to date, this study identified novel modifiable and nonmodifiable risk factors of microwave ablation failure. These results can lead to decreasing technical failure rates after microwave ablation.

Predicting response to neoadjuvant chemotherapy for colorectal liver metastasis using deep learning on prechemotherapy cross-sectional imaging.

BACKGROUND AND OBJECTIVES: Deep learning models (DLMs) are applied across domains of health sciences to generate meaningful predictions. DLMs make use of neural networks to generate predictions from discrete data inputs. This study employs DLM on prechemotherapy cross-sectional imaging to predict patients' response to neoadjuvant chemotherapy. METHODS: Adult patients with colorectal liver metastasis who underwent surgery after neoadjuvant chemotherapy were included. A DLM was trained on computed tomography images using attention-based multiple-instance learning. A logistic regression model incorporating clinical parameters of the Fong clinical risk score was used for comparison. Both model performances were benchmarked against the Response Evaluation Criteria in Solid Tumors criteria. A receiver operating curve was created and resulting area under the curve (AUC) was determined. RESULTS: Ninety-five patients were included, with 33,619 images available for study inclusion. Ninety-five percent of patients underwent 5-fluorouracil-based chemotherapy with oxaliplatin and/or irinotecan. Sixty percent of the patients were categorized as chemotherapy responders (30% reduction in tumor diameter). The DLM had an AUC of 0.77. The AUC for the clinical model was 0.41. CONCLUSIONS: Image-based DLM for prediction of response to neoadjuvant chemotherapy in patients with colorectal cancer liver metastases was superior to a clinical-based model. These results demonstrate potential to identify nonresponders to chemotherapy and guide select patients toward earlier curative resection.

Outcomes after Surgical Microwave Ablation for the Treatment of Colorectal Liver Metastasis.

BACKGROUND: Colorectal cancer (CRC) is the third most common cause of cancer mortality worldwide. Of these, approximately 25% will have liver metastasis. We performed 394 microwave ablations (MWAs) and analyzed outcomes for survival and ablation failure. STUDY DESIGN: We conducted a retrospective review of patients who underwent a surgical MWA at a single-center high-volume institution from October 2006 to September 2022 using a prospectively maintained database. The primary outcome was overall survival. RESULTS: A total of 394 operations were performed on 328 patients with 842 tumors undergoing MWA. Median tumor size was 1.5 cm (range 0.4 to 7.0 cm), with the median number of tumors ablated per operation being 1 (range 1 to 11). A laparoscopic approach was used 77.9% of the time. Concomitant procedures were performed 63% of the time, most commonly hepatectomy (22.3%), cholecystectomy (17.5%), and colectomy (6.6%). Clavien-Dindo grade III or IV complication occurred in 12 patients (3.6%), and all of these patients had undergone concomitant procedures. Mortality within 30 days occurred in 4 patients (1.2%). The rate of incomplete ablation was 1.5% per tumor. Local recurrence occurred at a rate of 6.3% per tumor. Black patients were found to have a higher incidence of incomplete ablation and local recurrence. One-year survival probability was 91% (95% CI 87.9 to 94.3), with a mean overall survival of 57.6 months (95% CI 49.9 to 65.4 months). CONCLUSIONS: Surgical MWA offers a low-morbidity approach to treatment of colorectal liver metastasis, with low rate of failure. This large series reviews the outcomes of MWA as definitive treatment for colorectal liver metastasis.

Surgical microwave ablation for the treatment of hepatocellular carcinoma in 791 operations.

INTRODUCTION: Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality and often arises in the setting of cirrhosis. The present series reviews outcomes following 791 operations. METHODS: Retrospective review surgical MWA for HCC from March 2007 through December 2022 at a high-volume institution was performed using a prospective database. Primary outcome was overall survival. RESULTS: A total of 791 operations in 623 patients and 1156 HCC tumors were treated with surgical MWA. Median tumor size was 2 cm (range 0.25-10 cm) with an average of 1 tumor ablated per operation (range 1-7 tumors). Nearly 90 % of patients had cirrhosis with a median MELD score of 8 (IQR = 6-11). Mortality within 30 days occurred in 13 patients (1.6 %). Per tumor, the rate of incomplete ablation was 2.25 % and local recurrence was 2.95 %. Previous ablation and tumor size were risk factors for recurrence. One-year overall survival was 82.0 % with a median overall survival of 36.5 months (95 % CI 15.7-93.7) and median disease-free survival of 15.9 months (range 5.7-37.3 months). CONCLUSION: Surgical MWA offers a low-morbidity approach for treatment of HCC, affording low rates of incomplete ablation and local recurrence.

Robotic Revision of Hepaticojejunostomy for Benign Biliary Stricture.

Surgical revision of biliary enteric anastomoses (BEA) can be a challenging undertaking and a robotic platform may provide advantages that address many of the technical obstacles. We present our technical approach and outcomes for patients undergoing robotic revision of BEA for benign strictures. A retrospective review was performed for robot-assisted benign BEA revision at our institution. Operative details, perioperative metrics, and outcomes are reported. Four patients underwent anastomotic revision following previously failed non-operative management. There were no intraoperative complications, mean length of stay was 4-days, and all patients experienced resolution of presenting clinical signs and symptoms. No patients required reoperation and there was no mortality. Postoperative outcomes were consistent with findings reported for other interventional modalities. Based on our experience we conclude robotic intervention in this context is safe and improves the technical feasibility of this complex procedure.

Robotic Resection of Gallbladder Cancer: A Single-Center Retrospective Comparative Study to Open Resection.

BACKGROUND: Minimally invasive surgery is gaining support for resection of gallbladder cancer (GBC). This study aims to compare operative and early outcomes of robotic resection (RR) to open resection (OR) from a single institution performing a high volume of robotic HPB surgery. METHODS: Twenty patients with GBC underwent RR from January 2013 to August 2019. Outcomes were compared to a historical control of 23 patients with OR. Radical cholecystectomy for suspected GBC and completion operations for incidental GBC after routine cholecystectomy were both included. RESULTS: Robotic resection had lower blood loss compared to OR (150 vs 350 mL, P = .002) and shorter postoperative length of stay (2.5 vs 6 days, P < .001), while median operative time was similar (193 vs 208 min, P = .604). There were no statistical differences in 30-day major complications or readmissions. No 30-day mortalities occurred. There was no statistical difference in survival trend (P = .438) or median lymph node harvest (5 vs 3, P = .189) for RR compared to OR. CONCLUSION: Robotic resection of GBC is safe and efficient, with lower length of hospital stay and blood loss compared to OR. Technical benefits of robotic-assisted surgery may prove advantageous though larger studies are still needed.

Enhanced Recovery After Surgery® in Octogenarians Undergoing Hepatopancreatobiliary Surgery.

Advances in perioperative care have increased the frequency of surgical intervention performed on the very elderly (≥80 years). This study aims to investigate the impact of Enhanced Recovery After Surgery (ERAS) on outcomes for octogenarians after major hepatopancreatobiliary (HPB) surgery. Patients ≥80 years old in a single HPB ERAS program (September 2015-July 2018) were prospectively tracked in the ERAS Interactive Audit System (EIAS). Postoperative length of stay (LOS) as well as 30-day major complications, readmissions, and mortality were compared to a pre-ERAS octogenarian control. Since ERAS implementation, octogenarians comprised 7.3% (27 of 370) of patients who underwent pancreaticoduodenectomy (n=17), distal pancreatectomy (n=7), or hepatectomy (n=3). Thirty-day readmissions decreased after ERAS implementation (50% to 15%, P=.037). Thirty-day major complications, mortality, and LOS were similar with 64% median protocol compliance. ERAS for octogenarians in HPB surgery is safe and may contribute to more sustainable recovery resulting in reduced readmissions.

Textbook outcomes and benchmarks of minimally invasive left lateral sectionectomy across North America.

BACKGROUND: Minimally invasive approach represents the gold standard for the resection of the left lateral section of the liver. Recently, the American Minimally Invasive Liver Resection (AMILES) registry has become available to track outcomes of laparoscopic and robotic liver resection in the Americas. The aim of the present study is to determine the benchmark performance of MILLS throughout the AMILES database. METHODS: The AMILES registry was interrogated for cases of minimally invasive left lateral sectionectomies (MILLS). Centers with best practices according to the achievement of textbook outcomes (TOs) were identified and were used to define benchmark performances. RESULTS: Seven institutions from US and Canada entered 1665 minimally invasive liver resections, encompassing 203 MILLS. Overall, 49% of cases of MILLS satisfied contemporarily all textbook outcomes. While all centers obtained TOs with different rates of success, the outcomes of the top-ranking centers were used for benchmarking. Benchmark performance metrics of MILLS across North America are: conversion rate ≤ 3.7%, blood loss ≤ 200 ml, OR time ≤ 199 min, transfusion rate ≤ 4.5%, complication rate ≤ 7.9%, LOS ≤ 4 days. CONCLUSION: Benchmark performances of MILLS have been defined on a large multi-institutional database in North America. As more institutions join the collaboration and more prospective cases accrue, benchmark for additional procedures and approaches will be defined.

AHPBA guidelines for managing VTE prophylaxis and anticoagulation for pancreatic surgery.

BACKGROUND: Major abdominal surgery and malignancy lead to a hypercoagulable state, with a risk of venous thromboembolism (VTE) of approximately 3% after pancreatic surgery. No guidelines exist to assist surgeons in managing VTE prophylaxis or anticoagulation in patients undergoing elective pancreatic surgery for malignancy or premalignant lesions. A systematic review specific to VTE prophylaxis and anticoagulation after resectional pancreatic surgery is herein provided. METHODS: Six topic areas are reviewed: pre- and perioperative VTE prophylaxis, early postoperative VTE prophylaxis, extended outpatient VTE prophylaxis, management of chronic anticoagulation, anti-coagulation after vascular reconstruction, and treatment of VTE. A Medline and PubMED search was completed with systematic medical literature review for each topic. Level of evidence was graded and strength of recommendation ranked according to the GRADE (Grades of Recommendation Assessment, Development and Evaluation) system for practice guidelines. RESULTS: Levels of evidence and strength of recommendations are presented. DISCUSSION: While strong data exist to guide management of chronic anticoagulation and treatment of VTE, data for anticoagulation after reconstruction is inconclusive and support for perioperative chemoprophylaxis with pancreatic surgery is similarly limited. The risk of post-pancreatectomy hemorrhage often exceeds that of thrombosis. The role of universal chemoprophylaxis must therefore be examined critically, particularly in the preoperative setting.

Long-Term Clinical Outcomes of an Antibiotic-Coated Non-Cross-linked Porcine Acellular Dermal Graft for Abdominal Wall Reconstruction for High-Risk and Contaminated Wounds.

BACKGROUND: Abdominal wall reconstruction in high-risk and contaminated cases remains a challenging surgical dilemma. We report long-term clinical outcomes for a rifampin-/minocycline-coated acellular dermal graft (XenMatrix™ AB) in complex abdominal wall reconstruction for patients with a prior open abdomen or contaminated wounds. METHODS: Patients undergoing abdominal wall reconstruction at our institution at high risk for surgical site occurrence and reconstructed with XenMatrix™ AB with intent-to-treat between 2014 and 2017 were included. Demographics, operative characteristics, and outcomes were collected. The primary outcome was hernia recurrence. The secondary outcomes included length of stay, surgical site occurrence, readmission, morbidity, and mortality. RESULTS: Twenty-two patients underwent abdominal wall reconstruction using XenMatrix™ AB during the study period. Two patients died while inpatient from progression of their comorbid diseases and were excluded. Sixty percent of patients had an open abdomen at the time of repair. All patients were from modified Ventral Hernia Working Group class 2 or 3. There were a total of four 30-day infectious complications including superficial cellulitis/fat necrosis (15%) and one intraperitoneal abscess (5%). No patients required reoperation or graft excision. Median clinical follow-up was 38.2 months with a mean of 35.2 +/- 18.5 months. Two asymptomatic recurrences and one symptomatic recurrence were noted during this period with one planning for elective repair of an eventration. Follow-up was extended by phone interview which identified no additional recurrences at a median of 45.5 and mean of 50.5 +/-12.7 months. CONCLUSION: We present long-term outcomes for patients with high-risk and contaminated wounds who underwent abdominal wall reconstruction reinforced with XenMatrix™ AB to achieve early, permanent abdominal closure. Acceptable outcomes were noted.

Robotic-assisted completion cholecystectomy: A safe and effective approach to a challenging surgical scenario - A single center retrospective cohort study.

BACKGROUND: Reoperation following a previous subtotal or aborted cholecystectomy presents a challenging surgical scenario that has traditionally required an open completion cholecystectomy. The aim of this study was to describe an institutional experience with a robotic-assisted approach to completion cholecystectomy. METHODS: A database was retrospectively audited to identify all patients who underwent robotic-assisted cholecystectomy performed by two hepatopancreatobiliary surgeons at a single centre from 2010 to 2019. RESULTS: Twenty six patients who underwent a robotic-assisted completion cholecystectomy were identified. Median operative time was 142 min (48-247 min) with a blood loss of 50 cc (0-500 cc). Minor complications (Clavien-Dindo ≤ II 90 days) occurred in three patients (11.5%) with no major complication or mortality reported. Median hospital length of stay was 1 day (0-6 days) with one patient readmitted. CONCLUSION: This study represents to our knowledge the largest series of robotic-assisted completion cholecystectomies to date. The robotic approach appears to be a safe and effective procedure associated with a low morbidity and high success rate.

The technical aspects of ex vivo hepatectomy with autotransplantation: a systematic review and meta-analysis.

PURPOSE: Ex vivo hepatectomy is the incorporation of liver transplant techniques in the non-transplant setting, providing opportunity for locally advanced tumors found conventionally unresectable. Because the procedure is rare and reports in the literature are limited, we sought to perform a systematic review and meta-analysis investigating technical variations of ex vivo hepatectomies. METHODS: In the literature, there is a split in those performing the procedure between venovenous bypass (VVB) and temporary portacaval shunts (PCS). Of the 253 articles identified on the topic of ex vivo resection, 37 had sufficient data to be included in our review. RESULTS: The majority of these procedures were performed for hepatic alveolar echinococcosis (69%) followed by primary and secondary hepatic malignancies. In 18 series, VVB was used, and in 18, a temporary PCS was performed. Comparing these two groups, intraoperative variables and morbidity were not statistically different, with a cumulative trend in favor of PCS. Ninety-day mortality was significantly lower in the PCS group compared to the VVB group (p=0.03). CONCLUSION: In order to better elucidate these differences between technical approaches, a registry and consensus statement are needed.

The treatment sequence may matter in patients undergoing pancreatoduodenectomy for early stage pancreatic cancer in the era of modern chemotherapy.

BACKGROUND: The aim of this study was to investigate outcomes associated with neoadjuvant chemotherapy in patients undergoing pancreatoduodenectomy for early stage pancreatic adenocarcinoma in the era of modern chemotherapy. METHODS: The National Cancer Database (2010-2016) was queried for patients with clinical stage 0-2 pancreatic adenocarcinoma who underwent pancreatoduodenectomy. Patients who underwent up-front pancreatoduodenectomy were propensity matched to patients who received neoadjuvant chemotherapy. Postoperative outcomes, pathologic outcomes, and overall survival were compared. RESULTS: A total of 2036 patients were in each group. Neoadjuvant chemotherapy was associated with shorter length of stay, lower 30-day readmission rate, and lower 30 and 90-day mortality rates (all p < 0.05). Neoadjuvant chemotherapy was associated with lower rates of positives nodes and positive resection margins (all p < 0.0001). Neoadjuvant chemotherapy was associated with longer survival (26.8 vs. 22.1months, p < 0.0001). Patients who received neoadjuvant chemotherapy followed by surgery and adjuvant therapy had the longest OS, followed by neoadjuvant + surgery, surgery + adjuvant therapy, and surgery alone (29.8 vs. 25.6 vs. 23.9 vs. 13.1 months; p < 0.0001). CONCLUSIONS: Neoadjuvant chemotherapy is associated with improved postoperative outcomes, oncologic outcomes, and overall survival in patients with early stage pancreatic adenocarcinoma. Neoadjuvant chemotherapy should be considered in all patients with early stage pancreatic adenocarcinoma.

Use of Artificial Intelligence Deep Learning to Determine the Malignant Potential of Pancreatic Cystic Neoplasms With Preoperative Computed Tomography Imaging.

BACKGROUND: Society consensus guidelines are commonly used to guide management of pancreatic cystic neoplasms (PCNs). However, downsides of these guidelines include unnecessary surgery and missed malignancy. The aim of this study was to use computed tomography (CT)-guided deep learning techniques to predict malignancy of PCNs. MATERIALS AND METHODS: Patients with PCNs who underwent resection were retrospectively reviewed. Axial images of the mucinous cystic neoplasms were collected and based on final pathology were assigned a binary outcome of advanced neoplasia or benign. Advanced neoplasia was defined as adenocarcinoma or intraductal papillary mucinous neoplasm with high-grade dysplasia. A convolutional neural network (CNN) deep learning model was trained on 66% of images, and this trained model was used to test 33% of images. Predictions from the deep learning model were compared to Fukuoka guidelines. RESULTS: Twenty-seven patients met the inclusion criteria, with 18 used for training and 9 for model testing. The trained deep learning model correctly predicted 3 of 3 malignant lesions and 5 of 6 benign lesions. Fukuoka guidelines correctly classified 2 of 3 malignant lesions as high risk and 4 of 6 benign lesions as worrisome. Following deep learning model predictions would have avoided 1 missed malignancy and 1 unnecessary operation. DISCUSSION: In this pilot study, a deep learning model correctly classified 8 of 9 PCNs and performed better than consensus guidelines. Deep learning can be used to predict malignancy of PCNs; however, further model improvements are necessary before clinical use.

Impact of Multidisciplinary Audit of Enhanced Recovery After Surgery (ERAS)® Programs at a Single Institution.

BACKGROUND: As Enhanced Recovery After Surgery (ERAS®) programs expand across numerous subspecialties, growth and sustainability on a system level becomes increasingly important and may benefit from reporting multidisciplinary and financial data. However, the literature on multidisciplinary outcome analysis in ERAS is sparse. This study aims to demonstrate the impact of multidisciplinary ERAS auditing in a hospital system. Additionally, we describe developing a financial metric for use in gaining support for system-wide ERAS adoption and sustainability. METHODS: Data from HPB, colorectal and urology ERAS programs at a single institution were analyzed from a prospective ERAS Interactive Audit System (EIAS) database from September 2015 to June 2019. Clinical 30-day outcomes for the ERAS cohort (n = 1374) were compared to the EIAS pre-ERAS control (n = 311). Association between improved ERAS compliance and improved outcomes were also assessed for the ERAS cohort. The potential multidisciplinary financial impact was estimated from hospital bed charges. RESULTS: Multidisciplinary auditing demonstrated a significant reduction in postoperative length of stay (LOS) (1.5 days, p < 0.001) for ERAS patients in aggregate and improved ERAS compliance was associated with reduced LOS (coefficient - 0.04, p = 0.004). Improved ERAS compliance in aggregate also significantly associated with improved 30-day survival (odds ratio 1.04, p = 0.001). Multidisciplinary analysis also demonstrated a potential financial impact of 44% savings (p < 0.001) by reducing hospital bed charges across all specialties. CONCLUSIONS: Multidisciplinary auditing of ERAS programs may improve ERAS program support and expansion. Analysis across subspecialties demonstrated associations between improved ERAS compliance and postoperative LOS as well as 30-day survival, and further suggested a substantial combined financial impact.

Treatment of spontaneously ruptured hepatocellular carcinoma: use of laparoscopic microwave ablation and washout.

BACKGROUND: Ruptured, or bleeding, hepatocellular carcinoma (rHCC) is a relatively rare disease presentation associated with high acute mortality rates. This study sought to evaluate outcomes following laparoscopic microwave ablation (MWA) and washout in rHCC. METHODS: A retrospective single-center review was performed to identify patients with rHCC (2008-2018). The treatment algorithm consisted of transarterial embolization (TAE) or trans-arterial chemoembolization (TACE) followed by laparoscopic MWA and washout. RESULTS: Fifteen patients with rHCC were identified (n = 5 single lesion, n = 5 multifocal disease, n = 5 extrahepatic metastatic disease). Median tumor size was 83 mm (range 5-228 mm), and 10 of 15 underwent TAE or TACE followed by laparoscopic MWA/washout. One patient required additional treatment for bleeding after MWA with repeat TAE. Thirty-day mortality was 6/15. For those patients discharged (n = 9), additional treatments included chemotherapy (n = 5), TACE (n = 3), and/or partial lobectomy (n = 2). Median follow-up was 18.2 months and median survival was 431 days (range 103-832) (one-year survival n = 7; two-year survival n = 4; three-year survival n = 3). Six patients had post-operative imaging from which one patient demonstrated recurrence. CONCLUSION: Using laparoscopic MWA with washout may offer advantage in the treatment of ruptured HCC. It not only achieves hemostasis but also could have oncologic benefit by targeting local tumor and decreasing peritoneal carcinomatosis risk.

Conversion therapy for intrahepatic cholangiocarcinoma and tumor downsizing to increase resection rates: A systematic review.

Intrahepatic cholangiocarcinoma (ICC) is a devastating malignant neoplasm with dismal outcomes. Several therapeutic modalities have been used with variable success to downsize these tumors for resection. Neoadjuvant therapy such as chemoembolization and radioembolization offer promising options to manage tumor burden prior to resection. A systematic review of the literature was performed with a focus on conversion therapy for ICC and tumor downsizing to increase resection rates among patients who have an initially unresectable tumor. Of 132 patients with initially unresectable ICC, we identified 27 who underwent conversion therapy with surgical resection. Adequate tumor downsizing was achieved with chemotherapy, chemoembolization, radioembolization, or combination thereof. Although negative tumor margins were possible in some patients, recurrence rates and survival outcomes were inconsistently reported. Twenty-three of 27 patients were alive at last reported follow-up. Conversion therapy for initially unresectable ICC may offer adequate tumor downsizing for resection.

Pure and Hybrid Deep Learning Models can Predict Pathologic Tumor Response to Neoadjuvant Therapy in Pancreatic Adenocarcinoma: A Pilot Study.

BACKGROUND: Neoadjuvant therapy may improve survival of patients with pancreatic adenocarcinoma; however, determining response to therapy is difficult. Artificial intelligence allows for novel analysis of images. We hypothesized that a deep learning model can predict tumor response to NAC. METHODS: Patients with pancreatic cancer receiving neoadjuvant therapy prior to pancreatoduodenectomy were identified between November 2009 and January 2018. The College of American Pathologists Tumor Regression Grades 0-2 were defined as pathologic response (PR) and grade 3 as no response (NR). Axial images from preoperative computed tomography scans were used to create a 5-layer convolutional neural network and LeNet deep learning model to predict PRs. The hybrid model incorporated decrease in carbohydrate antigen 19-9 (CA19-9) of 10%. Accuracy was determined by area under the curve. RESULTS: A total of 81 patients were included in the study. Patients were divided between PR (333 images) and NR (443 images). The pure model had an area under the curve (AUC) of .738 (P < .001), whereas the hybrid model had an AUC of .785 (P < .001). CA19-9 decrease alone was a poor predictor of response with an AUC of .564 (P = .096). CONCLUSIONS: A deep learning model can predict pathologic tumor response to neoadjuvant therapy for patients with pancreatic adenocarcinoma and the model is improved with the incorporation of decreases in serum CA19-9. Further model development is needed before clinical application.

An objective approach to evaluate novice robotic surgeons using a combination of kinematics and stepwise cumulative sum (CUSUM) analyses.

BACKGROUND: Current evaluation methods for robotic-assisted surgery (ARCS or GEARS) are limited to 5-point Likert scales which are inherently time-consuming and require a degree of subjective scoring. In this study, we demonstrate a method to break down complex robotic surgical procedures using a combination of an objective cumulative sum (CUSUM) analysis and kinematics data obtained from the da Vinci® Surgical System to evaluate the performance of novice robotic surgeons. METHODS: Two HPB fellows performed 40 robotic-assisted hepaticojejunostomy reconstructions to model a portion of a Whipple procedure. Kinematics data from the da Vinci® system was recorded using the dV Logger® while CUSUM analyses were performed for each procedural step. Each kinematic variable was modeled using machine learning to reflect the fellows' learning curves for each task. Statistically significant kinematics variables were then combined into a single formula to create the operative robotic index (ORI). RESULTS: The inflection points of our overall CUSUM analysis showed improvement in technical performance beginning at trial 16. The derived ORI model showed a strong fit to our observed kinematics data (R2 = 0.796) with an ability to distinguish between novice and intermediate robotic performance with 89.3% overall accuracy. CONCLUSIONS: In this study, we demonstrate a novel approach to objectively break down novice performance on the da Vinci® Surgical System. We identified kinematics variables associated with improved overall technical performance to create an objective ORI. This approach to robotic operative evaluation demonstrates a valuable method to break down complex surgical procedures in an objective, stepwise fashion. Continued research into objective methods of evaluation for robotic surgery will be invaluable for future training and clinical implementation of the robotic platform.