Is a Picture Worth a Thousand Words? A Scoping Review of the Impact of Visual Aids on Patients Undergoing Surgery.

OBJECTIVE: While graphics are commonly used by clinicians to communicate information to patients, the impact of using visual media on surgical patients is not understood. This review seeks to understand the current landscape of research analyzing impact of using visual aids to communicate with patients undergoing surgery, as well as gaps in the present literature. DESIGN: A comprehensive literature search was performed across 4 databases. Search terms included: visual aids, diagrams, graphics, surgery, patient education, informed consent, and decision making. Inclusion criteria were (i) full-text, peer-reviewed articles in English; (ii) evaluation of a nonelectronic visual aid(s); and (iii) surgical patient population. RESULTS: There were 1402 articles identified; 21 met study criteria. Fifteen were randomized control trials and 6 were prospective cohort studies. Visual media assessed comprised of diagrams as informed consent adjuncts (n = 6), graphics for shared decision-making conversations (n = 3), other preoperative educational graphics (n = 8), and postoperative educational materials (n = 4). There was statistically significant improvement in patient comprehension, with an increase in objective knowledge recall (7.8%-29.6%) using illustrated educational materials (n = 10 of 15). Other studies noted increased satisfaction (n = 4 of 6), improvement in shared decision-making (n = 2 of 4), and reduction in patient anxiety (n = 3 of 6). For behavioral outcomes, visual aids improved postoperative medication compliance (n = 2) and lowered postoperative analgesia requirements (n = 2). CONCLUSIONS: The use of visual aids to enhance the surgical patient experience is promising in improving knowledge retention, satisfaction, and reducing anxiety. Future studies ought to consider visual aid format, and readability, as well as patient language, race, and healthcare literacy.

Sleeve Gastrectomy Reduces the Need for Liver Transplantation in Patients with Obesity and Non-Alcoholic Steatohepatitis: a Predictive Model.

BACKGROUND: Non-alcoholic steatohepatitis (NASH) is one of the leading indications for liver transplantation (LT) in the United States. As with the current obesity epidemic, the incidence of NASH continues to rise. However, the impact of broad utilization of bariatric surgery (BS) for patients with NASH is unknown, particularly in regard to mitigating the need for LT. METHODS: Markov decision analysis was performed to simulate the lives of 20,000 patients with obesity and concomitant NASH who were deemed ineligible to be waitlisted for LT unless they achieved a body mass index (BMI) < 35 kg/m2. Life expectancy following medical weight management (MWM) and sleeve gastrectomy (SG) were estimated. Base case patients were defined as having NASH without fibrosis and a pre-intervention BMI of 45 kg/m2. Sensitivity analysis of initial BMI was performed. RESULTS: Simulated base case analysis patients who underwent SG gained 14.3 years of life compared to patients who underwent MWM. One year after weight loss intervention, 9% of simulated MWM patients required LT compared to only 5% of SG patients. Survival benefit for SG was observed above a BMI of 32.2 kg/m2. CONCLUSION: In this predictive model of 20,000 patients with obesity and concomitant NASH, surgical weight loss is associated with a reduction in the progression of NASH, thereby reducing the need for LT. A reduced BMI threshold of 32 kg/m2 for BS may offer survival benefit for patients with obesity and NASH.

Remote triage practices in general surgery patients from freestanding emergency departments: A 6-year analysis.

BACKGROUND: Freestanding emergency departments have risen in popularity as a means to expand access to care. Although some evaluation of freestanding emergency department utility in specific patient populations exists, management of surgical patients via remote triage and disposition has not been previously described. We report our experience with remote triage to discharge home, level I trauma center, or community hospital admission for general surgery patients who present to an affiliated freestanding emergency department. METHODS: A retrospective cohort study of patients presenting to freestanding emergency departments requiring surgical consultation between 2016 and 2021 was conducted. Outcomes included disposition, length of stay, surgical intervention, 30-day mortality, and readmission. Undertriage and overtriage rates were calculated and defined as the following: (1) discharge undertriage-discharge home with 30-day emergency department visit/readmission; 2) transfer undertriage-transfers to community hospital requiring transfer to trauma center; and (3) overtriage-admissions <24 hours without surgery. RESULTS: Of 1,105 patients, 15% were discharged home, 27% were transferred to trauma centers, and 58% were transferred to community hospitals. Patients admitted to trauma centers were older and had higher acuity pathology, whereas patients admitted to community hospitals had higher operative rates with shorter lengths of stay, operating room time, 30-day readmission, and mortality. Transfer undertriage was 0.9% (n = 6), with only 1 patient requiring transfer from a community hospital to a trauma center for disease acuity. Discharge undertriage was 12% (n = 20) due to worsening or persistent pathology. Overtriage was 5.5% (n = 52), with most having a partial small bowel obstruction or ambiguous diagnostic imaging requiring observation. CONCLUSION: Remote surgery triage at freestanding emergency departments, without an in-person examination, demonstrated both low undertriage and overtriage rates, reflecting appropriate triage practices.

Virtual triage from freestanding emergency departments: a propensity score-weighted analysis of short-term outcomes in emergency general surgery.

BACKGROUND: Freestanding emergency departments (FSEDs) have generated improved hospital metrics, including decreased ED wait times and increased patient selection. Patient outcomes and process safety have not been evaluated. This study investigates the safety of FSED virtual triage in the emergency general surgery (EGS) patient population. METHODS AND PROCEDURES: A retrospective review evaluated all adult EGS patients admitted to a community hospital between January 2016 and December 2021 who either presented at a FSED and received virtual evaluation from a surgical team (fEGS) or presented at the community hospital emergency department and received in-person evaluation from the same surgical group (cEGS). Patients' demographics, acute care utilization history, and clinical characteristics at the onset of the index visit were used to build a propensity score model and stabilized Inverse Probability of Treatment Weights (IPTW) were used to create a weighted sample. Multivariable regression models were then employed to the weighted sample to evaluate the treatment effect of virtual triage compared to in-person evaluation on short-term outcomes, including length of stay (LOS) and 30-day readmission and mortality. Variables which occurred during the index visit (such as surgery duration and type of surgery) were adjusted for in the multivariable analyses. RESULTS: Of 1962 patients, 631 (32.2%) were initially evaluated virtually (fEGS) and 1331 (67.8%) underwent an in-person evaluation (cEGS). Baseline characteristics demonstrated significant differences between the cohorts in gender, race, payer status, BMI, and CCI score. Baseline risks were well balanced in the IPTW-weighted sample (SD range 0.002-0.18). Multivariable analysis found no significant differences between the balanced cohorts in 30-day readmission, 30-day mortality, and LOS (p > 0.05 for all). CONCLUSION: Patients who undergo virtual triage have similar outcomes to those who undergo in-person triage for EGS diagnoses. Virtual triage at FSED for these EGS patients may be an efficient and safe means for initial evaluation.

Future liver remnant augmentation preceding ex vivo hepatectomy with IVC replacement: a strategy to achieve R0 margins.

PURPOSE: Ex vivo hepatectomy with autotransplantation (EHAT) provides opportunity for R0 resection. As EHAT outcomes after future liver remnant (FLR) augmentation techniques are not well documented, we examine results of EHAT after augmentation for malignant tumors. METHODS: Retrospective analysis of six cases of EHAT was performed. Of these, four occurred after preoperative FLR augmentation between 2018 and 2022. RESULTS: Six patients were offered EHAT of 26 potential candidates. Indications for resection were involvement of hepatic vein outflow and inferior vena cava (IVC) with metastatic colorectal carcinoma (n = 3), cholangiocarcinoma (n = 2), or leiomyosarcoma (n = 1). Five patients were treated with neoadjuvant chemotherapy and four had preoperative liver augmentation. One hundred percent of cases achieved R0 resection. Of the augmented cases, three patients are alive after median follow-up of 28 months. Postoperative mortality due to liver failure was 25% (n = 1). CONCLUSIONS: For select patients with locally advanced tumors involving all hepatic veins and the IVC for whom conventional resection is not an option, EHAT provides opportunity for R0 resection. In addition, in patients with inadequate FLR volume, further operative candidacy with acceptable results can be achieved by combined liver augmentation techniques. To better characterize outcomes in this small subset, a registry is needed.

Just What the Doctor Ordered: Missed Ordering of Venous Thromboembolism Chemoprophylaxis Is Associated With Increased VTE Events in High-risk General Surgery Patients.

OBJECTIVE: To define the impact of missed ordering of venous thromboembolism (VTE) chemoprophylaxis in high-risk general surgery populations. BACKGROUND: The primary cause of preventable death in surgical patients is VTE. Although guidelines and validated risk calculators assist in dosing recommendations, there remains considerable variability in ordering and adherence to recommended dosing. METHODS: All adult inpatients who underwent a general surgery procedure between 2016 and 2019 and were entered into Atrium Health National Surgical Quality Improvement Program registry were identified. Patients at high risk for VTE (2010 Caprini score ≥5) and without bleeding history and/or acute renal failure were included. Primary outcome was 30-day postoperative VTE. Electronic medical record identified compliance with "perfect" VTE chemoprophylaxis orders (pVTE): no missed orders and no inadequate dose ordering. Multivariable analysis examined association between pVTE and 30-day VTE events. RESULTS: A total of 19,578 patients were identified of which 4252 were high-risk inpatients. Hospital compliance of pVTE was present in 32.4%. pVTE was associated with shorter postoperative length of stay and lower perioperative red blood cell transfusions. There was 50% reduced odds of 30-day VTE event with pVTE (odds ratio: 0.50; 95% CI, 0.30-0.80) and 55% reduction in VTE event/mortality (odds ratio: 0.45; 95% CI, 0.31-0.63). After controlling for relevant covariates, pVTE remained significantly associated with decreased odds of VTE event and VTE event/mortality. CONCLUSIONS: pVTE ordering in high-risk general surgery patients was associated with 42% reduction in odds of postoperative 30-day VTE. Comprehending factors contributing to missed or suboptimal ordering and development of quality improvement strategies to reduce them are critical to improving outcomes.

The physiology of failure: Identifying risk factors for mortality in emergency general surgery patients using a regional health system integrated electronic medical record.

BACKGROUND: Emergency general surgery (EGS) patients have increased mortality risk compared with elective counterparts. Recent studies on risk factors have largely used national data sets limited to administrative data. Our aim was to examine risk factors in an integrated regional health system EGS database, including clinical and administrative data, hypothesizing that this novel process would identify clinical variables as important risk factors for mortality. METHODS: Our nine-hospital health system's billing data were queried for EGS International Classification of Disease codes between 2013 and 2018. Codes were grouped by diagnosis, and urgent or emergent encounters were included and merged with electronic medical record clinical data. Outcomes assessed were inpatient and 1-year mortality. Standard and multivariable statistics evaluated factors associated with mortality. RESULTS: There were 253,331 EGS admissions with 3.6% inpatient mortality rate. Patients who suffered inpatient and 1-year mortality were older, more likely to be underweight, and have neutropenia or elevated lactate. On multivariable analysis for inpatient mortality: age (odds ratio [OR], 1.7-6.7), underweight body mass index (OR, 1.6), transfer admission (OR, 1.8), leukopenia (OR, 2.0), elevated lactate (OR, 1.8), and ventilator requirement (OR, 7.1) remained associated with increased risk. Adjusted analysis for 1-year mortality demonstrated similar findings, with highest risk associated with older age (OR, 2.8-14.6), underweight body mass index (OR, 2.3), neutropenia (OR, 2.0), and tachycardia (OR, 1.7). CONCLUSION: After controlling for patient and disease characteristics available in administrative databases, clinical variables remained significantly associated with mortality. This novel yet simple process allows for easy identification of clinical data points imperative to the study of EGS diagnoses that are critical in understanding factors that impact mortality. LEVEL OF EVIDENCE: Prognostic and Epidemiologic; Level III.

Novel Way to Improve Satisfaction, Comprehension, and Anxiety in Caregivers: A Randomized Trial Exploring the Use of Comprehensive, Illustrated Children's Books for Pediatric Surgical Populations.

BACKGROUND: Surgery generates anxiety and stress, which can negatively impact informed consent and postoperative outcomes. This study assessed whether educational, illustrated children's books improve comprehension, satisfaction, and anxiety of caregivers in pediatric surgical populations. METHODS: A prospective randomized trial was initiated at a tertiary care children's hospital. All patients ≤ 18 years old with caregiver and diagnosis of 1) uncomplicated appendicitis (English or Spanish speaking); 2) ruptured appendicitis; 3) pyloric stenosis; 4) need for gastrostomy tube; or 5) umbilical hernia were eligible. Conventional consent was obtained followed by completion of 17 validated survey questions addressing apprehension, satisfaction, and comprehension. Randomization (2:1) occurred after consent and before operative intervention with the experimental group (EG) receiving an illustrated comprehensive children's book outlining anatomy, pathophysiology, hospital course, and postoperative care. A second identical survey was completed before discharge. Primary outcomes were caregiver apprehension, satisfaction, and comprehension. RESULTS: Eighty caregivers were included (55: EG, 25: control group [CG]). There were no significant differences in patient or caregiver demographics between groups. The baseline survey demonstrated no difference in comprehension, satisfaction, or apprehension between groups (all p values NS). After intervention, EG had significant improvement in 14 of 17 questions compared with CG (all p < 0.05). When tabulated by content, there was significant improvement in comprehension (p = 0.0009), satisfaction (p < 0.0001), and apprehension (p < 0.0001). CONCLUSION: The use of illustrated educational children's books to explain pathophysiology and surgical care is a novel method to improve comprehension, satisfaction, and anxiety of caregivers. This could benefit informed consent, understanding, and postoperative outcomes.

Deconstructing the roadmap to surgical residency: a national survey of residents illuminates factors associated with recruitment success as well as applicants' needs and beliefs.

Purpose: As applications increase and residency becomes more competitive, applicants and programs will be challenged by increased demands on recruitment, metric assessment, and rank determination. Studies have investigated program opinions; however, this survey sought to illuminate the process from an applicant's perspective. Methods: An anonymous survey was distributed to past or current surgery residents nationwide using social media and program director emails. Regression analyses were performed to assess factors correlating with percentage of programs which offered the applicant an interview. Results: There were 223 respondents who applied to an average of 61 programs (± 40) with 16 (± 11) interviews offered. Applicants believed that programs were most interested in (1) personality, (2) letter of recommendation (LOR) writers, and (3) medical school reputation. Top factors considered by applicants in ranking were resident culture, location, program reputation, and autonomy. Bivariate analysis found factors that decreased percent of interview invites to be Asian race, whereas factors that increased interview invites included age, year of match, surgery clerkship grade, medicine clerkship grade, AOA status, honor surgery rotation, gold humanism (GHHS) status, phone call for interview made, and step scores (all p < 0.05). AOA status, step scores, honor surgery rotation, year of match, and Asian race remained significant after multivariate analysis. Conclusions: National surveys illuminate how applicants approach the application process and what programs and applicants appear to value. This information provides insight and guidance to candidates and programs as the process of matching becomes more challenging with surging application numbers, changes in testing parameters and virtual interviews. Supplementary Information: The online version contains supplementary material available at 10.1007/s44186-022-00070-9.

Emergency general surgery transfer to lower acuity facility: The role of right-sizing care in emergency general surgery regionalization.

BACKGROUND: Regionalization of emergency general surgery (EGS) has primarily focused on expediting care of high acuity patients through interfacility transfers. In contrast, triaging low-risk patients to a nondesignated trauma facility has not been evaluated. This study evaluates a 16-month experience of a five-surgeon team triaging EGS patients at a tertiary care, Level I trauma center (TC) to an affiliated community hospital 1.3 miles away. METHODS: All EGS patients who presented to the Level I TC emergency department from January 2020 to April 2021 were analyzed. Patients were screened by EGS surgeons covering both facilities for transfer appropriateness including hemodynamics, resource need, and comorbidities. Patients were retrospectively evaluated for disposition, diagnosis, comorbidities, length of stay, surgical intervention, and 30-day mortality and readmission. RESULTS: Of 987 patients reviewed, 31.5% were transferred to the affiliated community hospital, 16.1% were discharged home from the emergency department, and 52.4% were admitted to the Level I TC. Common diagnoses were biliary disease (16.8%), bowel obstruction (15.7%), and appendicitis (14.3%). Compared with Level I TC admissions, Charlson Comorbidity Index was lower (1.89 vs. 4.45, p < 0.001) and length of stay was shorter (2.23 days vs. 5.49 days, p < 0.001) for transfers. Transfers had a higher rate of surgery (67.5% vs. 50.1%, p < 0.001) and lower readmission and mortality (8.4% vs. 15.3%, p = 0.004; 0.6% vs. 5.0%, p < 0.001). Reasons not to transfer were emergency evaluation, comorbidity burden, operating room availability, and established care. No transfers required transfer back to higher care (under-triage). Bed days saved at the Level I TC were 693 (591 inpatients). Total operating room minutes saved were 24,008 (16,919, between 7:00 am and 5:00 pm). CONCLUSION: Transfer of appropriate patients maintains high quality care and outcomes, while improving operating room and bed capacity and resource utilization at a tertiary care, Level I TC. Emergency general surgery regionalization should consider triage of both high-risk and low-risk patients. LEVEL OF EVIDENCE: Prospective comparative cohort study, Level II.

Curriculum matrix development for a hepato-pancreato-biliary robotic surgery fellowship.

Robotic surgery is being increasingly used for complex benign and malignant hepato-pancreato-biliary (HPB) cases. As use of robotics increases, fellowships to excel in complex robotic procedures will be sought after. With this dedicated training, attending surgeon positions can be obtained that can incorporate and teach this skill set. Unfortunately, there are no evidence-based approaches for constructing a curriculum for an HPB robotic surgery fellowship. This paper describes a technique to develop a structured curriculum to ensure competence and fulfil the learning and practice needs for robotic HPB fellows.

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.

Objective evaluation of technical dexterity in robotic hepaticojejunostomy: Assessment of hepatopancreatobiliary fellows using cumulative sum analytics.

BACKGROUND: The development of technical dexterity is a critical for surgeons in training. This study describes and assesses the feasibility of an objective method for the evaluation of procedure-specific technical dexterity in hepatopancreatobiliary (HPB) surgery using cumulative sum (CUSUM) analysis. METHODS: Dry-lab HPB procedures were divided into procedural steps with binary outcomes (success or failure). Two HPB fellows completed 20 dry lab hepaticojejunostomy (HJ) procedures. Participant progress was tracked over time with CUSUM analytics to establish a learning curve for procedural proficiency. RESULTS: The CUSUM charts for 20 consecutive dry-lab HJ procedures were analysed. A learning curve was created and used to identify areas of weakness to facilitate improvement in technical proficiency. CONCLUSIONS: CUSUM is effective tool for objective evaluation of technical dexterity offering both simplicity and adaptability. We demonstrate its use and feasibility for surgical education and plan to expand its' application to assess residents performing general surgery procedures.

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.

Palliative Care in Emergency General Surgery Patients: Reduced Inpatient Mortality And Increased Discharge to Hospice.

BACKGROUND: Admissions due to emergency general surgery (EGS) are on the rise, and patients who undergo emergency surgery are at increased risk of mortality. We hypothesized that utilization of palliative care and discharge to hospice in the EGS population have increased over time and that this is associated with a decrease in inpatient mortality. METHODS: Using the 2002-2011 nationwide inpatient sample and American Association for the Surgery of Trauma-defined EGS diagnosis codes, we identified patients ≥18 years old with an EGS admission. Demographics, hospitalization characteristics, mortality, use of palliative care services, and discharge to hospice were queried. All Patient Refined-Diagnosis Related Group risk of mortality was used to categorize those with an extreme likelihood of dying (ELD). Multivariable logistic regression was used to investigate the association between palliative care consult and discharge to hospice. RESULTS: Of the included patients, 0.3% received palliative care and 0.2% were discharged to hospice. Over time, rates of palliative care and hospice discharge increased while inpatient mortality decreased. In the 4% of patients with ELD, 3% received palliative care, 5% were transitioned to hospice care, and 22% suffered inpatient mortality. Controlling for patient characteristics, utilization of palliative care services was associated with increased odds of discharge to hospice compared to inpatient mortality (OR = 1.78 all patients and OR = 2.04 for ELD). CONCLUSIONS: Despite the known increased risks associated with emergency surgical diagnoses, palliative care services remain infrequently utilized in the EGS population. This may be an opportunity for lessening suffering, improving patient-concordant care and outcomes, and reducing nonbeneficial and unwanted care.

Effect of Margin Status on Survival After Resection of Hilar Cholangiocarcinoma in the Modern Era of Adjuvant Therapies.

INTRODUCTION: Studies have shown that for patients with hilar cholangiocarcinoma (HC), survival is associated with negative resection margins (R0). This requires increasingly proximal resection, putting patients at higher risk for complications, which may delay chemotherapy. For patients with microscopically positive resection margins (R1), the use of modern adjuvant therapies may offset the effect of R1 resection. METHODS: Patients at our institution with HC undergoing curative-intent resection between January 2008 and July 2019 were identified by retrospective record review. Demographic data, operative details, tumor characteristics, postoperative outcomes, recurrence, survival, and follow-up were recorded. Patients with R0 margin were compared to those with R1 margin. Patients with R2 resection were excluded. RESULTS: Seventy-five patients underwent attempted resection with 34 (45.3%) cases aborted due to metastatic disease or locally advanced disease. Forty-one (54.7%) patients underwent curative-intent resection with R1 rate of 43.9%. Both groups had similar rates of adjuvant therapy (56.5% vs. 61.1%, P = .7672). Complication rates and 30 mortality were similar between groups (all P > .05). Both groups had similar median recurrence-free survival (R0 29.2 months vs. R1 27.8 months, P = .540) and median overall survival (R0 31.2 months vs. R1 38.8 months, P = .736) with similar median follow-up time (R0 29.9 months vs. R1 28.5 months, P = .8864). CONCLUSIONS: At our institution, patients undergoing hepatic resection for HC with R1 margins have similar recurrence-free and overall survival to those with R0 margins. Complications and short-term mortality were similar. This may indicate that with use of modern adjuvant therapies obtaining an R0 resection is not an absolute mandate.

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.