Recalibration of a Deep Learning Model for Low-Dose Computed Tomographic Images to Inform Lung Cancer Screening Intervals.

Importance: Annual low-dose computed tomographic (LDCT) screening reduces lung cancer mortality, but harms could be reduced and cost-effectiveness improved by reusing the LDCT image in conjunction with deep learning or statistical models to identify low-risk individuals for biennial screening. Objective: To identify low-risk individuals in the National Lung Screening Trial (NLST) and estimate, had they been assigned a biennial screening, how many lung cancers would have been delayed 1 year in diagnosis. Design, Setting, and Participants: This diagnostic study included participants with a presumed nonmalignant lung nodule in the NLST between January 1, 2002, and December 31, 2004, with follow-up completed on December 31, 2009. Data were analyzed for this study from September 11, 2019, to March 15, 2022. Exposures: An externally validated deep learning algorithm that predicts malignancy in current lung nodules using LDCT images (Lung Cancer Prediction Convolutional Neural Network [LCP-CNN]; Optellum Ltd) was recalibrated to predict 1-year lung cancer detection by LDCT for presumed nonmalignant nodules. Individuals with presumed nonmalignant lung nodules were hypothetically assigned annual vs biennial screening based on the recalibrated LCP-CNN model, Lung Cancer Risk Assessment Tool (LCRAT + CT [a statistical model combining individual risk factors and LDCT image features]), and the American College of Radiology recommendations for lung nodules, version 1.1 (Lung-RADS). Main Outcomes and Measures: Primary outcomes included model prediction performance, the absolute risk of a 1-year delay in cancer diagnosis, and the proportion of people without lung cancer assigned a biennial screening interval vs the proportion of cancer diagnoses delayed. Results: The study included 10 831 LDCT images from patients with presumed nonmalignant lung nodules (58.7% men; mean [SD] age, 61.9 [5.0] years), of whom 195 were diagnosed with lung cancer from the subsequent screen. The recalibrated LCP-CNN had substantially higher area under the curve (0.87) than LCRAT + CT (0.79) or Lung-RADS (0.69) to predict 1-year lung cancer risk (P < .001). If 66% of screens with nodules were assigned to biennial screening, the absolute risk of a 1-year delay in cancer diagnosis would have been lower for recalibrated LCP-CNN (0.28%) than LCRAT + CT (0.60%; P = .001) or Lung-RADS (0.97%; P < .001). To delay only 10% of cancer diagnoses at 1 year, more people would have been safely assigned biennial screening under LCP-CNN than LCRAT + CT (66.4% vs 40.3%; P < .001). Conclusions and Relevance: In this diagnostic study evaluating models of lung cancer risk, a recalibrated deep learning algorithm was most predictive of 1-year lung cancer risk and had least risk of 1-year delay in cancer diagnosis among people assigned biennial screening. Deep learning algorithms could prioritize people for workup of suspicious nodules and decrease screening intensity for people with low-risk nodules, which may be vital for implementation in health care systems.

Presence of refractory GERD-like symptoms following laparoscopic fundoplication is rarely indicative of true recurrent GERD.

BACKGROUND: Laparoscopic fundoplication (LF) is the gold standard for gastroesophageal reflux disease (GERD). Recurrent GERD is a known complication; however, the incidence of recurrent GERD-like symptoms and long-term fundoplication failure is rarely reported. Our objective was to identify the rate of recurrent pathologic GERD in patients with GERD-like symptoms following fundoplication. We hypothesized that patients with recurrent GERD-like symptoms refractory to medical management do not have evidence of fundoplication failure as indicated by a positive ambulatory pH study. METHODS: This is a retrospective cohort study of 353 consecutive patients undergoing LF for GERD between 2011 and 2017. Baseline demographics, objective testing, GERD-HRQL scores, and follow-up data were collected in a prospective database. Patients with return visits to clinic following routine post-operative visits were identified (n = 136, 38.5%), and those with a primary complaint of GERD-like symptoms (n = 56, 16%) were included. The primary outcome was the proportion of patients with a positive post-operative ambulatory pH study. Secondary outcomes included proportion of patients with symptoms managed with acid-reducing medications, time to return to clinic, and need for reoperation. P values < 0.05 were considered significant. RESULTS: Fifty-six (16%) patients returned during the study period for an evaluation of recurrent GERD-like symptoms with a median interval of 51.2 (26.2-74.7) months. Twenty-four patients (42.9%) were successfully managed expectantly or with acid-reducing medications. Thirty two (57.1%) presented with GERD-like symptoms and failure of management with medical acid suppression and underwent repeat ambulatory pH testing. Of these, only 5 (9%) were found to have a DeMeester score of > 14.7, and three (5%) underwent recurrent fundoplication. CONCLUSION: Following LF, the incidence of GERD-like symptoms refractory to PPI therapy is much higher than the incidence of recurrent pathologic acid reflux. Few patients with recurrent GI symptoms require surgical revision. Evaluation, including objective reflux testing, is critical to evaluating these symptoms.

Expectations versus reality: trainee participation on the robotic console in academic surgery.

INTRODUCTION: Trainees underestimate the amount of operative autonomy they receive, whereas faculty overestimate; this has not been studied in robotics. We aimed to assess the perceptions and expectations of our general surgery trainees and faculty on robotic console participation in academic surgery. METHODS: A survey was administered to general surgery robotic faculty and trainees eligible to sit at the console. Participants estimated the average percentage of trainee console participation time (CPT) per case for robotic cholecystectomies (CCY) and inguinal hernia repairs (IHR) from January to June 2019. Trainees were additionally asked what CPT they expected according to their training level (novice or senior). Expected CPTs were compared to actual CPTs extracted from robotic console logs during the same time frame. RESULTS: Survey response rate was 80% for faculty (4 of 5) and 65% for trainees (15 of 23). Novices expected a higher CPT than they perceived in CCY (42.8% ± 14.8% vs 19.0% ± 17.2%, p = 0.03) and IHR (36.1% ± 17.6% vs. 10.7% ± 13.7%, p = 0.01), but in actuality, they did more CPT than perceived (by 34.9% in CCY, p < 0.01; 14% in IHR, p = 0.10). Senior trainees accurately perceived their CPT in IHR, but expected a higher CPT by 15.9% (p = 0.04). In CCY, seniors perceived a 23.8% higher CPT than in reality (p = 0.04). Faculty generally overperceived trainee CPT by 12.8-16.3% (p > 0.05). Compared to faculty, novices perceived lower CPTs in both CCY by 29.9% (p = 0.16) and IHR by 26.8% (p = 0.07), but seniors tended to agree with the faculty-perceived CPTs (p > 0.05). CONCLUSION: Our robotic trainees expect to do more on the console than they perceive. Faculty think they allow their trainees more participation than in reality. Compared to faculty perception, novice trainees perceive a much lower level of trainee participation than senior trainees do. Expectation setting and standardizing learning curves are important for robotic surgery training.

Tobacco use is not associated with increased risk of recurrent reflux 5 years after laparoscopic anti-reflux surgery.

INTRODUCTION: Tobacco use worsens gastrointestinal reflux disease (GERD). Smoking cessation improves GERD symptoms, but its impact on the efficacy of laparoscopic anti-reflux surgery (LARS) is unclear. In this retrospective cohort study, we hypothesized that non-smokers would demonstrate greater long-term improvements in disease-specific quality of life than active smokers. METHODS: Data were maintained in an IRB-approved prospective database, and patients were stratified according to tobacco use. Postoperative follow-up occurred in clinic and long-term follow-up via telephone interview. Outcomes measured were gastroesophageal health-related quality of life (GERD-HRQL) and GERD symptom scale (GERSS) scores, proton pump inhibitor (PPI) cessation, and satisfaction with surgery. RESULTS: Two hundred and thirty-five patients underwent primary LARS, and 31 (13%) were active smokers with 18 median pack-years [10-30]. Baseline PPI use (96% vs. 94%, p = 0.64), presence of a hiatal hernia (79% vs. 68%, p = 0.13), esophagitis (28% vs. 45%, p = 0.13), and DeMeester score (41.9 vs. 33.6, p = 0.47) were similar. Baseline GERD-HRQL and GERSS scores and their post-surgical decreases were also similar between groups. PPI cessation was achieved in 92% of non-smokers and 94% of smokers (p = 0.79), and GERD-HRQL scores decreased to 4 [1-7] and 5 [0-12], respectively (p = 0.53). After 59 [25-74] months, GERD-HRQL scores were 5 [2-11] and 2 [0-13] (p = 0.61) and PPI cessation was maintained in 69% and 79% of patients (p = 0.59). Satisfaction with surgery was similar between smokers and non-smokers (88% vs. 87%, p = 0.85). Female gender was significantly associated with increased improvements in GERD-HRQL (p < 0.01) and GERSS scores (p = 0.04) postoperatively but not at long-term follow-up. Patients without a hiatal hernia were less likely to achieve long-term PPI cessation compared to those with a hernia (OR 0.23, p < 0.01). CONCLUSIONS: After 5 years, smokers demonstrate similar symptom resolution, PPI cessation rates, and satisfaction with surgery as non-smokers. Active smoking does not appear to negatively impact long-term symptomatic outcomes of LARS.

Use of the MMPI-2 personality profile in predicting 30-day ED-visits and readmissions following primary bariatric surgery.

BACKGROUND: Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is used for psychological screening of bariatric surgery (BS) candidates. To date, no studies have analyzed the relationship between MMPI-2 and early returns to hospital. The aim of this study was to determine whether high T scores on the MMPI-2 clinical scales were associated with early return to hospital after primary bariatric surgery. METHODS: Patients who completed an MMPI-2 evaluation, undergoing primary BS from 2014 to 2016 were evaluated. T score for the tested scales were collected and stratified into a high T score (T > 65) vs not (T < 65). The optimal 'cut-point' (specific number of high T scores predicting likelihood for 30-day ED-visit/hospital readmission) was calculated using Youden's Index (J) = Max(c) [sensitivity (c) + specificity (c) - 1], where c = number of scales with a T score > 65. Patients were stratified based on the optimal cut-point which was determined to be ≥ 4 high T scores. Univariate and multivariate logistic regression analyses were used to identify differences between groups and predictors for early ED-visits and hospital readmissions. RESULTS: 375 patients had psychological evaluations available for review. Patients were divided into those with ≥ 4 high T scores (Scr(≥4); n = 86) versus not (Scr (<4); n = 289). Multivariate analysis showed Scr(≥ 4) (aOR 2.99, CI 1.20-7.47; p = 0.019), bipolar disorder (aOR 4.82, CI 1.25-18.83; p = 0.022), and urgent hospital complications (aOR 6.81, CI 2.02-22.91; p = 0.002), were significant independent predictors of 30-day readmissions. Early ED-visits were significantly predicted by public insurance (aOR 3.30, CI 1.22-8.91; p = 0.019), but the effect of the Scr(≥4) profile (aOR 2.42, CI 0.97-6.09; p = 0.06), while influential, did not reach significance. CONCLUSION: Differences in personality traits may be associated increased 30-day readmissions following primary bariatric surgery. Our study represents a novel application of the MMPI-2.

A longitudinal cadaver-based simulation curriculum creates sustainable increases in resident confidence and exposure to fundamental techniques: Results of a 5-year program evaluation.

INTRODUCTION: For the past five years, our surgical residency program has led a cadaver-based simulation course focused on fundamental surgical maneuvers. This study aimed to quantify the impact of this course on resident exposure to surgical skills and longitudinal impact on resident education. METHODS: General surgery residents participated in an annual cadaver-based simulation curriculum. Participants completed surveys regarding improvements in knowledge and confidence; these results were stratified between course iterations (P1: 2 years, 2014-15; P2: 3 years, 2016-2018). RESULTS: Residents reported a sustained increase in knowledge of anatomy and technical dissection, confidence in performing operative skills independently, and exposure to operative skills that were otherwise not encountered in clinical rotations. Junior residents demonstrated an increase in gaining skills they would otherwise not achieve (87% vs. 98%, p = 0.028) and confidence to safely perform these procedures in the clinical setting (94% vs. 100%, p = 0.077). CONCLUSION: This annual, longitudinal cadaver-based skills course focused on fundamental maneuvers demonstrates a sustained impact in resident and faculty surgical confidence in resident's operative skills as a component of a longitudinal simulation curriculum to enhance competency-based promotion.

Can a Structured, Video-Based Cadaver Curriculum Demonstrating Proficiency Enhance Resident Operative Autonomy?

INTRODUCTION: Autonomy is of foremost concern in the current era of surgical residency, and it is especially important to trainees when considering their surgical education. Factors impacting trainee independence include the restriction of clinical work hours and the development of advanced minimally invasive techniques such as robotics, which requires separate technical education outside of conventional surgical education. Moreover, when residents are left to learn fundamental exposures via their clinical experience alone, they run the risk of not being exposed to some fundamental skills based on case volume and type. The Department of Surgery at Emory University developed a cadaver-based simulation curriculum to standardize exposure to fundamental operative skills and enhance proficiency outside the operating room, with the larger aim of improving resident autonomy. METHODS: Residents were assigned to small groups led by a chief resident with an even distribution of postgraduate year (PGY) levels. Each group participated in core surgical exposures and fundamental maneuvers on a cadaver over a 6-hour session. Residents were tested on skills according to their PGY level about 1 month after the course. Testing included recitation of the skill in an oral boards format, highlighting major steps, followed by performance of the skill. All steps were video-recorded with no resident identifiers. These were reviewed by 2 independent, blinded faculty examiners who assigned proficiency grades to each resident video. RESULTS: Three hundred and thirty-three individual procedure evaluations were done over the 5-year period. Senior residents (PGY3-5) had 86% pass rate while junior residents (PGY1-2) had 70% pass rate. Overall, 21% of residents failed to achieve competence in their assigned skills. Junior residents were less likely to achieve competence compared to senior residents. Faculty graders had improved congruence in grading as the course progressed through the 5 years. The most recent 2 years had >80% congruence in faculty grading compared to less than 50% congruence in the first 2 years. 81% of attendings agreed this course positively influenced the granting of autonomy in the operating room. CONCLUSIONS: A cadaveric skills course focused on fundamental maneuvers with objective confirmation of achieving competency is a viable adjunct to clinical operative experience. Video-recorded evaluation, of these fundamental skills improved both resident and attending confidence in trainee operative skill.

Visceral Kaposi Sarcoma Presenting as Small Bowel Intussusception: A Rare Presentation and Call to Action.

Surgical emergencies related to visceral involvement of Kaposi sarcoma (KS) are rare complications of the disease. In this report, we describe a case of visceral KS causing small bowel intussusception in a young, previously undiagnosed human immunodeficiency virus (HIV)-positive patient. Southern surgeons should be particularly attentive to HIV/AIDS-related disease as a cause of surgical pathology, particularly in the southeast, and can play a significant advocacy role for improved access to HIV/AIDS diagnostic and treatment services.