Gender differences in autonomy and performance assessments in a national cohort of vascular surgery trainees.

OBJECTIVE: Gender disparities in surgical training and assessment are described in the general surgery literature. Assessment disparities have not been explored in vascular surgery. We sought to investigate gender disparities in operative assessment in a national cohort of vascular surgery integrated residents (VIRs) and fellows (VSFs). METHODS: Operative performance and autonomy ratings from the Society for Improving Medical Professional Learning (SIMPL) application database were collected for all vascular surgery participating institutions from 2018 to 2023. Logistic generalized linear mixed models were conducted to examine the association of faculty and trainee gender on faculty and self-assessment of autonomy and performance. Data were adjusted for post-graduate year and case complexity. Random effects were included to account for clustering effects due to participant, program, and procedure. RESULTS: One hundred three trainees (n = 63 VIRs; n = 40 VSFs; 63.1% men) and 99 faculty (73.7% men) from 17 institutions (n = 12 VIR and n = 13 VSF programs) contributed 4951 total assessments (44.4% by faculty, 55.6% by trainees) across 235 unique procedures. Faculty and trainee gender were not associated with faculty ratings of performance (faculty gender: odds ratio [OR], 0.78; 95% confidence interval [CI], 0.27-2.29; trainee gender: OR, 1.80; 95% CI, 0.76-0.43) or autonomy (faculty gender: OR, 0.99; 95% CI, 0.41-2.39; trainee gender: OR, 1.23; 95% CI, 0.62-2.45) of trainees. All trainees self-assessed at lower performance and autonomy ratings as compared with faculty assessments. However, women trainees rated themselves significantly lower than men for both autonomy (OR, 0.57; 95% CI, 0.43-0.74) and performance (OR, 0.40; 95% CI, 0.30-0.54). CONCLUSIONS: Although gender was not associated with differences in faculty assessment of performance or autonomy among vascular surgery trainees, women trainees perceive themselves as performing with lower competency and less autonomy than their male colleagues. These findings suggest utility for exploring gender differences in real-time feedback delivered to and received by trainees and targeted interventions to align trainee self-perception with actual operative performance and autonomy to optimize surgical skill acquisition.

Re-Aiming Equity Evaluation in Clinical Decision Support: A Scoping Review of Equity Assessments in Surgical Decision Support Systems.

OBJECTIVE: We critically evaluated the surgical literature to explore the prevalence and describe how equity assessments occur when using clinical decision support systems. BACKGROUND: Clinical decision support (CDS) systems are increasingly used to facilitate surgical care delivery. Despite formal recommendations to do so, equity evaluations are not routinely performed on CDS systems and underrepresented populations are at risk of harm and further health disparities. We explored surgical literature to determine frequency and rigor of CDS equity assessments and offer recommendations to improve CDS equity by appending existing frameworks. METHODS: We performed a scoping review up to Augus 25, 2021 using PubMed and Google Scholar for the following search terms: clinical decision support, implementation, RE-AIM, Proctor, Proctor's framework, equity, trauma, surgery, surgical. We identified 1415 citations and 229 abstracts met criteria for review. A total of 84 underwent full review after 145 were excluded if they did not assess outcomes of an electronic CDS tool or have a surgical use case. RESULTS: Only 6% (5/84) of surgical CDS systems reported equity analyses, suggesting that current methods for optimizing equity in surgical CDS are inadequate. We propose revising the RE-AIM framework to include an Equity element (RE 2 -AIM) specifying that CDS foundational analyses and algorithms are performed or trained on balanced datasets with sociodemographic characteristics that accurately represent the CDS target population and are assessed by sensitivity analyses focused on vulnerable subpopulations. CONCLUSION: Current surgical CDS literature reports little with respect to equity. Revising the RE-AIM framework to include an Equity element (RE 2 -AIM) promotes the development and implementation of CDS systems that, at minimum, do not worsen healthcare disparities and possibly improve their generalizability.

Postoperative Overtriage to an Intensive Care Unit Is Associated With Low Value of Care.

OBJECTIVE: We test the hypothesis that for low-acuity surgical patients, postoperative intensive care unit (ICU) admission is associated with lower value of care compared with ward admission. BACKGROUND: Overtriaging low-acuity patients to ICU consumes valuable resources and may not confer better patient outcomes. Associations among postoperative overtriage, patient outcomes, costs, and value of care have not been previously reported. METHODS: In this longitudinal cohort study, postoperative ICU admissions were classified as overtriaged or appropriately triaged according to machine learning-based patient acuity assessments and requirements for immediate postoperative mechanical ventilation or vasopressor support. The nearest neighbors algorithm identified risk-matched control ward admissions. The primary outcome was value of care, calculated as inverse observed-to-expected mortality ratios divided by total costs. RESULTS: Acuity assessments had an area under the receiver operating characteristic curve of 0.92 in generating predictions for triage classifications. Of 8592 postoperative ICU admissions, 423 (4.9%) were overtriaged. These were matched with 2155 control ward admissions with similar comorbidities, incidence of emergent surgery, immediate postoperative vital signs, and do not resuscitate order placement and rescindment patterns. Compared with controls, overtraiged admissions did not have a lower incidence of any measured complications. Total costs for admission were $16.4K for overtriage and $15.9K for controls ( P =0.03). Value of care was lower for overtriaged admissions [2.9 (2.0-4.0)] compared with controls [24.2 (14.1-34.5), P <0.001]. CONCLUSIONS: Low-acuity postoperative patients who were overtriaged to ICUs had increased total costs, no improvements in outcomes, and received low-value care.

Artificial Intelligence-enabled Decision Support in Surgery: State-of-the-art and Future Directions.

OBJECTIVE: To summarize state-of-the-art artificial intelligence-enabled decision support in surgery and to quantify deficiencies in scientific rigor and reporting. BACKGROUND: To positively affect surgical care, decision-support models must exceed current reporting guideline requirements by performing external and real-time validation, enrolling adequate sample sizes, reporting model precision, assessing performance across vulnerable populations, and achieving clinical implementation; the degree to which published models meet these criteria is unknown. METHODS: Embase, PubMed, and MEDLINE databases were searched from their inception to September 21, 2022 for articles describing artificial intelligence-enabled decision support in surgery that uses preoperative or intraoperative data elements to predict complications within 90 days of surgery. Scientific rigor and reporting criteria were assessed and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines. RESULTS: Sample size ranged from 163-2,882,526, with 8/36 articles (22.2%) featuring sample sizes of less than 2000; 7 of these 8 articles (87.5%) had below-average (<0.83) area under the receiver operating characteristic or accuracy. Overall, 29 articles (80.6%) performed internal validation only, 5 (13.8%) performed external validation, and 2 (5.6%) performed real-time validation. Twenty-three articles (63.9%) reported precision. No articles reported performance across sociodemographic categories. Thirteen articles (36.1%) presented a framework that could be used for clinical implementation; none assessed clinical implementation efficacy. CONCLUSIONS: Artificial intelligence-enabled decision support in surgery is limited by reliance on internal validation, small sample sizes that risk overfitting and sacrifice predictive performance, and failure to report confidence intervals, precision, equity analyses, and clinical implementation. Researchers should strive to improve scientific quality.

Defining critical illness using immunological endotypes in patients with and without sepsis: a cohort study.

BACKGROUND: Sepsis is a heterogenous syndrome with limited therapeutic options. Identifying immunological endotypes through gene expression patterns in septic patients may lead to targeted interventions. We investigated whether patients admitted to a surgical intensive care unit (ICU) with sepsis and with high risk of mortality express similar endotypes to non-septic, but still critically ill patients using two multiplex transcriptomic metrics obtained both on admission to a surgical ICU and at set intervals. METHODS: We analyzed transcriptomic data from 522 patients in two single-site, prospective, observational cohorts admitted to surgical ICUs over a 5-year period ending in July 2020. Using an FDA-cleared analytical platform (nCounter FLEX®, NanoString, Inc.), we assessed a previously validated 29-messenger RNA transcriptomic classifier for likelihood of 30-day mortality (IMX-SEV-3) and a 33-messenger RNA transcriptomic endotype classifier. Clinical outcomes included all-cause mortality, development of chronic critical illness, and secondary infections. Univariate and multivariate analyses were performed to assess for true effect and confounding. RESULTS: Sepsis was associated with a significantly higher predicted and actual hospital mortality. At enrollment, the predominant endotype for both septic and non-septic patients was adaptive, though with significantly different distributions. Inflammopathic and coagulopathic septic patients, as well as inflammopathic non-septic patients, showed significantly higher frequencies of secondary infections compared to those with adaptive endotypes (p < 0.01). Endotypes changed during ICU hospitalization in 57.5% of patients. Patients who remained adaptive had overall better prognosis, while those who remained inflammopathic or coagulopathic had worse overall outcomes. For severity metrics, patients admitted with sepsis and a high predicted likelihood of mortality showed an inflammopathic (49.6%) endotype and had higher rates of cumulative adverse outcomes (67.4%). Patients at low mortality risk, whether septic or non-septic, almost uniformly presented with an adaptive endotype (100% and 93.4%, respectively). CONCLUSION: Critically ill surgical patients express different and evolving immunological endotypes depending upon both their sepsis status and severity of their clinical course. Future studies will elucidate whether endotyping critically ill, septic patients can identify individuals for targeted therapeutic interventions to improve patient management and outcomes.

Continuous ECG monitoring should be the heart of bedside AI-based predictive analytics monitoring for early detection of clinical deterioration.

The idea that we can detect subacute potentially catastrophic illness earlier by using statistical models trained on clinical data is now well-established. We review evidence that supports the role of continuous cardiorespiratory monitoring in these predictive analytics monitoring tools. In particular, we review how continuous ECG monitoring reflects the patient and not the clinician, is less likely to be biased, is unaffected by changes in practice patterns, captures signatures of illnesses that are interpretable by clinicians, and is an underappreciated and underutilized source of detailed information for new mathematical methods to reveal.

Aligning Patient Acuity With Resource Intensity After Major Surgery: A Scoping Review.

OBJECTIVE: Develop unifying definitions and paradigms for data-driven methods to augment postoperative resource intensity decisions. SUMMARY BACKGROUND DATA: Postoperative level-of-care assignments and frequency of vital sign and laboratory measurements (ie, resource intensity) should align with patient acuity. Effective, data-driven decision-support platforms could improve value of care for millions of patients annually, but their development is hindered by the lack of salient definitions and paradigms. METHODS: Embase, PubMed, and Web of Science were searched for articles describing patient acuity and resource intensity after inpatient surgery. Study quality was assessed using validated tools. Thirty-five studies were included and assimilated according to PRISMA guidelines. RESULTS: Perioperative patient acuity is accurately represented by combinations of demographic, physiologic, and hospital-system variables as input features in models that capture complex, non-linear relationships. Intraoperative physiologic data enriche these representations. Triaging high-acuity patients to low-intensity care is associated with increased risk for mortality; triaging low-acuity patients to intensive care units (ICUs) has low value and imparts harm when other, valid requests for ICU admission are denied due to resource limitations, increasing their risk for unrecognized decompensation and failure-to-rescue. Providing high-intensity care for low-acuity patients may also confer harm through unnecessary testing and subsequent treatment of incidental findings, but there is insufficient evidence to evaluate this hypothesis. Compared with data-driven models, clinicians exhibit volatile performance in predicting complications and making postoperative resource intensity decisions. CONCLUSION: To optimize value, postoperative resource intensity decisions should align with precise, data-driven patient acuity assessments augmented by models that accurately represent complex, non-linear relationships among risk factors.

Clinical Trajectories of Acute Kidney Injury in Surgical Sepsis: A Prospective Observational Study.

OBJECTIVE: To characterize endothelial function, inflammation, and immunosuppression in surgical patients with distinct clinical trajectories of AKI and to determine the impact of persistent kidney injury and renal non-recovery on clinical outcomes, resource utilization, and long-term disability and survival. SUMMARY OF BACKGROUND DATA: AKI is associated with increased healthcare costs and mortality. Trajectories that account for duration and recovery of AKI have not been described for sepsis patients, who are uniquely vulnerable to renal dysfunction. METHODS: This prospective observational study included 239 sepsis patients admitted and enrolled between January 2015 and July 2017. Kidney Disease: Improving Global Outcomes (KDIGO) and Acute Disease Quality Initiative (ADQI) criteria were used to classify subjects as having no AKI, rapidly reversed AKI, persistent AKI with renal recovery, or persistent AKI without renal recovery. Serial biomarker profiles, clinical outcomes, resource utilization, and long-term physical performance status and survival were compared among AKI trajectories. RESULTS: Sixty-two percent of the study population developed AKI. Only one-third of AKI episodes rapidly reversed within 48 hours; the remaining had persistent AKI, among which 57% did not have renal recovery by discharge. One-year survival and proportion of subjects fully active 1 year after sepsis was lowest among patients with persistent AKI compared with other groups. Long-term mortality hazard rates were 5-fold higher for persistent AKI without renal recovery compared with no AKI. CONCLUSIONS: Among critically ill surgical sepsis patients, persistent AKI and the absence of renal recovery are associated with distinct early and sustained immunologic and endothelial biomarker signatures and decreased long-term physical function and survival.

Nutritional impact of omega 3 fatty acids and metabolites in acute and chronic critical illness.

PURPOSE OF REVIEW: Lipids have been utilized historically as a calorie dense means to ensure delivery of essential fatty acids (FA). Since the development of mixed lipid emulsion and investigation of immunomodulatory formulas, there has been an awakening that not all lipids are created equal. This narrative review focuses on contemporary evidence in the utilization of lipids (namely omega 3 fatty acids) in both acute and chronic critical illness. RECENT FINDINGS: Though randomized control trials and meta-analyses provide little guidance regarding clinical practice for patients suffering from chronic critical illness, available literature suggests the potential to use lipid formulas to decrease the inflammatory cycle that drives catabolism. Additionally, this review will address the expanding evidence that specialized pro-resolving mediators (SPMs) may be the future of immunomodulating inflammation in acute and chronic critical illness and the persistent inflammation, immunosuppression, and catabolic syndrome (PICS). SUMMARY: Although societal guidelines, expert consensus, and literature support the use of omega 3 fatty acids in the acute critically ill population, more research is needed regarding omega 3 fatty acids for chronic critical illness and PICS.

Early Biomarker Signatures in Surgical Sepsis.

INTRODUCTION: Sepsis has complex, time-sensitive pathophysiology and important phenotypic subgroups. The objective of this study was to use machine learning analyses of blood and urine biomarker profiles to elucidate the pathophysiologic signatures of subgroups of surgical sepsis patients. METHODS: This prospective cohort study included 243 surgical sepsis patients admitted to a quaternary care center between January 2015 and June 2017. We applied hierarchical clustering to clinical variables and 42 blood and urine biomarkers to identify phenotypic subgroups in a development cohort. Clinical characteristics and short-term and long-term outcomes were compared between clusters. A naïve Bayes classifier predicted cluster labels in a validation cohort. RESULTS: The development cohort contained one cluster characterized by early organ dysfunction (cluster I, n = 18) and one cluster characterized by recovery (cluster II, n = 139). Cluster I was associated with higher Acute Physiologic Assessment and Chronic Health Evaluation II (30 versus 16, P < 0.001) and SOFA scores (13 versus 5, P < 0.001), greater prevalence of chronic cardiovascular and renal disease (P < 0.001) and septic shock (78% versus 17%, P < 0.001). Cluster I had higher mortality within 14 d of sepsis onset (11% versus 1.5%, P = 0.001) and within 1 y (44% versus 20%, P = 0.032), and higher incidence of chronic critical illness (61% versus 30%, P = 0.001). The Bayes classifier achieved 95% accuracy and identified two clusters that were similar to development cohort clusters. CONCLUSIONS: Machine learning analyses of clinical and biomarker variables identified an early organ dysfunction sepsis phenotype characterized by inflammation, renal dysfunction, endotheliopathy, and immunosuppression, as well as poor short-term and long-term clinical outcomes.

Safety of laparoscopic inguinal hernia repair in the setting of antithrombotic therapy.

INTRODUCTION: There are a paucity of data regarding the safety of laparoscopic inguinal hernia repair in patients on antiplatelet and anticoagulant therapy (APT/ACT). We aim to compare the postoperative outcomes of laparoscopic (LIHR) vs. open repair of inguinal hernias (OIHR) in patients on APT/ACT. METHOD: We conducted a retrospective cohort study using the Vizient Clinical DataBase. We included adults receiving APT/ACT who underwent outpatient, elective, and primary inguinal hernia repair between 2017 and 2019. Subgroup analysis was performed on patients receiving aspirin, non-aspirin antiplatelet, and anticoagulant therapy. Mixed-effects logistic regression was used to assess both the effect of APT/ACT on the probability of receiving LIHR vs OIHR and their respective outcomes. RESULT: A total of 142,052 repairs were included, of which 21,441 (15%) were performed on patients receiving APT/ACT. Mean age was 69 years (± 10.5) and 93% were male. 19% of hernias were bilateral. 40% of operations were performed at teaching hospitals. On multivariable analysis, patients on non-aspirin antiplatelet or anticoagulant therapy were more likely to receive an open procedure (Odds Ratio (OR) = 1.2; 95% Confidence Intervals (CI) [1.1, 1.4] and OR = 1.4; CI [1.3, 1.5], respectively). LIHR was associated with a lower rate of length of stay > 1 day (OR = 0.65; CI [0.5, 0.9]). Rates of 30-day postoperative hematoma, transfusions, stroke, myocardial infarction, deep venous thrombosis, pulmonary embolism, readmission, and emergency department visits were similar between the two operative approaches. CONCLUSION: Patients on APT/ACT represent a substantial proportion of those undergoing inguinal hernia repair. Non-aspirin antiplatelet or anticoagulant therapy are independent predictors of choosing an open repair. Laparoscopic repair appears to be safe in patients receiving APT/ACT under current perioperative management patterns.

Metabolomic Profiling for Diagnosis and Prognostication in Surgery: A Scoping Review.

OBJECTIVE: This review assimilates and critically evaluates available literature regarding the use of metabolomic profiling in surgical decision-making. BACKGROUND: Metabolomic profiling is performed by nuclear magnetic resonance spectroscopy or mass spectrometry of biofluids and tissues to quantify biomarkers (ie, sugars, amino acids, and lipids), producing diagnostic and prognostic information that has been applied among patients with cardiovascular disease, inflammatory bowel disease, cancer, and solid organ transplants. METHODS: PubMed was searched from 1995 to 2019 to identify studies investigating metabolomic profiling of surgical patients. Articles were included and assimilated into relevant categories per PRISMA-ScR guidelines. Results were summarized with descriptive analytical methods. RESULTS: Forty-seven studies were included, most of which were retrospective studies with small sample sizes using various combinations of analytic techniques and types of biofluids and tissues. Results suggest that metabolomic profiling has the potential to effectively screen for surgical diseases, suggest diagnoses, and predict outcomes such as postoperative complications and disease recurrence. Major barriers to clinical adoption include a lack of high-level evidence from prospective studies, heterogeneity in study design regarding tissue and biofluid procurement and analytical methods, and the absence of large, multicenter metabolome databases to facilitate systematic investigation of the efficacy, reproducibility, and generalizability of metabolomic profiling diagnoses and prognoses. CONCLUSIONS: Metabolomic profiling research would benefit from standardization of study design and analytic approaches. As technologies improve and knowledge garnered from research accumulates, metabolomic profiling has the potential to provide personalized diagnostic and prognostic information to support surgical decision-making from preoperative to postdischarge phases of care.

Optimal Antibiotic Duration for Bloodstream Infections Secondary to Intraabdominal Infection.

BACKGROUND: Bloodstream infections (BSIs) secondary to intraabdominal infections (IAIs) are common in the intensive care unit (ICU). The Surgical Infection Society guidelines recommend treatment duration after achieving source control in patients with secondary bacteremia; however, literature supporting this recommendation is limited. The purpose of this study was to compare outcomes in patients who received shorter versus extended duration of antibiotics for bacteremia secondary to IAI. MATERIALS AND METHODS: A retrospective cohort analysis was conducted in adult surgical ICU patients (n = 42) with BSIs and source control procedure(s) for IAI. The primary outcome was recurrent IAI. Secondary outcomes included surgical site infections (SSIs), Clostridium difficile infections (CDIs), secondary fungal infections, and in-hospital mortality. RESULTS: Forty-two patients met inclusion criteria and were divided into groups according to antimicrobial duration; 12 patients received <7 d, and 30 patients received >7 d of antibiotics. There were no differences in baseline characteristics between the two cohorts except for the presence of sepsis [4/12 (33.3%) versus 27/30 (90.0%); P = 0.001]. Thirty-one percent (13/42) of all organisms isolated from blood cultures were gram-negative bacteria, 12/42 (28.6%) were MDROs, and 2/42 (4.8%) patients experienced a culture mismatch in which cultured bacteria were not susceptible to empiric antibiotic therapy. Rates of recurrent IAI were similar between the two cohorts [1/12 (8.3%) versus 4/30 (13.3%), P = 0.554]. CONCLUSIONS: Among surgical ICU patients with BSI secondary to IAI, cessation of antibiotic therapy within 7 d of source control was not associated with an increased incidence of recurrent IAI.

The Morbidity and Mortality Assessment Tool (MMAT): Design and Proof of Concept.

BACKGROUND: Morbidity and Mortality Conference (MMC) is a traditional quality assessment tool among surgical residencies to evaluate complications within their care and to integrate and educate best practices. Unfortunately, it is difficult to validate and measure loop closure as a result of MMC. This may result in repeating past mistakes or worse, becoming a meeting of anecdotal experiences. The goal of this study is to present results from a morbidity and mortality assessment tool (MMAT) which provides a means of measuring and tracking factors related to those surgical complications discussed within the MMC. MATERIALS AND METHODS: Three years of MMC presentations were organized into a database and further divided and organized into variables which included case by the class year of the presenting resident, service line, month the case was presented, and potential contributing factors. Contributing factors considered for each case were categorized as: Systems-Based, Direct Patient Care, and Interpersonal Communication. Contributing factors were assigned to cases by a review committee consisting of residents and faculty members. RESULTS: The lack of knowledge, technical error, lack of experience, lack of supervision, failure to communicate with team members were present in greater than 10% of the presented cases. There was a "July Effect" in the Trauma service, where there was a statistically significant difference between the percentage of cases presented that involved Failure to Communicate errors when compared with the remainder of the year. CONCLUSIONS: MMAT allows longitudinal collection of data from each MMC to recognize patterns that facilitate improvements in systems of care and institutional memory.

Mediators of Prolonged Hematopoietic Progenitor Cell Mobilization After Severe Trauma.

BACKGROUND: This study investigated the molecular mediators of prolonged hematopoietic progenitor cell mobilization a trauma and chronic stress and the role of propranolol in modifying this response. METHODS: Sprague-Dawley rats were randomized to lung contusion (LC), LC plus hemorrhagic shock (LCHS), or LCHS with daily restraint stress (LCHS/CS). Propranolol was administered daily. Bone marrow (BM) and lung expression of high mobility group box 1 (HMGB1), granulocyte colony-stimulating factor (G-CSF), neutrophil elastase, stromal cell-derived factor 1 (SDF-1)/CXR4, and vascular cell adhesion protein 1 (VCAM-1)/very late antigen-4 were measured by real-time polymerase chain reaction. RESULTS: Bone marrow HMGB1, G-CSF, and neutrophil elastase expression were significantly elevated two- to four-fold after LCHS/CS, and all were decreased with the use of propranolol. SDF-1 and VCAM-1 were both significantly decreased after LCHS/CS. CONCLUSIONS: The increased expression of HMGB1 and G-CSF and decreased expression of BM anchoring molecules, SDF-1 and VCAM-1, after LCHS/CS, likely mediates prolonged hematopoietic progenitor cell mobilization. Propranolol's ability to reduce HMGB1, G-CSF, and neutrophil elastase expression suggests that the mobilization of hematopoietic progenitor cells was driven by persistent hypercatecholaminemia.

Advances in artificial intelligence and deep learning systems in ICU-related acute kidney injury.

PURPOSE OF REVIEW: Acute kidney injury (AKI) affects nearly 60% of all patients admitted to ICUs. Large volumes of clinical, monitoring and laboratory data produced in ICUs allow the application of artificial intelligence analytics. The purpose of this article is to assimilate and critically evaluate recently published literature regarding artificial intelligence applications for predicting, diagnosing and subphenotyping AKI among critically ill patients. RECENT FINDINGS: Among recent studies regarding artificial intelligence implementations for predicting, diagnosing and subphenotyping AKI among critically ill patients, there are many promising models, but few had external validation, clinical interpretability and high predictive performance. Deep learning techniques leveraging multimodal clinical data show great potential to provide continuous, accurate, early predictions of AKI risk, which could be implemented clinically to optimize preventive and early therapeutic management strategies. SUMMARY: Use of consensus criteria, standard definitions and common data models could facilitate access to machine learning-ready data sets for external validation. The lack of interpretability, explainability, fairness and transparency of artificial intelligence models hinder their entrustment and clinical implementation; compliance with standardized reporting guidelines can mitigate these challenges.

Objective predictors of intern performance.

BACKGROUND: Residency programs select medical students for interviews and employment using metrics such as the United States Medical Licensing Examination (USMLE) scores, grade-point average (GPA), and class rank/quartile. It is unclear whether these metrics predict performance as an intern. This study tested the hypothesis that performance on these metrics would predict intern performance. METHODS: This single institution, retrospective cohort analysis included 244 graduates from four classes (2015-2018) who completed an Accreditation Council for Graduate Medical Education (ACGME) certified internship and were evaluated by program directors (PDs) at the end of the year. PDs provided a global assessment rating and ratings addressing ACGME competencies (response rate = 47%) with five response options: excellent = 5, very good = 4, acceptable = 3, marginal = 2, unacceptable = 1. PDs also classified interns as outstanding = 4, above average = 3, average = 2, and below average = 1 relative to other interns from the same residency program. Mean USMLE scores (Step 1 and Step 2CK), third-year GPA, class rank, and core competency ratings were compared using Welch's ANOVA and follow-up pairwise t-tests. RESULTS: Better performance on PD evaluations at the end of intern year was associated with higher USMLE Step 1 (p = 0.006), Step 2CK (p = 0.030), medical school GPA (p = 0.020) and class rank (p = 0.016). Interns rated as average had lower USMLE scores, GPA, and class rank than those rated as above average or outstanding; there were no significant differences between above average and outstanding interns. Higher rating in each of the ACGME core competencies was associated with better intern performance (p < 0.01). CONCLUSIONS: Better performance as an intern was associated with higher USMLE scores, medical school GPA and class rank. When USMLE Step 1 reporting changes from numeric scores to pass/fail, residency programs can use other metrics to select medical students for interviews and employment.

Added Value of Intraoperative Data for Predicting Postoperative Complications: The MySurgeryRisk PostOp Extension.

BACKGROUND: Models that predict postoperative complications often ignore important intraoperative events and physiological changes. This study tested the hypothesis that accuracy, discrimination, and precision in predicting postoperative complications would improve when using both preoperative and intraoperative data input data compared with preoperative data alone. METHODS: This retrospective cohort analysis included 43,943 adults undergoing 52,529 inpatient surgeries at a single institution during a 5-y period. Random forest machine learning models in the validated MySurgeryRisk platform made patient-level predictions for seven postoperative complications and mortality occurring during hospital admission using electronic health record data and patient neighborhood characteristics. For each outcome, one model trained with preoperative data alone; one model trained with both preoperative and intraoperative data. Models were compared by accuracy, discrimination (expressed as area under the receiver operating characteristic curve), precision (expressed as area under the precision-recall curve), and reclassification indices. RESULTS: Machine learning models incorporating both preoperative and intraoperative data had greater accuracy, discrimination, and precision than models using preoperative data alone for predicting all seven postoperative complications (intensive care unit length of stay >48 h, mechanical ventilation >48 h, neurologic complications including delirium, cardiovascular complications, acute kidney injury, venous thromboembolism, and wound complications), and in-hospital mortality (accuracy: 88% versus 77%; area under the receiver operating characteristic curve: 0.93 versus 0.87; area under the precision-recall curve: 0.21 versus 0.15). Overall reclassification improvement was 2.4%-10.0% for complications and 11.2% for in-hospital mortality. CONCLUSIONS: Incorporating both preoperative and intraoperative data significantly increased the accuracy, discrimination, and precision of machine learning models predicting postoperative complications and mortality.

Intelligent, Autonomous Machines in Surgery.

Surgeons perform two primary tasks: operating and engaging patients and caregivers in shared decision-making. Human dexterity and decision-making are biologically limited. Intelligent, autonomous machines have the potential to augment or replace surgeons. Rather than regarding this possibility with denial, ire, or indifference, surgeons should understand and steer these technologies. Closer examination of surgical innovations and lessons learned from the automotive industry can inform this process. Innovations in minimally invasive surgery and surgical decision-making follow classic S-shaped curves with three phases: (1) introduction of a new technology, (2) achievement of a performance advantage relative to existing standards, and (3) arrival at a performance plateau, followed by replacement with an innovation featuring greater machine autonomy and less human influence. There is currently no level I evidence demonstrating improved patient outcomes using intelligent, autonomous machines for performing operations or surgical decision-making tasks. History suggests that if such evidence emerges and if the machines are cost effective, then they will augment or replace humans, initially for simple, common, rote tasks under close human supervision and later for complex tasks with minimal human supervision. This process poses ethical challenges in assigning liability for errors, matching decisions to patient values, and displacing human workers, but may allow surgeons to spend less time gathering and analyzing data and more time interacting with patients and tending to urgent, critical-and potentially more valuable-aspects of patient care. Surgeons should steer these technologies toward optimal patient care and net social benefit using the uniquely human traits of creativity, altruism, and moral deliberation.