Artificial Intelligence versus Surgeon Gestalt in Predicting Risk of Emergency General Surgery.

Venous thromboembolism (VTE) is the most preventable cause of morbidity and mortality in US hospitals, and approximately 2.5% of emergency general surgery (EGS) patients will be diagnosed with a VTE event. Emergency general surgery patients are at increased risk of morbidity and mortality because of the nature of acute surgical conditions and the challenges related to prophylaxis. MEDLINE, Embase, and the Cochrane Database of Collected Reviews were searched from January 1, 1990, through December 31, 2015. Nearly all operatively and nonoperatively treated EGS patients have a moderate to high risk of developing a VTE, and individual risk should be assessed at admission. Pharmacologic prophylaxis in the form of unfractionated or low-molecular-weight heparin should be considered unless an absolute contraindication, such as bleeding, exists. Patients should receive the first dose at admission to the hospital, and administration should continue until discharge without missed doses. Certain patient populations, such as those with malignant tumors, may benefit from prolonged VTE prophylaxis after discharge. Mechanical prophylaxis should be considered in all patients, particularly if pharmacologic prophylaxis is contraindicated. Studies that specifically target improved adherence with VTE prophylaxis in EGS patients suggest that efficacy and quality improvement initiatives should be undertaken from a system and institutional perspective. Operatively and nonoperatively treated EGS patients are at a comparatively high risk of VTE. Despite gaps in existing literature with respect to this increasing patient population, successful best practices can be applied. Best practices include assessment of VTE risk, optimal prophylaxis, and physician, nurse, and patient education regarding the use of mechanical and pharmacologic VTE prophylaxis and institutional policies.

Creation and Implementation of an Emergency General Surgery Registry Modeled after the National Trauma Data Bank

Background:A large number of patients live with undiagnosed HIV and/or hepatitis C despite broadened national screening guidelines. European studies, however, suggest many patients falsely believe they have been screened during a prior hospitalization. This study aims to define current perceptions among trauma and emergency general surgery (EGS) patients regarding HIV and hepatitis C screening practices.Methods:Prospective survey administered to adult (>18 years old) acute care surgery service (trauma and EGS) patients at a Level 1 academic trauma center. The survey consisted of 13 multiple choice questions: demographics, whether admission tests included HIV and hepatitis C at index and prior hospital visits and whether receiving no result indicated a negative result, prior primary care screening. Response percentages calculated in standard fashion.Results:One hundred and twenty-five patients were surveyed: 80 trauma and 45 EGS patients. Overall, 32% and 29.6% of patients believed they were screened for HIV and hepatitis C at admission. There was no significant difference in beliefs between trauma and EGS. Sixty-eight percent of patients had a hospital visit within 10 years of these, 49.3% and 44.1% believe they had been screened for HIV and hepatitis C. More EGS patients believed they had a prior screen for both conditions. Among patients who believed they had a prior screen and did not receive any results, 75.9% (HIV) and 80.8% (hepatitis C) believed a lack of results meant they were negative. Only 28.9% and 23.6% of patients had ever been offered outpatient HIV and hepatitis C screening.Conclusions:A large portion of patients believe they received admission or prior hospitalization HIV and/or hepatitis C screening and the majority interpreted a lack of results as a negative diagnosis. Due to these factors, routine screening of trauma/EGS patients should be considered to conform to patient expectations and national guidelines, increase diagnosis and referral for medical management, and decrease disease transmission.

Fragmentation of care (FOC) is defined as readmission to a nonindex hospital and can be associated with worse outcomes among emergency general surgery (EGS) patients. This study assessed hospital characteristics associated with patterns of FOC in EGS patients and the effect of interfacility transfer after FOC on mortality. Using the Nationwide Readmissions Database 2019, we included patients 18 years or older who were admitted to an index hospital with an EGS diagnosis managed operatively and had an emergency readmission within 90 days. Patients were classified into 4 patterns of fragmentation: no FOC, FOC, defragmented care (readmitted to a nonindex hospital, then transferred back), and hyperfragmented care (readmitted to a nonindex hospital, then transferred to another nonindex hospital). Patient severity of illness was defined using severity of illness categories of All Patient-Refined Diagnosis Related Groups (APR-DRG). Multivariable logistic regression analyses were used to identify hospital characteristics associated with FOC and the effect of FOC on 90-day mortality. Of 26,100 patients included, 82.5% had no FOC, 16.4% had FOC, 0.3% had defragmented care, and 0.9% had hyperfragmented care. Index hospital characteristics associated with FOC were small size (by number of beds), private investor or government owned, and low EGS volumes. Controlling for hospital and patient characteristics, FOC (odds ratio 1.22; 95% CI 1.01 to 1.47; p = 0.045) was significantly associated with increased 90-day mortality, driven by those with major or extreme severity of illness of APR-DRGs (odds ratio 1.36; 95% CI 1.13 to 1.64; p = 0.001). FOC is associated with worse outcomes in EGS patients. This study highlights the impact of FOC and the hospital characteristics associated with FOC, highlighting the need for further investigation into EGS systems of care.

Electrocardiogram (ECG)-enabled stethoscope (ECG-Scope) acquires a single-lead ECGs during cardiac auscultation and may facilitate real-time screening for pathologies not routinely identified by cardiac auscultation alone. We previously demonstrated an artificial intelligence (AI) algorithm can identify left ventricular dysfunction (LVSD) [defined as ejection fraction (EF) ≤ 40%] with an area under the curve (AUC) of 0.91 using a 12-lead ECG. One hundred patients referred for clinically indicated echocardiography were prospectively recruited. ECG-Scope recordings with the patient supine and sitting were obtained in multiple electrode locations at the time of the echocardiogram. The AI algorithm for the detection of LVSD was retrained using single leads from ECG-12 and validated against ECG-Scope to determine accuracy for low EF detection (≤35%, <40%, or <50%). We evaluated the algorithm with respect to body position and lead location. Amongst 100 patients (aged 61.3 ± 13.8; 61% male, BMI: 30.0 ± 5.4), eight had EF≤40%, and six had EF 40-50%. The best single recording position was V2 with the patient supine [AUC: 0.88 (CI: 0.80-0.97) for EF≤35%, 0.85 (CI: 0.75-0.95) for EF≤40%, and 0.81 (CI: 0.71-0.90) for EF < 50%]. When using an AI model to select the recording automatically, AUC was 0.91 (CI: 0.84-0.97) for EF≤35%, 0.89 (CI: 0.83-0.96) for EF≤40%, and 0.84 (CI: 0.73-0.94) for EF < 50%. An AI algorithm applied to an ECG-enabled stethoscope recording in standard auscultation positions reliably detected the presence of a low EF in this prospective study of patients referred for echocardiography. The ability to screen patients with a possible low EF during routine physical examination may facilitate rapid detection of LVSD.

Using the Geriatric Emergency Perioperative Risk Index Derived From Artificial Intelligence Algorithms to Predict Outcomes of Geriatric Emergency General Surgery.

Explaining the excess morbidity of emergency general surgery: packed red blood cell and fresh frozen plasma transfusion practices are associated with major complications in nonmassively transfused patients

Aims Determine the desired future operative role of dedicated Emergency General Surgery (EGS) surgeons in the United Kingdom (UK) and the Republic of Ireland (ROI), and how to improve the attractiveness of an EGS post. Methods A live interactive online survey, designed and endorsed by ASGBI, was completed by attendees at two ASGBI conferences in 2023. The study was conducted according to CHERRIES and exempt from REC according to HRA. Consent was obtained from study participants, who were asked using a Likert scale which operations a dedicated EGS surgeon should perform and how an EGS post could be more attractive. Results The survey was completed by 156/371 eligible participants, of which 132 responses were analysed once duplicate responses (n=16), and non-UK/ROI respondents (n=8), were excluded. Participants included consultants (34%), surgical trainees (42%), and other doctors (23%), 61% were male and 39% were female, aged between 25 and 69. Respondents stated the desired future operative role of EGS surgeons included most EGS operations, including for torso trauma. Few believed EGS surgeons should perform emergency colonic resection for cancer (49%) or inflammatory bowel disease (37%), gastric volvulus (34%), bariatric complications (26%), or Boerhaaves (28%). The vast majority agreed that EGS posts could be improved by including elective general and subspecialty operating lists, continuing professional development, minimally-invasive (laparoscopic/robotic) opportunities, rest days post-on-call, family-friendly hours, EGS ambulatory and follow-up clinics, and major trauma responsibility. Conclusions EGS surgeons should operate on the majority of EGS patients. Job plans that include elective operating, professional development, and patient follow-up may improve recruitment and retention.

Data from the trauma patient population suggests handsewn (HS) anastomoses are superior to stapled (ST). A recent retrospective study in emergency general surgery (EGS) patients had similar findings. The aim of the current study was to evaluate HS and ST anastomoses in EGS patients undergoing urgent/emergent operations. The study was sponsored by the American Association for the Surgery of Trauma Multi-Institutional Studies Committee. Patients undergoing urgent/emergent bowel resection for EGS pathology were prospectively enrolled from July 22, 2013 to December 31, 2015. Patients were grouped by HS/ST anastomoses, and variables were collected. The primary outcome was anastomotic failure. Similar to other studies, anastomotic failure was evaluated at the anastomosis level. Multivariable logistic regression was performed controlling for age and risk factors for anastomotic failure. Fifteen institutions enrolled a total of 595 patients with 649 anastomoses (253 HS and 396 ST). Mean age was 61 years, 51% were men, 7% overall mortality. Age and sex were the same between groups. The overall anastomotic failure rate was 12.5%. The HS group had higher lactate, lower albumin, and were more likely to be on vasopressors. Hospital and intensive care unit days, as well as mortality, were greater in the HS group. Anastomotic failure rates and operative time were equivalent for HS and ST. On multivariate regression, the presence of contamination at initial resection (odds ratio, 1.965; 95% confidence interval, 1.183-3.264) and the patient being managed with open abdomen (odds ratio, 2.529; 95% confidence interval, 1.492-4.286) were independently associated with anastomotic failure, while the type of anastomosis was not. EGS patients requiring bowel resection and anastomosis are at high risk for anastomotic failure. The current study illustrates an apparent bias among acute care surgeons to perform HS techniques in higher-risk patients. Despite the individualized application of technique for differing patient populations, the risk of anastomotic failure was equivalent when comparing HS and ST anastomoses. Therapeutic study, level II.

Access to the operating room (OR) is variable among emergency general surgery (EGS) services, with some having dedicated EGS ORs, and others only a shared queue. Currently in Canada, only a limited number of acute care surgery services have dedicated daytime operating room (OR) access; hence, we aimed to describe the burden of after-hours EGS operating in Canada and differences associated with OR access. In this multicentre retrospective cohort study, we used data from a previously conducted study designed to evaluate nonappendiceal, nonbiliary disease across 8 Canadian hospitals. We performed a secondary analysis to describe booking priorities and timing of operative interventions, compare sites with and without access to a dedicated EGS daytime OR, and identify differences in morbidity and mortality based on timing of operative intervention. Among 1244 patients, operations were performed during weekday daytime in 521 cases (41.9%), in the evening in 279 (22.4%), on the weekend in 293 (23.6%) and overnight in 151 (12.1%). Operating room booking priority was more than 2 hours to 8 hours in 657 cases (52.8%), more than 8 hours to 24 hours in 334 (26.9%) and more than 24 hours to 48 hours in 253 (20.3%). Substantial variation in booking priority was observed for the same preoperative diagnoses. Sites with dedicated EGS ORs performed a greater proportion of cases during daytime versus overnight compared to sites without dedicated EGS ORs (198/237 [83.5%] v. 323/435 [74.2%], p = 0.006). No significant differences in outcome were found between cases performed during the daytime, evening and overnight. We found considerable variation in OR booking priority within the same preoperative diagnoses among EGS patients in Canada. Sites with dedicated EGS ORs performed more cases during weekday daytime compared to sites without dedicated EGS ORs; however, this study showed no evidence of compromised outcomes based on OR timing.

Given the high mortality and morbidity of emergency general surgery (EGS), designing and implementing effective quality assessment tools is imperative. Currently accepted EGS risk scores are limited by the need for manual extraction, which is time-intensive and costly. We developed an automated institutional electronic health record (EHR)-linked EGS registry that calculates a modified Emergency Surgery Score (mESS) and a modified Predictive OpTimal Trees in Emergency Surgery Risk (POTTER) score and demonstrated their use in benchmarking outcomes. The EHR-linked EGS registry was queried for patients undergoing emergent laparotomies from 2018 to 2023. Data captured included demographics, admission and discharge data, diagnoses, procedures, vitals, and laboratories. The mESS and modified POTTER (mPOTTER) were calculated based off previously defined variables, with estimation of subjective variables using diagnosis codes and other abstracted treatment variables. This was validated against ESS and the POTTER risk calculators by chart review. Observed versus expected (O:E) 30-day mortality and complication ratios were generated. The EGS registry captured 177 emergent laparotomies. There were 32 deaths (18%) and 79 complications (45%) within 30 days of surgery. For mortality, the mean difference between the mESS and ESS risk predictions for mortality was 3% (SD, 10%) with 86% of mESS predictions within 10% of ESS. The mean difference between the mPOTTER and POTTER was -2% (SD, 11%) with 76% of mPOTTER predictions within 10% of POTTER. Observed versus expected ratios by mESS and ESS were 1.45 and 1.86, respectively, and for mPOTTER and POTTER, they were 1.45 and 1.30, respectively. There was similarly good agreement between automated and manual risk scores in predicting complications. Our study highlights the effective implementation of an institutional EHR-linked EGS registry equipped to generate automated quality metrics. This demonstrates potential in enhancing the standardization and assessment of EGS care while mitigating the need for extensive human resources investment. Prognostic and Epidemiologic Study; Level IV.

To compare predictive accuracy of frailty instruments operationalizable in electronic data for prognosticating outcomes among older adults undergoing emergency general surgery (EGS). Older patients undergoing EGS are at higher risk of perioperative morbidity and mortality. Preoperative frailty is a common and strong perioperative risk factor in this population. Despite this, existing barriers preclude routine preoperative frailty assessment. We conducted a retrospective cohort study of adults above 65 undergoing EGS from 2012 to 2018 using Institute for Clinical Evaluative Sciences (ICES) provincial healthcare data in Ontario, Canada. We compared 4 frailty instruments: Frailty Index (FI), Hospital Frailty Risk Score (HFRS), Risk Analysis Index-Administrative (RAI), ACG Frailty-defining diagnoses indicator (ACG). We compared predictive accuracy beyond baseline risk models (age, sex, American Society of Anesthesiologists' score, procedural risk). Predictive performance was measured using discrimination, calibration, explained variance, net reclassification index and Brier score (binary outcomes); using explained variance, root mean squared error and mean absolute prediction error (continuous outcomes). Primary outcome was 30-day mortality. Secondary outcomes were 365-day mortality, nonhome discharge, days alive at home, length of stay, and 30-day and 365-day health systems cost. A total of 121,095 EGS patients met inclusion criteria. Of these, 11,422 (9.4%) experienced death 30 days postoperatively. Addition of FI, HFRS, and RAI to the baseline model led to improved discrimination, net reclassification index, and R2 ; RAI demonstrated the largest improvements. Adding 4 frailty instruments to typically assessed preoperative risk factors demonstrated strong predictive performance in accurately prognosticating perioperative outcomes. These findings can be considered in developing automated risk stratification systems among older EGS patients.

Prostate Cancer Risk Prediction and Online Calculation Based on Machine Learning Algorithm

Artificial intelligence (AI) has demonstrated high accuracy in detecting lymph node (LN) metastases in treatment-naïve invasive breast cancer. However, its performance post-neoadjuvant chemotherapy (NACT) remains underexplored, where therapy-related morphologic alterations complicate assessment. To evaluate an AI algorithm (aetherAI) for detecting LN metastases in post-NACT invasive breast cancer. This retrospective study included 72 patients (58 [80.5%] post-NACT; 14 [19.5%] treatment-naïve), yielding 526 whole slide images and 1290 LNs. AI-generated LN contours were reviewed and modified, if needed. LN status was independently assessed by 4 pathologists and the AI algorithm. Discordant cases were resolved by consensus/immunohistochemistry to define ground truth, against which AI was compared. AI accurately contoured 1089 of 1290 LNs (84.4%). Pathologists showed 94.42% agreement, with 5.58% discordance. Against consensus ground truth, AI achieved 94.06% sensitivity, 90.14% specificity, 91.47% accuracy, and area under the curve (AUC) of 0.92. Performance was superior in treatment-naïve LNs (sensitivity, 98.95%; specificity, 98.08%; AUC, 0.99) compared with post-NACT LNs (sensitivity, 92.71%; specificity, 88.36%; AUC, 0.91) (P < .001). Within the post-NACT group, significantly reduced accuracy was associated with interobserver discordance (P < .001) and smaller metastases (micrometastasis, isolated tumor cells) (P < .001), but not noted in treatment-naïve cases. Most false results were attributable to therapy-induced changes. Unlike most prior studies restricted to treatment-naïve LNs, this post-NACT cohort demonstrates that NACT-related alterations reduce AI sensitivity and accuracy. Our findings suggest that incorporating post-NACT nodes into algorithm training will be essential for reliable clinical translation of AI-assisted LN evaluation.

IntroductionPrior work has linked severe psychiatric illness (SPI) with greater postoperative mortality and complications following several elective operations. However, this relationship has not been defined in the setting of emergency general surgery (EGS).MethodsThe 2016–2021 United States Nationwide Readmissions Database was used to identify all non-elective adult EGS hospitalizations performed within 48 h of admission. SPI was defined using ICD-10 codes for bipolar disorder and schizophrenia. Patients with severe psychiatric illness comprised the SPI cohort (others: Non-SPI). Multivariable linear and logistic regression models were developed to evaluate the independent association of severe psychiatric illness with in-hospital mortality, perioperative complications, and costs.ResultsOf 2,124,284 EGS patients, 52,130 (2.5 %) were categorized as SPI. On adjusted analysis, SPI was associated with greater odds of in-hospital mortality (Adjusted Odds Ratio [AOR] 1.21; 95 % Confidence Interval [CI]:1.11–1.31), hospitalization costs (+$2304; 95 %CI: +1950, +2658), as well as a 3-fold increase in relative risk of non-home discharge (AOR 3.18, 95 % CI: 3.05–3.31).ConclusionsAmong EGS patients, severe psychiatric illness was linked with inferior clinical and financial outcomes. Improved psychiatric screening and care may allow for early intervention and targeted postoperative care, potentially mitigating complications and costs for these vulnerable patients.