Psychological Distress After Inpatient Noncardiac Surgery: A Secondary Analysis of the Measurement of Exercise Tolerance Before Surgery Prospective Cohort Study.

OBJECTIVE: To describe the incidence and natural progression of psychological distress after major surgery. BACKGROUND: The recovery process after surgery imposes physical and mental burdens that put patients at risk of psychological distress. Understanding the natural course of psychological distress after surgery is critical to supporting the timely and tailored management of high-risk individuals. METHODS: We conducted a secondary analysis of the "Measurement of Exercise Tolerance before Surgery" multicentre cohort study (Canada, Australia, New Zealand, and the UK). Measurement of Exercise Tolerance before Surgery recruited adult participants (≥40 years) undergoing elective inpatient noncardiac surgery and followed them for 1 year. The primary outcome was the severity of psychological distress measured using the anxiety-depression item of EQ-5D-3L. We used cumulative link mixed models to characterize the time trajectory of psychological distress among relevant patient subgroups. We also explored potential predictors of severe and/or worsened psychological distress at 1 year using multivariable logistic regression models. RESULTS: Of 1546 participants, moderate-to-severe psychological distress was reported by 32.6% of participants before surgery, 27.3% at 30 days after surgery, and 26.2% at 1 year after surgery. Psychological distress appeared to improve over time among females [odds ratio (OR): 0.80, 95% CI: 0.65-0.95] and patients undergoing orthopedic procedures (OR: 0.73, 95% CI: 0.55-0.91), but not among males (OR: 0.87, 95% CI: 0.87-1.07) or patients undergoing nonorthopedic procedures (OR: 0.95, 95% CI: 0.87-1.04). Among the average middle-aged adult, there were no time-related changes (OR: 0.94, 97% CI: 0.75-1.13), whereas the young-old (OR: 0.89, 95% CI: 0.79-0.99) and middle-old (OR: 0.87, 95% CI: 0.73-1.01) had small improvements. Predictors of severe and/or worsened psychological distress at 1 year were younger age, poor self-reported functional capacity, smoking history, and undergoing open surgery. CONCLUSIONS: One-third of adults experience moderate to severe psychological distress before major elective noncardiac surgery. This distress tends to persist or worsen over time among select patient subgroups.

Predicting Outcomes Following Endovascular Abdominal Aortic Aneurysm Repair Using Machine Learning.

OBJECTIVE: To develop machine learning (ML) models that predict outcomes following endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA). BACKGROUND: EVAR carries non-negligible perioperative risks; however, there are no widely used outcome prediction tools. METHODS: The National Surgical Quality Improvement Program targeted database was used to identify patients who underwent EVAR for infrarenal AAA between 2011 and 2021. Input features included 36 preoperative variables. The primary outcome was 30-day major adverse cardiovascular event (composite of myocardial infarction, stroke, or death). Data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 ML models were trained using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve. Model robustness was evaluated with calibration plot and Brier score. Subgroup analysis was performed to assess model performance based on age, sex, race, ethnicity, and prior AAA repair. RESULTS: Overall, 16,282 patients were included. The primary outcome of 30-day major adverse cardiovascular event occurred in 390 (2.4%) patients. Our best-performing prediction model was XGBoost, achieving an area under the receiver operating characteristic curve (95% CI) of 0.95 (0.94-0.96) compared with logistic regression [0.72 [0.70-0.74)]. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.06. Model performance remained robust on all subgroup analyses. CONCLUSIONS: Our newer ML models accurately predict 30-day outcomes following EVAR using preoperative data and perform better than logistic regression. Our automated algorithms can guide risk mitigation strategies for patients being considered for EVAR.

Ambulatory Cardiology or General Internal Medicine Assessment Prior to Scheduled Major Vascular Surgery is Associated with Improved Outcomes.

OBJECTIVE: To characterize the association between ambulatory cardiology or general internal medicine (GIM) assessment prior to surgery and outcomes following scheduled major vascular surgery. BACKGROUND: Cardiovascular risk assessment and management prior to high-risk surgery remains an evolving area of care. METHODS: This is population-based retrospective cohort study of all adults who underwent scheduled major vascular surgery in Ontario, Canada, April 1, 2004-March 31, 2019. Patients who had an ambulatory cardiology and/or GIM assessment within 6 months prior to surgery were compared to those who did not. The primary outcome was 30-day mortality. Secondary outcomes included: composite of 30-day mortality, myocardial infarction or stroke; 30-day cardiovascular death; 1-year mortality; composite of 1-year mortality, myocardial infarction or stroke; and 1-year cardiovascular death. Cox proportional hazard regression using inverse probability of treatment weighting (IPTW) was used to mitigate confounding by indication. RESULTS: Among 50,228 patients, 20,484 (40.8%) underwent an ambulatory assessment prior to surgery: 11,074 (54.1%) with cardiology, 8,071 (39.4%) with GIM and 1,339 (6.5%) with both. Compared to patients who did not, those who underwent an assessment had a higher Revised Cardiac Risk Index (N with Index over 2= 4,989[24.4%] vs. 4,587[15.4%], P<0.001) and more frequent pre-operative cardiac testing (N=7,772[37.9%] vs. 6,113[20.6%], P<0.001) but, lower 30-day mortality (N=551[2.7%] vs. 970[3.3%], P<0.001). After application of IPTW, cardiology or GIM assessment prior to surgery remained associated with a lower 30-day mortality (weighted Hazard Ratio [95%CI] = 0.73 [0.65-0.82]) and a lower rate of all secondary outcomes. CONCLUSIONS: Major vascular surgery patients assessed by a cardiology or GIM physician prior to surgery have better outcomes than those who are not. Further research is needed to better understand potential mechanisms of benefit.

Using Machine Learning (XGBoost) to Predict Outcomes After Infrainguinal Bypass for Peripheral Artery Disease.

OBJECTIVE: To develop machine learning (ML) algorithms that predict outcomes after infrainguinal bypass. BACKGROUND: Infrainguinal bypass for peripheral artery disease carries significant surgical risks; however, outcome prediction tools remain limited. METHODS: The Vascular Quality Initiative database was used to identify patients who underwent infrainguinal bypass for peripheral artery disease between 2003 and 2023. We identified 97 potential predictor variables from the index hospitalization [68 preoperative (demographic/clinical), 13 intraoperative (procedural), and 16 postoperative (in-hospital course/complications)]. The primary outcome was 1-year major adverse limb event (composite of surgical revision, thrombectomy/thrombolysis, or major amputation) or death. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained 6 ML models using preoperative features. The primary model evaluation metric was the area under the receiver operating characteristic curve (AUROC). The top-performing algorithm was further trained using intraoperative and postoperative features. Model robustness was evaluated using calibration plots and Brier scores. RESULTS: Overall, 59,784 patients underwent infrainguinal bypass, and 15,942 (26.7%) developed 1-year major adverse limb event/death. The best preoperative prediction model was XGBoost, achieving an AUROC (95% CI) of 0.94 (0.93-0.95). In comparison, logistic regression had an AUROC (95% CI) of 0.61 (0.59-0.63). Our XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs (95% CI's) of 0.94 (0.93-0.95) and 0.96 (0.95-0.97), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.08 (preoperative), 0.07 (intraoperative), and 0.05 (postoperative). CONCLUSIONS: ML models can accurately predict outcomes after infrainguinal bypass, outperforming logistic regression.

Using machine learning to predict outcomes following suprainguinal bypass.

OBJECTIVE: Suprainguinal bypass for peripheral artery disease (PAD) carries important surgical risks; however, outcome prediction tools remain limited. We developed machine learning (ML) algorithms that predict outcomes following suprainguinal bypass. METHODS: The Vascular Quality Initiative database was used to identify patients who underwent suprainguinal bypass for PAD between 2003 and 2023. We identified 100 potential predictor variables from the index hospitalization (68 preoperative [demographic/clinical], 13 intraoperative [procedural], and 19 postoperative [in-hospital course/complications]). The primary outcomes were major adverse limb events (MALE; composite of untreated loss of patency, thrombectomy/thrombolysis, surgical revision, or major amputation) or death at 1 year following suprainguinal bypass. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). The best performing algorithm was further trained using intra- and postoperative data. Model robustness was evaluated using calibration plots and Brier scores. Performance was assessed on subgroups based on age, sex, race, ethnicity, rurality, median Area Deprivation Index, symptom status, procedure type, prior intervention for PAD, concurrent interventions, and urgency. RESULTS: Overall, 16,832 patients underwent suprainguinal bypass, and 3136 (18.6%) developed 1-year MALE or death. Patients with 1-year MALE or death were older (mean age, 64.9 vs 63.5 years; P < .001) with more comorbidities, had poorer functional status (65.7% vs 80.9% independent at baseline; P < .001), and were more likely to have chronic limb-threatening ischemia (67.4% vs 47.6%; P < .001) than those without an outcome. Despite being at higher cardiovascular risk, they were less likely to receive acetylsalicylic acid or statins preoperatively and at discharge. Our best performing prediction model at the preoperative stage was XGBoost, achieving an AUROC of 0.92 (95% confidence interval [CI], 0.91-0.93). In comparison, logistic regression had an AUROC of 0.67 (95% CI, 0.65-0.69). Our XGBoost model maintained excellent performance at the intra- and postoperative stages, with AUROCs of 0.93 (95% CI, 0.92-0.94) and 0.98 (95% CI, 0.97-0.99), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.12 (preoperative), 0.11 (intraoperative), and 0.10 (postoperative). Of the top 10 predictors, nine were preoperative features including chronic limb-threatening ischemia, previous procedures, comorbidities, and functional status. Model performance remained robust on all subgroup analyses. CONCLUSIONS: We developed ML models that accurately predict outcomes following suprainguinal bypass, performing better than logistic regression. Our algorithms have potential for important utility in guiding perioperative risk mitigation strategies to prevent adverse outcomes following suprainguinal bypass.

Incidence and risk factors of chronic postoperative pain in same-day surgery: A prospective cohort study.

BACKGROUND: The amount of same-day surgery has increased markedly worldwide in recent decades, but there remains limited evidence on chronic postsurgical pain in this setting. METHODS: We assessed pain 90 days after ambulatory surgery in an international, multicentre prospective cohort study of patients ≥45 years old with comorbidities or ≥65 years old. Pain was assessed using the Brief Pain Inventory. Chronic postsurgical pain was defined as a change ≥1 in self-rated average pain at the surgical site between baseline and 90 days, and moderate to severe chronic postsurgical pain as a score ≥4 in self-rated average pain at the surgical site at 90 days. Risk factors for chronic postsurgical pain were identified using multivariable logistic regression. RESULTS: Between November 2021 and January 2023, a total of 2054 participants were included, and chronic postsurgical pain occurred in 12% of participants, of whom 93.1% had new chronic pain at the surgical site (i.e., participants without pain prior to surgery). Moderate to severe chronic postsurgical pain occurred in 9% of overall participants. Factors associated with chronic postsurgical pain were: active smoking (OR 1.82; 95% CI 1.20 to 2.76), orthopaedic surgery (OR 4.7; 95% CI 2.24 to 9.7), plastic surgery (OR 4.3; 95% CI 1.97 to 9.2), breast surgery (OR 2.74; 95% CI 1.29 to 5.8), vascular surgery (OR 2.71; 95% CI 1.09 to 6.7), and ethnicity (i.e., Hispanic/Latino ethnicity OR 3.41; 95% CI 1.68 to 6.9 and First Nations/Native persons OR 4.0; 95% CI 1.05 to 15.4). CONCLUSIONS: Persistent postsurgical pain after same-day surgery is common, usually moderate to severe in nature, and occurs mostly in patients without chronic pain prior to surgery.

Self-reported fitness as a measure of perioperative cardiovascular risk: tension between subjective and objective assessments persists.

Despite recent high-quality international studies, the optimal sum and sequence of subjective and objective assessments that build the complex picture of fitness for surgery remains to be defined. Physicians' subjective assessment of patient fitness after a typical unstructured interview has poor prognostic accuracy in predicting the risk of major cardiovascular events after noncardiac surgery. How does self-reported fitness assessed by structured questionnaire compare as an indicator of perioperative cardiovascular risk? Here we discuss the latest evidence in this evolving and fundamental aspect of perioperative care.

Postoperative troponin surveillance to detect myocardial infarction: an observational cohort modelling study.

BACKGROUND: Clinical presentation of postoperative myocardial infarction (POMI) is often silent. Several international guidelines recommend routine troponin surveillance in patients at risk. We compared how these different guidelines select patients for surveillance after noncardiac surgery with our established risk stratification model. METHODS: We used outcome data from two prospective studies: Measurement of Exercise Tolerance before Surgery (METS) and Troponin Elevation After Major non-cardiac Surgery (TEAMS). We compared the major American, Canadian, and European guideline recommendations for troponin surveillance with our established risk stratification model. For each guideline and model, we quantified the number of patients requiring monitoring, % POMI detected, sensitivity, specificity, diagnostic odds ratio, and number needed to screen (NNS). RESULTS: METS and TEAMS contributed 2350 patients, of whom 319 (14%) had myocardial injury, 61 (2.5%) developed POMI, and 14 (0.6%) died. Our risk stratification model selected fewer patients for troponin monitoring (20%), compared with the Canadian (78%) and European (79%) guidelines. The sensitivity to detect POMI was highest with the Canadian and European guidelines (0.85; 95% confidence interval [CI] 0.74-0.92). Specificity was highest using the American guidelines (0.91; 95% CI 0.90-0.92). Our risk stratification model had the best diagnostic odds ratio (2.5; 95% CI 1.4-4.2) and a lower NNS (21 vs 35) compared with the guidelines. CONCLUSIONS: Most postoperative myocardial infarctions were detected by the Canadian and European guidelines but at the cost of low specificity and a higher number of patients undergoing screening. Patient selection based on our risk stratification model was optimal.

Postoperative myocardial injury phenotypes and self-reported disability in patients undergoing noncardiac surgery: a multicentre observational study.

BACKGROUND: Postoperative myocardial injury (PMI) comprises a spectrum of mechanisms resulting in troponin release. The impact of different PMI phenotypes on postoperative disability remains unknown. METHODS: This was a multicentre prospective cohort study including patients aged ≥50 yr undergoing elective major noncardiac surgery. Patients were stratified in five groups based on the occurrence of PMI and clinical information on postoperative adverse events: PMI classified as myocardial infarction (MI; according to fourth definition), PMI plus adverse event other than MI, clinically silent PMI (PMI without adverse events), adverse events without PMI, and neither PMI nor an adverse event (reference). The primary endpoint was 6-month self-reported disability (assessed by WHO Disability Assessment Schedule 2.0 [WHODAS]). Disability-free survival was defined as WHODAS ≤16%. RESULTS: We included 888 patients of mean age 69 (range 53-91) yr, of which 356 (40%) were women; 151 (17%) patients experienced PMI, and 625 (71%) experienced 6-month disability-free survival. Patients with PMI, regardless of its phenotype, had higher preoperative disability scores than patients without PMI (difference in WHODAS; ß: 3.3, 95% confidence interval [CI]: 0.5-6.2), but scores remained stable after surgery (ß: 1.2, 95% CI: -3.2-5.6). Before surgery, patients with MI (n=36, 4%) were more disabled compared with patients without PMI and no adverse events (ß: 5.5, 95% CI: 0.3-10.8). At 6 months, patients with MI and patients without PMI but with adverse events worsened in disability score (ß: 11.2, 95% CI: 2.3-20.2; ß: 8.1, 95% CI: 3.0-13.2, respectively). Patients with clinically silent PMI did not change in disability score at 6 months (ß: 1.39, 95% CI: -4.50-7.29, P=0.642). CONCLUSIONS: Although patients with postoperative myocardial injury had higher preoperative self-reported disability, disability scores did not change at 6 months after surgery. However, patients experiencing myocardial infarction worsened in disability score after surgery.

Days alive and out of hospital for adult female and male cardiac surgery patients: a population-based cohort study.

BACKGROUND: Research shows women experience higher mortality than men after cardiac surgery but information on sex-differences during postoperative recovery is limited. Days alive and out of hospital (DAH) combines death, readmission and length of stay, and may better quantify sex-differences during recovery. This main objective is to evaluate (i) how DAH at 30-days varies between sex and surgical procedure, (ii) DAH responsiveness to patient and surgical complexity, and (iii) longer-term prognostic value of DAH. METHODS: We evaluated 111,430 patients (26% female) who underwent one of three types of cardiac surgery (isolated coronary artery bypass [CABG], isolated non-CABG, combination procedures) between 2009 - 2019. Primary outcome was DAH at 30 days (DAH30), secondary outcomes were DAH at 90 days (DAH90) and 180 days (DAH180). Data were stratified by sex and surgical group. Unadjusted and risk-adjusted analyses were conducted to determine the association of DAH with patient-, surgery-, and hospital-level characteristics. Patients were divided into two groups (below and above the 10th percentile) based on the number of days at DAH30. Proportion of patients below the 10th percentile at DAH30 that remained in this group at DAH90 and DAH180 were determined. RESULTS: DAH30 were lower for women compared to men (22 vs. 23 days), and seen across all surgical groups (isolated CABG 23 vs. 24, isolated non-CABG 22 vs. 23, combined surgeries 19 vs. 21 days). Clinical risk factors including multimorbidity, socioeconomic status and surgical complexity were associated with lower DAH30 values, but women showed lower values of DAH30 compared to men for many factors. Among patients in the lowest 10th percentile at DAH30, 80% of both females and males remained in the lowest 10th percentile at 90 days, while 72% of females and 76% males remained in that percentile at 180 days. CONCLUSION: DAH is a responsive outcome to differences in patient and surgical risk factors. Further research is needed to identify new care pathways to reduce disparities in outcomes between male and female patients.

Intraoperative Hypotension in Patients Having Major Noncardiac Surgery Under General Anesthesia: A Systematic Review of Blood Pressure Optimization Strategies.

INTRODUCTION: Intraoperative hypotension is associated with increased risks of postoperative complications. Consequently, a variety of blood pressure optimization strategies have been tested to prevent or promptly treat intraoperative hypotension. We performed a systematic review to summarize randomized controlled trials that evaluated the efficacy of blood pressure optimization interventions in either mitigating exposure to intraoperative hypotension or reducing risks of postoperative complications. METHODS: Medline, Embase, PubMed, and Cochrane Controlled Register of Trials were searched from database inception to August 2, 2023, for randomized controlled trials (without language restriction) that evaluated the impact of any blood pressure optimization intervention on intraoperative hypotension and/or postoperative outcomes. RESULTS: The review included 48 studies (N = 46,377), which evaluated 10 classes of blood pressure optimization interventions. Commonly assessed interventions included hemodynamic protocols using arterial waveform analysis, preoperative withholding of antihypertensive medications, continuous blood pressure monitoring, and adjuvant agents (vasopressors, anticholinergics, anticonvulsants). These same interventions reduced intraoperative exposure to hypotension. Conversely, low blood pressure alarms had an inconsistent impact on exposure to hypotension. Aside from limited evidence that higher prespecified intraoperative blood pressure targets led to a reduced risk of complications, there were few data suggesting that these interventions prevented postoperative complications. Heterogeneity in interventions and outcomes precluded meta-analysis. CONCLUSIONS: Several different blood pressure optimization interventions show promise in reducing exposure to intraoperative hypotension. Nonetheless, the impact of these interventions on clinical outcomes remains unclear. Future trials should assess promising interventions in samples sufficiently large to identify clinically plausible treatment effects on important outcomes.

Improved Re-estimation of Perioperative Cardiac Risk Using the Surgical Apgar Score: A Retrospective Cohort Study.

OBJECTIVE: To assess whether the Surgical Apgar Score (SAS) improves re-estimation of perioperative cardiac risk. BACKGROUND: The SAS is a novel risk index that integrates three relevant and easily measurable intraoperative parameters (blood loss, heart rate, mean arterial pressure) to predict outcomes. The incremental prognostic value of the SAS when used in combination with standard preoperative risk indices is unclear. METHODS: We conducted a retrospective cohort study of adults (18 years and older) who underwent elective noncardiac surgery at a quaternary care hospital in Canada (2009-2014). The primary outcome was postoperative acute myocardial injury. The SAS (range 0-10) was calculated based on intraoperative estimated blood loss, lowest mean arterial pressure, and lowest heart rate documented in electronic medical records. Incremental prognostic value of the SAS when combined with the Revised Cardiac Risk Index was assessed based on discrimination (c-statistic), reclassification (integrated discrimination improvement, net reclassification index), and clinical utility (decision curve analysis). RESULTS: The cohort included 16,835 patients, of whom 607 (3.6%) patients had acute postoperative myocardial injury. Addition of the SAS to the Revised Cardiac Risk Index improved risk estimation based on the integrated discrimination improvement [2.0%; 95% confidence interval (CI): 1.5%-2.4%], continuous net reclassification index (54%; 95% CI: 46%-62%), and c-index, which increased from 0.68 (95% CI: 0.65-0.70) to 0.75 (95% CI: 0.73-0.77). On decision curve analysis, addition of the SAS to the Revised Cardiac Risk Index resulted in a higher net benefit at all decision thresholds. CONCLUSIONS: When combined with a validated preoperative risk index, the SAS improved the accuracy of cardiac risk assessment for noncardiac surgery. Further research is needed to delineate how intraoperative data can better guide postoperative decision-making.

Surgeon Postoperative Opioid Prescribing Intensity and Risk of Persistent Opioid Use Among Opioid-naive Adult Patients: A Population-based Cohort Study.

OBJECTIVE: The aim of this study was to determine the relationship between surgeon opioid prescribing intensity and subsequent persistent opioid use among patients undergoing surgery. SUMMARY BACKGROUND DATA: The extent to which different postoperative prescribing practices lead to persistent opioid use among surgical patients is poorly understood. METHODS: Retrospective population-based cohort study assessing opioid-naive adults who underwent 1 of 4 common surgeries. For each surgical procedure, the surgeons' opioid prescribing intensity was categorized into quartiles based on the median daily dose of morphine equivalents of opioids dispensed within 7 days of the surgical visit for all the surgeons' patients. The primary outcome was persistent opioid use in the year after surgery, defined as 180 days or more of opioids supplied within the year after the index date excluding prescriptions filled within 30 days of the index date. Secondary outcomes included a refill for an opioid within 30 days and emergency department visits and hospitalizations within 1 year. RESULTS: Among 112,744 surgical patients, patients with surgeons in the highest intensity quartile (Q4) were more likely to fill an opioid prescription within 7 days after surgery compared with those in the lowest quartile (Q1) (83.3% Q4 vs 65.4% Q1). In the primary analysis, the incidence of persistent opioid use in the year after surgery was rare in both highest and lowest quartiles (0.3% Q4 vs 0.3% Q1), adjusted odds ratio (AOR) of 1.18, 95% CI 0.83-1.66). However, multiple analyses using stricter definitions of persistent use that included the requirement of a prescription filled within 7 days of discharge after surgery showed a significant association with surgeon quartile (up to an AOR 1.36, 95% CI 1.25, 1.47). Patients in Q4 were more likely to refill a prescription within 30 days (4.8% Q4 vs 4.0% Q1, AOR 1.14, 95% CI 1.04-1.24). CONCLUSIONS: Surgeons' overall prescribing practices may contribute to persistent opioid use and represent a target for quality improvement. However, the association was highly sensitive to the definition of persistent use used.

Predicting outcomes following open revascularization for aortoiliac occlusive disease using machine learning.

OBJECTIVE: Open surgical treatment options for aortoiliac occlusive disease carry significant perioperative risks; however, outcome prediction tools remain limited. Using machine learning (ML), we developed automated algorithms that predict 30-day outcomes following open aortoiliac revascularization. METHODS: The National Surgical Quality Improvement Program (NSQIP) targeted vascular database was used to identify patients who underwent open aortoiliac revascularization for atherosclerotic disease between 2011 and 2021. Input features included 38 preoperative demographic/clinical variables. The primary outcome was 30-day major adverse limb event (MALE; composite of untreated loss of patency, major reintervention, or major amputation) or death. The 30-day secondary outcomes were individual components of the primary outcome, major adverse cardiovascular event (MACE; composite of myocardial infarction, stroke, or death), individual components of MACE, wound complication, bleeding, other morbidity, non-home discharge, and unplanned readmission. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score. Variable importance scores were calculated to determine the top 10 predictive features. Performance was assessed on subgroups based on age, sex, race, ethnicity, symptom status, procedure type, and urgency. RESULTS: Overall, 9649 patients were included. The primary outcome of 30-day MALE or death occurred in 1021 patients (10.6%). Our best performing prediction model for 30-day MALE or death was XGBoost, achieving an AUROC of 0.95 (95% confidence interval [CI], 0.94-0.96). In comparison, logistic regression had an AUROC of 0.79 (95% CI, 0.77-0.81). For 30-day secondary outcomes, XGBoost achieved AUROCs between 0.87 and 0.97 (untreated loss of patency [0.95], major reintervention [0.88], major amputation [0.96], death [0.97], MACE [0.95], myocardial infarction [0.88], stroke [0.93], wound complication [0.94], bleeding [0.87], other morbidity [0.96], non-home discharge [0.90], and unplanned readmission [0.91]). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.05. The strongest predictive feature in our algorithm was chronic limb-threatening ischemia. Model performance remained robust on all subgroup analyses of specific demographic/clinical populations. CONCLUSIONS: Our ML models accurately predict 30-day outcomes following open aortoiliac revascularization using preoperative data, performing better than logistic regression. They have potential for important utility in guiding risk-mitigation strategies for patients being considered for open aortoiliac revascularization to improve outcomes.

Using machine learning to predict outcomes following carotid endarterectomy.

OBJECTIVE: Prediction of outcomes following carotid endarterectomy (CEA) remains challenging, with a lack of standardized tools to guide perioperative management. We used machine learning (ML) to develop automated algorithms that predict outcomes following CEA. METHODS: The Vascular Quality Initiative (VQI) database was used to identify patients who underwent CEA between 2003 and 2022. We identified 71 potential predictor variables (features) from the index hospitalization (43 preoperative [demographic/clinical], 21 intraoperative [procedural], and 7 postoperative [in-hospital complications]). The primary outcome was stroke or death at 1 year following CEA. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). After selecting the best performing algorithm, additional models were built using intra- and postoperative data. Model robustness was evaluated using calibration plots and Brier scores. Performance was assessed on subgroups based on age, sex, race, ethnicity, insurance status, symptom status, and urgency of surgery. RESULTS: Overall, 166,369 patients underwent CEA during the study period. In total, 7749 patients (4.7%) had the primary outcome of stroke or death at 1 year. Patients with an outcome were older with more comorbidities, had poorer functional status, and demonstrated higher risk anatomic features. They were also more likely to undergo intraoperative surgical re-exploration and have in-hospital complications. Our best performing prediction model at the preoperative stage was XGBoost, achieving an AUROC of 0.90 (95% confidence interval [CI], 0.89-0.91). In comparison, logistic regression had an AUROC of 0.65 (95% CI, 0.63-0.67), and existing tools in the literature demonstrate AUROCs ranging from 0.58 to 0.74. Our XGBoost models maintained excellent performance at the intra- and postoperative stages, with AUROCs of 0.90 (95% CI, 0.89-0.91) and 0.94 (95% CI, 0.93-0.95), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.15 (preoperative), 0.14 (intraoperative), and 0.11 (postoperative). Of the top 10 predictors, eight were preoperative features, including comorbidities, functional status, and previous procedures. Model performance remained robust on all subgroup analyses. CONCLUSIONS: We developed ML models that accurately predict outcomes following CEA. Our algorithms perform better than logistic regression and existing tools, and therefore, have potential for important utility in guiding perioperative risk mitigation strategies to prevent adverse outcomes.

Using machine learning to predict outcomes following open abdominal aortic aneurysm repair.

OBJECTIVE: Prediction of outcomes following open abdominal aortic aneurysm (AAA) repair remains challenging with a lack of widely used tools to guide perioperative management. We developed machine learning (ML) algorithms that predict outcomes following open AAA repair. METHODS: The Vascular Quality Initiative (VQI) database was used to identify patients who underwent elective open AAA repair between 2003 and 2023. Input features included 52 preoperative demographic/clinical variables. All available preoperative variables from VQI were used to maximize predictive performance. The primary outcome was in-hospital major adverse cardiovascular event (MACE; composite of myocardial infarction, stroke, or death). Secondary outcomes were individual components of the primary outcome, other in-hospital complications, and 1-year mortality and any reintervention. We split our data into training (70%) and test (30%) sets. Using 10-fold cross-validation, six ML models were trained using preoperative features (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score. The top 10 predictive features in our final model were determined based on variable importance scores. Performance was assessed on subgroups based on age, sex, race, ethnicity, rurality, median area deprivation index, proximal clamp site, prior aortic surgery, and concomitant procedures. RESULTS: Overall, 12,027 patients were included. The primary outcome of in-hospital MACE occurred in 630 patients (5.2%). Compared with patients without a primary outcome, those who developed in-hospital MACE were older with more comorbidities, demonstrated poorer functional status, had more complex aneurysms, and were more likely to require concomitant procedures. Our best performing prediction model for in-hospital MACE was XGBoost, achieving an AUROC of 0.93 (95% confidence interval, 0.92-0.94). Comparatively, logistic regression had an AUROC of 0.71 (95% confidence interval, 0.70-0.73). For secondary outcomes, XGBoost achieved AUROCs between 0.84 and 0.94. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.05. These findings highlight the excellent predictive performance of the XGBoost model. The top three predictive features in our algorithm for in-hospital MACE following open AAA repair were: (1) coronary artery disease; (2) American Society of Anesthesiologists classification; and (3) proximal clamp site. Model performance remained robust on all subgroup analyses. CONCLUSIONS: Open AAA repair outcomes can be accurately predicted using preoperative data with our ML models, which perform better than logistic regression. Our automated algorithms can help guide risk-mitigation strategies for patients being considered for open AAA repair to improve outcomes.

Machine learning to predict outcomes following endovascular abdominal aortic aneurysm repair.

BACKGROUND: Endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA) carries important perioperative risks; however, there are no widely used outcome prediction tools. The aim of this study was to apply machine learning (ML) to develop automated algorithms that predict 1-year mortality following EVAR. METHODS: The Vascular Quality Initiative database was used to identify patients who underwent elective EVAR for infrarenal AAA between 2003 and 2023. Input features included 47 preoperative demographic/clinical variables. The primary outcome was 1-year all-cause mortality. Data were split into training (70 per cent) and test (30 per cent) sets. Using 10-fold cross-validation, 6 ML models were trained using preoperative features with logistic regression as the baseline comparator. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score. RESULTS: Some 63 655 patients were included. One-year mortality occurred in 3122 (4.9 per cent) patients. The best performing prediction model for 1-year mortality was XGBoost, achieving an AUROC (95 per cent c.i.) of 0.96 (0.95-0.97). Comparatively, logistic regression had an AUROC (95 per cent c.i.) of 0.69 (0.68-0.71). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.04. The top 3 predictive features in the algorithm were 1) unfit for open AAA repair, 2) functional status, and 3) preoperative dialysis. CONCLUSIONS: In this data set, machine learning was able to predict 1-year mortality following EVAR using preoperative data and outperformed standard logistic regression models.

Nonopioid Analgesic Prescriptions Filled after Surgery among Older Adults in Ontario, Canada: A Population-based Cohort Study.

BACKGROUND: The objective was to assess changes over time in prescriptions filled for nonopioid analgesics for older postoperative patients in the immediate postdischarge period. The authors hypothesized that the number of patients who filled a nonopioid analgesic prescription increased during the study period. METHODS: The authors performed a population-based cohort study using linked health administrative data of 278,366 admissions aged 66 yr or older undergoing surgery between fiscal year 2013 and 2019 in Ontario, Canada. The primary outcome was the percentage of patients with new filled prescriptions for nonopioid analgesics within 7 days of discharge, and the secondary outcome was the analgesic class. The authors assessed whether patients filled prescriptions for a nonopioid only, an opioid only, both opioid and nonopioid prescriptions, or a combination opioid/nonopioid. RESULTS: Overall, 22% (n = 60,181) of patients filled no opioid prescription, 2% (n = 5,534) filled a nonopioid only, 21% (n = 59,608) filled an opioid only, and 55% (n = 153,043) filled some combination of opioid and nonopioid. The percentage of patients who filled a nonopioid prescription within 7 days postoperatively increased from 9% (n = 2,119) in 2013 to 28% (n = 13,090) in 2019, with the greatest increase for acetaminophen: 3% (n = 701) to 20% (n = 9,559). The percentage of patients who filled a combination analgesic prescription decreased from 53% (n = 12,939) in 2013 to 28% (n = 13,453) in 2019. However, the percentage who filled both an opioid and nonopioid prescription increased: 4% (n = 938) to 21% (n = 9,880) so that the overall percentage of patients who received both an opioid and a nonopioid remained constant over time 76% (n = 18,642) in 2013 to 75% (n = 35,391) in 2019. CONCLUSIONS: The proportion of postoperative patients who fill prescriptions for nonopioid analgesics has increased. However, rather than a move to use of nonopioids alone for analgesia, this represents a shift away from combination medications toward separate prescriptions for opioids and nonopioids.

Survival and health care costs after inpatient elective surgery: comparison of patients with and without chronic obstructive pulmonary disease.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is common among surgical patients, and patients with COPD have higher risk for complications and death within 30 days after surgery. We sought to describe the longer-term postoperative survival and costs of patients with COPD compared with those without COPD within 1 year after inpatient elective surgery. METHODS: In this retrospective population-based cohort study, we used linked health administrative databases to identify all patients undergoing inpatient elective surgery in Ontario, Canada, from 2005 to 2019. We ascertained COPD status using validated definitions. We followed participants for 1 year after surgery to evaluate survival and costs to the health system. We quantified the association of COPD with survival (Cox proportional hazards models) and costs (linear regression model with log-transformed costs) with partial adjustment (for sociodemographic factors and procedure type) and full adjustment (also adjusting for comorbidities). We assessed for effect modification by frailty, cancer and procedure type. RESULTS: We included 932 616 patients, of whom 170 482 (18%) had COPD. With respect to association with risk of death, COPD had a partially adjusted hazard ratio (HR) of 1.61 (95% confidence interval [CI] 1.58-1.64), and a fully adjusted HR of 1.26 (95% CI 1.24-1.29). With respect to impact on health system costs, COPD was associated with a partially adjusted relative increase of 13.1% (95% CI 12.7%-13.4%), and an increase of 4.6% (95% CI 4.3%-5.0%) with full adjustment. Frailty, cancer and procedure type (such as orthopedic and lower abdominal surgery) modified the association between COPD and outcomes. INTERPRETATION: Patients with COPD have decreased survival and increased costs in the year after surgery. Frailty, cancer and the type of surgical procedure modified associations between COPD and outcomes, and must be considered when risk-stratifying surgical patients with COPD.

Factors influencing withdrawal of life-supporting treatment in cervical spinal cord injury: a large multicenter observational cohort study.

BACKGROUND: Traumatic spinal cord injury (SCI) leads to profound neurologic sequelae, and the provision of life-supporting treatment serves great importance among this patient population. The decision for withdrawal of life-supporting treatment (WLST) in complete traumatic SCI is complex with the lack of guidelines and limited understanding of practice patterns. We aimed to evaluate the individual and contextual factors associated with the decision for WLST and assess between-center differences in practice patterns across North American trauma centers for patients with complete cervical SCI. METHODS: This retrospective multicenter observational cohort study utilized data derived from the American College of Surgeons Trauma Quality Improvement Program database between 2017 and 2020. The study included adult patients (> 16 years) with complete cervical SCI. We constructed a multilevel mixed effect logistic regression model to adjust for patient, injury and hospital factors influencing WLST. Factors associated with WLST were estimated through odds ratios with 95% confidence intervals. Hospital variability was characterized using the median odds ratio. Unexplained residual variability was assessed through the proportional change in variation between models. RESULTS: We identified 5070 patients with complete cervical SCI treated across 477 hospitals, of which 960 (18.9%) had WLST. Patient-level factors associated with significantly increased likelihood of WLST were advanced age, male sex, white race, prior dementia, low presenting Glasgow Coma Scale score, having a pre-hospital cardiac arrest, SCI level of C3 or above, and concurrent severe injury to the head or thorax. Patient-level factors associated with significantly decreased likelihood of WLST included being racially Black or Asian. There was significant variability across hospitals in the likelihood for WLST while accounting for case-mix, hospital size, and teaching status (MOR 1.51 95% CI 1.22-1.75). CONCLUSIONS: A notable proportion of patients with complete cervical SCI undergo WLST during their in-hospital admission. We have highlighted several factors associated with this decision and identified considerable variability between hospitals. Further work to standardize WLST guidelines may improve equity of care provided to this patient population.