Patient Preoperative Risk Factors Research Articles

Machine-learning models for predicting surgical site infections using patient pre-operative risk and surgical procedure factors

BackgroundSurgical site infections (SSIs) are a significant health care problem as they can cause increased medical costs and increased morbidity and mortality. Assessing a patient's preoperative risk factors can improve risk stratification and help guide the surgical decision-making process. Previous efforts to use preoperative risk factors to predict the occurrence of SSIs have relied upon traditional statistical modeling approaches. The aim of this paper is to develop and validate, using state-of-the-art machine learning (ML) approaches, classification models for the occurrence of SSI to improve upon previous models. MethodsIn this work, using the American College of Surgeons’ National Surgical Quality Improvement Program (ACS NSQIP) database, the performances (eg prediction accuracy) of 7 different ML approaches (Logistic Regression (LR), Naïve Bayesian (NB), Random Forest (RF), Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Network (ANN), and Deep Neural Network (DNN)) were compared. The performance of these models was evaluated using the area under the curve, accuracy, precision, sensitivity, and F1-score metrics. ResultsOverall, 2,882,526 surgical procedures were identified in the study for the SSI predictive models’ development. The results indicate that the DNN model offers the best predictive performance with 10-fold compared to the other 6 approaches considered (area under the curve = 0.8518, accuracy = 0.8518, precision = 0.8517, sensitivity = 0.8527, F1-score = 0.8518). Emergency case surgeries, American Society of Anesthesiologists (ASA) Index of 4 (ASA_4), BMI, Vascular surgeries, and general surgeries were most significant influencing features towards developing an SSI. ConclusionsEqually important is that the commonly used LR approach for SSI prediction displayed mediocre performance. The results are encouraging as they suggest that the prediction performance for SSIs can be improved using modern ML approaches.

[Perioperative airway management of nasal endoscopic surgery].

Objective:To evaluate the perioperative airway management process of nasal endoscopic surgery, and find clinical evidence for accelerating recovery and reducing respiratory complications. Methods:The perioperative airway management process for nasal endoscopic surgery was developed according to the patient's preoperative risk factors and preoperative pulmonary function. 512 patients who entered the airway management process from March 2019 to May 2020 were included. The improvement of pulmonary function and the occurrence of adverse respiratory events during the perioperative period were analyzed. Results:265 of 512 patients showed abnormal pulmonary function, including 203 cases with ventilatory dysfunction; 103 cases with positive bronchial provocation test; 59 cases with positive bronchodilation test. Patients with abnormal lung function were treated with aerosol inhalation for 3 to 5 days before surgery, the pulmonary function indicators were greatly improved（P<0.01）. After individualized airway management, patients were then treated with surgery, and there was no perioperative dyspnea event. Conclusion:Perioperative airway management can improve pulmonary function and reduce the risk of nasal endoscopic surgery.

Cardiac Surgery in Patients With Opioid Use Disorder: An Analysis of 1.7 Million Surgeries

Adverse repercussions associated with the current opioid epidemic have been documented in many surgical specialties. This study evaluated the impact of opioid use disorder (OUD) on in-hospital cardiac surgery outcomes by using a large national database. Using the National Inpatient Sample, this study isolated patients undergoing coronary artery bypass grafting, valve repair, or valve replacement from 2009 to 2014. Patients were stratified by OUD status. Multivariable analysis was performed to evaluate the association between opioid use and postoperative outcomes. Patients were further stratified by surgery type. Overall, 1,743,161 patients underwent cardiac surgery, and 6960 patients had OUD (0.4%). Mean age was 47.2 and 65.8 years among those with and without OUD, respectively. Although in-hospital mortality did not differ among these groups, patients with OUD had a significantly higher incidence of stroke (8.3% vs 2.8%) and acute kidney injury (21.4% vs 16.2%), longer hospital stays (18 days vs 10 days), and higher hospitalization costs ($81,238 vs $58,654; all P < .01). However, after adjusting for patient and hospital-level factors, OUD was associated only with a longer hospital length of stay (2.2 days; 95% confidence interval, 1.19 to 3.20) compared with non-opioid users. OUD among cardiac surgery patients is associated with prolonged hospitalization and increased risk of postoperative morbidity, mainly driven by the patient's preoperative risk factors. Strategies to minimize these risk factors at the prehospitalization level is warranted to curb the opioid epidemic and improve overall outcomes in this vulnerable population.

Examining the validity of the ACS-NSQIP Risk Calculator in plastic surgery: lack of input specificity, outcome variability and imprecise risk calculations

American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) created the Surgical Risk Calculator, to allow physicians to offer patients a risk-adjusted 30-day surgical outcome prediction. This tool has...

Aortic Valve Replacement and Concomitant Coronary Artery Bypass: Assessing the Impact of Multiple Grafts

The impact of multivessel coronary artery disease and multivessel coronary artery bypass grafting on outcomes after combined aortic valve replacement and coronary artery bypass grafting (AVR-CABG) has not been sufficiently evaluated. We retrospectively reviewed all patients who underwent AVR-CABG at our institution between January 2000 and December 2004. Patients with any previous or concomitant procedures were excluded. The Kaplan-Meier method was used to calculate survival and freedom from postoperative repeat revascularization. Predictors of mortality were determined by Cox regression analysis. The study cohort consisted of 233 AVR-CABG patients. Mean follow-up was 2.2 +/- 1.7 years with one patient lost to follow-up. Preoperative clinical characteristics were well-matched between patients who received one (n = 86), two (n = 81), or three or four (n = 66) bypass grafts. Operative mortality was 9.3%, 11.1%, and 7.6%, respectively (p = 0.76). Patients in all groups demonstrated significant improvement in New York Heart Association (NYHA) status (p < 0.01). Freedom from postoperative repeat revascularization for all patients after five years was 96.8% and did not differ among groups (p = 0.93). Five-year survival for each group was 63.6%, 72.4%, and 63.9%, respectively (p = 0.91). Emergent operation, ejection fraction less than 0.30, operative age greater than 65 years, NYHA class III/IV, and chronic obstructive pulmonary disease were significant predictors of mortality. The number of stenosed vessels, the number of bypass grafts, incomplete revascularization, and the presence of aortic stenosis or aortic insufficiency did not predict mortality. For patients undergoing AVR-CABG, the number of bypass grafts does not adversely affect survival. Rather, a patient's preoperative risk factors are a better predictor of outcome.

Preoperative and Intraoperative Risk Factors Associated With 30-Day Morbidity Following Urological Surgery: The National Surgical Quality Improvement Program

The National Surgical Quality Improvement Program was established in 1994 to measure and enhance of the quality of surgical care. Since its inception, 30-day postoperative morbidity after major surgery in the Veterans Administration has decreased by 45%. We performed a pilot study to assess risk factors associated with 30-day morbidity using National Surgical Quality Improvement Program indicators in a private setting. A total of 643 subjects were accrued by a trained surgical reviewer from our institution between December 2003 and December 2004. Patient preoperative, intraoperative and nonNational Surgical Quality Improvement Program data elements were abstracted from clinical records. Bivariate relationships between preoperative risk factors, intraoperative process measures and 30-day postoperative morbidity were determined. Multiple variable logistic regression analysis was used to identify patient preoperative and intraoperative risk factors associated with 30-day morbidity. On multivariate analysis the preoperative risk factors history of congestive heart failure (OR 14.42, 95% CI 2.66 to 78.30), diabetes with end organ damage (OR 12.56, 95% CI 2.09 to 75.53), angioplasty (OR 2.75, 95% CI 1.27 to 5.93), quadriplegia (OR 4.39, 95% CI 1.36 to 14.14), low albumin (OR 2.49, 95% CI 1.18 to 5.24) and hydronephrosis (OR 2.51, 95% CI 1.11 to 5.69) were statistically significant predictors of 30-day postoperative morbidity. Intraoperative process measures, that is operative time in hours (OR 1.24, 95% CI 1.12 to 1.37) and transfusion (OR 3.11, 95% CI 1.49 to 6.50), were also significant contributors to postoperative morbidity. We found that patient preoperative risk factors and intraoperative process measures were important determinants of postoperative morbidity. Implementation of the National Surgical Quality Improvement Program in urology may be used to measure processes of care that are associated with surgical outcomes and, thereby, improve the quality of urological care.

The association of complication type with mortality and prolonged stay after cardiac surgery with cardiopulmonary bypass.

Outcome after cardiac surgery varies depending on complication type. We therefore sought to determine the association between complication type, mortality, and length of stay in a large series of patients undergoing cardiac surgery with cardiopulmonary bypass (CPB). Multivariate logistic regression was used to test for differences between complication types in mortality and prolonged length of stay (>10 days) while controlling for preoperative and intraoperative risk factors. In 2609 consecutive cardiac surgical patients requiring CPB, the mortality rate was 3.6%; 36.5% had one or more complications, and 15.7% experienced an adverse outcome (death or prolonged length of stay). Multivariate logistic regression demonstrated that complication type was significantly associated with adverse outcome (P < 0.001) independent of Parsonnet score and CPB time (c-index = 0.80). The development of noncardiac complications only (Group NC) and cardiac complications with other organ involvement (Group B) significantly increased mortality and hospital and intensive care unit length of stay (P < 0.001) when compared with cardiac complications only (Group C). The incidences of adverse outcome in Groups C, NC, and B were 15%, 43%, and 67%, respectively; the mortality rates were 3%, 7%, and 20%, respectively. All these intergroup comparisons were significantly different (adjusted P < 0.05). Complications involving organs other than the heart appear to be more deleterious than cardiac complications alone, underscoring the need for strategies to reduce noncardiac complications. Complications, particularly when they involve organs other than just the heart, increase mortality and prolong the length of hospital stay after heart surgery, independent of a patient's preoperative risk factors and the duration of cardiopulmonary bypass. Strategies aimed at preventing damage to other organs during cardiac surgery need to be improved.