Predictive Model For Complications Research Articles

Predictive Model for Complications in Minimally Invasive Hysterectomy: A Retrospective Single Cohort Analysis

<h3>Study Objective</h3> To create a predictive model score by estimating rate of intra-operative and post-operative complications and by defining the risk of 30-day major post-operative complications (Clavien-Dindo≥3) according to the presence of 10 different variables analyzed at the time of minimally invasive (MI) hysterectomy. <h3>Design</h3> A single center single-arm retrospective study. Data of consecutive patients who have undergone MI hysterectomy between May 2018 and April 2021 have been analyzed. Perioperative surgical outcomes, occurrence of intra- and post-operative complications, 30 days-readmissions were registered. Univariate and multivariable analyses were performed. <h3>Setting</h3> Miulli's Hospital surgery rooms (Bari). <h3>Patients or Participants</h3> 445 patients were included in the study. <h3>Interventions</h3> A standardized surgical approach of MI hysterectomy was adopted by laparoscopic or robotic-lps assisted approach. <h3>Measurements and Main Results</h3> The majority of patients developed a minor event. Major complications (grade III) were observed in 14 patients. The percentage of intraoperative and postoperative complications (Clavien-Dindo classification) was: 44 patients (9.88%) showed a related surgical complication. None showed a grade IV or V complication. Univariate analysis was performed on patients who had developed intra- or post-operative complications from those who did not experience complications. BMI (p-value 0.045) and surgeon's experience (p-value 0.015) were found to be associated with a different surgery time. No one clinic-pathological characteristics were found to be correlated with complications. About major postoperative complications, it was found a statistically significant association for the variables BMI (p-value 0,006), previous surgery (p-value 0,015), and surgeon experience (<i>p</i>-value 0,035). The three significant variables were inserted in a reproducible predictive model in order to stratify the risk of post-operative complications: the score of our predictive model was directly proportional to the severity of complications. <h3>Conclusion</h3> MI hysterectomy should be considered as a safe approach, and it's associated with a low risk of intra- and post-operative complications. Overall, we developed a risk assessment tool including factors not previously considered in the Literature.

Training prediction models for individual risk assessment of postoperative complications after surgery for colorectal cancer.

The occurrence of postoperative complications and anastomotic leakage are major drivers of mortality in the immediate phase after colorectal cancer surgery. We trained prediction models for calculating patients' individual risk of complications based only on preoperatively available data in a multidisciplinary team setting. Knowing prior to surgery the probability of developing a complication could aid in improving informed decision-making by surgeon and patient and individualize surgical treatment trajectories. All patients over 18years of age undergoing any resection for colorectal cancer between January 1, 2014 and December 31, 2019 from the nationwide Danish Colorectal Cancer Group database were included. Data from the database were converted into Observational Medical Outcomes Partnership Common Data Model maintained by the Observation Health Data Science and Informatics initiative. Multiple machine learning models were trained to predict postoperative complications of Clavien-Dindo grade ≥ 3B and anastomotic leakage within 30days after surgery. Between 2014 and 2019, 23,907 patients underwent resection for colorectal cancer in Denmark. A Clavien-Dindo complication grade ≥ 3B occurred in 2,958 patients (12.4%). Of 17,190 patients that received an anastomosis, 929 experienced anastomotic leakage (5.4%). Among the compared machine learning models, Lasso Logistic Regression performed best. The predictive model for complications had an area under the receiver operating characteristic curve (AUROC) of 0.704 (95%CI 0.683-0.724) and an AUROC of 0.690 (95%CI 0.655-0.724) for anastomotic leakage. The prediction of postoperative complications based only on preoperative variables using a national quality assurance colorectal cancer database shows promise for calculating patient's individual risk. Future work will focus on assessing the value of adding laboratory parameters and drug exposure as candidate predictors. Furthermore, we plan to assess the external validity of our proposed model.

Computed-tomography-based predictive scores of surgical complications to help decision-making in enrolling obese patients in kidney transplantation list.

This study aimed at developing scores predicting surgical complications in obese transplant recipients, based on preoperative computed tomography (CT) parameters. All consecutive patients with a body mass index (BMI) ≥ 30 kg/m2 who underwent kidney transplantation between 2012 and 2019 were included. The preoperative CT parameters were assessed: total fatty surface (TFS), subcutaneous fatty surface (SFS), iliac vessel to skin distance (VSD), and abdominal perimeter (AP). Per- and postoperative complications (vascular, urinary, parietal, and digestive complications) within 30 days were listed. Predictive models of surgical complications were generated based on the results of the logistic regression. Among the 163 patients included, 53 (32.5%) experienced surgical complications. The AP was a risk factor for complications in multivariate analysis (OR: 1.050; 95% CI: 1.016-1.087; p = 0.03). Two predictive models of complications were created based on the statistical analysis: a one-variable model based on AP (sensitivity 86.8%, specificity 41.8%, area under the curve (AUC) 65.3, with a cutoff value of 107 cm) and a five-variable model based on BMI, TFS, SFS, VSD, and AP (sensitivity 73.6%, specificity 57.3%, AUC 66.2). These models, based on patient morphometric measurements, could allow predicting the occurrence of surgical complications in obese candidates for kidney transplantation.

Development of novel prognostic models for predicting complications of urethroplasty.

To identify predictors of thirty-day perioperative complications after urethroplasty and create a model to predict patients at increased risk. We selected all patients recorded in the National Surgery Quality Improvement Program (NSQIP) from 2005 to 2015 who underwent urethroplasty, determined by Current Procedural Terminology (CPT) codes. The primary outcome of interest was a composite 30-day complication rate. To develop predictive models of urethroplasty complications we used random forest and logistic regression with tenfold cross-validation employing demographic, comorbidity, laboratory, and wound characteristics as candidate predictors. Models were selected based on the receiver operating characteristic (ROC) curve, with the primary measure of performance being the area under curve (AUC). We identified 1135 patients who underwent urethroplasty and met inclusion criteria. The mean age was 53years with 84% being male. The overall incidence of complications was 8.6% (n = 98). Patients who experienced a complication more commonly had diabetes, a preoperative blood transfusion, preoperative sepsis, lower hematocrit and albumin, as well as a longer operative time (p < 0.05). LASSO logistic and random forest logistic models for predicting urethroplasty complications had an AUC (95% CI) 0.73 (0.58-0.87), and 0.48 (0.33-0.68), respectively. The variables that were determined to be most important and included in the predictive models were operative time, age, American Society of Anesthesiologists (ASA) classification and preoperative laboratory values (white blood cell count, hematocrit, creatinine, platelets). Our predictive models of complications perform well and may allow for improved preoperative counseling and risk stratification in the surgical management of urethral stricture.

Predictive Models of Complications after Endovascular Treatment of Abdominal Aortic Aneurysms

Predictive Models of Complications after Endovascular Treatment of Abdominal Aortic Aneurysms

Craniotomy for Glioma Resection: A Predictive Model

Regulatory agencies are standardizing quality metrics on the basis of which surgical procedures will be evaluated. We attempted to create a predictive model of perioperative complications in patients undergoing craniotomies for glioma resection. We performed a retrospective cohort study involving patients who underwent craniotomies for glioma resection from 2005-2011 and were registered in the National Inpatient Sample (NIS) database. A predictive model for complications was developed and validated. Overall, 21,384 patients underwent glioma resection. The respective inpatient postoperative risks were 1.6% for death, 25.8% for discharge to rehabilitation, 4.0% for treated hydrocephalus, 0.7% for cardiac complications, 0.5% for respiratory complications, 0.8% for deep wound infection, 0.6% for deep venous thrombosis (DVT), 3.1% for pulmonary embolus (PE), and 1.3% for acute renal failure (ARF). Predictive models for individual complications were developed on the basis of a logistic regression analysis and subsequently validated in a bootstrapped sample. The models demonstrated good discrimination with areas under the curve (AUC) of 0.71, 0.71, 0.69, 0.71, 0.74, 0.70, 0.73, 0.64, and 0.81 for postoperative risk of death, discharge to rehabilitation, hydrocephalus, cardiac complications, respiratory complications, deep wound infection, DVT, PE, and ARF, respectively. Additionally, the Hosmer-Lemeshow test was used to assess the calibration of all models. The presented models can assist in the preoperative estimation of the complication risk for glioma patients and be used as an adjunct for outcome benchmarking in this population.