Risk prediction models for postoperative pulmonary complications (PPCs) can assist healthcare professionals in assessing the likelihood of PPCs occurring after surgery, thereby supporting rapid decision-making. This study evaluated the merits, limitations, and challenges of these models, focusing on model types, construction methods, performance, and clinical applications. The findings indicate that current risk prediction models for PPCs following lung cancer surgery demonstrate a certain level of predictive effectiveness. However, there are notable deficiencies in study design, clinical implementation, and reporting transparency. Future research should prioritize large-scale, prospective, multi-center studies that utilize multiomics approaches to ensure robust data for accurate predictions, ultimately facilitating clinical translation, adoption, and promotion.
Acute kidney injury (AKI) is a complication with high morbidity and mortality after cardiac surgery. In order to predict the incidence of AKI after cardiac surgery, many risk prediction models have been established worldwide. We made a detailed introduction to the composing features, clinical application and predictive capability of 14 commonly used models. Among the 14 risk prediction models, age, congestive heart failure, hypertension, left ventricular ejection fraction, diabetes, cardiac valve surgery, coronary artery bypass grafting (CABG) combined with cardiac valve surgery, emergency surgery, preoperative creatinine, preoperative estimated glomerular filtration rate (eGFR), preoperative New York Heart Association (NYHA) score>Ⅱ, previous cardiac surgery, cadiopulmonary bypass (CPB) time and low cardiac output syndrome (LCOS) are included in many risks prediction models (>3 times). In comparison to Mehta and SRI models, Cleveland risk prediction model shows the best discrimination for the prediction of renal replacement therapy (RRT)-AKI and AKI in the European. However, in Chinese population, the predictive ability of the above three risk prediction models for RRT-AKI and AKI is poor.
ObjectiveTo analyze the influencing factors of preoperative frailty in elderly esophageal cancer patients, and to construct and evaluate a nomogram model. MethodsElderly esophageal cancer patients hospitalized in the First Affiliated Hospital of Navy Medical University from January 2022 to August 2024 were selected. General information of the patients was collected, and the patients were divided into a frailty group and a non-frailty group according to the frailty score. Single-factor and multi-factor logistic regression analysis were used to screen the independent influencing factors of preoperative frailty in elderly esophageal cancer patients, and the nomogram model was constructed and evaluated accordingly. ResultsA total of 332 elderly esophageal cancer patients were included, including 256 males and 76 females, with an average age of (69.71±5.95) years. The incidence of preoperative frailty was 45.48%. Multivariate logistic regression analysis showed that age≥70 years [OR=2.986, 95%CI (1.796, 4.965), P<0.001], basic diseases≥2 types [OR=3.129, 95%CI (1.794, 5.457), P=0.012], self-care ability [OR=2.599, 95%CI (1.512, 4.467), P<0.001], and depression [OR=3.784, 95%CI (2.135, 6.706), P=0.005] were independent risk factors for preoperative frailty in elderly esophageal cancer patients. The nomogram model was constructed as follows: Z=−2.038+1.094×(age≥70 years)+1.141×(basic diseases≥2 types)+0.955×self-care ability (non-independent)+1.331×depression, with an area under the receiver operating characteristic curve of 0.802 [95%CI (0.756, 0.847)], and the sensitivity and specificity were 78.8% and 65.2%, respectively. The goodness of fit test showed that the model had good discrimination and calibration (χ2=6.64, P=0.355). ConclusionAge≥70 years, basic diseases≥2 types, self-care ability (non-independent), and depression are independent risk factors for preoperative frailty in elderly esophageal cancer patients. The constructed nomogram model shows good predictive performance and can identify elderly esophageal cancer patients with preoperative frailty, providing a reference for the formulation of corresponding intervention measures.
ObjectiveTo systematically evaluate the risk prediction model of anastomotic fistula after radical resection of esophageal cancer, and to provide objective basis for selecting a suitable model. MethodsA comprehensive search was conducted on Chinese and English databases including CNKI, Wanfang, VIP, CBM, PubMed, EMbase, Web of Science, The Cochrane Library for relevant studies on the risk prediction model of anastomotic fistula after radical resection of esophageal cancer from inception to April 30, 2023. Two researchers independently screened literatures and extracted data information. PROBAST tool was used to assess the risk of bias and applicability of included literatures. Meta-analysis was performed on the predictive value of common predictors in the model with RevMan 5.3 software. ResultsA total of 18 studies were included, including 11 Chinese literatures and 7 English literatures. The area under the curve (AUC) of the prediction models ranged from 0.68 to 0.954, and the AUC of 10 models was >0.8, indicating that the prediction performance was good, but the risk of bias in the included studies was high, mainly in the field of research design and data analysis. The results of the meta-analysis on common predictors showed that age, history of hypertension, history of diabetes, C-reactive protein, history of preoperative chemotherapy, hypoproteinemia, peripheral vascular disease, pulmonary infection, and calcification of gastric omental vascular branches are effective predictors for the occurrence of anastomotic leakage after radical surgery for esophageal cancer (P<0.05). ConclusionThe study on the risk prediction model of anastomotic fistula after radical resection of esophageal cancer is still in the development stage. Future studies can refer to the common predictors summarized by this study, and select appropriate methods to develop and verify the anastomotic fistula prediction model in combination with clinical practice, so as to provide targeted preventive measures for patients with high-risk anastomotic fistula as soon as possible.
Objective To systematically evaluate risk prediction models for acute exacerbation of chronic obstructive pulmonary disease (COPD), and provide a reference for early clinical identification. Methods The literature on the risk prediction models of acute exacerbation of COPD published by CNKI, VIP, Cochrane, Embase and Web of Science database was searched in Chinese and English from inception to April 2022, and relevant studies were collected on the development of risk prediction models for acute exacerbations of COPD. After independent screening of the literature and extraction of information by two independent researchers, the quality of the included literature was evaluated using the PROBASTA tool. Results Five prospective studies, one retrospective case-control study and seven retrospective cohort studies were included, totally 13 papers containing 24 models. Twelve studies (92.3%) reported the area under the receiver operator characteristic curve ranging 0.66 to 0.969. Only five studies reported calibrated statistics, and three studies were internally and externally validated. The overall applicability of 13 studies was good, but there was a high risk of bias, mainly in the area of analysis. Conclusions The existing predictive risk models for acute exacerbations of COPD are unsatisfactory, with wide variation in model performance, inappropriate and incomplete inclusion of predictors, and a need for better ways to develop and validate high-quality predictive models. Future research should refine the study design and study report, and continue to update and validate existing models. Secondly medical staff should develop and implement risk stratification strategies for acute exacerbations of COPD based on predicted risk classification results in order to reduce the frequency of acute exacerbations and to facilitate the rational allocation of medical resources.
Objective To investigate the key risk factors for low anterior resection syndrome (LARS) within 6 months after rectal cancer surgery and to construct a risk prediction model based on the random forest algorithm, providing a reference for early clinical intervention. Methods A retrospective study was conducted on patients who underwent rectal cancer surgery at the West China Hospital of Sichuan University from January 2020 to August 2021. A prediction model for the occurrence of LARS within 6 months after rectal cancer surgery was constructed using the random forest algorithm. The dataset was divided into a training set and a test set in an 8∶2 ratio. The model performance was evaluated by accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC), Brier score, and decision curve analysis (DCA). Results A total of 394 patients were enrolled. Among the 394 patients, 106 developed LARS within 6 months after surgery, with an incidence rate of 26.9%. According to the importance ranking in the random forest algorithm, the key predictive factors were: distance from the inferior tumor margin to the dentate line, body mass index (BMI), tumor size, time to first postoperative flatus, operation time, age, neoadjuvant therapy, and TNM stage. The prediction model constructed using these key factors achieved the accuracy of 73.4%, sensitivity of 75.0%, specificity of 72.7%, AUC (95% confidence interval) of 0.801 (0.685, 0.916), and the Brier score of 0.198. DCA showed that the model provided favorable clinical benefit when the threshold probability was between 25% and 64%. Conclusions The results of this study suggest that patients with a shorter distance from the tumor to the dentate line, higher BMI, and larger tumor size are at higher risk of developing LARS. The risk prediction model constructed in this study demonstrates a good predictive performance and may provide a useful reference for early identification of high-risk patients after rectal cancer surgery.
ObjectiveTo summarize the current status and update of the use of medical imaging in risk prediction of pancreatic fistula following pancreaticoduodenectomy (PD).MethodA systematic review was performed based on recent literatures regarding the radiological risk factors and risk prediction of pancreatic fistula following PD.ResultsThe risk prediction of pancreatic fistula following PD included preoperative, intraoperative, and postoperative aspects. Visceral obesity was the independent risk factor for clinically relevant postoperative pancreatic fistula (CR-POPF). Radiographically determined sarcopenia had no significant predictive value on CR-POPF. Smaller pancreatic duct diameter and softer pancreatic texture were associated with higher incidence of pancreatic fistula. Besides the surgeons’ subjective intraoperative perception, quantitative assessment of the pancreatic texture based on medical imaging had been reported as well. In addition, the postoperative laboratory results such as drain amylase and serum lipase level on postoperative day 1 could also be used for the evaluation of the risk of pancreatic fistula.ConclusionsRisk prediction of pancreatic fistula following PD has considerable clinical significance, it leads to early identification and early intervention of the risk factors for pancreatic fistula. Medical imaging plays an important role in this field. Results from relevant studies could be used to optimize individualized perioperative management of patients undergoing PD.
ObjectiveTo construct a demand model for electronic medical record (EMR) data quality in regards to the lifecycle in machine learning (ML)-based disease risk prediction, to guide the implementation of EMR data quality assessment. MethodsReferring to the lifecycle in ML-based predictive model, we explored the demand for EMR data quality. First, we summarized the key data activities involved in each task on predicting disease risk with ML through a literature review. Second, we mapped the data activities in each task to the associated requirements. Finally, we clustered those requirements into four dimensions. ResultsWe constructed a three-layer structured ring to represent the demand model for EMR data quality in ML-based disease risk prediction research. The inner layer shows the seven main tasks in ML-based predictive models: data collection, data preprocessing, feature representation, feature selection and extraction, model training, model evaluation and optimization, and model deployment. The middle layer is the key data activities in each task; and the outer layer represents four dimensions of data quality requirements: operability, completeness, accuracy, and timeliness. ConclusionThe proposed model can guide real-world EMR data governance, improve its quality management, and promote the generation of real-world evidence.
Objective To explore the risk factors of chronic postoperative inguinal pain (CPIP) after transabdominal preperitoneal hernia repair (TAPP), establish and verify the risk prediction model, and then evaluate the prediction effectiveness of the model. Methods The clinical data of 362 patients who received TAPP surgery was retrospectively analyzed and divided into model group (n=300) and validation group (n=62). The risk factors of CPIP in the model group were screened by univariate analysis and multivariate logistic regression analysis, and the risk prediction model was established and tested. Results The incidence of CPIP at 6 months after operation was 27.9% (101/362). Univariate analysis showed that gender (χ2= 12.055, P=0.001), age (t=–4.566, P<0.01), preoperative pain (χ2=44.686, P<0.01) and early pain at 1 week after operation (χ2=150.795, P<0.01) were related to CPIP. Multivariate logistic regression analysis showed that gender, age, preoperative pain, early pain at 1 week after operation, and history of lower abdominal surgery were independent risk predictors of CPIP. The area under curve (AUC) of the receiver operating characteristic (ROC) of the risk prediction model was calculated to be 0.933 [95%CI (0.898, 0.967)], and the optimal cut-off value was 0.129, while corresponding specificity and sensitivity were 87.6% and 91.5% respectively. The prediction accuracy, specificity and sensitivity of the model were 91.9% (57/62), 90.7% and 94.7%, respectively when the validation group data were substituted into the prediction model. Conclusion Female, age≤64 years old, preoperative pain, early pain at 1 week after operation and without history of lower abdominal surgery are independent risk factors for the incidence of CPIP after TAPP, and the risk prediction model established on this basis has good predictive efficacy, which can further guide the clinical practice.
ObjectiveTo systematically evaluate the risk prediction models for postoperative delirium in adults with cardiac surgery. MethodsThe SinoMed, CNKI, Wanfang, VIP, PubMed, EMbase, Web of Science, and Cochrane Library databases were searched to collect studies on risk prediction models for postoperative delirium in cardiac surgery published up to January 29, 2025. Two researchers screened the literature according to inclusion and exclusion criteria, used the PROBAST bias tool to assess the quality of the literature, and conducted a meta-analysis of common predictors in the model using Stata 17.0 software. ResultsA total of 21 articles were included, establishing 45 models with 28733 patients. Age, cardiopulmonary bypass time, history of diabetes, history of cerebrovascular disease, and gender were the top five common predictors. The area under the curve (AUC) of the 45 models ranged from 0.6 to 0.926. Fourteen out of the 21 studies had good applicability, while the applicability of the remaining seven was unclear; 20 studies had a high risk of bias. Meta-analysis showed that the incidence of postoperative delirium in adults with cardiac surgery was 18.6% [95%CI (15.7%, 21.6%)], and age [OR=1.04 (1.04, 1.05), P<0.001], history of cerebrovascular disease [OR=1.76 (1.46, 2.06), P<0.001], gender [OR=1.73 (1.43, 2.03), P<0.001], minimum mental state examination score [OR=1.00 (0.82, 1.17), P<0.001], and length of ICU stay [OR=5.59 (4.29, 6.88), P<0.001] weer independent influencing factors of postoperative delirium after cardiac surgery. ConclusionThe risk prediction models for postoperative delirium after cardiac surgery have good predictive performance, but there is a high overall risk of bias. In the future, large-sample, multicenter, high-quality prospective clinical studies should be conducted to construct the optimal risk prediction model for postoperative delirium in adults with cardiac surgery, aiming to identify and prevent the occurrence of postoperative delirium as early as possible.