ObjectiveTo investigate the factors associated with unplanned readmission within 30 days after discharge in adult patients who underwent coronary artery bypass grafting (CABG) and to develop and validate a risk prediction model. MethodsA retrospective analysis was conducted on the clinical data of patients who underwent isolated CABG at the Nanjing First Hospital between January 2020 and June 2024. Data from January 2020 to August 2023 were used as a training set, and data from September 2023 to June 2024 were used as a validation set. In the training set, patients were divided into a readmission group and a non-readmission group based on whether they had unplanned readmission within 30 days post-discharge. Clinical data between the two groups were compared, and logistic regression was performed to identify independent risk factors for unplanned readmission. A risk prediction model and a nomogram were constructed, and internal validation was performed to assess the model’s performance. The validation set was used for validation. ResultsA total of 2 460 patients were included, comprising 1 787 males and 673 females, with a median age of 70 (34, 89) years. The training set included 1 932 patients, and the validation set included 528 patients. In the training set, there were statistically significant differences between the readmission group (79 patients) and the non-readmission group (1 853 patients) in terms of gender, age, carotid artery stenosis, history of myocardial infarction, preoperative anemia, and heart failure classification (P<0.05). The main causes of readmission were poor wound healing, postoperative pulmonary infections, and new-onset atrial fibrillation. Multivariable logistic regression analysis revealed that females [OR=1.659, 95%CI (1.022, 2.692), P=0.041], age [OR=1.042, 95%CI (1.011, 1.075), P=0.008], carotid artery stenosis [OR=1.680, 95%CI (1.130, 2.496), P=0.010], duration of first ICU stay [OR=1.359, 95%CI (1.195, 1.545), P<0.001], and the second ICU admission [OR=4.142, 95%CI (1.507, 11.383), P=0.006] were independent risk factors for unplanned readmission. In the internal validation, the area under the curve (AUC) was 0.806, and the net benefit rate of the clinical decision curve analysis (DCA) was >3%. In the validation set, the AUC was 0.732, and the DCA net benefit rate ranged from 3% to 48%. ConclusionFemales, age, carotid artery stenosis, duration of first ICU stay, and second ICU admission are independent risk factors for unplanned readmission within 30 days after isolated CABG. The constructed nomogram demonstrates good predictive power.
Objective To predict the patients who can benefit from local surgery for bone-only metastatic breast cancer (bMBC). Methods Patients newly diagnosed with bMBC between 2010 and 2019 in SEER database were randomly divided into a training set and a validation set at a ratio of 7∶3. The Cox proportional hazards model was used to analyze the independent prognostic factors of overall survival in the training set, and the variables were screened and the prognostic prediction model was constructed. The concordance index (C-index), time-dependent clinical receiver operating characteristic curve and area under the curve (AUC), calibration curve and decision curve analysis (DCA) were used to evaluate the discrimination, calibration and clinical applicability of the model in the training set and validation set, respectively. The model was used to calculate the patient risk score and classify the patients into low-, medium- and high-risk groups. Survival analysis was used to compare the survival difference between surgical and non-surgical patients in different risk groups. Results A total of 2057 patients were enrolled with a median age of 45 years (interquartile range 47-62 years) and a median follow-up of 32 months (interquartile range 16-53 months). Totally 865 patients (42.1%) died. Multivariate Cox proportional hazards model analysis showed that the overall survival of patients with surgery was better than that of patients without surgery [hazard ratio=0.51, 95% confidence interval (0.43, 0.60), P<0.001]. Chemotherapy, marital status, molecular subtype, age, pathological type and histological grade were independent prognostic factors for overall survival (P<0.05), and a prognostic prediction model was constructed based on the independent prognostic factors. The C-index was 0.702 in the training set and 0.703 in the validation set. The 1-, 3-, and 5-year AUCs of the training set and validation set were 0.734, 0.727, 0.731 and 0.755, 0.737, 0.708, respectively. The calibration curve showed that the predicted survival rates of 1, 3, and 5 years in the training set and the validation set were highly consistent with the actual survival rates. DCA showed that the prediction model had certain clinical applicability in the training set and the validation set. Patients were divided into low-, medium- and high-risk subgroups according to their risk scores. The results of log-rank test showed that local surgery improved overall survival in the low-risk group (training set: P=0.013; validation set: P=0.024), but local surgery did not improve overall survival in the medium-risk group (training set: P=0.45; validation set: P=0.77) or high-risk group (training set: P=0.56; validation set: P=0.94). Conclusions Local surgery can improve the overall survival of some patients with newly diagnosed bMBC. The prognostic stratification model based on clinicopathological features can evaluate the benefit of local surgery in patients with newly diagnosed bMBC.
This study introduced the construction of individualized risk assessment model based on Bayesian networks, comparing with traditional regression-based logistic models using practical examples. It evaluates the model's performance and demonstrates its implementation in the R software, serving as a valuable reference for researchers seeking to understand and utilize Bayesian network models.
ObjectiveTo reveal the scientific output and trends in pulmonary nodules/early-stage lung cancer prediction models. MethodsPublications on predictive models of pulmonary nodules/early lung cancer between January 1, 2002 and June 3, 2023 were retrieved and extracted from CNKI, Wanfang, VIP and Web of Science database. CiteSpace 6.1.R3 and VOSviewer 1.6.18 were used to analyze the hotspots and theme trends. ResultsA marked increase in the number of publications related to pulmonary nodules/early-stage lung cancer prediction models was observed. A total of 12581 authors from 2711 institutions in 64 countries/regions published 2139 documents in 566 academic journals in English. A total of 282 articles from 1256 authors were published in 176 journals in Chinese. The Chinese and English journals which published the most pulmonary nodules/early-stage lung cancer prediction model-related papers were Journal of Clinical Radiology and Frontiers in Oncology, respectively. Chest was the most frequently cited journal. China and the United States were the leading countries in the field of pulmonary nodules/early-stage lung cancer prediction models. The institutions represented by Fudan University had significant academic influence in the field. Analysis of keywords revealed that multi-omics, nomogram, machine learning and artificial intelligence were the current focus of research. ConclusionOver the last two decades, research on risk-prediction models for pulmonary nodules/early-stage lung cancer has attracted increasing attention. Prognosis, machine learning, artificial intelligence, nomogram, and multi-omics technologies are both current hotspots and future trends in this field. In the future, in-depth explorations using different omics should increase the sensitivity and accuracy of pulmonary nodules/early-stage lung cancer prediction models. More high-quality future studies should be conducted to validate the efficacy and safety of pulmonary nodules/early-stage lung cancer prediction models further and reduce the global burden of lung cancer.
Objective To develop and validate a prediction model to assess the risk of depression in patients with chronic kidney disease (CKD) based on National Health and Nutrition Examination Survey (NHANES) database. Methods Data on patients with CKD were selected from the NHANES between 2005 and 2018. Participants were randomly divided into a training set and a validation set in a 7∶3 ratio for model development and validation, respectively. Multivariable logistic regression was used in the training set to identify independent risk factors associated with depression in CKD patients, with stepwise selection applied to determine the final predictors. Model performance was assessed using receiver operating characteristic curve (ROC), calibration plots, and decision curve analysis (DCA). Internal validation was performed through bootstrap resampling, and a predictive model was ultimately established. Results A total of 4413 CKD patients were included, including 2112 males (47.86%) and 2301 females (52.14%). Among them, 3089 patients were assigned to the training set and 1324 to the validation set. In the training set, 332 patients (10.75%) presented with depressive symptoms, while 143 patients (10.80%) in the validation set had depressive symptoms. Multivariate logistic regression analysis showed that other hispanic, current smoking, and sleep disorders were risk factors (P＜0.05). Male, middle or high-income, high school grad/ged or above, married or widowed were protective factors (P＜0.05). Finally, 7 variables were included to construct a prediction model, including gender, poverty income ratio, education level, marital status, smoking status, body mass index, and sleep disorders. The ROC curve showed that the AUC=0.773 [95% confidence interval (0.747, 0.799)] in the training set, the internal validation was evaluated by 1000 Bootstrap resampling methods, and the corrected C-index=0.763. The validation set AUC=0.778 [95% confidence interval (0.740, 0.815)], showed good discrimination ability. The calibration curve showed that the model’s predicted probability was highly consistent with the actual occurrence. Decision curve analysis showed that the model provided a significant net benefit for clinical decision-making at a threshold probability of 20%～50%. Conclusions The prediction model constructed in this study can effectively predict the risk of depression in patients with CKD and can provide guidance for early screening and personalized intervention for high-risk groups. However, the external validation and localization of the model still needed further research.
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.
ObjectivesTo compare different formula calculated dosages with the actual doses of warfarin from patients in Beijing Hospital so as to investigate suitable warfarin dosing models for Chinese patients.MethodsOne hundred and three Chinese patients with long-term prescription of warfarin were randomly selected from Beijing Hospital from July 2012 to May 2013. The CYP2C9 and VKROC1 genotypes and basic statistical information were collected. SPSS 18.0 software was used to compare the differences between different formula calculated dosages and the actual dosages of warfarin.ResultsFive genotypes were found in 103 patients, including: CYP2C9 AA genotype + VKORC1 AA genotype (n=72, 69.9%), CYP2C9 AA genotype + VKORC1 AG genotype (n=17, 16.5%), CYP2C9 AC genotype + VKORC1 AA genotype (n=10, 9.7%), CYP2C9 AC genotype + VKORC1 AG genotype (n=3, 2.9%) and CYP2C9 AA genotype + VKORC1 GG genotype (n=1, 1%). Compared with the actual dosages of warfarin, the degree of coincidence was highest for dosages calculated by Jeffrey’s formula.Conclusions Using Jeffrey’s formula to calculate warfarin dosages may be more suitable for Chinese patients with using long-term warfarin. Due to limited sample size, prospective and large sample size studies are required to verify the above conclusion.
ObjectiveCombined with long non-coding RNA (lncRNA) to find a regression model that can be used to predict the survival rate of patients with colon cancer before operation.MethodsThe clinical information and gene expression information of patients with colon cancer were downloaded by using TCGA database. The differentially expressed lncRNAs in tumor and paracancerous tissues were screened out, and then combined with the clinical information of patients to construct Cox proportional hazard regression model.ResultsA total of 26 kinds of lncRNAs with statistical difference in gene expression between paracancerous tissues and tumor tissues were selected (P<0.05). Through repeated screening and comparison of prediction efficiency, the prediction model was finally selected, which was constructed by patients’ age, M stage, N stage, and three kinds of lncRNAs (ZFAS1, SNHG25, and SNHG7) gene expression level: age [HR=4.00, 95%CI: (1.48, 10.84), P=0.006], M stage [HR=3.96, 95%CI: (2.23, 7.04), P<0.001], N stage [HR=1.87, 95%CI: (1.24, 2.84), P=0.003], ZFAS1 gene expression level [HR=0.60, 95%CI: (0.41, 0.86), P=0.006], SNHG25 gene expression level [HR=0.85, 95%CI: (0.73, 1.00), P=0.045], and SNHG7 gene expression level [HR=2.32, 95%CI: (1.53, 3.52), P<0.001] were all independent risk factors for postoperative survival of patients with colon cancer. The area under the ROC curves for predicting 1, 3, and 5-year overall survival were 0.802, 0.828, and 0.771, respectiely, which had a good prediction ability.ConclusionThe predictive model constructed by the combination of ZFAS1, SNHG25, SNHG7 genes expression level with M stage, N stage, and age can better predict the overall survival rate of patients before operation, which can effectively guide clinical decision-making and choose the most suitable treatment method for patients.
ObjectiveTo investigate the effect of CYP2C9 and APOE on the dose of stable warfarin and model prediction in Hainan population.MethodsFrom August 2016 to July 2018, 368 patients who required heart valve replacement and agreed to take warfarin anticoagulation at the second department of cardiothoracic surgery in our hospital were enrolled, including 152 males aged 48.5–70.5 (60.03±10.18) years and 216 females aged 43.5–65.6 (54.24±11.35) years. CYP2C9 and APOE were amplified by polymerase chain reaction. The gene fragment was sequenced by the Single Nucleotide Polymorphisms (SNP) site. The patients' age, sex, weight, history of smoking and drinking, and the dose of stable warfarin were recorded. Regression analysis of these clinical data was made to construct a dose prediction model.ResultsAmong 368 patients, CYP2C9 genotype test results showed 301 patients (81.8%) with *1*1 genotype, and 67 patients (18.2%) with *1*3 type. For different CYP2C9 genotype patients, the difference was statistically significant in the dose of stable warfarin (P<0.05). The results of APOE genotype showed 93 patients (25.3%) with E2 genotype, 221 patients (60.1%) with E3 genotype, and 54 patients (14.7%) with E4 genotype; the dose of stable warfarin in patients with different APOE genotypes was statistically significant (P<0.05). Multiple regression analysis showed that patients' age, body weight, and CYP2C9 and APOE genotypes were correlated with the dose of stable warfarin. The correlation coefficient R2 was 0.572, and the prediction model was statistically significant (P<0.05).ConclusionCYP2C9 and APOE gene polymorphisms exist in Hainan population. There is significant difference in the dose of stable warfarin among different genotypes of patients. The model to predict stable warfarin can partly explain the difference of warfarin among different patients.
ObjectiveTo explore the influencing factors of cancer-specific survival of patients with large hepatocellular carcinoma, and draw a nomogram to predict the cancer-specific survival rate of large hepatocellular carcinoma patients.MethodsThe clinicopathological data of patients with large hepatocellular carcinoma during the period from 1975 to 2017 in the Surveillance, Epidemiology, and End Results (SEER) database were searched and randomly divided into training group and validation group at 1∶1. Using the training data, the Cox proportional hazard regression model was used to explore the influencing factors of cancer-specific survival and construct the nomogram; finally, the receiver operating characteristic curve (ROC curve) and the calibration curve were drawn to verify the nomogram internally and externally.ResultsThe results of the multivariate Cox proportional hazard regression model showed that the degree of liver cirrhosis, tumor differentiation, tumor diameter, T stage, M stage, surgery, and chemotherapy were independent influencing factors that affect the specific survival of patients with large hepatocellular carcinoma (P<0.05), and then these factors were enrolled into the nomogram of the prediction model. The areas under the 1, 3, and 5-year curves of the training group were 0.800, 0.827, and 0.814, respectively; the areas under the 1, 3, and 5-year curves of the validation group were 0.800, 0.824, and 0.801, respectively. The C index of the training group was 0.779, and the verification group was 0.777. The calibration curve of the training group and the verification group was close to the ideal curve of the actual situation.ConclusionThe nomogram of the prediction model drawn in this study can be used to predict the specific survival of patients with large hepatocellular carcinoma in the clinic.