Elucidating predictors of preoperative acute heart failure in elderly patients with hip fractures through machine learning and SHAP analysis: a retrospective cohort study

Elucidating predictors of preoperative acute heart failure in elderly patients with hip fractures through machine learning and SHAP analysis: a retrospective cohort study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Elucidating predictors of preoperative acute heart failure in elderly patients with hip fractures through machine learning and SHAP analysis: a retrospective cohort study Qili Yu, Mingming Fu, Zhiyong Hou, Zhiqian Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4274769/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Apr, 2025 Read the published version in BMC Geriatrics → Version 1 posted 10 You are reading this latest preprint version Abstract Background Acute heart failure has become a significant challenge in elderly patients with hip fractures. Timely identification and assessment of preoperative acute heart failure have become key factors in reducing surgical risks and improving outcomes. Objective This study aims to precisely predict the risk of acute heart failure in elderly patients with hip fractures before surgery through machine learning techniques and SHapley Additive exPlanations (SHAP), providing a scientific basis for clinicians to optimize patient management strategies and reduce adverse events. Methods A retrospective study design was employed, selecting patients admitted for hip surgery in the Department of Geriatric Orthopedics at the Third Hospital of Hebei Medical University from January 2018 to December 2022 as research subjects. Data were analyzed using logistic regression, random forests, support vector machines, AdaBoost, XGBoost, and GBM machine learning methods combined with SHAP analysis to interpret relevant factors and assess the risk of acute heart failure. Results A total of 2,631 patients were included in the final cohort, with an average age of 79.3 ± 7.7. 33.7% of patients experienced acute heart failure before surgery. A predictive model for preoperative acute heart failure in elderly hip fracture patients was established through multivariate logistics regression: Logit(P) = -2.262–0.315 × Sex + 0.673 × Age + 0.556 × Coronary heart disease + 0.908 × Pulmonary infection + 0.839 × Ventricular arrhythmia + 2.058 × Acute myocardial infarction + 0.442 × Anemia + 0.496 × Hypokalemia + 0.588 × Hypoalbuminemia, with a model nomogram established and an AUC of 0.767 (0.723–0.799). Predictive models were also established using five machine learning methods, with GBM performing optimally, achieving an AUC of 0.757 (0.721–0.792). SHAP analysis revealed the importance of all variables, identifying acute myocardial infarction as the most critical predictor and further explaining the interactions between significant variables. Conclusion This study successfully developed a predictive model based on machine learning that accurately predicts the risk of acute heart failure in elderly patients with hip fractures before surgery. The application of SHAP enhanced the model's interpretability, providing a powerful tool for clinicians to identify high-risk patients and take appropriate preventive and therapeutic measures in preoperative management. Heart failure Hip fracture Preoperative Machine learning SHAP Prediction model Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Introduction With the acceleration of global population aging, hip fractures in the elderly have become a significant public health challenge. Since records began in 1990, over 1.6 million people worldwide have suffered from hip fractures. It is predicted that, over time, especially among the elderly, the incidence of hip fractures will show a gradually increasing trend. By 2050, the number of individuals affected is expected to rise to at least 4.5 million [10; 22]. These fractures not only increase the mortality and disability rates of patients, severely affecting the quality of life, but also impose a significant burden on the healthcare system, including the costs of surgical treatment, long-term rehabilitation, and the subsequent socio-economic burdens. In particular, the risk of acute heart failure before surgery in elderly patients with hip fractures has significantly increased, becoming a key complication closely associated with high mortality rates and prolonged hospital stays, further exacerbating the risk of postoperative complications, including infections and re-fractures.[ 4 ]. Therefore, developing effective prediction and prevention strategies is crucial for improving the treatment outcomes of this patient group. In clinical practice, we have observed that most surgeons tend to overlook the assessment of heart failure biomarkers such as Brain Natriuretic Peptide (BNP) or N-Terminal pro-B-Type Natriuretic Peptide (NT-proBNP) in the preoperative evaluation of elderly patients with hip fractures. This oversight could miss patients who have developed acute heart failure, thus failing to intervene timely in this potentially high-risk state[ 7 ]. Machine learning offers a new perspective and approach by analyzing vast amounts of patient data to predict complications that may arise after a hip fracture, such as preoperative acute heart failure, thereby providing a scientific basis for clinical decision-making, optimizing patient management strategies, and reducing the incidence of adverse events[19; 21]. This study utilizes machine learning methods and SHAP values aimed at precisely predicting the risk of acute heart failure before surgery in elderly patients with hip fractures. By analyzing clinical data to reveal the complex associations between patient characteristics, laboratory test results, and preoperative complications, this research offers a new perspective and method. It not only enhances the accuracy of predictions but also provides actionable data support for doctors, optimizing patient management strategies, and reducing the occurrence of adverse events. Therefore, the establishment and use of this study's model can alert physicians to conduct a more comprehensive preoperative assessment, including the measurement of BNP or NT-proBNP, thus identifying those high-risk patients. Such an integrated preoperative approach can not only reduce surgical risks and postoperative complications but also shorten hospital stays and potentially lower mortality rates. It provides a safer and more effective treatment plan for elderly patients with hip fractures, significantly improving their prognosis and ultimately achieving the goal of improving the clinical outcomes of elderly patients with hip fractures. Materials and methods 2.1 Study Design and Patients This retrospective study selected inpatients who underwent hip surgery at the Department of Geriatric Orthopedics, Hebei Medical University Third Hospital, from January 2018 to December 2022, as the study subjects. The inclusion criteria were: (1) aged 65 years and older; (2) hip fractures confirmed by radiographic examinations such as X-rays; (3) patients with complete medical records, laboratory test results, and other necessary medical documents. Exclusion criteria were lack of complete medical records, laboratory test results, or other necessary medical documents, and patients who did not meet the diagnostic criteria for hip fractures. 2.2 Ethical Statement This study, based on the retrospective analysis of existing case data, ensured that all patient data collection and analysis were conducted anonymously to protect patient privacy. Furthermore, the study was in compliance with the Declaration of Helsinki and had been approved and supported by the Institutional Review Board of Hebei Medical University Third Hospital (Approval No.: 2021-087-1). 2.3 Disease Definition Acute Heart Failure (AHF) is a condition where there is a sudden decrease in the heart's ability to pump blood, leading to the body's circulatory volume being insufficient to meet metabolic demands. According to the European Society of Cardiology, AHF is caused by acute changes in the structure or function of the heart, accompanied by increased filling pressures and/or a significant reduction in ejection fraction. Common symptoms include shortness of breath, pulmonary congestion, and inadequate organ perfusion. A key biochemical marker for diagnosing AHF is a significant increase in serum BNP or NT-proBNP levels [15]. Different thresholds of BNP and NT-proBNP are used for diagnosing AHF to accommodate patients of varying ages. Specifically, the diagnostic threshold for BNP is ≥300 pg/mL for patients of all ages. For NT-proBNP, the thresholds are age-stratified: >450 pg/mL for patients under 55 years of age; >900 pg/mL for those between 55 to 75 years; and >1800 pg/mL for patients 75 years and older. These elevated markers are instrumental in confirming the diagnosis of AHF[11]. In clinical practice, differentiating between an acute exacerbation of heart failure and chronic heart failure is key, focusing on changes in clinical symptoms and acute variations in BNP or NT-proBNP levels. Through this approach, physicians can more accurately diagnose acute heart failure, thereby providing appropriate treatment for patients. 2.4 Data Collection This retrospective study is based on data collected from patients who underwent hip surgery in the Department of Geriatric Orthopedics at Hebei Medical University Third Hospital, from January 2018 to December 2022. Data on patients prior to surgery were gathered through the medical record system, including details on heart failure status, sex, age, admission time, and comorbidities such as hypertension, Old cerebral infarction, coronary artery disease, diabetes, chronic obstructive pulmonary disease (COPD), cancer, arrhythmias, pulmonary infections, ventricular arrhythmias, acute myocardial infarction, acute cerebrovascular disease, stress hyperglycemia, stress ulcers, urinary tract infections, anemia, hypokalemia, hyponatremia, hypoalbuminemia, and lower limb venous thrombosis. By analyzing this data, we aim to gain a deeper understanding of the risk factors for acute heart failure in elderly patients with hip fractures before surgery, thereby providing precise intervention suggestions for clinical practice. 2.5 Model Establishment 2.5.1 Dataset Configuration and Variable Selection In this study, we utilized the logistic regression method to predict the occurrence of acute heart failure in elderly patients with hip fractures before surgery. Initially, the collected data were divided into training and validation sets at a ratio of 7:3 to ensure the adequacy of the training process and the independence of the evaluation process. To identify risk factors significantly associated with acute heart failure and avoid overfitting, we employed the Least Absolute Shrinkage and Selection Operator (LASSO) method for selection [18; 24]. Furthermore, we utilized a multivariate logistic regression model to assess the relationship between these factors and acute heart failure, ensuring that only statistically significant predictors were included in the final model. By constructing a nomogram of the model, we made the prediction outcomes and the contributions of various variables both intuitive and easy to understand. 2.5.2 Multimodel Development and Validation Process Beyond the basic logistic regression model, we delved into five additional machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM), Adaptive Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost) and gradient boosting machine (GBM). The models were trained using a 5-fold cross-validation method, which helped us to more accurately evaluate their performance on unseen data [14]. Given the imbalance between positive and negative samples in our dataset, we incorporated the Synthetic Minority Over-sampling Technique (SMOTE) focusing on the minority class samples at the boundary to improve the sample distribution and optimize model performance. 2.5.3 Model Interpretability and Comprehensive Evaluation During the model evaluation phase, we not only focus on the model's discriminative ability, assessed by calculating the Area Under the Curve (AUC) value, but also on the model's calibration through the Hosmer-Lemeshow test. Additionally, Decision Curve Analysis (DCA) is applied to compute the net benefit at different thresholds, comprehensively evaluating the model's practical value in clinical decision-making [25]. The Clinical Impact Curve (CIC) is utilized to visualize the benefit values brought by different thresholds. To enhance the model's interpretability, we employed SHapley Additive exPlanations (SHAP) analysis, which demonstrates the contribution of different variables at the individual level. Furthermore, every observation in the dataset can be explicated with designated SHAP values. Through the aforementioned strategy, our goal is to construct a model that is both precise and highly interpretable, providing a powerful tool for clinicians. This will enable them to identify the risk of acute heart failure before surgery more promptly when treating elderly patients with hip fractures. 2.6 Statistical Analysis In this study, our aim was to reveal the relationship between acute heart failure and various risk factors in elderly patients with hip fractures. Initially, we analyzed the baseline information of participants through descriptive statistics. The normality of continuous variables was verified using the Kolmogorov-Smirnov test. Variables fitting a normal distribution were described in terms of mean ± standard deviation, while non-normally distributed data were represented by median and interquartile range. The distribution characteristics of categorical variables were presented in frequencies and percentages. Variance Inflation Factor (VIF) and tolerance were calculated to assess potential collinearity among parameters, with a VIF below 5 and tolerance above 0.1 considered as standards indicating no significant collinearity. All statistical analyses were conducted using SPSS 24.0 and R language. The level of statistical significance was set at P < 0.05. Results 3.1 Patient Baseline Characteristics Between January 2018 and December 2022, a total of 4,170 elderly patients with hip fractures were included in our study. After screening, 1,539 patients were excluded, leaving 2,631 patients in the final analysis. The excluded patients comprised 1,077 with non-hip fractures, 328 non-surgical patients, and 134 with incomplete data (Figure 1). Table 1 presents the baseline clinical characteristics of the overall sample and compares those between the acute heart failure (AHF) group and the non-AHF group among elderly patients with hip fractures. Overall, the mean age of the patients was 79.3±7.7, with 766 males (29.1%) and 1,865 females (70.9%). Among them, 888 patients (33.7%) experienced acute heart failure before surgery. There were statistically significant differences in gender distribution, age, and age groups (<75 years and ≥75 years) between the two groups (p<0.05). Regarding comorbidities, the prevalence of coronary artery disease and arrhythmias was significantly higher in the AHF group compared to the non-AHF group (p<0.05). Additionally, preoperative complications such as pulmonary infection, ventricular arrhythmias, and acute myocardial infarction also showed a higher incidence in the AHF group, with significant statistical differences (p<0.05). 3.2 Univariate Analysis of Laboratory Data and Ultrasound Examinations Table 2 displays the preoperative laboratory and lower limb venous ultrasound characteristics of elderly patients with hip fractures. The incidences of anemia, hypokalemia, hyponatremia, and hypoalbuminemia were significantly higher in the AHF group compared to the non-AHF group, showing significant statistical differences (p<0.05). However, there was no significant difference in the incidence of lower limb venous thrombosis between the two groups. 3.3 Development and Validation of Nomograms Using R, patients were randomly divided into a training set and a test set in a 7:3 ratio, with 1,843 patients in the training set and 788 in the test set. Initial analysis with LASSO regression on the training set selected 17 variables out of 22 (Figures 2A and 2B). Subsequent multivariable logistic regression analysis identified gender, age, coronary heart disease, pulmonary infection, ventricular arrhythmia, acute myocardial infarction, anemia, hypokalemia, and hypoalbuminemia as independent risk factors for the occurrence of acute heart failure before surgery in elderly patients with hip fractures (Tables 3 and Figure 3). Based on these independent risk factors, we developed a nomogram model to predict the probability of pre-surgical acute heart failure in elderly patients with hip fractures (Figure 4). The predictive model is given by Logit(P) = -2.262 - 0.315 × Sex + 0.673 × Age + 0.556 × Coronary heart disease + 0.908 × Pulmonary infection + 0.839 × Ventricular arrhythmia + 2.058 × Acute myocardial infarction + 0.442 × Anemia + 0.496 × Hypokalemia + 0.588 × Hypoalbuminemia. The variance inflation factor (VIF) was calculated for each variable in the model, indicating all predictor variables had VIF values well below the threshold of 5, specifically: Sex 1.01, Age 1.01, Coronary heart disease 1.01, Pulmonary infection 1.01, Ventricular arrhythmia 1.02, Acute myocardial infarction 1.01, Anemia 1.11, Hypokalemia 1.02, Hypoalbuminemia 1.12. The nomogram was evaluated through 1,000 bootstrap resampling, and the results showed that the calibration curve deviated only slightly from the perfect prediction line, indicating good agreement between the model's predictions and the actual observations (Figure 5). Comparing the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) between the training and validation datasets, the AUC for the training set was 0.761 (0.740-0.786), and for the test set, it was 0.767 (0.723-0.799) (Figure 6). Moreover, the nomogram model's corrected C-statistic obtained through bootstrap resampling was 0.776, demonstrating good performance in internal validation. This means that the model has strong discriminative ability and can accurately predict the risk of acute heart failure in patients. Decision Curve Analysis (DCA) indicates significant clinical decision-making value with a probability range of 8%-90% in the training set (Figure 7A) and 9%-86% in the validation set (Figure 7B). Additionally, the Clinical Impact Curve (CIC) demonstrates the effect of different threshold settings on the number of patients predicted by the model (Figures 7C and 7D). This further suggests the model has substantial application potential, especially in predicting the risk of acute heart failure in elderly patients after hip fractures. The model provides a powerful tool to more precisely predict the likelihood of acute heart failure, thereby guiding clinicians towards more appropriate preventative and therapeutic measures. Implementing clinical interventions based on this model's predictions can effectively optimize patient management, likely leading to positive impacts on patient health outcomes. 3.4 Development of Predictive Models Using Machine Learning Methods All raw data was preprocessed prior to being input into the machine learning model, including cleaning and transformation steps, to ensure data integrity and high quality for accurate handling and analysis by the machine learning algorithms. The features with the highest importance scores in standardization were Acute Myocardial Infarction, Ventricular Arrhythmia, Pulmonary Infection, and Anemia (Figure 8A and Table 4). Correlations between variables were also calculated and are displayed in Figure (Figure 8B). Models were evaluated using various machine learning methods, with the Area Under the Curve (AUC) values obtained as follows: RF 0.746 (0.710—0.782), SVM 0.714 (0.676-0.752), AdaBoost 0.735 (0.699-0.772), XGBoost 0.747 (0.711-0.783), GBM 0.757 (0.721 - 0.792), with GBM showing the best AUC among the models (Figure 9). Accuracy, sensitivity, precision, and F1 score were also calculated for each model, with GBM showing the best performance in terms of accuracy (73%) and sensitivity (95.6%) (Table 5). SHAP analysis was conducted to understand the impact of multiple features on the predictive model for acute heart failure in elderly patients with hip fractures before surgery (Figure 10). The Feature Importance Plot shows each observation as a dot, with the SHAP value on the x-axis indicating the impact of the feature on the model's output. Positive values indicate contributions that increase risk, while negative values indicate contributions that decrease risk. The color gradient from purple to yellow represents feature values from low to high. It is observed that the SHAP values for Acute Myocardial Infarction are distributed in the positive region, with several higher positive points indicating that the presence of acute myocardial infarction significantly increases the risk of acute heart failure. Conversely, Hyponatremia shows both positive and negative SHAP values, concentrated near zero, suggesting a relatively small or individual-dependent impact on the prediction. However, the SHAP values for COPD are mainly in the negative region, possibly indicating a lower risk of acute heart failure in patients with COPD in this model. Through individual-level predictive behavior analysis using the SHAP algorithm, the model revealed key variables influencing the risk of acute heart failure for four patients, showing the contribution of each factor to the prediction and identifying Acute Myocardial Infarction as the main variable affecting all patients. Its SHAP value was significantly higher than other features, and we also found that Anemia, Ventricular Arrhythmia, and Pulmonary Infection play important roles in increasing the risk of heart failure (Figures 11A-D). The SHAP values of these variables provide positive contributions, reinforcing their importance in risk assessment, consistent with the overall trends in the Feature Importance Plot. By constructing multivariate dependence plots (Figure 12), we suggest interactions between variable features, such as between Acute Myocardial Infarction and Anemia, where scatter plots reveal their association in predicting the risk of acute heart failure. With an increase in the feature value of acute myocardial infarction, a significant rise in SHAP values is observed, especially at higher feature values of acute myocardial infarction, where we see a cluster of yellow dots in the upper right corner of the graph. These yellow dots represent higher values of anemia, implying that in the context of high values of acute myocardial infarction, anemia's predictive contribution to the risk of acute heart failure increases. Conversely, when the feature values of acute myocardial infarction are lower, the dots, mostly shown in purple and concentrated in the lower left corner of the graph, represent a smaller predictive contribution to heart failure risk. This pattern indicates that anemia has a lesser predictive impact on the risk of heart failure in patients with a lower degree of myocardial infarction. Thus, the scatter plot shows that the interaction between acute myocardial infarction and anemia in predicting the risk of acute heart failure is non-linear and modulated by the combined influence of these two feature values. Discussion Hip fractures are a common type of fracture among the elderly population, significantly impacting the quality of life of patients. Therefore, timely surgical intervention is crucial for restoring normal life functions and independence in patients [1]. However, the occurrence of acute heart failure (AHF) preoperatively in elderly patients with hip fractures is a common and serious complication. This complication not only increases surgical risks but may also prolong hospital stays, elevate medical costs, and make postoperative recovery more challenging, even leading to patient mortality[5; 12]. In elderly patients with hip fractures, predicting the risk of preoperative acute heart failure (AHF) is key to improving patient outcomes and reducing medical costs [8]. Recent studies have shown that predictive models constructed using multivariate logistic regression models and machine learning methods can identify high-risk patients. It has been found that advanced age (≥70 years), hypertension, anemia, hypoalbuminemia, and surgical duration exceeding 120 minutes are risk factors for heart failure in elderly patients with hip fractures. Understanding these risk factors provides important references for the perioperative management of elderly patients with hip fractures[27]. However, we have developed a logistic regression model and five machine learning models through retrospective studies to predict the likelihood of acute heart failure preoperatively in elderly patients with hip fractures and employed SHAP to offer an explanation of feature vector importance and the interactions among vectors in machine learning models, enhancing model transparency and interpretability. This provides clinicians with a quantitative tool to assess the risk of acute heart failure preoperatively in elderly patients with hip fractures, allowing for more targeted preventive and therapeutic measures in preoperative management, thereby improving patient outcomes. With the advent of the big data era, machine learning models have gained increasing attention due to their ability to handle large datasets, identify complex nonlinear relationships, and interactive effects. This technology has shown immense potential in medical fields such as heart failure, where analyzing big data from electronic health records can not only identify subtypes of heart failure but also improve risk prediction, offering possibilities for personalized medicine[17]. A recent article published in The Lancet highlights the advantages of machine learning methods. Amitava Banerjee's team utilized machine learning to classify and predict outcomes of heart failure by analyzing large electronic health record datasets, clarifying classifications of heart failure patients, utilizing polygenic risk scores to measure their relevancy, and explaining potential biological mechanisms between different heart failure subtypes[3]. For the perioperative assessment of patients with hip fractures, a research team developed a predictive model to evaluate the risk of acute heart failure perioperatively. This model is based on multivariate logistic regression analysis, covering factors such as respiratory diseases, history of heart disease, and ASA scores[23]. However, it's noteworthy that our research differs from previous studies as we focus more on predicting acute heart failure preoperatively in elderly patients with hip fractures. Our study found that, in terms of AUC value, LR (0.761) has a higher value compared to GBM (0.757), suggesting that the LR model performs better overall in distinguishing patients with or without AHF. Although there is little difference in accuracy between the two, GBM exhibits higher performance in sensitivity, precision, and F1 score, indicating that GBM is more reliable in identifying true cases and has higher positive predictive accuracy. LR, on the other hand, shows lower performance in these metrics, potentially leading to a higher misdiagnosis rate and lower overall predictive accuracy. Therefore, in clinical research, among models constructed using logistic regression and five machine learning methods, GBM might be more suitable as a diagnostic model for preoperative acute heart failure in elderly patients with hip fractures. However, the specific choice of model still requires comprehensive consideration based on the actual application scenarios and needs. In recent research, SHAP values have played a crucial role in interpreting complex machine learning models in the field of heart failure, helping to identify key predictive factors that could impact patient outcomes[26]. For instance, studies have utilized SHAP values to highlight the importance of different clinical variables in predicting the 3-year all-cause mortality rate among patients with chronic heart failure, providing clinicians with valuable model interpretations[20]. In assessing the risk of acute heart failure preoperatively in elderly patients with hip fractures, our machine learning model combined with SHAP values offers more objective and effective support for clinical decision-making. This approach allows us to quantify the contribution of each clinical feature variable to the prediction model, which is particularly valuable in handling multivariate and complex medical data. In our study, through the analysis combining machine learning models with SHAP, we found that acute myocardial infarction, ventricular arrhythmia, pulmonary infection, and anemia are the four most important feature variables affecting the model's predictions. These factors are closely related to the occurrence of AHF. Acute myocardial infarction, a significant manifestation of cardiovascular disease, directly relates to a sharp decline in cardiac function, which is particularly important in the elderly population as it may exacerbate existing cardiac burdens[13]. Ventricular arrhythmias could be an early warning of insufficient cardiac pump function, and in high-risk populations, it may precede heart failure [16]. Pulmonary infections can increase cardiac load, especially in elderly patients with hip fractures requiring high cardiac output, potentially exacerbating existing cardiac conditions [6]. Anemia, by reducing oxygen-carrying capacity, can affect the cardiac oxygenation status, thereby increasing the cardiac workload [2]. Through multiple variable partial dependence plot analyses of feature variables in our study, we observed that although each variable contributes uniquely to the risk prediction of heart failure, they are not isolated. There may be interactions among them, meaning the presence of certain variable combinations could increase or decrease the risk of AHF. For example, an anemic condition could exacerbate the risk of heart failure caused by arrhythmias. This finding is consistent with other studies, such as Richard J. and colleagues, who found that patients with chronic kidney disease (CKD) and end-stage renal disease (ESRD) often have anemia and electrolyte imbalances, which may promote electrical instability, induce reentrant arrhythmias, and ultimately lead to congestive heart failure or even induce sudden cardiac death[9]. By identifying these key predictive factors, we can better understand and interpret the results of model predictions. These insights remind clinicians to promptly identify and focus on these key risk factors when assessing the surgical risk of elderly patients with hip fractures. More targeted strengthening of cardiac protection and monitoring, optimizing preoperative management strategies, improving overall treatment effectiveness, improving long-term prognosis, and preventing adverse events are advised. Limitations Although this study has constructed a risk prediction model for preoperative acute heart failure in elderly patients with hip fractures using machine learning methods, it still faces several limitations. First, the selection and scope of samples are restricted, as the study is based on the data of elderly femoral fracture patients from a specific hospital. This selection may limit the general applicability of the study results, especially under different regional and medical conditions. Second, this retrospective study may have missed some patients with heart failure not included in the sample, introducing a certain bias in sample selection. Third, the interpretability of machine learning models remains a concern. Despite the increased interpretability through SHAP analysis, machine learning models are often considered "black box" models, which may limit their application in clinical decision-making. Fourth, there may be important predictive variables not included in the model that could significantly affect the risk of heart failure. Fifth, although the study was divided into training and validation sets and cross-validation was performed, it still belongs to internal validation without external validation, which is a limitation. Conclusion In summary, we have constructed a prediction model for preoperative acute heart failure in elderly patients with hip fractures using LR and five machine learning methods, among which GBM exhibited the best performance in terms of AUC, sensitivity, precision, and F1 score. Additionally, the application of SHAP analysis has enhanced the interpretability of the model, providing clinicians with an effective assessment method, significantly improving the scientific accuracy and precision of preoperative evaluation and decision-making by clinicians. Our research not only offers a new methodological perspective but also brings new thoughts and exploration directions to the fields of heart failure and orthopedic research, demonstrating the significant role of the big data era in advancing medical science development. Declarations Data availability The datasets utilized in the present study are contained within the internal network of the Third Hospital of Hebei Medical University. Due to existing data privacy policies, these datasets are not publicly accessible. However, they can be made available from the corresponding author upon reasonable request. Acknowledgments We are grateful to all those who took part in or assisted with this study project. Ethics approval and consent to participate The ethical review board of the Third Hospital of Hebei Medical University evaluated and sanctioned this research protocol, ensuring adherence to the Helsinki Declaration. The approval was granted under the reference number 2021–087-1. Due to the retrospective nature of data gathering in this study, the board also provided a waiver for informed consent. Prior to analysis, all patient data were anonymized to protect privacy. Consent for publication Not applicable. Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Funding None Author contributions QLY conceived of the study and drafted the manuscript. MMF gathered and processed the data. ZQW and ZYH supervision, and revised the manuscript. All authors contributed to the article and approved the submitted version. References Alexiou, K.I., Roushias, A.,Varitimidis, S.E., (2018). 'Quality of life and psychological consequences in elderly patients after a hip fracture: a review'. Clinical Interventions in Aging, 13:143-150. Anand, I.S., (2008). 'Anemia and Chronic Heart Failure'. JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY, 52 (7):501-511. Banerjee, A., Dashtban, A.,Chen, S., (2023). 'Identifying subtypes of heart failure from three electronic health record sources with machine learning: an external, prognostic, and genetic validation study'. The Lancet. Digital health, 5 (6):e370-e379. Boddaert, J., Raux, M.,Khiami, F., (2014). 'Perioperative management of elderly patients with hip fracture'. ANESTHESIOLOGY, 121 (6):1336-1341. Carbone, L., Buzkova, P.,Fink, H.A., (2010). 'Hip fractures and heart failure: findings from the Cardiovascular Health Study'. EUROPEAN HEART JOURNAL, 31 (1):77-84. Drozd, M., Garland, E.,Walker, A.M.N., (2020). 'Infection-Related Hospitalization in Heart Failure With Reduced Ejection Fraction'. Circulation: Heart Failure, 13 (5). Duceppe, E., Patel, A.,Chan, M., (2020). 'Preoperative N-Terminal Pro-B-Type Natriuretic Peptide and Cardiovascular Events After Noncardiac Surgery: A Cohort Study'. ANNALS OF INTERNAL MEDICINE, 172 (2):96-104. Fu, M., Zhang, Y.,Guo, J., (2022). 'Application of integrated management bundle incorporating with multidisciplinary measures improved in-hospital outcomes and early survival in geriatric hip fracture patients with perioperative heart failure: a retrospective cohort study'. AGING CLINICAL AND EXPERIMENTAL RESEARCH, 34 (5):1149-1158. Glassock, R.J., Pecoits-Filho, R.,Barberato, S.H., (2009). 'Left Ventricular Mass in Chronic Kidney Disease and ESRD'. Clinical Journal of the American Society of Nephrology, 4 (Supplement 1):S79-S91. Gullberg, B., Johnell, O.,Kanis, J.A., (1997). 'World-wide projections for hip fracture'. OSTEOPOROSIS INTERNATIONAL, 7 (5):407-413. Ibrahim, N.E.,Januzzi, J.L., (2018). 'Established and Emerging Roles of Biomarkers in Heart Failure'. CIRCULATION RESEARCH, 123 (5):614-629. Kamijikkoku, S.,Yoshimura, Y., (2023). 'Concurrent Negative Impact of Undernutrition and Heart Failure on Functional and Cognitive Recovery in Hip Fracture Patients'. Nutrients, 15 (22):4800. Kochar, A., Doll, J.A.,Liang, L., (2022). 'Temporal Trends in Post Myocardial Infarction Heart Failure and Outcomes Among Older Adults'. JOURNAL OF CARDIAC FAILURE, 28 (4):531-539. Mahesh, T.R., Dhilip Kumar, V.,Vinoth Kumar, V., (2022). 'AdaBoost Ensemble Methods Using K-Fold Cross Validation for Survivability with the Early Detection of Heart Disease'. Computational Intelligence and Neuroscience, 2022:9005211-9005278. McDonagh, T.A., Metra, M.,Adamo, M., (2021). '2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure'. EUROPEAN HEART JOURNAL, 42 (36):3599-3726. Mcmurray, J., (2000). 'Beta-blockers, ventricular arrhythmias, and sudden death in heart failure: not as simple as it seems'. EUROPEAN HEART JOURNAL, 21 (15):1214-1215. Mohammad, M.A., (2023). 'Advancing heart failure research using machine learning'. The Lancet. Digital health, 5 (6):e331-e332. R., M.,R., R., (2016). 'LASSO: A feature selection technique in predictive modeling for machine learning'. 2016 IEEE International Conference on Advances in Computer Applications (ICACA) , 18-20. Shameer, K., Johnson, K.W.,Glicksberg, B.S., (2018). 'Machine learning in cardiovascular medicine: are we there yet?'. HEART, 104 (14):1156-1164. Sun, Z., Dong, W.,Shi, H., (2022). 'Comparing Machine Learning Models and Statistical Models for Predicting Heart Failure Events: A Systematic Review and Meta-Analysis'. Frontiers in Cardiovascular Medicine, 9:812276. Taleb, I., Kyriakopoulos, C.P.,Fong, R., (2024). 'Machine Learning Multicenter Risk Model to Predict Right Ventricular Failure After Mechanical Circulatory Support: The STOP-RVF Score'. JAMA Cardiology, 9 (3):272-282. Tewari, P., Sweeney, J.B.F.,Lemos, J.L., (2022). 'Evaluation of Systemwide Improvement Programs to Optimize Time to Surgery for Patients With Hip Fractures: A Systematic Review'. JAMA Network Open, 5 (9):e2231911. Tian, M., Li, W.,Wang, Y., (2023). 'Risk factors for perioperative acute heart failure in older hip fracture patients and establishment of a nomogram predictive model'. Journal of Orthopaedic Surgery and Research, 18 (1). Tibshirani, R., (1996). 'Regression Shrinkage and Selection Via the Lasso'. Journal of the Royal Statistical Society: Series B (Methodological), 58 (1):267-288. Van Calster, B., Wynants, L.,Verbeek, J.F.M., (2018). 'Reporting and Interpreting Decision Curve Analysis: A Guide for Investigators'. EUROPEAN UROLOGY, 74 (6):796-804. Wang, K., Tian, J.,Zheng, C., (2021). 'Interpretable prediction of 3-year all-cause mortality in patients with heart failure caused by coronary heart disease based on machine learning and SHAP'. COMPUTERS IN BIOLOGY AND MEDICINE, 137:104813. You, F., Ma, C.,Sun, F., (2021). 'The risk factors of heart failure in elderly patients with hip fracture: what should we care'. BMC MUSCULOSKELETAL DISORDERS, 22 (1):832. Tables TABLE 1 Baseline clinical characteristics of hip fracture patients classified by acute heart failure Variables Total(N=2631) Non-acute heart failure(N=1743) Acute heart failure(N=888) p-value Gender, N (%) Male 766(29.1%) 534(30.6%) 232(26.1%) 0.016 Female 1865(70.9%) 1209(69.4%) 656(73.9%) Age, mean±SD (years) 79.3±7.7 78.2±7.8 81.6±7.0 ＜ 0.001 Age group, N (%) ＜ 75years 746(28.4%) 588(33.7%) 158(17.8%) ＜ 0.001 ≥ 75 years 1885(71.6%) 1155(66.3%) 730(82.2%) Admission time ＜ 48hours 1813(68.9%) 1212(69.5%) 601(67.7%) 0.331 ≥ 48hours 818(31.1%) 531(30.5%) 287(32.3%) Comorbidity N (%) Hypertension Yes 1370(52.1%) 900(51.6%) 470(52.9%) 0.53 No 1261(47.9%) 843(48.4%) 418(47.1%) Old cerebral infarction Yes 1207(45.9%) 782(44.9%) 425(47.9%) 0.145 No 1424(54.1%) 961(55.1%) 463(52.1%) Coronary heart disease Yes 705(26.8%) 426(24.4%) 279(31.4%) ＜ 0.001 No 1926(73.2%) 1317(55.1%) 609(68.6%) Diabetes Yes 662(25.2%) 439(25.2%) 223(25.1%) 0.967 No 1969(74.8%) 1304(74.8%) 665(74.9%) COPD Yes 300(11.4%) 173(9.9%) 127(4.3%) 0.001 No 2331(88.6%) 1570(90.1%) 761(85.7%) Cancer Yes 107(4.1%) 68(3.9%) 39(4.4%) 0.547 No 2524(95.9%) 1675(96.1%) 849(95.6%) Arrhythmia Yes 175(6.7%) 102(5.9%) 73(8.2%) 0.021 No 2456(93.3%) 1641(94.1%) 815(91.8%) Complications Pulmonary infection Yes 376(14.3%) 198(11.4%) 178(20.0%) ＜ 0.001 No 2255(85.7%) 1545(88.6%) 710(80.0%) Ventricular arrhythmia Yes 424(16.1%) 212(12.2%) 212(23.9%) ＜ 0.001 NO 2207(83.9%) 1531(87.8%) 676(76.1%) Acute myocardial infarction Yes 172(6.5%) 53(3.0%) 119(13.4%) ＜ 0.001 No 2459(93.5%) 1690(97.0%) 769(86.6%) Acute cerebrovascular disease Yes 250(9.5%) 150(8.6%) 100(11.3%) 0.029 No 2381(90.5%) 1593(91.4%) 778(88.7%) Stress hyperglycemia Yes 38(1.4%) 20(1.1%) 18(2.0%) 0.084 No 2593(98.6%) 1723(98.9%) 870(98.0%) Stress ulcer Yes 27(1.0%) 19(1.1%) 8(0.9%) 0.838 No 2604(99.0%) 1724(98.9%) 880(99.1%) Urinary tract infection Yes 115(4.4%) 65(3.7%) 50(5.6%) 0.027 No 2516(95.6%) 1678(96.3%) 838(94.4%) Values are presented as mean±standard deviation, median (interquartile range), or number (percentage) as appropriate, SD Standard deviation, COPD Chronic Obstructive Pulmonary Disease TABLE 2 The results of univariate analysis of laboratory data and ultrasound examination Variables Total(N=2631) Non-acute heart failure(N=1743) Acute heart failure(N=888) p-value Anemia Yes 802(30.5%) 476(27.3%) 326(36.7%) ＜ 0.001 No 1829(69.5%) 1267(72 .7 %) 562(63.3% ） Hypokalemia Yes 588(22.3%) 341(19.6%) 247(27.8%) ＜ 0.001 No 2043(77.7%) 1402(80.4%) 641(72.2%) Hyponatremia Yes 739(28.1%) 445(25.5%) 294(33.1%) ＜ 0.001 No 1892(71.9%) 1298(74.5%) 594(66.9%) Hypoalbuminemia Yes 863(32.8%) 534(30.6%) 329(37.0%) 0.001 No 1768(67.2%) 1209(69.4%) 559(63.0%) Lower extremity venous thrombosis Yes 850(32.3%) 577(33.1%) 273(30.7%) 0.234 No 1781(67.7%) 1166(66.9%) 615(69.3%) Values are presented as median (interquartile range), or number (percentage) as appropriate TABLE 3 Prediction factors of preoperative acute heart failure in geriatric patients with hip fracture B Odds ratio P value 95%CI Sex （ Male ） -0.315 0.730 0.014 0.565-0.938 Age （ ≥75 ） 0.673 1.960 ＜ 0.001 1.509-2.563 Coronary heart disease 0.556 1.745 ＜ 0.001 1.374-2.216 Pulmonary infection 0.908 2.480 ＜ 0.001 1.865-3.302 Ventricular arrhythmia 0.839 2.313 ＜ 0.001 1.758-3.304 Acute myocardial infarction 2.058 7.836 ＜ 0.001 4.787-13.044 Anemia 0.442 1.556 ＜ 0.001 1.223-1.977 Hypokalemia 0.496 1.642 ＜ 0.001 1.271-2.120 Hypoalbuminemia 0.588 1.800 ＜ 0.001 1.391-2.333 Constant -2.262 0.104 ＜ 0.001 Table 4. The exact data of importance of all the variables Variables Importance Normalized importance Acute myocardial infarction 100 0.13605442 Ventricular arrhythmia 57 0.07755102 Pulmonary infection 55 0.07482993 Age 48 0.06530612 Anemia 45 0.06122449 Hypokalemia 43 0.0585034 Hypoalbuminemia 42 0.05714286 Coronary heart disease 37 0.05034014 Hypertension 36 0.04897959 Old cerebral infarction 35 0.04761905 Hyponatremia 34 0.0462585 Admission time 33 0.04489796 Sex 32 0.04353741 Lower limb venous thrombosis 29 0.03945578 Diabetes 28 0.03809524 COPD 27 0.03673469 Acute cerebrovascular disease 22 0.02993197 Arrhythmia 16 0.02176871 Urinary tract infection 7 0.00952381 Cancer 6 0.00816327 Stress hyperglycemia 3 0.00408163 Stress ulcer 0 0 Table 5. Comparing the parameters of preoperative acute heart failure models for predicting hip fracture surgery AUC Accuracy Sensitivity Precision F1 RF 0.746 72.0% 87.5% 74.9% 80.7% SVM 0.714 71.7% 90.6% 73.1% 80.9% AdaBoost 0.735 72.6% 87.2% 44.0% 80.8% XGBoost 0.747 73.1% 87.9% 75.9% 81.4% GBM 0.757 73.0% 95.6% 72.7% 82.6% LR 0.761 72.5% 79.5% 24.4% 37.3% AUC area under the curve of ROC, RF r andom Forest, SVM support vector machine, AdaBoost adaptive boosting, XGBoost extreme gradient boosting, GBM gradient boosting machine, LR logistic regression; Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 23 Apr, 2025 Read the published version in BMC Geriatrics → Version 1 posted Editorial decision: Revision requested 28 Oct, 2024 Reviews received at journal 01 Oct, 2024 Reviews received at journal 23 Sep, 2024 Reviewers agreed at journal 23 Sep, 2024 Reviewers agreed at journal 23 Sep, 2024 Reviewers invited by journal 05 Jul, 2024 Editor assigned by journal 18 Jun, 2024 Editor invited by journal 17 Apr, 2024 Submission checks completed at journal 17 Apr, 2024 First submitted to journal 16 Apr, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4274769","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":292255811,"identity":"cd9aeec6-5ff2-471b-baf1-72924ddb5b41","order_by":0,"name":"Qili Yu","email":"","orcid":"","institution":"Third Hospital of Hebei Medical University","correspondingAuthor":false,"prefix":"","firstName":"Qili","middleName":"","lastName":"Yu","suffix":""},{"id":292255813,"identity":"77c86152-5ad7-4824-a1e7-4838231bebaa","order_by":1,"name":"Mingming Fu","email":"","orcid":"","institution":"Third Hospital of Hebei Medical University","correspondingAuthor":false,"prefix":"","firstName":"Mingming","middleName":"","lastName":"Fu","suffix":""},{"id":292255815,"identity":"4a05e9ca-14f5-4b7c-9452-ea58bd2ac60e","order_by":2,"name":"Zhiyong Hou","email":"","orcid":"","institution":"Third Hospital of Hebei Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhiyong","middleName":"","lastName":"Hou","suffix":""},{"id":292255816,"identity":"d84322ea-dbee-4fdf-9f66-a3739b199a6f","order_by":3,"name":"Zhiqian Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9ElEQVRIiWNgGAWjYBACefnHBw4k/pOQ45dgYIMIHSCgxbAhLfHBBzYbY8kZxGphOJBjbDiDLS1xww1itTA2HEuT5uE5nLj5dvOxRzfbGOT4biQwfi7Ao4WdsfmYNI/EYeNtd46lG+e2MRhL3khglp6Bz5ZmNqAtBodlt93IMZMGagG6MIGNmQefy47xmEnzJBxm3Dwj/xtISz1hLWd4gN4/kKa4QSKHDaQlwYCQFmBoJT742GBjLHEjzdw455yE4cwzD5ul8WmRl2AGRmUDMCpnJD97nFNmI893PPngZ7wOQwMSDKCAJ0HDKBgFo2AUjAJsAAD32k9JiwdroAAAAABJRU5ErkJggg==","orcid":"","institution":"Third Hospital of Hebei Medical University","correspondingAuthor":true,"prefix":"","firstName":"Zhiqian","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2024-04-16 09:13:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4274769/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4274769/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12877-025-05920-x","type":"published","date":"2025-04-23T15:57:36+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":55320194,"identity":"31608ab7-81a3-41d2-b3ea-6a4d33bc6f80","added_by":"auto","created_at":"2024-04-25 16:06:07","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1346531,"visible":true,"origin":"","legend":"\u003cp\u003eThe patient flow chart in our study\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/389848208cf75d1aa116c0b0.png"},{"id":55318850,"identity":"b1a48e27-df9f-419a-a028-5ec7dd0902b1","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":207409,"visible":true,"origin":"","legend":"\u003cp\u003eData statistics and clinical feature selection using the LASSO binary logistic regression model.\u003c/p\u003e\n\u003cp\u003e(A) Optimal parameter (lambda) selection in the LASSO model used fivefold cross-validation via minimum criteria. The partial likelihood deviance (binomial deviance) curve was plotted versus log(lambda). (B) LASSO coefficient profiles of the 22 features. A coefficient profile plot was produced against the log(lambda) sequence.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/6f2c449619e95d1095c7a913.png"},{"id":55318856,"identity":"90cb6014-d3d0-4dad-9d9d-501e2ba8a0be","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1309578,"visible":true,"origin":"","legend":"\u003cp\u003eForest plot showing the relationship between risk factors and the occurrence of preoperative acute heart failure in elderly patients with hip fracture\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/e2023f7d007ba42fc142c5cd.png"},{"id":55318852,"identity":"290b06cb-c039-48e9-bf05-0e628df3376b","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":252138,"visible":true,"origin":"","legend":"\u003cp\u003eA nomogram model for predicting the occurrence of preoperative acute heart failure in elderly patients with hip fractures\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/ca92a2f15063ec3cf11fcb1f.png"},{"id":55318854,"identity":"06e98459-f918-471d-81d2-b026ca0e929d","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1146468,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration curves of the acute heart failure nomogram prediction in the cohort.\u003c/p\u003e\n\u003cp\u003ePanel A shows the calibration curve for the training dataset, and Panel B shows the curve for the test dataset.\u003c/p\u003e\n\u003cp\u003eThe x-axis represents the predicted acute heart failure risk.\u003c/p\u003e\n\u003cp\u003eThe y-axis represents the actual diagnosed acute heart failure. The diagonal dotted line represents a perfect prediction by an ideal model. The solid line represents the performance of the nomogram, of which a closer fit to the diagonal dotted line represents a better prediction.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/28a23d40d3c281310c5c2bcb.png"},{"id":55318857,"identity":"b222ac86-36bb-463f-a75f-f046b8ff98e9","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":250292,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of the ROC curve for the predictive values of preoperative acute heart conditions. The blue curve represents the ROC for the training set, with an area under the curve (AUC) of 0.761 (95% CI: 0.740–0.786), illustrating the model's performance on the dataset used for model development. The red curve represents the ROC for the validation set, with an AUC of 0.767 (95% CI: 0.723–0.799), indicating the model's performance on a separate dataset used to test the model. The dashed diagonal line represents the line of no discrimination, which a purely random classifier would achieve. The closer the ROC curve is to the top left corner, the higher the test's overall accuracy\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/c636e9e4d6906b404f7b1b60.png"},{"id":55318853,"identity":"541a2252-ec72-4a41-8da9-e14cdb2a14b0","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":264688,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curve analysis (DCA) and Clinical Impact Curves (CIC) for the acute heart failure nomogram.\u003c/p\u003e\n\u003cp\u003eA and B depict the DCA for the training and test datasets respectively, with the y-axis measuring net benefit. The blue line in each represents the performance of the acute heart failure risk nomogram. The grey solid line assumes all patients have acute heart failure, and the grey dashed line assumes no patients have the condition.\u003c/p\u003e\n\u003cp\u003eC and D show the CIC for the training and test datasets respectively, with the y-axis indicating the number of patients. In C and D, the solid blue line represents high-risk patients as identified by the nomogram, and the dashed red line indicates the actual patients with heart failure. These graphs suggest that the nomogram provides a positive net benefit for clinical decision-making within a probability threshold range.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/63bf733302bbd5dda5be6784.png"},{"id":55318861,"identity":"c5e51023-36fe-4cc7-91c2-e10d3f87100d","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":2206718,"visible":true,"origin":"","legend":"\u003cp\u003eVariable Importance and Correlation Matrix from Preprocessed Data in Machine Learning Model Analysis.\u003c/p\u003e\n\u003cp\u003eA displays the variable importance scores, with the most significant features for the model's standardization being Acute Myocardial Infarction, Ventricular Arrhythmia, Pulmonary Infection, and Anemia.\u003c/p\u003e\n\u003cp\u003eB shows the correlation matrix of the variables, with red indicating a strong positive correlation, blue a strong negative correlation, and white indicating no correlation. These visualizations provide the relationships between different clinical variables.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/03831e654be358631ac6af8b.png"},{"id":55320195,"identity":"cb0174f2-9e32-40f0-b056-60e94832c3c9","added_by":"auto","created_at":"2024-04-25 16:06:07","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":185064,"visible":true,"origin":"","legend":"\u003cp\u003eReceiver Operating Characteristic (ROC) curves for various machine learning models in the evaluation of the dataset.\u003c/p\u003e\n\u003cp\u003eThe curves compare the sensitivity (true positive rate) and 1 - specificity (false positive rate) across different thresholds for Random Forest (RF), Support Vector Machine (SVM), AdaBoost, Extreme Gradient Boosting (XGBoost), and Gradient Boosting Machine (GBM). Area Under the Curve (AUC) values are displayed in the legend, with GBM showing the highest AUC of 0.757.\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/c3edd20b25a89f26ab6de218.png"},{"id":55320196,"identity":"abcb844d-e3d3-4c8a-b9d3-427a419c4466","added_by":"auto","created_at":"2024-04-25 16:06:07","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":261927,"visible":true,"origin":"","legend":"\u003cp\u003eSHAP Value Analysis for Predictive Modeling of Acute Heart Failure in Elderly Patients with Hip Fractures.\u003c/p\u003e\n\u003cp\u003eThis Feature Importance Plot visualizes the impact of individual features on the prediction of acute heart failure risk. Each dot represents an observation, plotted against its SHAP value on the x-axis. The direction and magnitude of these SHAP values indicate whether the feature increases (positive value) or decreases (negative value) the risk of acute heart failure according to the model. The color gradient signifies the value of the feature, ranging from low (purple) to high (yellow).\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/0fa1bd0ada6794b8d85fb2ad.png"},{"id":55318859,"identity":"a25966c1-cee6-481f-b9cc-00eefc25e11f","added_by":"auto","created_at":"2024-04-25 15:58:07","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":2059886,"visible":true,"origin":"","legend":"\u003cp\u003eSHAP Value Distributions for Individual Predictive Analysis Across Four Patients (A-D).\u003c/p\u003e\n\u003cp\u003eThese plots display the influence of various clinical features on the model's prediction of acute heart failure risk for each patient. In each subfigure, the x-axis represents the SHAP value, indicating the impact level of each feature. Features with higher SHAP values contribute more significantly to the prediction. Across all patients, Acute Myocardial Infarction is consistently the most influential variable with the highest SHAP values, indicating a strong association with increased heart failure risk.\u003c/p\u003e","description":"","filename":"floatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/04e027f86d2b022f2bcd00e0.png"},{"id":55320197,"identity":"864163ef-aa91-49ba-938a-8cf0a7894065","added_by":"auto","created_at":"2024-04-25 16:06:07","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":303061,"visible":true,"origin":"","legend":"\u003cp\u003eMultivariate Dependence Plots Demonstrating Feature Interactions in Acute Heart Failure Risk Prediction.\u003c/p\u003e\n\u003cp\u003eEach plot illustrates the relationship between a specific feature and SHAP values, which quantify the impact on the model’s output. The color gradient, from purple to yellow, shows the value of one feature relative to another, with yellow indicating higher values. The plots reveal non-linear interactions between features, indicating complex relationships that are crucial for understanding the model's predictions.\u003c/p\u003e","description":"","filename":"floatimage13.png","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/041f49290f96962399f8a726.png"},{"id":81569838,"identity":"d4c7a856-5b25-4636-829a-9aecb37f0f2d","added_by":"auto","created_at":"2025-04-28 16:11:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":12673854,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4274769/v1/cc19030f-8fd4-4883-a5a7-c25040ebec62.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Elucidating predictors of preoperative acute heart failure in elderly patients with hip fractures through machine learning and SHAP analysis: a retrospective cohort study","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWith the acceleration of global population aging, hip fractures in the elderly have become a significant public health challenge. Since records began in 1990, over 1.6\u0026nbsp;million people worldwide have suffered from hip fractures. It is predicted that, over time, especially among the elderly, the incidence of hip fractures will show a gradually increasing trend. By 2050, the number of individuals affected is expected to rise to at least 4.5\u0026nbsp;million [10; 22]. These fractures not only increase the mortality and disability rates of patients, severely affecting the quality of life, but also impose a significant burden on the healthcare system, including the costs of surgical treatment, long-term rehabilitation, and the subsequent socio-economic burdens. In particular, the risk of acute heart failure before surgery in elderly patients with hip fractures has significantly increased, becoming a key complication closely associated with high mortality rates and prolonged hospital stays, further exacerbating the risk of postoperative complications, including infections and re-fractures.[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Therefore, developing effective prediction and prevention strategies is crucial for improving the treatment outcomes of this patient group.\u003c/p\u003e \u003cp\u003eIn clinical practice, we have observed that most surgeons tend to overlook the assessment of heart failure biomarkers such as Brain Natriuretic Peptide (BNP) or N-Terminal pro-B-Type Natriuretic Peptide (NT-proBNP) in the preoperative evaluation of elderly patients with hip fractures. This oversight could miss patients who have developed acute heart failure, thus failing to intervene timely in this potentially high-risk state[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Machine learning offers a new perspective and approach by analyzing vast amounts of patient data to predict complications that may arise after a hip fracture, such as preoperative acute heart failure, thereby providing a scientific basis for clinical decision-making, optimizing patient management strategies, and reducing the incidence of adverse events[19; 21]. This study utilizes machine learning methods and SHAP values aimed at precisely predicting the risk of acute heart failure before surgery in elderly patients with hip fractures. By analyzing clinical data to reveal the complex associations between patient characteristics, laboratory test results, and preoperative complications, this research offers a new perspective and method. It not only enhances the accuracy of predictions but also provides actionable data support for doctors, optimizing patient management strategies, and reducing the occurrence of adverse events.\u003c/p\u003e \u003cp\u003eTherefore, the establishment and use of this study's model can alert physicians to conduct a more comprehensive preoperative assessment, including the measurement of BNP or NT-proBNP, thus identifying those high-risk patients. Such an integrated preoperative approach can not only reduce surgical risks and postoperative complications but also shorten hospital stays and potentially lower mortality rates. It provides a safer and more effective treatment plan for elderly patients with hip fractures, significantly improving their prognosis and ultimately achieving the goal of improving the clinical outcomes of elderly patients with hip fractures.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study Design and Patients\u003c/h2\u003e \u003cp\u003eThis retrospective study selected inpatients who underwent hip surgery at the Department of Geriatric Orthopedics, Hebei Medical University Third Hospital, from January 2018 to December 2022, as the study subjects. The inclusion criteria were: (1) aged 65 years and older; (2) hip fractures confirmed by radiographic examinations such as X-rays; (3) patients with complete medical records, laboratory test results, and other necessary medical documents. Exclusion criteria were lack of complete medical records, laboratory test results, or other necessary medical documents, and patients who did not meet the diagnostic criteria for hip fractures.\u003c/p\u003e \u003c/div\u003e\u003cp\u003e\u003cstrong\u003e2.2 Ethical Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study, based on the retrospective analysis of existing case data, ensured that all patient data collection and analysis were conducted anonymously to protect patient privacy. Furthermore, the study was in compliance with the Declaration of Helsinki and had been approved and supported by the Institutional Review Board of Hebei Medical University Third Hospital (Approval No.: 2021-087-1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3 Disease Definition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAcute Heart Failure (AHF) is a condition where there is a sudden decrease in the heart\u0026apos;s ability to pump blood, leading to the body\u0026apos;s circulatory volume being insufficient to meet metabolic demands.\u0026nbsp;According to the European Society of Cardiology, AHF is caused by acute changes in the structure or function of the heart, accompanied by increased filling pressures and/or a significant reduction in ejection fraction. Common symptoms include shortness of breath, pulmonary congestion, and inadequate organ perfusion.\u0026nbsp;A key biochemical marker for diagnosing AHF is a significant increase in serum BNP or NT-proBNP levels\u0026nbsp;[15].\u0026nbsp;Different thresholds of BNP and NT-proBNP are used for diagnosing AHF to accommodate patients of varying ages. Specifically, the diagnostic threshold for BNP is \u0026ge;300 pg/mL for patients of all ages. For NT-proBNP, the thresholds are age-stratified: \u0026gt;450 pg/mL for patients under 55 years of age; \u0026gt;900 pg/mL for those between 55 to 75 years; and \u0026gt;1800 pg/mL for patients 75 years and older. These elevated markers are instrumental in confirming the diagnosis of AHF[11].\u0026nbsp;In clinical practice, differentiating between an acute exacerbation of heart failure and chronic heart failure is key, focusing on changes in clinical symptoms and acute variations in BNP or NT-proBNP levels. Through this approach, physicians can more accurately diagnose acute heart failure, thereby providing appropriate treatment for patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.4 Data Collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective study is based on data collected from patients who underwent hip surgery in the Department of Geriatric Orthopedics at Hebei Medical University Third Hospital, from January 2018 to December 2022. Data on patients prior to surgery were gathered through the medical record system, including details on heart failure status, sex, age, admission time, and comorbidities such as hypertension, Old cerebral infarction, coronary artery disease, diabetes, chronic obstructive pulmonary disease (COPD), cancer, arrhythmias, pulmonary infections, ventricular arrhythmias, acute myocardial infarction, acute cerebrovascular disease, stress hyperglycemia, stress ulcers, urinary tract infections, anemia, hypokalemia, hyponatremia, hypoalbuminemia, and lower limb venous thrombosis. By analyzing this data, we aim to gain a deeper understanding of the risk factors for acute heart failure in elderly patients with hip fractures before surgery, thereby providing precise intervention suggestions for clinical practice.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5 Model Establishment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5.1 Dataset Configuration and Variable Selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, we utilized the logistic regression method to predict the occurrence of acute heart failure in elderly patients with hip fractures before surgery. Initially, the collected data were divided into training and validation sets at a ratio of 7:3 to ensure the adequacy of the training process and the independence of the evaluation process. To identify risk factors significantly associated with acute heart failure and avoid overfitting, we employed the Least Absolute Shrinkage and Selection Operator (LASSO) method for selection\u0026nbsp;[18; 24].\u0026nbsp;Furthermore, we utilized a multivariate logistic regression model to assess the relationship between these factors and acute heart failure, ensuring that only statistically significant predictors were included in the final model. By constructing a nomogram of the model, we made the prediction outcomes and the contributions of various variables both intuitive and easy to understand.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5.2 Multimodel Development and Validation Process\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBeyond the basic logistic regression model, we delved into five additional machine learning algorithms, including Random Forest (RF), Support Vector Machine (SVM), Adaptive Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost) and gradient boosting machine (GBM). The models were trained using a 5-fold cross-validation method, which helped us to more accurately evaluate their performance on unseen data\u0026nbsp;[14].\u0026nbsp;Given the imbalance between positive and negative samples in our dataset, we incorporated the Synthetic Minority Over-sampling Technique (SMOTE) focusing on the minority class samples at the boundary to improve the sample distribution and optimize model performance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5.3 Model Interpretability and Comprehensive Evaluation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring the model evaluation phase, we not only focus on the model\u0026apos;s discriminative ability, assessed by calculating the Area Under the Curve (AUC) value, but also on the model\u0026apos;s calibration through the Hosmer-Lemeshow test. Additionally, Decision Curve Analysis (DCA) is applied to compute the net benefit at different thresholds, comprehensively evaluating the model\u0026apos;s practical value in clinical decision-making\u0026nbsp;[25].\u0026nbsp;The Clinical Impact Curve (CIC) is utilized to visualize the benefit values brought by different thresholds. To enhance the model\u0026apos;s interpretability, we employed SHapley Additive exPlanations (SHAP) analysis, which demonstrates the contribution of different variables at the individual level. Furthermore, every observation in the dataset can be explicated with designated SHAP values.\u003c/p\u003e\n\u003cp\u003eThrough the aforementioned strategy, our goal is to construct a model that is both precise and highly interpretable, providing a powerful tool for clinicians. This will enable them to identify the risk of acute heart failure before surgery more promptly when treating elderly patients with hip fractures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.6 Statistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, our aim was to reveal the relationship between acute heart failure and various risk factors in elderly patients with hip fractures. Initially, we analyzed the baseline information of participants through descriptive statistics. The normality of continuous variables was verified using the Kolmogorov-Smirnov test. Variables fitting a normal distribution were described in terms of mean \u0026plusmn; standard deviation, while non-normally distributed data were represented by median and interquartile range. The distribution characteristics of categorical variables were presented in frequencies and percentages. Variance Inflation Factor (VIF) and tolerance were calculated to assess potential collinearity among parameters, with a VIF below 5 and tolerance above 0.1 considered as standards indicating no significant collinearity. All statistical analyses were conducted using SPSS 24.0 and R language. The level of statistical significance was set at P \u0026lt; 0.05.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e3.1 Patient Baseline Characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBetween January 2018 and December 2022, a total of 4,170 elderly patients with hip fractures were included in our study. After screening, 1,539 patients were excluded, leaving 2,631 patients in the final analysis. The excluded patients comprised 1,077 with non-hip fractures, 328 non-surgical patients, and 134 with incomplete data (Figure 1).\u003c/p\u003e\n\u003cp\u003eTable 1 presents the baseline clinical characteristics of the overall sample and compares those between the acute heart failure (AHF) group and the non-AHF group among elderly patients with hip fractures. Overall, the mean age of the patients was 79.3\u0026plusmn;7.7, with 766 males (29.1%) and 1,865 females (70.9%). Among them, 888 patients (33.7%) experienced acute heart failure before surgery. There were statistically significant differences in gender distribution, age, and age groups (\u0026lt;75 years and \u0026ge;75 years) between the two groups (p\u0026lt;0.05). Regarding comorbidities, the prevalence of coronary artery disease and arrhythmias was significantly higher in the AHF group compared to the non-AHF group (p\u0026lt;0.05). Additionally, preoperative complications such as pulmonary infection, ventricular arrhythmias, and acute myocardial infarction also showed a higher incidence in the AHF group, with significant statistical differences (p\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Univariate Analysis of Laboratory Data and Ultrasound Examinations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 2 displays the preoperative laboratory and lower limb venous ultrasound characteristics of elderly patients with hip fractures. The incidences of anemia, hypokalemia, hyponatremia, and hypoalbuminemia were significantly higher in the AHF group compared to the non-AHF group, showing significant statistical differences (p\u0026lt;0.05). However, there was no significant difference in the incidence of lower limb venous thrombosis between the two groups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Development and Validation of Nomograms\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUsing R, patients were randomly divided into a training set and a test set in a 7:3 ratio, with 1,843 patients in the training set and 788 in the test set. Initial analysis with LASSO regression on the training set selected 17 variables out of 22 (Figures 2A and 2B). Subsequent multivariable logistic regression analysis identified gender, age, coronary heart disease, pulmonary infection, ventricular arrhythmia, acute myocardial infarction, anemia, hypokalemia, and hypoalbuminemia as independent risk factors for the occurrence of acute heart failure before surgery in elderly patients with hip fractures (Tables 3 and Figure 3). Based on these independent risk factors, we developed a nomogram model to predict the probability of pre-surgical acute heart failure in elderly patients with hip fractures (Figure 4). The predictive model is given by Logit(P) = -2.262 - 0.315 \u0026times; Sex + 0.673 \u0026times; Age + 0.556 \u0026times; Coronary heart disease + 0.908 \u0026times; Pulmonary infection + 0.839 \u0026times; Ventricular arrhythmia + 2.058 \u0026times; Acute myocardial infarction + 0.442 \u0026times; Anemia + 0.496 \u0026times; Hypokalemia + 0.588 \u0026times; Hypoalbuminemia. The variance inflation factor (VIF) was calculated for each variable in the model, indicating all predictor variables had VIF values well below the threshold of 5, specifically: Sex 1.01, Age 1.01, Coronary heart disease 1.01, Pulmonary infection 1.01, Ventricular arrhythmia 1.02, Acute myocardial infarction 1.01, Anemia 1.11, Hypokalemia 1.02, Hypoalbuminemia 1.12.\u003c/p\u003e\n\u003cp\u003eThe nomogram was evaluated through 1,000 bootstrap resampling, and the results showed that the calibration curve deviated only slightly from the perfect prediction line, indicating good agreement between the model\u0026apos;s predictions and the actual observations (Figure 5). Comparing the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) between the training and validation datasets, the AUC for the training set was 0.761 (0.740-0.786), and for the test set, it was 0.767 (0.723-0.799) (Figure 6). Moreover, the nomogram model\u0026apos;s corrected C-statistic obtained through bootstrap resampling was 0.776, demonstrating good performance in internal validation. This means that the model has strong discriminative ability and can accurately predict the risk of acute heart failure in patients. Decision Curve Analysis (DCA) indicates significant clinical decision-making value with a probability range of 8%-90% in the training set (Figure 7A) and 9%-86% in the validation set (Figure 7B). Additionally, the Clinical Impact Curve (CIC) demonstrates the effect of different threshold settings on the number of patients predicted by the model (Figures 7C and 7D). This further suggests the model has substantial application potential, especially in predicting the risk of acute heart failure in elderly patients after hip fractures. The model provides a powerful tool to more precisely predict the likelihood of acute heart failure, thereby guiding clinicians towards more appropriate preventative and therapeutic measures. Implementing clinical interventions based on this model\u0026apos;s predictions can effectively optimize patient management, likely leading to positive impacts on patient health outcomes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4 Development of Predictive Models Using Machine Learning Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll raw data was preprocessed prior to being input into the machine learning model, including cleaning and transformation steps, to ensure data integrity and high quality for accurate handling and analysis by the machine learning algorithms. The features with the highest importance scores in standardization were Acute Myocardial Infarction, Ventricular Arrhythmia, Pulmonary Infection, and Anemia (Figure 8A and Table 4). Correlations between variables were also calculated and are displayed in Figure (Figure 8B).\u003c/p\u003e\n\u003cp\u003eModels were evaluated using various machine learning methods, with the Area Under the Curve (AUC) values obtained as follows: RF 0.746 (0.710\u0026mdash;0.782), SVM 0.714 (0.676-0.752), AdaBoost 0.735 (0.699-0.772), XGBoost 0.747 (0.711-0.783), GBM 0.757 (0.721 - 0.792), with GBM showing the best AUC among the models (Figure 9). Accuracy, sensitivity, precision, and F1 score were also calculated for each model, with GBM showing the best performance in terms of accuracy (73%) and sensitivity (95.6%) (Table 5).\u003c/p\u003e\n\u003cp\u003eSHAP analysis was conducted to understand the impact of multiple features on the predictive model for acute heart failure in elderly patients with hip fractures before surgery (Figure 10). The Feature Importance Plot shows each observation as a dot, with the SHAP value on the x-axis indicating the impact of the feature on the model\u0026apos;s output. Positive values indicate contributions that increase risk, while negative values indicate contributions that decrease risk. The color gradient from purple to yellow represents feature values from low to high. It is observed that the SHAP values for Acute Myocardial Infarction are distributed in the positive region, with several higher positive points indicating that the presence of acute myocardial infarction significantly increases the risk of acute heart failure. Conversely, Hyponatremia shows both positive and negative SHAP values, concentrated near zero, suggesting a relatively small or individual-dependent impact on the prediction. However, the SHAP values for COPD are mainly in the negative region, possibly indicating a lower risk of acute heart failure in patients with COPD in this model. Through individual-level predictive behavior analysis using the SHAP algorithm, the model revealed key variables influencing the risk of acute heart failure for four patients, showing the contribution of each factor to the prediction and identifying Acute Myocardial Infarction as the main variable affecting all patients. Its SHAP value was significantly higher than other features, and we also found that Anemia, Ventricular Arrhythmia, and Pulmonary Infection play important roles in increasing the risk of heart failure (Figures 11A-D). The SHAP values of these variables provide positive contributions, reinforcing their importance in risk assessment, consistent with the overall trends in the Feature Importance Plot.\u003c/p\u003e\n\u003cp\u003eBy constructing multivariate dependence plots (Figure 12), we suggest interactions between variable features, such as between Acute Myocardial Infarction and Anemia, where scatter plots reveal their association in predicting the risk of acute heart failure. With an increase in the feature value of acute myocardial infarction, a significant rise in SHAP values is observed, especially at higher feature values of acute myocardial infarction, where we see a cluster of yellow dots in the upper right corner of the graph. These yellow dots represent higher values of anemia, implying that in the context of high values of acute myocardial infarction, anemia\u0026apos;s predictive contribution to the risk of acute heart failure increases. Conversely, when the feature values of acute myocardial infarction are lower, the dots, mostly shown in purple and concentrated in the lower left corner of the graph, represent a smaller predictive contribution to heart failure risk. This pattern indicates that anemia has a lesser predictive impact on the risk of heart failure in patients with a lower degree of myocardial infarction. Thus, the scatter plot shows that the interaction between acute myocardial infarction and anemia in predicting the risk of acute heart failure is non-linear and modulated by the combined influence of these two feature values.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eHip fractures are a common type of fracture among the elderly population, significantly impacting the quality of life of patients. Therefore, timely surgical intervention is crucial for restoring normal life functions and independence in patients\u0026nbsp;[1].\u0026nbsp;However, the occurrence of acute heart failure (AHF) preoperatively in elderly patients with hip fractures is a common and serious complication. This complication not only increases surgical risks but may also prolong hospital stays, elevate medical costs, and make postoperative recovery more challenging, even leading to patient mortality[5; 12].\u0026nbsp;In elderly patients with hip fractures, predicting the risk of preoperative acute heart failure (AHF) is key to improving patient outcomes and reducing medical costs\u0026nbsp;[8].\u0026nbsp;Recent studies have shown that predictive models constructed using multivariate logistic regression models and machine learning methods can identify high-risk patients. It has been found that advanced age (\u0026ge;70 years), hypertension, anemia, hypoalbuminemia, and surgical duration exceeding 120 minutes are risk factors for heart failure in elderly patients with hip fractures. Understanding these risk factors provides important references for the perioperative management of elderly patients with hip fractures[27].\u0026nbsp;However, we have developed a logistic regression model and five machine learning models through retrospective studies to predict the likelihood of acute heart failure preoperatively in elderly patients with hip fractures and employed SHAP to offer an explanation of feature vector importance and the interactions among vectors in machine learning models, enhancing model transparency and interpretability. This provides clinicians with a quantitative tool to assess the risk of acute heart failure preoperatively in elderly patients with hip fractures, allowing for more targeted preventive and therapeutic measures in preoperative management, thereby improving patient outcomes.\u003c/p\u003e\n\u003cp\u003eWith the advent of the big data era, machine learning models have gained increasing attention due to their ability to handle large datasets, identify complex nonlinear relationships, and interactive effects. This technology has shown immense potential in medical fields such as heart failure, where analyzing big data from electronic health records can not only identify subtypes of heart failure but also improve risk prediction, offering possibilities for personalized medicine[17].\u0026nbsp;A recent article published in The Lancet highlights the advantages of machine learning methods. Amitava Banerjee\u0026apos;s team utilized machine learning to classify and predict outcomes of heart failure by analyzing large electronic health record datasets, clarifying classifications of heart failure patients, utilizing polygenic risk scores to measure their relevancy, and explaining potential biological mechanisms between different heart failure subtypes[3].\u0026nbsp;For the perioperative assessment of patients with hip fractures, a research team developed a predictive model to evaluate the risk of acute heart failure perioperatively. This model is based on multivariate logistic regression analysis, covering factors such as respiratory diseases, history of heart disease, and ASA scores[23].\u003c/p\u003e\n\u003cp\u003eHowever, it\u0026apos;s noteworthy that our research differs from previous studies as we focus more on predicting acute heart failure preoperatively in elderly patients with hip fractures. Our study found that, in terms of AUC value, LR (0.761) has a higher value compared to GBM (0.757), suggesting that the LR model performs better overall in distinguishing patients with or without AHF. Although there is little difference in accuracy between the two, GBM exhibits higher performance in sensitivity, precision, and F1 score, indicating that GBM is more reliable in identifying true cases and has higher positive predictive accuracy. LR, on the other hand, shows lower performance in these metrics, potentially leading to a higher misdiagnosis rate and lower overall predictive accuracy. Therefore, in clinical research, among models constructed using logistic regression and five machine learning methods, GBM might be more suitable as a diagnostic model for preoperative acute heart failure in elderly patients with hip fractures. However, the specific choice of model still requires comprehensive consideration based on the actual application scenarios and needs.\u003c/p\u003e\n\u003cp\u003eIn recent research, SHAP values have played a crucial role in interpreting complex machine learning models in the field of heart failure, helping to identify key predictive factors that could impact patient outcomes[26].\u0026nbsp;For instance, studies have utilized SHAP values to highlight the importance of different clinical variables in predicting the 3-year all-cause mortality rate among patients with chronic heart failure, providing clinicians with valuable model interpretations[20].\u0026nbsp;In assessing the risk of acute heart failure preoperatively in elderly patients with hip fractures, our machine learning model combined with SHAP values offers more objective and effective support for clinical decision-making. This approach allows us to quantify the contribution of each clinical feature variable to the prediction model, which is particularly valuable in handling multivariate and complex medical data.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn our study, through the analysis combining machine learning models with SHAP, we found that acute myocardial infarction, ventricular arrhythmia, pulmonary infection, and anemia are the four most important feature variables affecting the model\u0026apos;s predictions. These factors are closely related to the occurrence of AHF. Acute myocardial infarction, a significant manifestation of cardiovascular disease, directly relates to a sharp decline in cardiac function, which is particularly important in the elderly population as it may exacerbate existing cardiac burdens[13].\u0026nbsp;Ventricular arrhythmias could be an early warning of insufficient cardiac pump function, and in high-risk populations, it may precede heart failure\u0026nbsp;[16].\u0026nbsp;Pulmonary infections can increase cardiac load, especially in elderly patients with hip fractures requiring high cardiac output, potentially exacerbating existing cardiac conditions\u0026nbsp;[6].\u0026nbsp;Anemia, by reducing oxygen-carrying capacity, can affect the cardiac oxygenation status, thereby increasing the cardiac workload\u0026nbsp;[2].\u003c/p\u003e\n\u003cp\u003eThrough multiple variable partial dependence plot analyses of feature variables in our study, we observed that although each variable contributes uniquely to the risk prediction of heart failure, they are not isolated. There may be interactions among them, meaning the presence of certain variable combinations could increase or decrease the risk of AHF. For example, an anemic condition could exacerbate the risk of heart failure caused by arrhythmias. This finding is consistent with other studies, such as Richard J. and colleagues, who found that patients with chronic kidney disease (CKD) and end-stage renal disease (ESRD) often have anemia and electrolyte imbalances, which may promote electrical instability, induce reentrant arrhythmias, and ultimately lead to congestive heart failure or even induce sudden cardiac death[9]. By identifying these key predictive factors, we can better understand and interpret the results of model predictions. These insights remind clinicians to promptly identify and focus on these key risk factors when assessing the surgical risk of elderly patients with hip fractures. More targeted strengthening of cardiac protection and monitoring, optimizing preoperative management strategies, improving overall treatment effectiveness, improving long-term prognosis, and preventing adverse events are advised.\u003c/p\u003e"},{"header":"Limitations","content":"\u003cp\u003eAlthough this study has constructed a risk prediction model for preoperative acute heart failure in elderly patients with hip fractures using machine learning methods, it still faces several limitations. First, the selection and scope of samples are restricted, as the study is based on the data of elderly femoral fracture patients from a specific hospital. This selection may limit the general applicability of the study results, especially under different regional and medical conditions. Second, this retrospective study may have missed some patients with heart failure not included in the sample, introducing a certain bias in sample selection. Third, the interpretability of machine learning models remains a concern. Despite the increased interpretability through SHAP analysis, machine learning models are often considered \u0026quot;black box\u0026quot; models, which may limit their application in clinical decision-making. Fourth, there may be important predictive variables not included in the model that could significantly affect the risk of heart failure. Fifth, although the study was divided into training and validation sets and cross-validation was performed, it still belongs to internal validation without external validation, which is a limitation.\u003c/p\u003e\n"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, we have constructed a prediction model for preoperative acute heart failure in elderly patients with hip fractures using LR and five machine learning methods, among which GBM exhibited the best performance in terms of AUC, sensitivity, precision, and F1 score. Additionally, the application of SHAP analysis has enhanced the interpretability of the model, providing clinicians with an effective assessment method, significantly improving the scientific accuracy and precision of preoperative evaluation and decision-making by clinicians. Our research not only offers a new methodological perspective but also brings new thoughts and exploration directions to the fields of heart failure and orthopedic research, demonstrating the significant role of the big data era in advancing medical science development.\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets utilized in the present study are contained within the internal network of the Third Hospital of Hebei Medical University. Due to existing data privacy policies, these datasets are not publicly accessible. However, they can be made available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to all those who took part in or assisted with this study project.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe ethical review board of the Third Hospital of Hebei Medical University evaluated and sanctioned this research protocol, ensuring adherence to the Helsinki Declaration. The approval was granted under the reference number 2021\u0026ndash;087-1. Due to the retrospective nature of data gathering in this study, the board also provided a waiver for informed consent. Prior to analysis, all patient data were anonymized to protect privacy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eQLY conceived of the study and drafted the manuscript. MMF gathered and processed the data. ZQW and ZYH supervision, and revised the manuscript. All authors contributed to the article and approved the submitted version.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAlexiou, K.I., Roushias, A.,Varitimidis, S.E., (2018). \u0026apos;Quality of life and psychological consequences in elderly patients after a hip fracture: a review\u0026apos;. Clinical Interventions in Aging, 13:143-150.\u003c/li\u003e\n\u003cli\u003eAnand, I.S., (2008). \u0026apos;Anemia and Chronic Heart Failure\u0026apos;. JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY, 52 (7):501-511.\u003c/li\u003e\n\u003cli\u003eBanerjee, A., Dashtban, A.,Chen, S., (2023). \u0026apos;Identifying subtypes of heart failure from three electronic health record sources with machine learning: an external, prognostic, and genetic validation study\u0026apos;. The Lancet. Digital health, 5 (6):e370-e379.\u003c/li\u003e\n\u003cli\u003eBoddaert, J., Raux, M.,Khiami, F., (2014). \u0026apos;Perioperative management of elderly patients with hip fracture\u0026apos;. ANESTHESIOLOGY, 121 (6):1336-1341.\u003c/li\u003e\n\u003cli\u003eCarbone, L., Buzkova, P.,Fink, H.A., (2010). \u0026apos;Hip fractures and heart failure: findings from the Cardiovascular Health Study\u0026apos;. EUROPEAN HEART JOURNAL, 31 (1):77-84.\u003c/li\u003e\n\u003cli\u003eDrozd, M., Garland, E.,Walker, A.M.N., (2020). \u0026apos;Infection-Related Hospitalization in Heart Failure With Reduced Ejection Fraction\u0026apos;. Circulation: Heart Failure, 13 (5).\u003c/li\u003e\n\u003cli\u003eDuceppe, E., Patel, A.,Chan, M., (2020). \u0026apos;Preoperative N-Terminal Pro-B-Type Natriuretic Peptide and Cardiovascular Events After Noncardiac Surgery: A Cohort Study\u0026apos;. ANNALS OF INTERNAL MEDICINE, 172 (2):96-104.\u003c/li\u003e\n\u003cli\u003eFu, M., Zhang, Y.,Guo, J., (2022). \u0026apos;Application of integrated management bundle incorporating with multidisciplinary measures improved in-hospital outcomes and early survival in geriatric hip fracture patients with perioperative heart failure: a retrospective cohort study\u0026apos;. AGING CLINICAL AND EXPERIMENTAL RESEARCH, 34 (5):1149-1158.\u003c/li\u003e\n\u003cli\u003eGlassock, R.J., Pecoits-Filho, R.,Barberato, S.H., (2009). \u0026apos;Left Ventricular Mass in Chronic Kidney Disease and ESRD\u0026apos;. Clinical Journal of the American Society of Nephrology, 4 (Supplement 1):S79-S91.\u003c/li\u003e\n\u003cli\u003eGullberg, B., Johnell, O.,Kanis, J.A., (1997). \u0026apos;World-wide projections for hip fracture\u0026apos;. OSTEOPOROSIS INTERNATIONAL, 7 (5):407-413.\u003c/li\u003e\n\u003cli\u003eIbrahim, N.E.,Januzzi, J.L., (2018). \u0026apos;Established and Emerging Roles of Biomarkers in Heart Failure\u0026apos;. CIRCULATION RESEARCH, 123 (5):614-629.\u003c/li\u003e\n\u003cli\u003eKamijikkoku, S.,Yoshimura, Y., (2023). \u0026apos;Concurrent Negative Impact of Undernutrition and Heart Failure on Functional and Cognitive Recovery in Hip Fracture Patients\u0026apos;. Nutrients, 15 (22):4800.\u003c/li\u003e\n\u003cli\u003eKochar, A., Doll, J.A.,Liang, L., (2022). \u0026apos;Temporal Trends in Post Myocardial Infarction Heart Failure and Outcomes Among Older Adults\u0026apos;. JOURNAL OF CARDIAC FAILURE, 28 (4):531-539.\u003c/li\u003e\n\u003cli\u003eMahesh, T.R., Dhilip Kumar, V.,Vinoth Kumar, V., (2022). \u0026apos;AdaBoost Ensemble Methods Using K-Fold Cross Validation for Survivability with the Early Detection of Heart Disease\u0026apos;. Computational Intelligence and Neuroscience, 2022:9005211-9005278.\u003c/li\u003e\n\u003cli\u003eMcDonagh, T.A., Metra, M.,Adamo, M., (2021). \u0026apos;2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure\u0026apos;. EUROPEAN HEART JOURNAL, 42 (36):3599-3726.\u003c/li\u003e\n\u003cli\u003eMcmurray, J., (2000). \u0026apos;Beta-blockers, ventricular arrhythmias, and sudden death in heart failure: not as simple as it seems\u0026apos;. EUROPEAN HEART JOURNAL, 21 (15):1214-1215.\u003c/li\u003e\n\u003cli\u003eMohammad, M.A., (2023). \u0026apos;Advancing heart failure research using machine learning\u0026apos;. The Lancet. Digital health, 5 (6):e331-e332.\u003c/li\u003e\n\u003cli\u003eR., M.,R., R., (2016). \u0026apos;LASSO: A feature selection technique in predictive modeling for machine learning\u0026apos;. \u003cem\u003e2016 IEEE International Conference on Advances in Computer Applications (ICACA)\u003c/em\u003e, 18-20.\u003c/li\u003e\n\u003cli\u003eShameer, K., Johnson, K.W.,Glicksberg, B.S., (2018). \u0026apos;Machine learning in cardiovascular medicine: are we there yet?\u0026apos;. HEART, 104 (14):1156-1164.\u003c/li\u003e\n\u003cli\u003eSun, Z., Dong, W.,Shi, H., (2022). \u0026apos;Comparing Machine Learning Models and Statistical Models for Predicting Heart Failure Events: A Systematic Review and Meta-Analysis\u0026apos;. Frontiers in Cardiovascular Medicine, 9:812276.\u003c/li\u003e\n\u003cli\u003eTaleb, I., Kyriakopoulos, C.P.,Fong, R., (2024). \u0026apos;Machine Learning Multicenter Risk Model to Predict Right Ventricular Failure After Mechanical Circulatory Support: The STOP-RVF Score\u0026apos;. JAMA Cardiology, 9 (3):272-282.\u003c/li\u003e\n\u003cli\u003eTewari, P., Sweeney, J.B.F.,Lemos, J.L., (2022). \u0026apos;Evaluation of Systemwide Improvement Programs to Optimize Time to Surgery for Patients With Hip Fractures: A Systematic Review\u0026apos;. JAMA Network Open, 5 (9):e2231911.\u003c/li\u003e\n\u003cli\u003eTian, M., Li, W.,Wang, Y., (2023). \u0026apos;Risk factors for perioperative acute heart failure in older hip fracture patients and establishment of a nomogram predictive model\u0026apos;. Journal of Orthopaedic Surgery and Research, 18 (1).\u003c/li\u003e\n\u003cli\u003eTibshirani, R., (1996). \u0026apos;Regression Shrinkage and Selection Via the Lasso\u0026apos;. Journal of the Royal Statistical Society: Series B (Methodological), 58 (1):267-288.\u003c/li\u003e\n\u003cli\u003eVan Calster, B., Wynants, L.,Verbeek, J.F.M., (2018). \u0026apos;Reporting and Interpreting Decision Curve Analysis: A Guide for Investigators\u0026apos;. EUROPEAN UROLOGY, 74 (6):796-804.\u003c/li\u003e\n\u003cli\u003eWang, K., Tian, J.,Zheng, C., (2021). \u0026apos;Interpretable prediction of 3-year all-cause mortality in patients with heart failure caused by coronary heart disease based on machine learning and SHAP\u0026apos;. COMPUTERS IN BIOLOGY AND MEDICINE, 137:104813.\u003c/li\u003e\n\u003cli\u003eYou, F., Ma, C.,Sun, F., (2021). \u0026apos;The risk factors of heart failure in elderly patients with hip fracture: what should we care\u0026apos;. BMC MUSCULOSKELETAL DISORDERS, 22 (1):832.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eTABLE 1 Baseline clinical characteristics of hip fracture patients classified by acute heart failure\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal(N=2631)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNon-acute heart failure(N=1743)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute heart failure(N=888)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eGender, N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eMale\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e766(29.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e534(30.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e232(26.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.016\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eFemale\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1865(70.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1209(69.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e656(73.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge, mean\u0026plusmn;SD (years)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e79.3\u0026plusmn;7.7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e78.2\u0026plusmn;7.8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e81.6\u0026plusmn;7.0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge group, N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e75years\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e746(28.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e588(33.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e158(17.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026ge;\u003c/strong\u003e\u003cstrong\u003e75 years\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1885(71.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1155(66.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e730(82.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdmission time\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e48hours\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1813(68.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1212(69.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e601(67.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.331\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026ge;\u003c/strong\u003e\u003cstrong\u003e48hours\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e818(31.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e531(30.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e287(32.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eComorbidity N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypertension\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1370(52.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e900(51.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e470(52.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.53\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1261(47.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e843(48.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e418(47.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOld cerebral infarction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1207(45.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e782(44.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e425(47.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.145\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1424(54.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e961(55.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e463(52.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCoronary heart disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e705(26.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e426(24.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e279(31.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1926(73.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1317(55.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e609(68.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiabetes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e662(25.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e439(25.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e223(25.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.967\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1969(74.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1304(74.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e665(74.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCOPD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e300(11.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e173(9.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e127(4.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2331(88.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1570(90.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e761(85.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCancer\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e107(4.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e68(3.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e39(4.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.547\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2524(95.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1675(96.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e849(95.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eArrhythmia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e175(6.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e102(5.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e73(8.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.021\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2456(93.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1641(94.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e815(91.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eComplications\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePulmonary infection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e376(14.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e198(11.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e178(20.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2255(85.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1545(88.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e710(80.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVentricular arrhythmia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e424(16.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e212(12.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e212(23.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNO\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2207(83.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1531(87.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e676(76.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute myocardial infarction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e172(6.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e53(3.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e119(13.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2459(93.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1690(97.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e769(86.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute cerebrovascular disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e250(9.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e150(8.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e100(11.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.029\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2381(90.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1593(91.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e778(88.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eStress hyperglycemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e38(1.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e20(1.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e18(2.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.084\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2593(98.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1723(98.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e870(98.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eStress ulcer\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e27(1.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e19(1.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e8(0.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.838\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2604(99.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1724(98.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e880(99.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eUrinary tract infection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e115(4.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e65(3.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e50(5.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.027\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"40.816326530612244%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2516(95.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1678(96.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.3265306122449%\"\u003e\n \u003cp\u003e\u003cstrong\u003e838(94.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.183673469387756%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eValues are presented as mean\u0026plusmn;standard deviation, median (interquartile range), or number (percentage) as appropriate, SD Standard deviation, COPD Chronic Obstructive Pulmonary Disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"104%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eTABLE 2 The results of univariate analysis of laboratory data and ultrasound examination\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal(N=2631)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNon-acute heart failure(N=1743)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute heart failure(N=888)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAnemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e802(30.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e476(27.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e326(36.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1829(69.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1267(72\u003c/strong\u003e\u003cstrong\u003e.7\u003c/strong\u003e\u003cstrong\u003e%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp; \u0026nbsp;562(63.3%\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypokalemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e588(22.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e341(19.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e247(27.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e2043(77.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1402(80.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e641(72.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHyponatremia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e739(28.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e445(25.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e294(33.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1892(71.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1298(74.5%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e594(66.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypoalbuminemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e863(32.8%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e534(30.6%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e329(37.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1768(67.2%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1209(69.4%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e559(63.0%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eLower extremity venous thrombosis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e850(32.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e577(33.1%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e273(30.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.234\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.61855670103093%\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.402061855670103%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1781(67.7%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"30.927835051546392%\"\u003e\n \u003cp\u003e\u003cstrong\u003e1166(66.9%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.77319587628866%\"\u003e\n \u003cp\u003e\u003cstrong\u003e615(69.3%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.278350515463918%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eValues are presented as median (interquartile range), or number (percentage) as appropriate\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"5\"\u003e\n \u003cp\u003e\u003cstrong\u003eTABLE 3 Prediction factors of preoperative acute heart failure in geriatric patients with hip fracture\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cstrong\u003eB\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cstrong\u003eOdds ratio\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cstrong\u003e95%CI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003eMale\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e-0.315\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.730\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.014\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.565-0.938\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e\u0026ge;75\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.673\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.960\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.509-2.563\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCoronary heart disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.556\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.745\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.374-2.216\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePulmonary infection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.908\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.480\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.865-3.302\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eVentricular arrhythmia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.839\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.313\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.758-3.304\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute myocardial infarction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.058\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e7.836\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.787-13.044\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAnemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.442\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.556\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.223-1.977\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypokalemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.496\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.642\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.271-2.120\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypoalbuminemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.588\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.800\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.391-2.333\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.69387755102041%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eConstant\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.244897959183673%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e-2.262\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.104\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"41.45077720207254%\" colspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 4. The exact data of importance of all the variables\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"58.54922279792746%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eImportance\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eNormalized importance\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute myocardial infarction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.13605442\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVentricular arrhythmia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.07755102\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePulmonary infection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.07482993\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.06530612\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAnemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.06122449\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypokalemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.0585034\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypoalbuminemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.05714286\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCoronary heart disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.05034014\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypertension\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.04897959\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eOld cerebral infarction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.04761905\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHyponatremia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.0462585\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdmission time\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.04489796\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.04353741\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eLower limb venous thrombosis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.03945578\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eDiabetes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.03809524\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCOPD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.03673469\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcute cerebrovascular disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.02993197\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eArrhythmia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.02176871\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eUrinary tract infection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.00952381\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eCancer\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.00816327\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eStress hyperglycemia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0.00408163\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"53.06122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eStress ulcer\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"19.387755102040817%\" valign=\"bottom\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.551020408163264%\" colspan=\"2\" valign=\"bottom\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"100%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"6\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 5. Comparing the parameters of preoperative acute heart failure models for predicting hip fracture surgery\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAccuracy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSensitivity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrecision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003eF1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRF\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.746\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e72.0%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.5%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e74.9%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e80.7%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSVM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.714\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e71.7%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e90.6%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e73.1%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e80.9%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdaBoost\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.735\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e72.6%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.2%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e44.0%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e80.8%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\n \u003cp\u003e\u003cstrong\u003eXGBoost\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.747\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e73.1%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.9%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e75.9%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e81.4%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\n \u003cp\u003e\u003cstrong\u003eGBM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.757\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e73.0%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e95.6%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e72.7%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e82.6%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"22.448979591836736%\"\u003e\n \u003cp\u003e\u003cstrong\u003eLR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.285714285714286%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.761\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e72.5%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.346938775510203%\"\u003e\n \u003cp\u003e\u003cstrong\u003e79.5%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e24.4%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.306122448979592%\"\u003e\n \u003cp\u003e\u003cstrong\u003e37.3%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"100%\" colspan=\"6\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC area under the curve of ROC, RF r\u003c/strong\u003e\u003cstrong\u003eandom Forest, \u0026nbsp;SVM \u0026nbsp;support vector machine, \u0026nbsp;AdaBoost \u0026nbsp; \u0026nbsp; adaptive boosting, \u0026nbsp;XGBoost \u0026nbsp;extreme gradient boosting, GBM gradient boosting machine, LR logistic regression;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Heart failure, Hip fracture, Preoperative, Machine learning, SHAP, Prediction model","lastPublishedDoi":"10.21203/rs.3.rs-4274769/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4274769/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eAcute heart failure has become a significant challenge in elderly patients with hip fractures. Timely identification and assessment of preoperative acute heart failure have become key factors in reducing surgical risks and improving outcomes.\u003c/p\u003e\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eThis study aims to precisely predict the risk of acute heart failure in elderly patients with hip fractures before surgery through machine learning techniques and SHapley Additive exPlanations (SHAP), providing a scientific basis for clinicians to optimize patient management strategies and reduce adverse events.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA retrospective study design was employed, selecting patients admitted for hip surgery in the Department of Geriatric Orthopedics at the Third Hospital of Hebei Medical University from January 2018 to December 2022 as research subjects. Data were analyzed using logistic regression, random forests, support vector machines, AdaBoost, XGBoost, and GBM machine learning methods combined with SHAP analysis to interpret relevant factors and assess the risk of acute heart failure.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eA total of 2,631 patients were included in the final cohort, with an average age of 79.3\u0026thinsp;\u0026plusmn;\u0026thinsp;7.7. 33.7% of patients experienced acute heart failure before surgery. A predictive model for preoperative acute heart failure in elderly hip fracture patients was established through multivariate logistics regression: Logit(P) = -2.262\u0026ndash;0.315 \u0026times; Sex\u0026thinsp;+\u0026thinsp;0.673 \u0026times; Age\u0026thinsp;+\u0026thinsp;0.556 \u0026times; Coronary heart disease\u0026thinsp;+\u0026thinsp;0.908 \u0026times; Pulmonary infection\u0026thinsp;+\u0026thinsp;0.839 \u0026times; Ventricular arrhythmia\u0026thinsp;+\u0026thinsp;2.058 \u0026times; Acute myocardial infarction\u0026thinsp;+\u0026thinsp;0.442 \u0026times; Anemia\u0026thinsp;+\u0026thinsp;0.496 \u0026times; Hypokalemia\u0026thinsp;+\u0026thinsp;0.588 \u0026times; Hypoalbuminemia, with a model nomogram established and an AUC of 0.767 (0.723\u0026ndash;0.799). Predictive models were also established using five machine learning methods, with GBM performing optimally, achieving an AUC of 0.757 (0.721\u0026ndash;0.792). SHAP analysis revealed the importance of all variables, identifying acute myocardial infarction as the most critical predictor and further explaining the interactions between significant variables.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis study successfully developed a predictive model based on machine learning that accurately predicts the risk of acute heart failure in elderly patients with hip fractures before surgery. The application of SHAP enhanced the model's interpretability, providing a powerful tool for clinicians to identify high-risk patients and take appropriate preventive and therapeutic measures in preoperative management.\u003c/p\u003e","manuscriptTitle":"Elucidating predictors of preoperative acute heart failure in elderly patients with hip fractures through machine learning and SHAP analysis: a retrospective cohort study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-25 15:58:02","doi":"10.21203/rs.3.rs-4274769/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-28T15:56:42+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-01T11:15:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-23T21:35:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"185026666070049102230757370209590938471","date":"2024-09-23T15:16:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"41236140423548607919773308770570266322","date":"2024-09-23T07:17:10+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-07-05T07:12:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-06-18T13:52:31+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-04-17T08:52:51+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-04-17T08:50:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Geriatrics","date":"2024-04-16T09:11:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5ec79665-74ce-4903-a159-813a14b9f819","owner":[],"postedDate":"April 25th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-28T16:05:45+00:00","versionOfRecord":{"articleIdentity":"rs-4274769","link":"https://doi.org/10.1186/s12877-025-05920-x","journal":{"identity":"bmc-geriatrics","isVorOnly":false,"title":"BMC Geriatrics"},"publishedOn":"2025-04-23 15:57:36","publishedOnDateReadable":"April 23rd, 2025"},"versionCreatedAt":"2024-04-25 15:58:02","video":"","vorDoi":"10.1186/s12877-025-05920-x","vorDoiUrl":"https://doi.org/10.1186/s12877-025-05920-x","workflowStages":[]},"version":"v1","identity":"rs-4274769","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4274769","identity":"rs-4274769","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}