Machine learning models to predict in-hospital mortality in patients with rhabdomyolysis combined with acute kidney injury

Machine learning models to predict in-hospital mortality in patients with rhabdomyolysis combined with acute kidney injury | 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 Machine learning models to predict in-hospital mortality in patients with rhabdomyolysis combined with acute kidney injury Wenyan Zhang, Yamin Liu, Ziling Feng, Ni Xiong, Leyao Tang, Wenhang Chen, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6824634/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background. Rhabdomyolysis-associated acute kidney injury (RI-AKI) is a serious complication in critically ill patients and is associated with increased in-hospital mortality. However, limited research has focused on predictive modeling of in-hospital mortality among this population. Objective. To develop and evaluate machine learning (ML) models for predicting in-hospital mortality in critically ill patients with RI-AKI. Methods. Data were extracted from the MIMIC-IV and eICU Collaborative Research Databases. Patients with RI-AKI were identified, and relevant clinical variables—including demographics, vital signs, laboratory indicators, comorbidities/complications, and treatments within the first 24 hours of ICU admission—were collected. The combined dataset was randomly divided into training and testing sets in an 8:2 ratio. Feature selection was performed using the least absolute shrinkage and selection operator (LASSO) and random forest (RF). ML models were constructed using Extreme Gradient Boosting (XGBoost), RF, and logistic regression (LR). Model performance was assessed by area under the receiver operating characteristic curve (AUC), Brier score, sensitivity, specificity, and calibration. Results. Ten key predictors, including age, sodium, phosphorus, and coagulation markers, were identified. In the training set, the XGBoost model achieved the highest AUC (0.889; 95% CI: 0.872–0.908), outperforming RF (0.797) and LR (0.740). Brier scores were 0.122, 0.185, and 0.203, respectively. Similar results were observed in the testing set. Conclusions. The XGBoost model demonstrated superior performance in predicting in-hospital mortality among critically ill RI-AKI patients, indicating its potential value in clinical risk stratification. Further external validation is warranted. rhabdomyolysis acute kidney injury in-hospital mortality machine learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Highlights 1. Multicenter ICU data used for large-scale modeling and validation in patient outcomes. 2. Comprehensive assessment identifies risk factors for mortality in critical patients. 3. XGBoost outperforms in multiple constructed models. Introduction Symptoms of rhabdomyolysis include muscle ache, muscle debility, and urine of a tea hue, which can be brought on by various factors including trauma, strenuous exercise, seafood consumption, medication side effects (e.g., statins), infections, and poisoning [ 1 – 3 ] . Acute kidney injury linked to rhabdomyolysis (RI-AKI), primarily due to myoglobin's mechanical blockage of renal tubules, ranks as a frequent and grave complication among severely ill individuals with rhabdomyolysis [4]. It accounts for 13–50% of patients with rhabdomyolysis, with its mortality rate being 62% [ 5 ] . Individuals suffering from RI-AKI face a risk of death in the hospital that is 3 to 4 times higher than those not afflicted with AKI [ 6 – 8 ] . As a result, predicting the mortality rates among critically ill RI-AKI patients in hospitals is vital for swift and effective treatment. Machine Learning (ML) serves as an effective method for data analysis, merging statistical methods with computer science to scrutinize vast data sets and reveal concealed details. ML models such as Extreme Gradient Boosting (XGBoost) and Random Forest (RF) account for higher-order non-linear interactions between predictors and are able to automatically reconstruct the relationships between variables and outcomes in large datasets [ 9 ] . Earlier research focusing on ICU patients suffering from sepsis, cardiovascular conditions, and AKI has endeavored to forecast mortality rates within hospitals using ML models, demonstrating their superiority in discrimination, calibration, and overall advantages over conventional models [ 10 , 11 ] . For example, Li et al. [ 12 ] that machine learning models markedly enhanced the precision in predicting in-hospital deaths among patients with ventricular arrhythmias, in contrast to conventional scoring methods. Dong et al. [ 13 ] discovered that the ML model precisely forecasted moderate to severe Acute Kidney Injury (AKI) up to 48 hours sooner than existing diagnostic protocols. Furthermore, an earlier investigation into ICU patients suffering from rhabdomyolysis revealed the superiority of the ML model over both the Acute Physiology Score III and the Sequential Organ Failure Assessment (SOFA) system, facilitating the early detection of patients with a high mortality risk [ 14 ] . However, there is still a limited amount of research focused on predicting mortality rates within hospitals among RI-AKI patients through ML models, which could obstruct the immediate identification of those at a higher risk of death in hospitals Earlier studies focusing solely on patients with rhabdomyolysis or AKI indicated that demographic factors might account for in-hospital death rates (such as age and gender), laboratory parameters (such as serum creatine kinase [CK], creatinine, urea nitrogen, and routine blood markers), and therapeutic requirements (including methods like renal replacement therapy [RRT] and mechanical ventilation) [ 15 – 18 ] . Hospital mortality rates may be linked to coexisting conditions like sepsis and chronic kidney disease (CKD). [ 19 – 21 ] .Consequently, the objective of this research was to create machine learning models for precise forecasting of in-hospital death rates in ICU patients with RI-AKI, by thoroughly evaluating the impact of demographic factors, lab conditions, treatment needs, and concurrent diseases and complications. Methods 2.1 Data Source The research employed two extensive databases, namely the Medical Information Mart for Intensive Care IV (MIMIC IV) v2.0 and the eICU Collaborative Research Database (eICU-CRD) v2.0. The MIMIC IV database, accessible to the public and housed in a single facility, encompasses patient records from every ICU and emergency department entry at Beth Israel Deaconess Medical Center spanning 2008 to 2019 [ 22 ] . The eICU-CRD, a database available at no cost and spanning multiple centers, encompasses information from more than 200,000 patient entries in 335 ICUs across 208 U.S. hospitals, amassed from 2014 to 2015 [ 23 ] . The eICU-CRD, developed subsequent to MIMIC-IV, broadens research scope by integrating data from multiple medical centers [ 24 ] . Therefore, critically ill RI-AKI patients in both databases may exhibit similar clinical characteristics. Before starting this research, the NIH Protecting Human Research Participants online training course was completed, and approval was obtained for extracting data from MIMIC IV for research purposes (Record ID: 58637411). Approval was also given by the Institutional Review Boards of both Beth Israel Deaconess Medical Centre (BIDMC, Boston, Massachusetts, USA) and Massachusetts Institute of Technology (MIT, Cambridge, Massachusetts, USA). In this research, every patient's health data was anonymized before analysis, thereby eliminating the need for written informed consent [ 25 ] . This study was conducted in strict adherence to the guidelines for the development and reporting of machine learning prediction models in biomedical research [ 26 ] . 2.2 Study population The research encompassed adults diagnosed with rhabdomyolysis and Acute Kidney Injury. Criteria for inclusion included:1) a confirmed diagnosis of rhabdomyolysis and AKI via manual examination of the International Classification of Diseases, Ninth Revision (ICD-9) codes;2) initial admission to the ICU; 3)and being over 18 years old. Criteria for exclusion encompassed: 1)maximum CK levels below 1000 U·L-1;2) ICU duration of less than a day;3) or absent data exceeding 25% or lacking outcome variables. 2.3 Data extraction Patient data were extracted from the MIMIC-IV and eICU-CRD databases using Structured Query Language (SQL) with PostgreSQL 9.6, focusing on information collected within the first 24 hours after ICU admission. This research system collects multidimensional clinical data covering demographic characteristics, laboratory indicators, vital signs, comorbidities/complications, and treatment needs. Demographic characteristics include age, gender, and ethnicity; laboratory indicators are divided into five categories: (1) hematological parameters: including white blood cell count, red blood cell count, hemoglobin, platelet count, hematocrit, as well as absolute counts and percentages of neutrophils, lymphocytes, eosinophils, basophils, and monocytes; additionally, red blood cell morphological indicators such as mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), and red cell distribution width (RDW); (2) renal function indicators: including blood urea nitrogen and serum creatinine, used to assess kidney filtration and metabolic function; (3) liver function indicators: such as albumin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (ALP), reflecting the liver's synthetic and metabolic capacity; (4) blood glucose and electrolytes: including blood glucose, potassium, sodium, chloride, calcium, and phosphorus levels, used to monitor metabolic balance and homeostasis; (5) coagulation parameters: such as prothrombin time (PT), international normalized ratio (INR), and activated partial thromboplastin time (APTT), to assess the coagulation system status. Vital signs include temperature, respiratory rate, heart rate, systolic and diastolic blood pressure, to reflect the patient's physiological status in real-time. Comorbidities and complications include coronary heart disease (CHD), diabetes, hypertension, hyperlipidemia, malignant tumors, chronic lung disease, chronic kidney disease (CKD), and sepsis, to analyze the impact of underlying diseases on prognosis. Treatment requirements include renal replacement therapy (RRT), mechanical ventilation, corticosteroid use, diuretics, and vasoactive drugs, to assess disease severity and intervention measures. This comprehensive data framework provides a crucial basis for in-depth analysis of disease characteristics, predicting clinical outcomes, and optimizing treatment strategies. In-hospital mortality served as the result variable, with pertinent information being gathered from a pair of databases. 2.4 Statistical analysis Variables of a categorical nature were displayed in terms of their occurrence rates and proportions, and were analyzed across different groups using the χ 2 test or Fisher's exact test. Continuous variables were summarized as mean (standard deviation) or median (interquartile range), depending on their distribution. Between-group comparisons were conducted using Student's t-test or Mann-Whitney U test. When the percentage of missing data fell below 25%, a method of multiple imputation was utilized to generate five datasets. Specifically, the lack of continuous variable values was replaced with the mean of the estimated data, while missing values for categorical variables were replaced with the mode of the imputed data. Every statistical examination was bidirectional, with a P-value below 0.05 deemed to hold statistical significance. The statistical evaluations were performed utilizing the R software, version 4.3.2. Model Development Information from the MIMIC IV and eICU databases was combined and arbitrarily segmented into training (80% of the sample) and testing groups (20% of the sample). A random oversampling method was utilized to rectify the class disparity (survival/death) present in the training dataset. The following steps were undertaken to select predictors and build the machine learning-based predictive model. Predictors were chosen using the Least Absolute Shrinkage and Selection Operator (LASSO) and RF, with the common variables identified by both techniques serving as the model's predictors. By implementing L1 regularization on regression coefficients, LASSO facilitates the selection of variables and streamlines the model, rendering features with zero coefficients superfluous and removed from the fitting process, thus allowing for automated variable selection [ 27 ] . RF was employed to rank the importance of variables using the Mean Decrease Gini, and the top 25 most important variables were selected. Cross-validation was applied during predictor selection to identify the optimal value corresponding to the minimum cross-validation error. For developing models, three powerful machine learning algorithms were utilized: XGBoost, RF, and logistic regression (LR). An effective Gradient Boosting Decision Tree implementation, XGBoost offers robust performance, fast training, and the capacity to manage big datasets [ 28 ] . RF can train multiple decision trees to address the bias and variance issues of individual models, thereby enhancing the model's accuracy and robustness [ 29 ] . LR is a classical linear model that allows the direct interpretation of individual feature effects through the model’s coefficients [ 30 ] . For both XGBoost and RF models, a grid search technique was utilized to pinpoint the optimal hyperparameter combination, enhancing predictive precision. In an effort to avoid overfitting and enhance the models' applicability, they underwent a 10-fold cross-validation process, leading to the creation of the final model through multiple iterations [ 10 ] . Model evaluation and validation Quantitative assessment of the predictive model's differentiation capabilities was conducted using multiple metrics, including the area under the receiver operating characteristic curve (AUC), Youden's index, accuracy (ACC), sensitivity, specificity, F1 score, positive predictive value (PPV), and negative predictive value (NPV). It was considered acceptable to attain an AUC between 0.7 and 0.8, with values surpassing 0.8 [ 31 ] . Three models' forecasting accuracy was evaluated using the Delong test. Calibration curves were used to assess the predictive accuracy of the model, and the Brier score was calculated to quantify calibration. A lower Brier score (ranging from 0 to 1) indicates better agreement between predicted probabilities and observed outcomes. The SHapley Additive ExPlanations (SHAP) method was utilized to assess each feature's impact on the model's forecasted results and to represent the combined effects during interactions with other features. Enhancing the clarity of ML models was achieved by depicting each feature's incremental impact on an importance ranking graph and illustrating the influence of each feature on the results in a dependency plot. Results 3.1 Baseline characteristics of study participants Figure 1 depicts the method used to select participants for this research. From the MIMIC-IV and eICU-CRD databases, 855 RI-AKI records were gathered, encompassing 141 deaths within hospitals, with a mortality rate of 16.5% (95% CI: 15.1%-17.8%). Randomly, 684 and 171 patients were allocated to the training and test groups, with in-hospital death rates recorded at 15.5% (106/684) and 20.5% (35/171) for the training and test groups, respectively. Table 1 displays the fundamental features of both the training and test datasets. Apart from ALP levels and temperature, the two groups showed no notable disparities. Table 1 Baseline characteristics of the training and test sets. Training set (n = 684) Test set (n = 171) P value Demographic characteristics Age 58.5(43.9,71.9) 57.8(49.0,70.1) 0.842 Female (%) 339(49.6) 84(49.1) 0.918 Ethnicity (%) White 483(70.6) 114(66.7) 0.562 Black 41(6.0) 16(9.4) Asian 6(0.9) 1(0.6) Hispanic 14(2.0) 4(2.3) Other/Unknown 140(20.5) 36(21.0) Laboratory indicators CK (U/L) 8235.5(3034.3,12812.8) 7130.0(2208.0,12812.8) 0.437 WBC (10 9 /L) 13.1(9.0,18.40) 13.1(9.7,18.0) 0.543 RBC (10 12 /L) 4.0(3.5,4.6) 4.0(3.5,4.6) 0.743 Hemoglobin (g/dL) 12.1(10.4,13.8) 12.0(10.5,14.1) 0.923 Platelet (10 9 /L) 191.0(135.3,248.8) 185.0(133.0,250.0) 0.716 Hematocrit (%) 36.7 ± 0.3 36.7 ± 0.5 0.957 Neutrophil (10 9 /L) 11.9(8.0,14.3) 11.9(8.0,13.5) 0.661 Lymphocyte (10 9 /L) 1.2(0.7,1.4) 1.1(0.7,1.3) 0.597 Eosinophil (10 9 /L) 0.1(0,0.1) 0.9(0.2,1.0) 0.398 Basophil (10 9 /L) 0(0,0) 0(0,0) 0.312 Monocyte (10 9 /L) 0.9(0.5,1.1) 0.9(0.6,1.1) 0.303 NEUT% 80.0(79.0,86.6) 80.0(78.6,86.5) 0.918 LYMPH% 9.7(5.0,11.5) 10.0(5.0,11.0) 0.696 EO% 0.9(0.1,0.9) 0.9(0.2,1.0) 0.517 BAS% 0.3(0.2,0.3) 0.3(0.2,0.3) 0.213 MONO% 6.4(4.2,8.0) 6.4(4.0,7.4) 0.955 MCV (fl) 91.0(87.0,95.0) 91.0(87.0,96.2) 0.176 MCH (PG) 30.3(28.8,31.7) 30.4(29.0,32.1) 0.282 MCHC (g/dL) 33.1(32.2,34.1) 33.2(32.3,34.0) 0.968 RDW (%) 14.2(13.5,15.4) 14.4(13.5,15.5) 0.246 BUN (mg/dL) 33.1(32.2,34.1) 31.0(19.0,52.0) 0.134 SCr (µmol/L) 185.6(99.9,324.9) 168.0(106.1,309.4) 0.729 Albumin (g/dL) 3.1(2.7,3.5) 3.1(2.7,3.5) 0.868 AST (U/L) 247.5(99.0,738.8) 209.0(77.0,738.8) 0.681 ALT (U/L) 95.5(43.3,381.2) 108.0(42.0,381.2) 0.673 ALP (U/L) 81.0(59.0,96.7) 91.0(61.0,98.0) 0.038 LDH(U/L) 1274.0(798.3,1274.0) 1274.0(771.0,1274.0) 0.693 BG (mmol/L) 7.3(5.7,9.4) 7.4(5.9,11.1) 0.118 Potassium (mmol/L) 4.4(3.9,5.1) 4.4(3.7,5.3) 0.630 Sodium (mmol/L) 139.0(135.0,142.0) 138.0(134.0,142.0) 0.998 Chlorine (mmol/L) 103.0(99.0,108.0) 103.0(98.0,107.0) 0.687 Calcium (mmol/L) 2.0(1.8,2.2) 2.0(1.9,2.2) 0.239 Phosphorus(mmol/L) 1.5(1.0,1.8) 1.5(1.1,1.7) 0.696 PT (s) 15.0(12.7,16.5) 14.7(12.6,16.5) 0.977 INR 1.2(1.1,1.5) 1.3(1.1,1.5) 0.887 APTT (s) 33.0(27.8,36.6) 33.0(27.0,36.6) 0.865 Vital signs Temperature (℃) 36.8(36.4,37.3) 36.7(36.3,37.2) 0.044 Respiratory rate (times/min) 20.0(17.0,24.0) 20.0(16.0,24.0) 0.258 Heart rate (times/min) 97.1(82.3,110.0) 94.0(83.0,105.0) 0.235 SBP (mmHg) 124.0(105.0,141.0) 123.0(107.0,140.0) 0.689 DBP (mmHg) 71.0(58.0,83.0) 71.0(51.0,86.0) 0.750 Comorbidities Coronary heart disease (%) 86(12.6) 24(14.0) 0.702 Diabetes mellitus (%) 134(19.6) 41(24.0) 0.244 Hypertension (%) 180(26.3) 48(28.1) 0.699 Hypertriglyceridemia (%) 42(6.1) 13(7.6) 0.602 Malignancy(%) 26(3.8) 10(5.8) 0.284 Chronic pulmonary disease (%) 93(13.6) 23(13.4) 1.000 CKD (%) 94(13.7) 20(11.7) 0.531 Sepsis (%) 309(45.2) 76(44.4) 0.932 Treatment requirements RRT (%) 112(16.4) 22(12.9) 0.291 Mechanical ventilation (%) 194(28.4) 47(27.5) 0.850 Glucocorticoid (%) 135(19.6) 41(24.0) 0.245 Diuretic (%) 163(23.8) 39(22.8) 0.841 Vasopressor (%) 210(30.7) 54(31.6) 0.853 CK: serum creatine kinase; WBC: white blood cell; RBC: red blood cell; NEUT%: percentage of neutrophils; LYMPH%: percentage of lymphocytes; EO%: percentage of eosinophils; BAS%: percentage of basophils; MONO%: percentage of monocytes; MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; MCHC: mean corpuscular hemoglobin concentration; RDW: red cell distribution width; BUN: blood urea nitrogen; SCr: serum creatinine; AST: aspartate aminotransferase; ALT: alanine transaminase; BG: blood glucose; PT: prothrombin time; INR: international normalized ratio; APTT: activated partial thromboplastin time ; SBP: systolic blood pressure; DBP: diastolic blood pressure; CKD: chronic kidney disease; RRT: renal replacement therapy. In terms of demographics, such as age and ethnicity, lab parameters, such as MCHC, RDW, albumin, sodium, phosphorus, PT, INR, and APTT, vital statistics, such as respiratory and heart rates, comorbidities/complications, such as CHD and hypertriglyceridemia, and treatment requirements, such as RRT, mechanical ventilation, glucocorticoids, and vasopressors, a notable disparity was observed between the groups with survival and those without in the training dataset(Table 2 ). Table 2 Characteristics between the survival and non-survival groups in the training set. Survival (n = 578) Mortality (n = 106) P value Demographic characteristics Age 57.0 (42.1,69.7) 68.0 (55.8,80.1) < 0.001 Female (%) 295 (51.0) 56 (52.8) 0.464 Ethnicity (%) White 411 (71.1) 72 (67.9) 0.029 Black 38 (6.6) 3 (2.8) Asian 5 (0.9) 1 (1.0) Hispanic 14 (2.4) 0 Other/Unknown 110 (19.0) 30 (28.3) Laboratory indicators CK (U/L) 8497.5 (3185.8,12812.8) 5966.5 (2695.8,12812.8) 0.288 WBC (10 9 /L) 13.0 (9.0,18.1) 13.8 (9.6,19.8) 0.290 RBC (10 12 /L) 4.1 (3.5, 4.6) 3.9 (3.1,4.7) 0.169 Hemoglobin (g/dL) 12.1 (10.6,13.8) 12.0 (9.2,13.9) 0.165 Platelet (10 9 /L) 192.0 (140.8,246.3) 169.0 (103.8,267.3) 0.211 Hematocrit (%) 36.6 (32.1,41.6) 36.1 (28.7,42.0) 0.293 Neutrophil (10 9 /L) 11.9 (8.0,14.4) 11.9 (8.5,13.9) 0.939 Lymphocyte (10 9 /L) 1.2 (0.7,1.5) 1.1 (0.7,1.4) 0.552 Eosinophil (10 9 /L) 0.1 (0.0,0.1) 0.1 (0,0.1) 0.571 Basophil (10 9 /L) 0 (0,0) 0 (0,0) 0.453 Monocyte (10 9 /L) 0.9 (0.5,1.2) 0.8 (0.4,1.1) 0.075 NEUT% 80.0 (79.0,86.0) 80.0 (79.1,88.0) 0.382 LYMPH% 10.0 (5.2,11.7) 7.9 (4.4,10.6) 0.829 EO% 0.9 (0.1,0.9) 0.9 (0.1,0.9) 0.721 BAS% 0.3 (0.2,0.3) 0.3 (0.2,0.3) 0.399 MONO% 6.4 (4.5,8.0) 6.2 (3.5,7.7) 0.076 MCV (fl) 91.0 (87.0,95.0) 92.5 (88.0–97.0) 0.197 MCH (PG) 30.2 (28.7,31.7) 30.5 (29.0,31.7) 0.669 MCHC (g/dL) 33.1 (32.2,34.2) 33.0 (32.0,33.8) 0.004 RDW (%) 14.2 (13.5,15.2) 14.4 (13.4,15.9) 0.005 BUN (mg/dL) 33.0 (19.0,57.0) 35.5 (21.8,66.8) 0.324 SCr (µmol/L) 185.6 (97.2,336.8) 185.6 (116.9,291.7) 0.322 Albumin (g/dL) 3.1 (2.7,3.5) 3.0 (2.4,3.3) 0.004 AST (U/L) 245.0 (93.8,738.8) 271.0 (117.3,738.8) 0.690 ALT (U/L) 94.5 (43.0,381.2) 95.0 (46.0,381.2) 0.622 ALP (U/L) 80.0 (59.0,96.7) 83.5 (58.8-109.3) 0.400 BG (mmol/L) 7.2 (5.7,9.4) 7.3 (5.7,10.5) 0.658 Potassium (mmol/L) 4.4 (3.8,5.1) 4.5 (3.9,5.3) 0.417 Sodium (mmol/L) 138.4 (135.0,141.3) 139.0 (136.0,144.0) 0.026 Chlorine (mmol/L) 103.1 (99.0,108.0) 104.0 (100.0,109.0) 0.147 Calcium (mmol/L) 2.0 (1.8,2.2) 2.0 (1.9–2.2) 0.960 Phosphorus(mmol/L) 1.4 (1.0,1.7) 1.5 (1.1,2.1) 0.014 PT (s) 14.8 (12.6,16.5) 16.5 (13.4,18.6) 0.030 INR 1.3 (1.1,1.5) 1.4 (1.2,1.7) 0.028 * APTT (s) 32.9 (27.5,36.6) 34.7 (29.9,36.6) 0.012 Vital signs Temperature (℃) 36.8 (36.4,37.3) 36.9 (36.4.37.4) 0.967 Respiratory rate (times/min) 20 (16,24) 22 (18,27) 0.040 Heart rate (times/min) 97 (81,110) 101 (90.8,115.8) < 0.001 SBP (mmHg) 124.0 (105.0,141.0) 123.5 (101.8,138.5) 0.573 DBP (mmHg) 71.5 (58.0,84.0) 68.5 (53.8,82.0) 0.101 Comorbidities Coronary heart disease (%) 59 (10.2) 27 (25.5) < 0.001 Diabetes mellitus (%) 117 (20.2) 17 (16.0) 0.316 Hypertension (%) 155 (26.8) 25 (23.6) 0.487 Hypertriglyceridemia (%) 30 (5.2) 12 (11.3) 0.016 Malignancy(%) 20 (3.5) 6 (5.7) 0.416 Chronic pulmonary disease (%) 80 (13.8) 13 (12.3) 0.779 CKD (%) 77 (13.3) 17 (16.0) 0.455 Sepsis (%) 252 (43.6) 57 (53.8) 0.053 Treatment requirements RRT (%) 87 (15.1) 25 (23.6) 0.029 * Mechanical ventilation (%) 150 (26.0) 44 (41.5) < 0.001 * Glucocorticoid (%) 100 (17.3) 35 (33.0) < 0.001 Diuretic (%) 134 (23.2) 29 (27.3) 0.354 Vasopressor (%) 158 (27.3) 52 (49.1) < 0.001 CK: serum creatine kinase; WBC: white blood cell; RBC: red blood cell; NEUT%: percentage of neutrophils; LYMPH%: percentage of lymphocytes; EO%: percentage of eosinophils; BAS%: percentage of basophils; MONO%: percentage of monocytes; MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; MCHC: mean corpuscular hemoglobin concentration; RDW: red cell distribution width; BUN: blood urea nitrogen; SCr: serum creatinine; AST: aspartate aminotransferase; ALT: alanine transaminase; BG: blood glucose; PT: prothrombin time; INR: international normalized ratio; APTT: activated partial thromboplastin time ; SBP: systolic blood pressure; DBP: diastolic blood pressure; CKD: chronic kidney disease; RRT: renal replacement therapy. 3.2 Model development Fifteen feature variables with non-zero coefficients were selected from a total of fifty-eight variables by LASSO regularization (Fig. 2 A, B), and the top 25 most important variables selected by the RF analysis are shown in Fig. 3 . Ten shared variables including age, sodium, albumin, heart rate, phosphorus, MCHC, RDW, INR, APTT, and monocyte count found by both LASSO regularization and RF analysis were considered as the predictors to develop ML models. 3.3 Model evaluation and validation Table 3 displays the performance of ML models across both training and test datasets. Within the training dataset, AUC values for XGBoost, RF, and LR stood at 0.889 (95% CI: 0.872–0.908), 0.797 (95% CI: 0.772–0.822), and 0.740 (95% CI: 0.711–0.768), in that order. XGBoost's AUC values notably surpassed those of the RF and LR models, with both showing statistical significance ( P < 0.001) (Fig. 4 A). Within the test dataset, the AUC values for XGBoost, RF, and LR stood at 0.775 (95% CI: 0.721–0.830), 0.719 (95% CI: 0.625–0.813), and 0.702 (95% CI: 0.639–0.765), in that order (Fig. 4 B). When comparing the XGBoost model with the RF and LR models, the P values recorded were 0.0178 and 0.069, in that order. Figures 4 C and 4 D illustrate the calibration ability for the three models within the training and test datasets, respectively. Regarding precision in calibration, the XGBoost model surpassed the RF and LR models during both its training and test stages. In the dataset used for training, the recorded Brier scores for XGBoost, RF, and LR were 0.122, 0.185, and 0.203, in that order. Within the test dataset, XGBoost, RF, and LR achieved Brier scores of 0.214, 0.222, and 0.228, in that order. Table 3. Discrimination indicators of the predictive models in the training and test sets. AUC (95%CI) Youden index ACC (%) Sensitivity (%) Specificity (%) F1 score PPV (%) NPV (%) Training set XGBoost 0.889(0.872-0.908) 0.616 0.808 0.667 0.677 0.803 0.822 0.795 RF 0.797(0.772-0.822) 0.379 0.689 0.682 0.753 0.713 0.663 0.726 LR 0.740(0.711-0.768) 0.344 0.672 0.667 0.677 0.677 0.687 0.657 Test set XGBoost 0.775(0.721-0.830) 0.331 0.665 0.919 0.544 0.712 0.618 0.778 RF 0.719(0.625-0.813） 0.176 0.588 0.904 0.419 0.659 0.562 0.650 LR 0.702(0.639-0.765) 0.301 0.651 0.904 0.500 0.671 0.634 0.672 XGBoost: extremely gradient; RF: random forest; boosting; LR: logistic regression; AUC: area under the curve; CI: Confidence interval; ACC: accuracy; PPV: positive predictive value; NPV: negative predictive value. In Fig. 5 , the SHAP values are displayed to elucidate how the XGBoost model forecasts in-hospital death rates among critically ill RI-AKI patients in the training dataset. Figure 6 shows the most important feature variables individually through the SHAP dependency graph. Discussion The research, employing ML methods along with MIMIC-IV and eICU data, developed and validated the effectiveness of three models in predicting in-hospital mortality in RI-AKI critically ill patients. The XGBoost model was found to surpass the RF and LR models in both distinguishing and calibrating. Furthermore, factors such as age, sodium, phosphorus, monocyte count, MCHC, RDW, albumin, INR, APTT, and heart rate were pinpointed as predictive elements. As far as we are aware, this research was pioneering in creating ML-based models to forecast in-hospital deaths among critically ill RI-AKI patients. XGBoost is renowned for its expertise and flexibility as a gradient boosting tree algorithm, especially in managing structured data and classification tasks, earning recognition as the leading algorithm in machine learning and prediction competitions on Kaggle.com [ 32 ] . Numerous research works have demonstrated the superiority of the XGBoost algorithm over alternative ML algorithms in forecasting adverse clinical outcomes [ 30 , 33 ] . For instance, Zhang et al. [ 34 ] discovered the XGBoost model to be more efficient in differentiating volume sensitivity in oliguric AKI patients than the traditional logistic regression model, enabling accurate modifications in fluid treatmentl. Dong et al. [ 35 ] demonstrated the superior performance of XGBoost in identifying, calibrating, and clinically predicting mortality risk in patients with septic AKI. Additionally, an extensive meta-analysis showed that XGBoost outperforms LR and a variety of machine learning frameworks, including (artificial neural networks, support vector machines, and Bayesian networks ) in predicting AKI [ 36 ] . This study added significantly to the existing knowledge by supporting the role of XGBoost model in predicting in-hospital mortality in critically ill patients with RI-AKI. Consequently, it is recommended that medical decision-makers employ XGBoost for forecasting in-hospital death rates among critically ill RI-AKI patients in clinical settings The study uncovered a link between advanced age, elevated heart rates, and higher death rates among critically ill RI-AKI patients in the hospital. Age can affect physical condition and organ compensatory capacity, and hence playing an important role in survival outcomes. The susceptibility to the harmful effects of toxins released by muscle destruction may be increased in elderly patients due to the degenerative changes in kidney structure and function [ 37 , 38 ] . Similarly, an elevated heart rate may lead to a higher risk of death through increasing cardiac strain and disrupts hemodynamics [ 39 , 40 ] . Therefore, more attention should be paid to those with increased age and elevated heart rate. The kidneys are crucial for sustaining electrolyte equilibrium, with disruptions in sodium or phosphorus balance frequently occurring in patients with AKI. The study uncovered a link between increased levels of hypernatremia and hyperphosphatemia and increased in-hospital death rates among critically ill RI-AKI patients, aligning with earlier findings. For example, Lindner et al. [ 41 ] found that ICU-acquired hypernatremia independently affected mortality, with a mortality rate of 48% in individuals whose plasma sodium exceeded 150 mmol/L. Atlani et al. [ 42 ] identified hypernatraemia as the only separate risk element for AKI-related mortality in COVID-19 sufferers, noting that those who did not survive had notably elevated median serum sodium levels. Furthermore, Jung et al. [ 43 ] Research revealed that with a rise in serum phosphate levels by 1 mg/dl, the related mortality risk ratios at 28 and 90 days stood at 1.36 (1.20–1.54) and 1.32 (1.17–1.48), in that order. Consequently, vigilant observation of serum phosphate and sodium concentrations in ICU patients suffering from RI-AKI is crucial for categorizing and managing risk. Furthermore, the study uncovered a link between the count of monocytes, MCHC, RDW, albumin, INR, and APTT and the mortality rates in ICU patients with RI-AKI. The number of monocytes, RDW, and MCHC serve as crucial hematological markers linked to the inflammatory reaction. Previous studies showed that in critically sick patients with AKI, the RDW/albumin ratio accurately predicts overall death rates at 1, 3, and 12 months [ 44 ] . RDW reflects the heterogeneity of erythrocyte volume, and a rise in RDW typically correlates with intense inflammation and oxidative stress, potentially leading to a less favorable prognosis and aiding in effective risk categorization for patients with serious illnesses [ 45 , 46 ] . Albumin serves as an essential indicator of nutritional health and is crucial in sustaining colloid osmotic pressure, along with blood and cell stability. Additionally, reduced levels of albumin have been recognized as a standalone risk element for both the emergence and fatality of AKI [ 47 , 48 ] . Research indicates that upon admission, both INR and APTT are indicative of death risk in sepsis-related AKI patients, where higher INR and extended APTT correlate with higher in-hospital death rates in AKI sufferers [ 49 , 50 ] . Consequently, these indicators could act as focal points for interventions aimed at lowering in-hospital deaths among ICU patients suffering from RI-AKI. The strengths of this study included multiple ICU centres, a substantial number of participants, and a comprehensive assessment of possible predictors. However, several limitations should be acknowledged. Firstly, there was uncertainty regarding the external applicability due to the lack of external validation. Additionally, research indicates that the etiology of rhabdomyolysis could affect the prognosis of patients suffering from RI-AKI [ 51 ] , the absence of such data precluding exploring its role in this study.Ultimately, the prospective indicators examined in this research were evaluated within a day of admission. Consequently, the forecasting accuracy of variables assessed outside this timeframe is still ambiguous. Conclusions The research established and confirmed the effectiveness of three ML models in forecasting in-hospital death rates among critically ill individuals with RI-AKI. In contrast to the LR and RF models, the XGBoost model demonstrated superior precision in both differentiation and calibration, suggesting its potential value in clinical decision-making. Ten key predictors including age, sodium, phosphorus, and coagulation parameters were identified. Therefore, those with increased age and elevated heart rate should be given special consern, and sodium, phosphorus, and coagulation parameters may be served as the intervention targets for reducing hospital deaths among ICU patients suffering from RI-AKI. Additional validation from outside sources remains essential in clinical settings. Declarations ACKNOWLEDGEMENTS We express our thanks to the Medical Information Mart for Intensive Care IV (MIMIC IV) v2.0 and the eICU Collaborative Research Database (eICU-CRD) for their comprehensive data provision. Ethics approval declaration: This study protocol was reviewed and approved by Ethics Committee of Xiangya Hospital, Central South University, approval number 202403060. Consent to participate statement In this research, every patient's health data was anonymized before analysis, thereby eliminating the need for written informed consent and was approved by the Ethics Committee of Xiangya Hospital, Central South University. This study was conducted in strict adherence to the guidelines for the development and reporting of machine learning prediction models in biomedical research. CONFLICT OF INTEREST STATEMENT The writers affirm the absence of any conflicting interests. FUNDING The research received funding through a grant awarded by Hunan Province's Natural Science Foundation (2021JJ40972). AUTHORS’ CONTRIBUTIONS Wenyan Zhang: conceptualization; methodology; data acquisition and management; formal analysis; visualization; writing—original draft; writing—review and editing. Yamin Liu: methodology; data acquisition and management; software; writing-reviewing and editing. Ziling Feng: methodology; project administration; supervision; validation; writing-reviewing and editing. Ni Xiong: data acquisition and management; writing—review and editing; validation. Leyao Tang: formal analysis; methodology; software; writing-reviewing and editing. Xu Zhu: formal analysis; methodology; software; writing-reviewing and editing. Jing Xue: conceptualization, project administration,validation, visualization, writing-reviewing and editing. Wenhang Chen: conceptualization; methodology; writing-reviewing and editing; supervision; formal analysis and validation; funding acquisition; project administration. Wenjie Dai: methodology; writing-reviewing and editing; supervision; formal analysis and validation. All authors read and approved the final manuscript. DATA AVAILABILITY STATEMENT Should there be a justified request, the researchers are able to supply the supporting data for this study's findings. References Chavez L O, Leon M, Einav S, Varon J. Beyond muscle destruction: a systematic review of rhabdomyolysis for clinical practice [J]. Crit Care, 2016, 20(1): 135.http://doi.org/10.1186/s13054-016-1314-5 Tawhari M, Aldalaan A, Alanazi R, Aldharman S, Alnafisah T, Alawad N, et al. Clinical presentation and outcomes of patients with rhabdomyolysis: A tertiary care center experience [J]. Saudi Med J, 2024, 45(5): 510-7.http://doi.org/10.15537/smj.2024.45.5.20230560 Pei P, Li X Y, Lu S S, Liu Z, Wang R, Lu X C, Lu K. The Emergence, Epidemiology, and Etiology of Haff Disease [J]. Biomed Environ Sci, 2019, 32(10): 769-78.http://doi.org/10.3967/bes2019.096 Al-Ismaili Z, Piccioni M, Zappitelli M. Rhabdomyolysis: pathogenesis of renal injury and management [J]. Pediatr Nephrol, 2011, 26(10): 1781-8.http://doi.org/10.1007/s00467-010-1727-3 Xiao L, Ran X, Zhong Y, Le Y, Li S. Serum creatine kinase levels are not associated with an increased need for continuous renal replacement therapy in patients with acute kidney injury following rhabdomyolysis [J]. Ren Fail, 2022, 44(1): 893-901.http://doi.org/10.1080/0886022x.2022.2079523 McMahon G M, Zeng X, Waikar S S. A Risk Prediction Score for Kidney Failure or Mortality in Rhabdomyolysis [J]. JAMA Internal Medicine, 2013, 173(19): 1821-7.http://doi.org/10.1001/jamainternmed.2013.9774 Yang C W, Li S, Dong Y, Paliwal N, Wang Y. Epidemiology and the Impact of Acute Kidney Injury on Outcomes in Patients with Rhabdomyolysis [J]. J Clin Med, 2021, 10(9).http://doi.org/10.3390/jcm10091950 Simpson J P, Taylor A, Sudhan N, Menon D K, Lavinio A. Rhabdomyolysis and acute kidney injury: creatine kinase as a prognostic marker and validation of the McMahon Score in a 10-year cohort: A retrospective observational evaluation [J]. Eur J Anaesthesiol, 2016, 33(12): 906-12.http://doi.org/10.1097/eja.0000000000000490 Li F, Xin H, Zhang J, Fu M, Zhou J, Lian Z. Prediction model of in-hospital mortality in intensive care unit patients with heart failure: machine learning-based, retrospective analysis of the MIMIC-III database [J]. BMJ Open, 2021, 11(7): e044779.http://doi.org/10.1136/bmjopen-2020-044779 Li M, Han S, Liang F, Hu C, Zhang B, Hou Q, Zhao S. Machine Learning for Predicting Risk and Prognosis of Acute Kidney Disease in Critically Ill Elderly Patients During Hospitalization: Internet-Based and Interpretable Model Study [J]. J Med Internet Res, 2024, 26: e51354.http://doi.org/10.2196/51354 Nemati S, Holder A, Razmi F, Stanley M D, Clifford G D, Buchman T G. An Interpretable Machine Learning Model for Accurate Prediction of Sepsis in the ICU [J]. Crit Care Med, 2018, 46(4): 547-53.http://doi.org/10.1097/ccm.0000000000002936 Li L, Ding L, Zhang Z, Zhou L, Zhang Z, Xiong Y, et al. Development and Validation of Machine Learning-Based Models to Predict In-Hospital Mortality in Life-Threatening Ventricular Arrhythmias: Retrospective Cohort Study [J]. J Med Internet Res, 2023, 25: e47664.http://doi.org/10.2196/47664 Dong J, Feng T, Thapa-Chhetry B, Cho B G, Shum T, Inwald D P, et al. Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care [J]. Crit Care, 2021, 25(1): 288.http://doi.org/10.1186/s13054-021-03724-0 Liu C, Liu X, Mao Z, Hu P, Li X, Hu J, et al. Interpretable Machine Learning Model for Early Prediction of Mortality in ICU Patients with Rhabdomyolysis [J]. Med Sci Sports Exerc, 2021, 53(9): 1826-34.http://doi.org/10.1249/mss.0000000000002674 Kaewput W, Thongprayoon C, Petnak T, Cheungpasitporn W, Qureshi F, Boonpheng B, et al. Rhabdomyolysis among hospitalized patients for salicylate intoxication in the United States: Nationwide inpatient sample 2003-2014 [J]. PLoS One, 2021, 16(3): e0248242.http://doi.org/10.1371/journal.pone.0248242 McMahon G M, Zeng X, Waikar S S. A risk prediction score for kidney failure or mortality in rhabdomyolysis [J]. JAMA Intern Med, 2013, 173(19): 1821-8.http://doi.org/10.1001/jamainternmed.2013.9774 Gupta A, Thorson P, Penmatsa K R, Gupta P. Rhabdomyolysis: Revisited [J]. Ulster Med J, 2021, 90(2): 61-9 de Fallois J, Scharm R, Lindner T H, Scharf C, Petros S, Weidhase L. Kidney replacement and conservative therapies in rhabdomyolysis: a retrospective analysis [J]. BMC Nephrol, 2024, 25(1): 96.http://doi.org/10.1186/s12882-024-03536-8 Martinez T, Harrois A, Codorniu A, Mongardon N, Pissot M, Popoff B, et al. Evaluation of severe rhabdomyolysis on day 30 mortality in trauma patients admitted to intensive care: a propensity score analysis of the Traumabase registry [J]. Crit Care, 2024, 28(1): 382.http://doi.org/10.1186/s13054-024-05158-w Sun K, Shi Z, Abudureheman Y, Liu Q, Zhao Y, Zhang X, et al. Clinical and Epidemiological Characteristics of Rhabdomyolysis: A Retrospective Study [J]. Int J Clin Pract, 2023, 2023: 6396576.http://doi.org/10.1155/2023/6396576 Morin A G, Somme D, Corvol A. Rhabdomyolysis in older adults: outcomes and prognostic factors [J]. BMC Geriatr, 2024, 24(1): 46.http://doi.org/10.1186/s12877-023-04620-8 Johnson A E W, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, et al. MIMIC-IV, a freely accessible electronic health record dataset [J]. Sci Data, 2023, 10(1): 1.http://doi.org/10.1038/s41597-022-01899-x Pollard T J, Johnson A E W, Raffa J D, Celi L A, Mark R G, Badawi O. The eICU Collaborative Research Database, a freely available multi-center database for critical care research [J]. Sci Data, 2018, 5: 180178.http://doi.org/10.1038/sdata.2018.178 Liu Y, Zhu X, Xue J, Maimaitituerxun R, Chen W, Dai W. Machine learning models for mortality prediction in critically ill patients with acute pancreatitis-associated acute kidney injury [J]. Clin Kidney J, 2024, 17(10): sfae284.http://doi.org/10.1093/ckj/sfae284 Takkavatakarn K, Oh W, Chan L, Hofer I, Shawwa K, Kraft M, et al. Machine learning derived serum creatinine trajectories in acute kidney injury in critically ill patients with sepsis [J]. Crit Care, 2024, 28(1): 156.http://doi.org/10.1186/s13054-024-04935-x Luo W, Phung D, Tran T, Gupta S, Rana S, Karmakar C, et al. Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary View [J]. J Med Internet Res, 2016, 18(12): e323.http://doi.org/10.2196/jmir.5870 Bai J, Huang J H, Price C P E, Schauer J M, Suh L A, Harmon R, et al. Prognostic factors for polyp recurrence in chronic rhinosinusitis with nasal polyps [J]. J Allergy Clin Immunol, 2022, 150(2): 352-61.e7.http://doi.org/10.1016/j.jaci.2022.02.029 Dong B, Zhang H, Duan Y, Yao S, Chen Y, Zhang C. Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma [J]. J Transl Med, 2024, 22(1): 455.http://doi.org/10.1186/s12967-024-05203-w Fan Z, Jiang J, Xiao C, Chen Y, Xia Q, Wang J, et al. Construction and validation of prognostic models in critically Ill patients with sepsis-associated acute kidney injury: interpretable machine learning approach [J]. J Transl Med, 2023, 21(1): 406.http://doi.org/10.1186/s12967-023-04205-4 Yue S, Li S, Huang X, Liu J, Hou X, Zhao Y, et al. Machine learning for the prediction of acute kidney injury in patients with sepsis [J]. J Transl Med, 2022, 20(1): 215.http://doi.org/10.1186/s12967-022-03364-0 Yang D, Zhao L, Kang J, Wen C, Li Y, Ren Y, et al. Development and validation of a predictive model for acute kidney injury in patients with moderately severe and severe acute pancreatitis [J]. Clin Exp Nephrol, 2022, 26(8): 770-87.http://doi.org/10.1007/s10157-022-02219-8 Hou N, Li M, He L, Xie B, Wang L, Zhang R, et al. Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost [J]. J Transl Med, 2020, 18(1): 462.http://doi.org/10.1186/s12967-020-02620-5 Tseng P Y, Chen Y T, Wang C H, Chiu K M, Peng Y S, Hsu S P, et al. Prediction of the development of acute kidney injury following cardiac surgery by machine learning [J]. Crit Care, 2020, 24(1): 478.http://doi.org/10.1186/s13054-020-03179-9 Zhang Z, Ho K M, Hong Y. Machine learning for the prediction of volume responsiveness in patients with oliguric acute kidney injury in critical care [J]. Crit Care, 2019, 23(1): 112.http://doi.org/10.1186/s13054-019-2411-z Dong L, Liu P, Qi Z, Lin J, Duan M. Development and validation of a machine-learning model for predicting the risk of death in sepsis patients with acute kidney injury [J]. Heliyon, 2024, 10(9): e29985.http://doi.org/10.1016/j.heliyon.2024.e29985 Song X, Liu X, Liu F, Wang C. Comparison of machine learning and logistic regression models in predicting acute kidney injury: A systematic review and meta-analysis [J]. Int J Med Inform, 2021, 151: 104484.http://doi.org/10.1016/j.ijmedinf.2021.104484 O'Sullivan E D, Hughes J, Ferenbach D A. Renal Aging: Causes and Consequences [J]. J Am Soc Nephrol, 2017, 28(2): 407-20.http://doi.org/10.1681/asn.2015121308 Shen J, Chu Y, Wang C, Yan S. Risk factors for acute kidney injury after major abdominal surgery in the elderly aged 75 years and above [J]. BMC Nephrol, 2022, 23(1): 224.http://doi.org/10.1186/s12882-022-02822-7 Wang M, Wang X, Zhu B, Li W, Jiang Q, Zuo Y, et al. The effects of timing onset and progression of AKI on the clinical outcomes in AKI patients with sepsis: a prospective multicenter cohort study [J]. Ren Fail, 2023, 45(1): 1-10.http://doi.org/10.1080/0886022x.2022.2138433 Wahab A, Smith R J, Lal A, Flurin L, Malinchoc M, Dong Y, Gajic O. CHARACTERISTICS AND PREDICTORS OF PATIENTS WITH SEPSIS WHO ARE CANDIDATES FOR MINIMALLY INVASIVE APPROACH OUTSIDE OF INTENSIVE CARE UNIT [J]. Shock, 2023, 59(5): 702-7.http://doi.org/10.1097/shk.0000000000002112 Lindner G, Funk G-C. Hypernatremia in critically ill patients [J]. Journal of Critical Care, 2013, 28(2): 216.e11-.e20.http://doi.org/https://doi.org/10.1016/j.jcrc.2012.05.001 Atlani M, Kumar A, Pakhare A P, Singhai A, Gadwala R. Potential Association of Hypernatremia With Mortality in Patients With Acute Kidney Injury and COVID-19 [J]. Cureus, 2022, 14(7): e27530.http://doi.org/10.7759/cureus.27530 Jung S Y, Kim H, Park S, Jhee J H, Yun H R, Kim H, et al. Electrolyte and mineral disturbances in septic acute kidney injury patients undergoing continuous renal replacement therapy [J]. Medicine (Baltimore), 2016, 95(36): e4542.http://doi.org/10.1097/md.0000000000004542 Gao C, Peng L. Association and prediction of red blood cell distribution width to albumin ratio in all-cause mortality of acute kidney injury in critically ill patients [J]. Front Med (Lausanne), 2023, 10: 1047933.http://doi.org/10.3389/fmed.2023.1047933 Hong J, Hu X, Liu W, Qian X, Jiang F, Xu Z, et al. Impact of red cell distribution width and red cell distribution width/albumin ratio on all-cause mortality in patients with type 2 diabetes and foot ulcers: a retrospective cohort study [J]. Cardiovasc Diabetol, 2022, 21(1): 91.http://doi.org/10.1186/s12933-022-01534-4 Rabb H, Griffin M D, McKay D B, Swaminathan S, Pickkers P, Rosner M H, et al. Inflammation in AKI: Current Understanding, Key Questions, and Knowledge Gaps [J]. J Am Soc Nephrol, 2016, 27(2): 371-9.http://doi.org/10.1681/asn.2015030261 Yu M Y, Lee S W, Baek S H, Na K Y, Chae D W, Chin H J, Kim S. Hypoalbuminemia at admission predicts the development of acute kidney injury in hospitalized patients: A retrospective cohort study [J]. PLoS One, 2017, 12(7): e0180750.http://doi.org/10.1371/journal.pone.0180750 Thongprayoon C, Cheungpasitporn W, Chewcharat A, Mao M A, Thirunavukkarasu S, Kashani K B. Impacts of admission serum albumin levels on short-term and long-term mortality in hospitalized patients [J]. Qjm, 2020, 113(6): 393-8.http://doi.org/10.1093/qjmed/hcz305 Li X, Li X, Zhao W, Wang D. Development and validation of a nomogram for predicting in-hospital death in cirrhotic patients with acute kidney injury [J]. BMC Nephrol, 2024, 25(1): 175.http://doi.org/10.1186/s12882-024-03609-8 Zhou H, Liu L, Zhao Q, Jin X, Peng Z, Wang W, et al. Machine learning for the prediction of all-cause mortality in patients with sepsis-associated acute kidney injury during hospitalization [J]. Front Immunol, 2023, 14: 1140755.http://doi.org/10.3389/fimmu.2023.1140755 Folkestad T, Brurberg K G, Nordhuus K M, Tveiten C K, Guttormsen A B, Os I, Beitland S. Acute kidney injury in burn patients admitted to the intensive care unit: a systematic review and meta-analysis [J]. Crit Care, 2020, 24(1): 2.http://doi.org/10.1186/s13054-019-2710-4 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-6824634","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":474347944,"identity":"b70c9b8f-3ad5-466c-8c32-d76cf8d04e09","order_by":0,"name":"Wenyan Zhang","email":"","orcid":"","institution":"Xiangya School of Public Health, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Wenyan","middleName":"","lastName":"Zhang","suffix":""},{"id":474347945,"identity":"2ce8c1d7-ded5-46f3-9d04-9392a14b06e3","order_by":1,"name":"Yamin Liu","email":"","orcid":"","institution":"Xiangya School of Public Health, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Yamin","middleName":"","lastName":"Liu","suffix":""},{"id":474347946,"identity":"6cf762b8-5344-46e6-a48c-6834f694b81a","order_by":2,"name":"Ziling Feng","email":"","orcid":"","institution":"Xiangya School of Public Health, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Ziling","middleName":"","lastName":"Feng","suffix":""},{"id":474347947,"identity":"50d80692-533f-4dd8-b6cf-f6969cd846d2","order_by":3,"name":"Ni Xiong","email":"","orcid":"","institution":"Xiangya School of Public Health, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Ni","middleName":"","lastName":"Xiong","suffix":""},{"id":474347948,"identity":"85607eff-a75e-4806-b312-726dd5751bcd","order_by":4,"name":"Leyao Tang","email":"","orcid":"","institution":"Xiangya School of Public Health, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Leyao","middleName":"","lastName":"Tang","suffix":""},{"id":474347950,"identity":"0c46fbf3-66af-41a7-acc3-1fc6b9f88b76","order_by":5,"name":"Wenhang Chen","email":"","orcid":"","institution":"Xiangya Hospital, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Wenhang","middleName":"","lastName":"Chen","suffix":""},{"id":474347952,"identity":"10902dea-e5d1-4212-8923-bd9fc1ece42a","order_by":6,"name":"Xu Zhu","email":"","orcid":"","institution":"College of Integrated Traditional Chinese and Western Medicine, Hunan University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xu","middleName":"","lastName":"Zhu","suffix":""},{"id":474347953,"identity":"384975ff-57f9-414c-96bc-dbb222259a53","order_by":7,"name":"Jing Xue","email":"","orcid":"","institution":"Xiangya Hospital, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Xue","suffix":""},{"id":474347955,"identity":"218cad30-63b1-438c-a824-20fb08c6e282","order_by":8,"name":"Wenjie Dai","email":"","orcid":"","institution":"Xiangya School of Public Health, Central South University","correspondingAuthor":false,"prefix":"","firstName":"Wenjie","middleName":"","lastName":"Dai","suffix":""},{"id":474347956,"identity":"5dbfa149-ed1e-4b3c-83d7-5da1366a4ced","order_by":9,"name":"Jianzhou Yang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAv0lEQVRIiWNgGAWjYBACxgYwZcPAx0yiljQGNqK1QMFhBjai1TL3nzH8XPDrvDwbO4+ZdAGDnZxuA0GHnTGWntl327CNGahlBkOysdkBQloaewykeXtuM4K18DAcSNxGUEszj/Fv3p5z9iRoaQOp/HEgkQQtPWxl1rwNycltzGzF1jwGRPjFsP/w5ts8f+xs+/kPb7zNU2EnR1hLA4cB0HUgJpDBYEBAOQjIM7A/YGD4A2KCGKNgFIyCUTAKsAAAbrM3p8sSo3cAAAAASUVORK5CYII=","orcid":"","institution":"Changzhi Medical College","correspondingAuthor":true,"prefix":"","firstName":"Jianzhou","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-06-05 03:08:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6824634/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6824634/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85306167,"identity":"569f5869-35a1-4500-b0f0-adf4ca2d4a18","added_by":"auto","created_at":"2025-06-24 12:48:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":72054,"visible":true,"origin":"","legend":"\u003cp\u003eThe participant selection process for this study.\u003c/p\u003e\n\u003cp\u003eFigure legends: MIMIC-IV: The Medical Information Mart for Intensive Care-IV; eICU-CRD: The eICU Collaborative Research Database; RI-AKI: Rhabdomyolysis associated acute kidney injury.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/bd42c7e32d03329b6781f4c5.png"},{"id":85304123,"identity":"80d1cd99-f432-4645-a46b-021e5ba5ac69","added_by":"auto","created_at":"2025-06-24 12:32:56","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":102136,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/9f27075877cfd0cec77b9d8c.png"},{"id":85304125,"identity":"7b1add0c-118c-4f8b-8efc-49e41660e844","added_by":"auto","created_at":"2025-06-24 12:32:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":55189,"visible":true,"origin":"","legend":"\u003cp\u003eRF-based feature importance plot. Variables were screened by Random Forest and the top 25 variables were selected in order of feature importance.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/831f6e23ee50806cef1edf24.png"},{"id":85304128,"identity":"3f9db3d3-cb99-4552-b158-9320abf971dd","added_by":"auto","created_at":"2025-06-24 12:32:56","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":173739,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves and calibration plots of the prediction models for in-hospital mortality for critically ill patients with RI-AKI. (A) and (B) are the ROC curves of the prediction models in the training and test sets, respectively. (C) and (D) are the calibration curves and Brier scores of the predictive models in the training and test sets, respectively. The x-axis represents the predicted probability of the model, and the y-axis represents the true probability of the outcome. If the model is perfectly calibrated, the predicted probabilities and the observed probabilities should be identical, and the calibration curve should align with the diagonal (i.e., a straight line where Y = X). A lower Brier score indicates higher predictive accuracy of the model. XGBoost: extremely gradient; RF: random forest; boosting; LR: logistic regression; AUC: area under the curve.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/cf61ea6a15fbda9ccb22a611.png"},{"id":85307548,"identity":"64cc90b3-f573-4c40-9e65-7d24975fb63d","added_by":"auto","created_at":"2025-06-24 13:04:56","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":152370,"visible":true,"origin":"","legend":"\u003cp\u003eModel evaluation. (A) Feature importance based on SHAP values in the training set. Each line on the y-axis represents a feature, and the x-axis represents the corresponding SHAP value. Red (blue) dots indicate high (low) feature values. A higher SHAP value for a feature indicates a greater contribution to the patient's risk of mortality. (B) Feature importance ranked by the XGBoost model using the 'gain' metric, which quantifies the contribution of each feature to improving the model's predictive accuracy. SHAP: Shapley Additive Explanations; MCHC: mean corpuscular hemoglobin concentration; RDW: red cell distribution width; APTT. activated partial thromboplastin time; INR: international normalised ratio; abs_monocytes: monocyte count.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/95446d5b4f8949e587f9694e.png"},{"id":85304134,"identity":"e8cd4722-0c34-4ec9-8210-b644ee54055e","added_by":"auto","created_at":"2025-06-24 12:32:56","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":220843,"visible":true,"origin":"","legend":"\u003cp\u003ePartial SHAP dependency plots for feature importance in predicting in-hospital mortality using the XGBoost model in critically ill patients with RI-AKI. MCHC: mean haemoglobin concentration; RDW: red blood cell distribution width; APTT: activated partial thromboplastin time; XGBoost: extreme gradient boost.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/5ada75845cc5733aac4b2730.png"},{"id":96402520,"identity":"93c1a692-c0d5-48ad-8b69-48a1975cc2fa","added_by":"auto","created_at":"2025-11-20 16:23:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1816729,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6824634/v1/0c3f493a-0be9-4133-8cab-e8cd4cf722b7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine learning models to predict in-hospital mortality in patients with rhabdomyolysis combined with acute kidney injury","fulltext":[{"header":"Highlights","content":"\u003cp\u003e1. Multicenter ICU data used for large-scale modeling and validation in patient outcomes.\u003c/p\u003e\u003cp\u003e2. Comprehensive assessment identifies risk factors for mortality in critical patients.\u003c/p\u003e\u003cp\u003e3. XGBoost outperforms in multiple constructed models.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eSymptoms of rhabdomyolysis include muscle ache, muscle debility, and urine of a tea hue, which can be brought on by various factors including trauma, strenuous exercise, seafood consumption, medication side effects (e.g., statins), infections, and poisoning \u003csup\u003e[\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. Acute kidney injury linked to rhabdomyolysis (RI-AKI), primarily due to myoglobin's mechanical blockage of renal tubules, ranks as a frequent and grave complication among severely ill individuals with rhabdomyolysis [4]. It accounts for 13\u0026ndash;50% of patients with rhabdomyolysis, with its mortality rate being 62% \u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e. Individuals suffering from RI-AKI face a risk of death in the hospital that is 3 to 4 times higher than those not afflicted with AKI \u003csup\u003e[\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. As a result, predicting the mortality rates among critically ill RI-AKI patients in hospitals is vital for swift and effective treatment.\u003c/p\u003e \u003cp\u003eMachine Learning (ML) serves as an effective method for data analysis, merging statistical methods with computer science to scrutinize vast data sets and reveal concealed details. ML models such as Extreme Gradient Boosting (XGBoost) and Random Forest (RF) account for higher-order non-linear interactions between predictors and are able to automatically reconstruct the relationships between variables and outcomes in large datasets \u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. Earlier research focusing on ICU patients suffering from sepsis, cardiovascular conditions, and AKI has endeavored to forecast mortality rates within hospitals using ML models, demonstrating their superiority in discrimination, calibration, and overall advantages over conventional models \u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. For example, Li et al. \u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e that machine learning models markedly enhanced the precision in predicting in-hospital deaths among patients with ventricular arrhythmias, in contrast to conventional scoring methods. Dong et al. \u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e discovered that the ML model precisely forecasted moderate to severe Acute Kidney Injury (AKI) up to 48 hours sooner than existing diagnostic protocols. Furthermore, an earlier investigation into ICU patients suffering from rhabdomyolysis revealed the superiority of the ML model over both the Acute Physiology Score III and the Sequential Organ Failure Assessment (SOFA) system, facilitating the early detection of patients with a high mortality risk \u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. However, there is still a limited amount of research focused on predicting mortality rates within hospitals among RI-AKI patients through ML models, which could obstruct the immediate identification of those at a higher risk of death in hospitals\u003c/p\u003e \u003cp\u003eEarlier studies focusing solely on patients with rhabdomyolysis or AKI indicated that demographic factors might account for in-hospital death rates (such as age and gender), laboratory parameters (such as serum creatine kinase [CK], creatinine, urea nitrogen, and routine blood markers), and therapeutic requirements (including methods like renal replacement therapy [RRT] and mechanical ventilation) \u003csup\u003e[\u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. Hospital mortality rates may be linked to coexisting conditions like sepsis and chronic kidney disease (CKD).\u003csup\u003e[\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e.Consequently, the objective of this research was to create machine learning models for precise forecasting of in-hospital death rates in ICU patients with RI-AKI, by thoroughly evaluating the impact of demographic factors, lab conditions, treatment needs, and concurrent diseases and complications.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Data Source\u003c/h2\u003e \u003cp\u003eThe research employed two extensive databases, namely the Medical Information Mart for Intensive Care IV (MIMIC IV) v2.0 and the eICU Collaborative Research Database (eICU-CRD) v2.0. The MIMIC IV database, accessible to the public and housed in a single facility, encompasses patient records from every ICU and emergency department entry at Beth Israel Deaconess Medical Center spanning 2008 to 2019 \u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e. The eICU-CRD, a database available at no cost and spanning multiple centers, encompasses information from more than 200,000 patient entries in 335 ICUs across 208 U.S. hospitals, amassed from 2014 to 2015 \u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e. The eICU-CRD, developed subsequent to MIMIC-IV, broadens research scope by integrating data from multiple medical centers \u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e. Therefore, critically ill RI-AKI patients in both databases may exhibit similar clinical characteristics. Before starting this research, the NIH Protecting Human Research Participants online training course was completed, and approval was obtained for extracting data from MIMIC IV for research purposes (Record ID: 58637411). Approval was also given by the Institutional Review Boards of both Beth Israel Deaconess Medical Centre (BIDMC, Boston, Massachusetts, USA) and Massachusetts Institute of Technology (MIT, Cambridge, Massachusetts, USA). In this research, every patient's health data was anonymized before analysis, thereby eliminating the need for written informed consent \u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e. This study was conducted in strict adherence to the guidelines for the development and reporting of machine learning prediction models in biomedical research \u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2.2 Study population\u003c/h3\u003e\n\u003cp\u003eThe research encompassed adults diagnosed with rhabdomyolysis and Acute Kidney Injury. Criteria for inclusion included:1) a confirmed diagnosis of rhabdomyolysis and AKI via manual examination of the International Classification of Diseases, Ninth Revision (ICD-9) codes;2) initial admission to the ICU; 3)and being over 18 years old. Criteria for exclusion encompassed: 1)maximum CK levels below 1000 U\u0026middot;L-1;2) ICU duration of less than a day;3) or absent data exceeding 25% or lacking outcome variables.\u003c/p\u003e\n\u003ch3\u003e2.3 Data extraction\u003c/h3\u003e\n\u003cp\u003ePatient data were extracted from the MIMIC-IV and eICU-CRD databases using Structured Query Language (SQL) with PostgreSQL 9.6, focusing on information collected within the first 24 hours after ICU admission. This research system collects multidimensional clinical data covering demographic characteristics, laboratory indicators, vital signs, comorbidities/complications, and treatment needs. Demographic characteristics include age, gender, and ethnicity; laboratory indicators are divided into five categories: (1) hematological parameters: including white blood cell count, red blood cell count, hemoglobin, platelet count, hematocrit, as well as absolute counts and percentages of neutrophils, lymphocytes, eosinophils, basophils, and monocytes; additionally, red blood cell morphological indicators such as mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), and red cell distribution width (RDW); (2) renal function indicators: including blood urea nitrogen and serum creatinine, used to assess kidney filtration and metabolic function; (3) liver function indicators: such as albumin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (ALP), reflecting the liver's synthetic and metabolic capacity; (4) blood glucose and electrolytes: including blood glucose, potassium, sodium, chloride, calcium, and phosphorus levels, used to monitor metabolic balance and homeostasis; (5) coagulation parameters: such as prothrombin time (PT), international normalized ratio (INR), and activated partial thromboplastin time (APTT), to assess the coagulation system status. Vital signs include temperature, respiratory rate, heart rate, systolic and diastolic blood pressure, to reflect the patient's physiological status in real-time. Comorbidities and complications include coronary heart disease (CHD), diabetes, hypertension, hyperlipidemia, malignant tumors, chronic lung disease, chronic kidney disease (CKD), and sepsis, to analyze the impact of underlying diseases on prognosis. Treatment requirements include renal replacement therapy (RRT), mechanical ventilation, corticosteroid use, diuretics, and vasoactive drugs, to assess disease severity and intervention measures. This comprehensive data framework provides a crucial basis for in-depth analysis of disease characteristics, predicting clinical outcomes, and optimizing treatment strategies.\u003c/p\u003e \u003cp\u003eIn-hospital mortality served as the result variable, with pertinent information being gathered from a pair of databases.\u003c/p\u003e\n\u003ch3\u003e2.4 Statistical analysis\u003c/h3\u003e\n\u003cp\u003eVariables of a categorical nature were displayed in terms of their occurrence rates and proportions, and were analyzed across different groups using the \u0026chi;\u003csup\u003e2\u003c/sup\u003e test or Fisher's exact test. Continuous variables were summarized as mean (standard deviation) or median (interquartile range), depending on their distribution. Between-group comparisons were conducted using Student's t-test or Mann-Whitney U test. When the percentage of missing data fell below 25%, a method of multiple imputation was utilized to generate five datasets. Specifically, the lack of continuous variable values was replaced with the mean of the estimated data, while missing values for categorical variables were replaced with the mode of the imputed data. Every statistical examination was bidirectional, with a P-value below 0.05 deemed to hold statistical significance. The statistical evaluations were performed utilizing the R software, version 4.3.2.\u003c/p\u003e\n\u003ch3\u003eModel Development\u003c/h3\u003e\n\u003cp\u003eInformation from the MIMIC IV and eICU databases was combined and arbitrarily segmented into training (80% of the sample) and testing groups (20% of the sample). A random oversampling method was utilized to rectify the class disparity (survival/death) present in the training dataset. The following steps were undertaken to select predictors and build the machine learning-based predictive model.\u003c/p\u003e \u003cp\u003ePredictors were chosen using the Least Absolute Shrinkage and Selection Operator (LASSO) and RF, with the common variables identified by both techniques serving as the model's predictors. By implementing L1 regularization on regression coefficients, LASSO facilitates the selection of variables and streamlines the model, rendering features with zero coefficients superfluous and removed from the fitting process, thus allowing for automated variable selection \u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e. RF was employed to rank the importance of variables using the Mean Decrease Gini, and the top 25 most important variables were selected. Cross-validation was applied during predictor selection to identify the optimal value corresponding to the minimum cross-validation error.\u003c/p\u003e \u003cp\u003eFor developing models, three powerful machine learning algorithms were utilized: XGBoost, RF, and logistic regression (LR). An effective Gradient Boosting Decision Tree implementation, XGBoost offers robust performance, fast training, and the capacity to manage big datasets \u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e. RF can train multiple decision trees to address the bias and variance issues of individual models, thereby enhancing the model's accuracy and robustness \u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e. LR is a classical linear model that allows the direct interpretation of individual feature effects through the model\u0026rsquo;s coefficients \u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e. For both XGBoost and RF models, a grid search technique was utilized to pinpoint the optimal hyperparameter combination, enhancing predictive precision. In an effort to avoid overfitting and enhance the models' applicability, they underwent a 10-fold cross-validation process, leading to the creation of the final model through multiple iterations \u003csup\u003e[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eModel evaluation and validation\u003c/h2\u003e \u003cp\u003eQuantitative assessment of the predictive model's differentiation capabilities was conducted using multiple metrics, including the area under the receiver operating characteristic curve (AUC), Youden's index, accuracy (ACC), sensitivity, specificity, F1 score, positive predictive value (PPV), and negative predictive value (NPV). It was considered acceptable to attain an AUC between 0.7 and 0.8, with values surpassing 0.8 \u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e. Three models' forecasting accuracy was evaluated using the Delong test. Calibration curves were used to assess the predictive accuracy of the model, and the Brier score was calculated to quantify calibration. A lower Brier score (ranging from 0 to 1) indicates better agreement between predicted probabilities and observed outcomes.\u003c/p\u003e \u003cp\u003eThe SHapley Additive ExPlanations (SHAP) method was utilized to assess each feature's impact on the model's forecasted results and to represent the combined effects during interactions with other features. Enhancing the clarity of ML models was achieved by depicting each feature's incremental impact on an importance ranking graph and illustrating the influence of each feature on the results in a dependency plot.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Baseline characteristics of study participants\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e depicts the method used to select participants for this research. From the MIMIC-IV and eICU-CRD databases, 855 RI-AKI records were gathered, encompassing 141 deaths within hospitals, with a mortality rate of 16.5% (95% CI: 15.1%-17.8%). Randomly, 684 and 171 patients were allocated to the training and test groups, with in-hospital death rates recorded at 15.5% (106/684) and 20.5% (35/171) for the training and test groups, respectively. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e displays the fundamental features of both the training and test datasets. Apart from ALP levels and temperature, the two groups showed no notable disparities.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of the training and test sets.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraining set (n\u0026thinsp;=\u0026thinsp;684)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTest set (n\u0026thinsp;=\u0026thinsp;171)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDemographic characteristics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58.5(43.9,71.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e57.8(49.0,70.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.842\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e339(49.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84(49.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.918\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEthnicity (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e483(70.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e114(66.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.562\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlack\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e41(6.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16(9.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAsian\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6(0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1(0.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHispanic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14(2.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4(2.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther/Unknown\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e140(20.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36(21.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLaboratory indicators\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCK (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8235.5(3034.3,12812.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7130.0(2208.0,12812.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.437\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWBC (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.1(9.0,18.40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13.1(9.7,18.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.543\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRBC (10\u003csup\u003e12\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.0(3.5,4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.0(3.5,4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.743\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHemoglobin (g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.1(10.4,13.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.0(10.5,14.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.923\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlatelet (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e191.0(135.3,248.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e185.0(133.0,250.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHematocrit (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.957\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11.9(8.0,14.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.9(8.0,13.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.661\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.2(0.7,1.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.1(0.7,1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.597\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEosinophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.1(0,0.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9(0.2,1.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.398\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0(0,0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0(0,0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.312\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.9(0.5,1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9(0.6,1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.303\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNEUT%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80.0(79.0,86.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.0(78.6,86.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.918\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLYMPH%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.7(5.0,11.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10.0(5.0,11.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.696\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEO%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.9(0.1,0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.9(0.2,1.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.517\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBAS%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.3(0.2,0.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.3(0.2,0.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.213\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMONO%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.4(4.2,8.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6.4(4.0,7.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.955\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCV (fl)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e91.0(87.0,95.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e91.0(87.0,96.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.176\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCH (PG)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30.3(28.8,31.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.4(29.0,32.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.282\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCHC (g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.1(32.2,34.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33.2(32.3,34.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.968\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRDW (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14.2(13.5,15.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.4(13.5,15.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.246\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBUN (mg/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.1(32.2,34.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31.0(19.0,52.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.134\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSCr (\u0026micro;mol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e185.6(99.9,324.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e168.0(106.1,309.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.729\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin (g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.1(2.7,3.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.1(2.7,3.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.868\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAST (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e247.5(99.0,738.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e209.0(77.0,738.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.681\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALT (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e95.5(43.3,381.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e108.0(42.0,381.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.673\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALP (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e81.0(59.0,96.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e91.0(61.0,98.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLDH(U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1274.0(798.3,1274.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1274.0(771.0,1274.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.693\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBG (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.3(5.7,9.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7.4(5.9,11.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.118\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePotassium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.4(3.9,5.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.4(3.7,5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.630\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSodium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e139.0(135.0,142.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e138.0(134.0,142.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.998\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChlorine (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e103.0(99.0,108.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e103.0(98.0,107.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.687\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCalcium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.0(1.8,2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.0(1.9,2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.239\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePhosphorus(mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.5(1.0,1.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.5(1.1,1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.696\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePT (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15.0(12.7,16.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.7(12.6,16.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.977\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eINR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.2(1.1,1.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.3(1.1,1.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.887\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPTT (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.0(27.8,36.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33.0(27.0,36.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVital signs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTemperature (℃)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36.8(36.4,37.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e36.7(36.3,37.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.044\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRespiratory rate (times/min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20.0(17.0,24.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20.0(16.0,24.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.258\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeart rate (times/min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e97.1(82.3,110.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94.0(83.0,105.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.235\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSBP (mmHg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e124.0(105.0,141.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e123.0(107.0,140.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.689\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDBP (mmHg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e71.0(58.0,83.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71.0(51.0,86.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.750\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComorbidities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoronary heart disease (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e86(12.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24(14.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes mellitus (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e134(19.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41(24.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.244\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e180(26.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48(28.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.699\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertriglyceridemia (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42(6.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13(7.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.602\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMalignancy(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26(3.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10(5.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.284\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChronic pulmonary disease (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e93(13.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23(13.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCKD (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e94(13.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20(11.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.531\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSepsis (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e309(45.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76(44.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.932\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTreatment requirements\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRRT (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112(16.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22(12.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.291\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMechanical ventilation (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e194(28.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e47(27.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.850\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlucocorticoid (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e135(19.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41(24.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.245\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiuretic (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e163(23.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e39(22.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.841\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVasopressor (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e210(30.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54(31.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.853\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eCK: serum creatine kinase; WBC: white blood cell; RBC: red blood cell; NEUT%: percentage of neutrophils; LYMPH%: percentage of lymphocytes; EO%: percentage of eosinophils; BAS%: percentage of basophils; MONO%: percentage of monocytes; MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; MCHC: mean corpuscular hemoglobin concentration; RDW: red cell distribution width; BUN: blood urea nitrogen; SCr: serum creatinine; AST: aspartate aminotransferase; ALT: alanine transaminase; BG: blood glucose; PT: prothrombin time; INR: international normalized ratio; APTT: activated partial thromboplastin time ; SBP: systolic blood pressure; DBP: diastolic blood pressure; CKD: chronic kidney disease; RRT: renal replacement therapy.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\u003cp\u003eIn terms of demographics, such as age and ethnicity, lab parameters, such as MCHC, RDW, albumin, sodium, phosphorus, PT, INR, and APTT, vital statistics, such as respiratory and heart rates, comorbidities/complications, such as CHD and hypertriglyceridemia, and treatment requirements, such as RRT, mechanical ventilation, glucocorticoids, and vasopressors, a notable disparity was observed between the groups with survival and those without in the training dataset(Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCharacteristics between the survival and non-survival groups in the training set.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSurvival (n\u0026thinsp;=\u0026thinsp;578)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMortality (n\u0026thinsp;=\u0026thinsp;106)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDemographic characteristics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e57.0 (42.1,69.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68.0 (55.8,80.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e295 (51.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e56 (52.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.464\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEthnicity (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e411 (71.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e72 (67.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlack\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e38 (6.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (2.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAsian\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (1.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHispanic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14 (2.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther/Unknown\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e110 (19.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (28.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLaboratory indicators\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCK (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8497.5 (3185.8,12812.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5966.5 (2695.8,12812.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.288\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWBC (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.0 (9.0,18.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13.8 (9.6,19.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.290\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRBC (10\u003csup\u003e12\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.1 (3.5, 4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.9 (3.1,4.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.169\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHemoglobin (g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.1 (10.6,13.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12.0 (9.2,13.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.165\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlatelet (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e192.0 (140.8,246.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e169.0 (103.8,267.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.211\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHematocrit (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36.6 (32.1,41.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36.1 (28.7,42.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.293\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e11.9 (8.0,14.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11.9 (8.5,13.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.939\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.2 (0.7,1.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.1 (0.7,1.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.552\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEosinophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.1 (0.0,0.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.1 (0,0.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.571\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBasophil (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0 (0,0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0,0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.453\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonocyte (10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.9 (0.5,1.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.8 (0.4,1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.075\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNEUT%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80.0 (79.0,86.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.0 (79.1,88.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.382\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLYMPH%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10.0 (5.2,11.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.9 (4.4,10.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.829\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEO%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.9 (0.1,0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.9 (0.1,0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.721\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBAS%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.3 (0.2,0.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.3 (0.2,0.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.399\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMONO%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.4 (4.5,8.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.2 (3.5,7.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.076\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCV (fl)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e91.0 (87.0,95.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e92.5 (88.0\u0026ndash;97.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.197\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCH (PG)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30.2 (28.7,31.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.5 (29.0,31.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.669\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMCHC (g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.1 (32.2,34.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33.0 (32.0,33.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRDW (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14.2 (13.5,15.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.4 (13.4,15.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBUN (mg/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e33.0 (19.0,57.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35.5 (21.8,66.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.324\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSCr (\u0026micro;mol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e185.6 (97.2,336.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e185.6 (116.9,291.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.322\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin (g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.1 (2.7,3.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.0 (2.4,3.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAST (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e245.0 (93.8,738.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e271.0 (117.3,738.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.690\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALT (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e94.5 (43.0,381.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e95.0 (46.0,381.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.622\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALP (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80.0 (59.0,96.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e83.5 (58.8-109.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.400\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBG (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.2 (5.7,9.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.3 (5.7,10.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.658\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePotassium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.4 (3.8,5.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.5 (3.9,5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.417\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSodium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e138.4 (135.0,141.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e139.0 (136.0,144.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.026\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChlorine (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e103.1 (99.0,108.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e104.0 (100.0,109.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.147\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCalcium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.0 (1.8,2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.0 (1.9\u0026ndash;2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.960\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePhosphorus(mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.4 (1.0,1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.5 (1.1,2.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.014\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePT (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14.8 (12.6,16.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16.5 (13.4,18.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eINR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.3 (1.1,1.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.4 (1.2,1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.028\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPTT (s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32.9 (27.5,36.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34.7 (29.9,36.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVital signs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTemperature (℃)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36.8 (36.4,37.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36.9 (36.4.37.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.967\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRespiratory rate (times/min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (16,24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22 (18,27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeart rate (times/min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e97 (81,110)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e101 (90.8,115.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSBP (mmHg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e124.0 (105.0,141.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e123.5 (101.8,138.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.573\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDBP (mmHg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e71.5 (58.0,84.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68.5 (53.8,82.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.101\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComorbidities\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoronary heart disease (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e59 (10.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e27 (25.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes mellitus (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e117 (20.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (16.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.316\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e155 (26.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25 (23.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.487\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertriglyceridemia (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (5.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 (11.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMalignancy(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 (3.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6 (5.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.416\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChronic pulmonary disease (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80 (13.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (12.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.779\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCKD (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e77 (13.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (16.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.455\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSepsis (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e252 (43.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e57 (53.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.053\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTreatment requirements\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRRT (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e87 (15.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25 (23.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.029\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMechanical ventilation (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e150 (26.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44 (41.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlucocorticoid (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e100 (17.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35 (33.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiuretic (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e134 (23.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e29 (27.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.354\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVasopressor (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e158 (27.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e52 (49.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"6\"\u003eCK: serum creatine kinase; WBC: white blood cell; RBC: red blood cell; NEUT%: percentage of neutrophils; LYMPH%: percentage of lymphocytes; EO%: percentage of eosinophils; BAS%: percentage of basophils; MONO%: percentage of monocytes; MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; MCHC: mean corpuscular hemoglobin concentration; RDW: red cell distribution width; BUN: blood urea nitrogen; SCr: serum creatinine; AST: aspartate aminotransferase; ALT: alanine transaminase; BG: blood glucose; PT: prothrombin time; INR: international normalized ratio; APTT: activated partial thromboplastin time ; SBP: systolic blood pressure; DBP: diastolic blood pressure; CKD: chronic kidney disease; RRT: renal replacement therapy.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Model development\u003c/h2\u003e \u003cp\u003eFifteen feature variables with non-zero coefficients were selected from a total of fifty-eight variables by LASSO regularization (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B), and the top 25 most important variables selected by the RF analysis are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Ten shared variables including age, sodium, albumin, heart rate, phosphorus, MCHC, RDW, INR, APTT, and monocyte count found by both LASSO regularization and RF analysis were considered as the predictors to develop ML models.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Model evaluation and validation\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e displays the performance of ML models across both training and test datasets. Within the training dataset, AUC values for XGBoost, RF, and LR stood at 0.889 (95% CI: 0.872\u0026ndash;0.908), 0.797 (95% CI: 0.772\u0026ndash;0.822), and 0.740 (95% CI: 0.711\u0026ndash;0.768), in that order. XGBoost's AUC values notably surpassed those of the RF and LR models, with both showing statistical significance (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Within the test dataset, the AUC values for XGBoost, RF, and LR stood at 0.775 (95% CI: 0.721\u0026ndash;0.830), 0.719 (95% CI: 0.625\u0026ndash;0.813), and 0.702 (95% CI: 0.639\u0026ndash;0.765), in that order (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). When comparing the XGBoost model with the RF and LR models, the P values recorded were 0.0178 and 0.069, in that order.\u003c/p\u003e \u003cp\u003eFigures \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD illustrate the calibration ability for the three models within the training and test datasets, respectively. Regarding precision in calibration, the XGBoost model surpassed the RF and LR models during both its training and test stages. In the dataset used for training, the recorded Brier scores for XGBoost, RF, and LR were 0.122, 0.185, and 0.203, in that order. Within the test dataset, XGBoost, RF, and LR achieved Brier scores of 0.214, 0.222, and 0.228, in that order.\u003c/p\u003e \n\u003cp\u003eTable 3. Discrimination indicators of the predictive models in the training and test sets.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003eAUC (95%CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003eYouden index\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003eACC (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003eSensitivity (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003eSpecificity (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003eF1 score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003ePPV (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003eNPV (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eXGBoost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e0.889(0.872-0.908)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e0.616\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.808\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.667\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.677\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.803\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e0.822\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.795\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e0.797(0.772-0.822)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e0.379\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.689\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.682\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.753\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.713\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e0.663\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.726\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e0.740(0.711-0.768)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e0.344\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.672\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.667\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.677\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.677\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e0.687\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.657\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eTest set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eXGBoost\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e0.775(0.721-0.830)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e0.331\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.665\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.919\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.544\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.712\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e0.618\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.778\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e0.719(0.625-0.813）\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e0.176\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.588\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.904\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.419\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.659\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e0.562\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.650\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 10.6878%;\"\u003e\n \u003cp\u003eLR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 15.7672%;\"\u003e\n \u003cp\u003e0.702(0.639-0.765)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.582%;\"\u003e\n \u003cp\u003e0.301\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.651\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.904\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.7937%;\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.671\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.2646%;\"\u003e\n \u003cp\u003e0.634\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.3704%;\"\u003e\n \u003cp\u003e0.672\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eXGBoost: extremely gradient; RF: random forest; boosting; LR: logistic regression; AUC: area under the curve; CI: Confidence interval; ACC: accuracy; PPV: positive predictive value; NPV: negative predictive value.\u003c/p\u003e\n \u003cp\u003eIn Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, the SHAP values are displayed to elucidate how the XGBoost model forecasts in-hospital death rates among critically ill RI-AKI patients in the training dataset. Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows the most important feature variables individually through the SHAP dependency graph.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe research, employing ML methods along with MIMIC-IV and eICU data, developed and validated the effectiveness of three models in predicting in-hospital mortality in RI-AKI critically ill patients. The XGBoost model was found to surpass the RF and LR models in both distinguishing and calibrating. Furthermore, factors such as age, sodium, phosphorus, monocyte count, MCHC, RDW, albumin, INR, APTT, and heart rate were pinpointed as predictive elements. As far as we are aware, this research was pioneering in creating ML-based models to forecast in-hospital deaths among critically ill RI-AKI patients.\u003c/p\u003e \u003cp\u003eXGBoost is renowned for its expertise and flexibility as a gradient boosting tree algorithm, especially in managing structured data and classification tasks, earning recognition as the leading algorithm in machine learning and prediction competitions on Kaggle.com \u003csup\u003e[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e. Numerous research works have demonstrated the superiority of the XGBoost algorithm over alternative ML algorithms in forecasting adverse clinical outcomes \u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e. For instance, Zhang et al. \u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e discovered the XGBoost model to be more efficient in differentiating volume sensitivity in oliguric AKI patients than the traditional logistic regression model, enabling accurate modifications in fluid treatmentl. Dong et al. \u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e demonstrated the superior performance of XGBoost in identifying, calibrating, and clinically predicting mortality risk in patients with septic AKI. Additionally, an extensive meta-analysis showed that XGBoost outperforms LR and a variety of machine learning frameworks, including (artificial neural networks, support vector machines, and Bayesian networks ) in predicting AKI \u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e. This study added significantly to the existing knowledge by supporting the role of XGBoost model in predicting in-hospital mortality in critically ill patients with RI-AKI. Consequently, it is recommended that medical decision-makers employ XGBoost for forecasting in-hospital death rates among critically ill RI-AKI patients in clinical settings\u003c/p\u003e \u003cp\u003eThe study uncovered a link between advanced age, elevated heart rates, and higher death rates among critically ill RI-AKI patients in the hospital. Age can affect physical condition and organ compensatory capacity, and hence playing an important role in survival outcomes. The susceptibility to the harmful effects of toxins released by muscle destruction may be increased in elderly patients due to the degenerative changes in kidney structure and function \u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e. Similarly, an elevated heart rate may lead to a higher risk of death through increasing cardiac strain and disrupts hemodynamics \u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e. Therefore, more attention should be paid to those with increased age and elevated heart rate.\u003c/p\u003e \u003cp\u003eThe kidneys are crucial for sustaining electrolyte equilibrium, with disruptions in sodium or phosphorus balance frequently occurring in patients with AKI. The study uncovered a link between increased levels of hypernatremia and hyperphosphatemia and increased in-hospital death rates among critically ill RI-AKI patients, aligning with earlier findings. For example, Lindner et al. \u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]\u003c/sup\u003e found that ICU-acquired hypernatremia independently affected mortality, with a mortality rate of 48% in individuals whose plasma sodium exceeded 150 mmol/L. Atlani et al. \u003csup\u003e[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e identified hypernatraemia as the only separate risk element for AKI-related mortality in COVID-19 sufferers, noting that those who did not survive had notably elevated median serum sodium levels. Furthermore, Jung et al. \u003csup\u003e[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]\u003c/sup\u003e Research revealed that with a rise in serum phosphate levels by 1 mg/dl, the related mortality risk ratios at 28 and 90 days stood at 1.36 (1.20\u0026ndash;1.54) and 1.32 (1.17\u0026ndash;1.48), in that order. Consequently, vigilant observation of serum phosphate and sodium concentrations in ICU patients suffering from RI-AKI is crucial for categorizing and managing risk.\u003c/p\u003e \u003cp\u003eFurthermore, the study uncovered a link between the count of monocytes, MCHC, RDW, albumin, INR, and APTT and the mortality rates in ICU patients with RI-AKI. The number of monocytes, RDW, and MCHC serve as crucial hematological markers linked to the inflammatory reaction. Previous studies showed that in critically sick patients with AKI, the RDW/albumin ratio accurately predicts overall death rates at 1, 3, and 12 months \u003csup\u003e[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/sup\u003e. RDW reflects the heterogeneity of erythrocyte volume, and a rise in RDW typically correlates with intense inflammation and oxidative stress, potentially leading to a less favorable prognosis and aiding in effective risk categorization for patients with serious illnesses\u003csup\u003e[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]\u003c/sup\u003e. Albumin serves as an essential indicator of nutritional health and is crucial in sustaining colloid osmotic pressure, along with blood and cell stability. Additionally, reduced levels of albumin have been recognized as a standalone risk element for both the emergence and fatality of AKI \u003csup\u003e[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]\u003c/sup\u003e. Research indicates that upon admission, both INR and APTT are indicative of death risk in sepsis-related AKI patients, where higher INR and extended APTT correlate with higher in-hospital death rates in AKI sufferers \u003csup\u003e[\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]\u003c/sup\u003e. Consequently, these indicators could act as focal points for interventions aimed at lowering in-hospital deaths among ICU patients suffering from RI-AKI.\u003c/p\u003e \u003cp\u003eThe strengths of this study included multiple ICU centres, a substantial number of participants, and a comprehensive assessment of possible predictors. However, several limitations should be acknowledged. Firstly, there was uncertainty regarding the external applicability due to the lack of external validation. Additionally, research indicates that the etiology of rhabdomyolysis could affect the prognosis of patients suffering from RI-AKI \u003csup\u003e[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]\u003c/sup\u003e, the absence of such data precluding exploring its role in this study.Ultimately, the prospective indicators examined in this research were evaluated within a day of admission. Consequently, the forecasting accuracy of variables assessed outside this timeframe is still ambiguous.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThe research established and confirmed the effectiveness of three ML models in forecasting in-hospital death rates among critically ill individuals with RI-AKI. In contrast to the LR and RF models, the XGBoost model demonstrated superior precision in both differentiation and calibration, suggesting its potential value in clinical decision-making. Ten key predictors including age, sodium, phosphorus, and coagulation parameters were identified. Therefore, those with increased age and elevated heart rate should be given special consern, and sodium, phosphorus, and coagulation parameters may be served as the intervention targets for reducing hospital deaths among ICU patients suffering from RI-AKI. Additional validation from outside sources remains essential in clinical settings.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe express our thanks to the Medical Information Mart for Intensive Care IV \u0026nbsp; (MIMIC IV) v2.0 and the eICU Collaborative Research Database (eICU-CRD) for their comprehensive data provision.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval declaration:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study protocol was reviewed and approved by Ethics Committee of Xiangya Hospital, Central South University, approval number 202403060.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate statement\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this research, every patient\u0026apos;s health data was anonymized before analysis, thereby eliminating the need for written informed consent and was approved by the Ethics Committee of Xiangya Hospital, Central South University. This study was conducted in strict adherence to the guidelines for the development and reporting of machine learning prediction models in biomedical research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCONFLICT OF INTEREST STATEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe writers affirm the absence of any conflicting interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFUNDING\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research received funding through a grant awarded by Hunan Province\u0026apos;s Natural Science Foundation (2021JJ40972).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAUTHORS\u0026rsquo; CONTRIBUTIONS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWenyan Zhang: conceptualization; methodology; data acquisition and management; formal analysis; visualization; writing\u0026mdash;original draft; writing\u0026mdash;review and editing. Yamin Liu: methodology; data acquisition and management; software; writing-reviewing and editing. Ziling Feng: methodology; project administration; supervision; validation; writing-reviewing and editing. Ni Xiong: data acquisition and management; writing\u0026mdash;review and editing; validation. Leyao Tang: formal analysis; methodology; software; writing-reviewing and editing. Xu Zhu: formal analysis; methodology; software; writing-reviewing and editing. Jing Xue: conceptualization, project administration,validation, visualization, writing-reviewing and editing. Wenhang Chen: conceptualization; methodology; writing-reviewing and editing; supervision; formal analysis and validation; funding acquisition; project administration. Wenjie Dai: methodology; writing-reviewing and editing; supervision; formal analysis and validation. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDATA AVAILABILITY STATEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eShould there be a justified request, the researchers are able to supply the supporting data for this study\u0026apos;s findings.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eChavez L O, Leon M, Einav S, Varon J. Beyond muscle destruction: a systematic review of rhabdomyolysis for clinical practice [J]. Crit Care, 2016, 20(1): 135.http://doi.org/10.1186/s13054-016-1314-5\u003c/li\u003e\n\u003cli\u003eTawhari M, Aldalaan A, Alanazi R, Aldharman S, Alnafisah T, Alawad N, et al. Clinical presentation and outcomes of patients with rhabdomyolysis: A tertiary care center experience [J]. Saudi Med J, 2024, 45(5): 510-7.http://doi.org/10.15537/smj.2024.45.5.20230560\u003c/li\u003e\n\u003cli\u003ePei P, Li X Y, Lu S S, Liu Z, Wang R, Lu X C, Lu K. The Emergence, Epidemiology, and Etiology of Haff Disease [J]. Biomed Environ Sci, 2019, 32(10): 769-78.http://doi.org/10.3967/bes2019.096\u003c/li\u003e\n\u003cli\u003eAl-Ismaili Z, Piccioni M, Zappitelli M. Rhabdomyolysis: pathogenesis of renal injury and management [J]. Pediatr Nephrol, 2011, 26(10): 1781-8.http://doi.org/10.1007/s00467-010-1727-3\u003c/li\u003e\n\u003cli\u003eXiao L, Ran X, Zhong Y, Le Y, Li S. Serum creatine kinase levels are not associated with an increased need for continuous renal replacement therapy in patients with acute kidney injury following rhabdomyolysis [J]. Ren Fail, 2022, 44(1): 893-901.http://doi.org/10.1080/0886022x.2022.2079523\u003c/li\u003e\n\u003cli\u003eMcMahon G M, Zeng X, Waikar S S. A Risk Prediction Score for Kidney Failure or Mortality in Rhabdomyolysis [J]. JAMA Internal Medicine, 2013, 173(19): 1821-7.http://doi.org/10.1001/jamainternmed.2013.9774\u003c/li\u003e\n\u003cli\u003eYang C W, Li S, Dong Y, Paliwal N, Wang Y. Epidemiology and the Impact of Acute Kidney Injury on Outcomes in Patients with Rhabdomyolysis [J]. J Clin Med, 2021, 10(9).http://doi.org/10.3390/jcm10091950\u003c/li\u003e\n\u003cli\u003eSimpson J P, Taylor A, Sudhan N, Menon D K, Lavinio A. Rhabdomyolysis and acute kidney injury: creatine kinase as a prognostic marker and validation of the McMahon Score in a 10-year cohort: A retrospective observational evaluation [J]. Eur J Anaesthesiol, 2016, 33(12): 906-12.http://doi.org/10.1097/eja.0000000000000490\u003c/li\u003e\n\u003cli\u003eLi F, Xin H, Zhang J, Fu M, Zhou J, Lian Z. Prediction model of in-hospital mortality in intensive care unit patients with heart failure: machine learning-based, retrospective analysis of the MIMIC-III database [J]. BMJ Open, 2021, 11(7): e044779.http://doi.org/10.1136/bmjopen-2020-044779\u003c/li\u003e\n\u003cli\u003eLi M, Han S, Liang F, Hu C, Zhang B, Hou Q, Zhao S. Machine Learning for Predicting Risk and Prognosis of Acute Kidney Disease in Critically Ill Elderly Patients During Hospitalization: Internet-Based and Interpretable Model Study [J]. J Med Internet Res, 2024, 26: e51354.http://doi.org/10.2196/51354\u003c/li\u003e\n\u003cli\u003eNemati S, Holder A, Razmi F, Stanley M D, Clifford G D, Buchman T G. An Interpretable Machine Learning Model for Accurate Prediction of Sepsis in the ICU [J]. Crit Care Med, 2018, 46(4): 547-53.http://doi.org/10.1097/ccm.0000000000002936\u003c/li\u003e\n\u003cli\u003eLi L, Ding L, Zhang Z, Zhou L, Zhang Z, Xiong Y, et al. Development and Validation of Machine Learning-Based Models to Predict In-Hospital Mortality in Life-Threatening Ventricular Arrhythmias: Retrospective Cohort Study [J]. J Med Internet Res, 2023, 25: e47664.http://doi.org/10.2196/47664\u003c/li\u003e\n\u003cli\u003eDong J, Feng T, Thapa-Chhetry B, Cho B G, Shum T, Inwald D P, et al. Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care [J]. Crit Care, 2021, 25(1): 288.http://doi.org/10.1186/s13054-021-03724-0\u003c/li\u003e\n\u003cli\u003eLiu C, Liu X, Mao Z, Hu P, Li X, Hu J, et al. Interpretable Machine Learning Model for Early Prediction of Mortality in ICU Patients with Rhabdomyolysis [J]. Med Sci Sports Exerc, 2021, 53(9): 1826-34.http://doi.org/10.1249/mss.0000000000002674\u003c/li\u003e\n\u003cli\u003eKaewput W, Thongprayoon C, Petnak T, Cheungpasitporn W, Qureshi F, Boonpheng B, et al. Rhabdomyolysis among hospitalized patients for salicylate intoxication in the United States: Nationwide inpatient sample 2003-2014 [J]. PLoS One, 2021, 16(3): e0248242.http://doi.org/10.1371/journal.pone.0248242\u003c/li\u003e\n\u003cli\u003eMcMahon G M, Zeng X, Waikar S S. A risk prediction score for kidney failure or mortality in rhabdomyolysis [J]. JAMA Intern Med, 2013, 173(19): 1821-8.http://doi.org/10.1001/jamainternmed.2013.9774\u003c/li\u003e\n\u003cli\u003eGupta A, Thorson P, Penmatsa K R, Gupta P. Rhabdomyolysis: Revisited [J]. Ulster Med J, 2021, 90(2): 61-9\u003c/li\u003e\n\u003cli\u003ede Fallois J, Scharm R, Lindner T H, Scharf C, Petros S, Weidhase L. Kidney replacement and conservative therapies in rhabdomyolysis: a retrospective analysis [J]. BMC Nephrol, 2024, 25(1): 96.http://doi.org/10.1186/s12882-024-03536-8\u003c/li\u003e\n\u003cli\u003eMartinez T, Harrois A, Codorniu A, Mongardon N, Pissot M, Popoff B, et al. Evaluation of severe rhabdomyolysis on day 30 mortality in trauma patients admitted to intensive care: a propensity score analysis of the Traumabase registry [J]. Crit Care, 2024, 28(1): 382.http://doi.org/10.1186/s13054-024-05158-w\u003c/li\u003e\n\u003cli\u003eSun K, Shi Z, Abudureheman Y, Liu Q, Zhao Y, Zhang X, et al. Clinical and Epidemiological Characteristics of Rhabdomyolysis: A Retrospective Study [J]. Int J Clin Pract, 2023, 2023: 6396576.http://doi.org/10.1155/2023/6396576\u003c/li\u003e\n\u003cli\u003eMorin A G, Somme D, Corvol A. Rhabdomyolysis in older adults: outcomes and prognostic factors [J]. BMC Geriatr, 2024, 24(1): 46.http://doi.org/10.1186/s12877-023-04620-8\u003c/li\u003e\n\u003cli\u003eJohnson A E W, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, et al. MIMIC-IV, a freely accessible electronic health record dataset [J]. Sci Data, 2023, 10(1): 1.http://doi.org/10.1038/s41597-022-01899-x\u003c/li\u003e\n\u003cli\u003ePollard T J, Johnson A E W, Raffa J D, Celi L A, Mark R G, Badawi O. The eICU Collaborative Research Database, a freely available multi-center database for critical care research [J]. Sci Data, 2018, 5: 180178.http://doi.org/10.1038/sdata.2018.178\u003c/li\u003e\n\u003cli\u003eLiu Y, Zhu X, Xue J, Maimaitituerxun R, Chen W, Dai W. Machine learning models for mortality prediction in critically ill patients with acute pancreatitis-associated acute kidney injury [J]. Clin Kidney J, 2024, 17(10): sfae284.http://doi.org/10.1093/ckj/sfae284\u003c/li\u003e\n\u003cli\u003eTakkavatakarn K, Oh W, Chan L, Hofer I, Shawwa K, Kraft M, et al. Machine learning derived serum creatinine trajectories in acute kidney injury in critically ill patients with sepsis [J]. Crit Care, 2024, 28(1): 156.http://doi.org/10.1186/s13054-024-04935-x\u003c/li\u003e\n\u003cli\u003eLuo W, Phung D, Tran T, Gupta S, Rana S, Karmakar C, et al. Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary View [J]. J Med Internet Res, 2016, 18(12): e323.http://doi.org/10.2196/jmir.5870\u003c/li\u003e\n\u003cli\u003eBai J, Huang J H, Price C P E, Schauer J M, Suh L A, Harmon R, et al. Prognostic factors for polyp recurrence in chronic rhinosinusitis with nasal polyps [J]. J Allergy Clin Immunol, 2022, 150(2): 352-61.e7.http://doi.org/10.1016/j.jaci.2022.02.029\u003c/li\u003e\n\u003cli\u003eDong B, Zhang H, Duan Y, Yao S, Chen Y, Zhang C. Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma [J]. J Transl Med, 2024, 22(1): 455.http://doi.org/10.1186/s12967-024-05203-w\u003c/li\u003e\n\u003cli\u003eFan Z, Jiang J, Xiao C, Chen Y, Xia Q, Wang J, et al. Construction and validation of prognostic models in critically Ill patients with sepsis-associated acute kidney injury: interpretable machine learning approach [J]. J Transl Med, 2023, 21(1): 406.http://doi.org/10.1186/s12967-023-04205-4\u003c/li\u003e\n\u003cli\u003eYue S, Li S, Huang X, Liu J, Hou X, Zhao Y, et al. Machine learning for the prediction of acute kidney injury in patients with sepsis [J]. J Transl Med, 2022, 20(1): 215.http://doi.org/10.1186/s12967-022-03364-0\u003c/li\u003e\n\u003cli\u003eYang D, Zhao L, Kang J, Wen C, Li Y, Ren Y, et al. Development and validation of a predictive model for acute kidney injury in patients with moderately severe and severe acute pancreatitis [J]. Clin Exp Nephrol, 2022, 26(8): 770-87.http://doi.org/10.1007/s10157-022-02219-8\u003c/li\u003e\n\u003cli\u003eHou N, Li M, He L, Xie B, Wang L, Zhang R, et al. Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost [J]. J Transl Med, 2020, 18(1): 462.http://doi.org/10.1186/s12967-020-02620-5\u003c/li\u003e\n\u003cli\u003eTseng P Y, Chen Y T, Wang C H, Chiu K M, Peng Y S, Hsu S P, et al. Prediction of the development of acute kidney injury following cardiac surgery by machine learning [J]. Crit Care, 2020, 24(1): 478.http://doi.org/10.1186/s13054-020-03179-9\u003c/li\u003e\n\u003cli\u003eZhang Z, Ho K M, Hong Y. Machine learning for the prediction of volume responsiveness in patients with oliguric acute kidney injury in critical care [J]. Crit Care, 2019, 23(1): 112.http://doi.org/10.1186/s13054-019-2411-z\u003c/li\u003e\n\u003cli\u003eDong L, Liu P, Qi Z, Lin J, Duan M. Development and validation of a machine-learning model for predicting the risk of death in sepsis patients with acute kidney injury [J]. Heliyon, 2024, 10(9): e29985.http://doi.org/10.1016/j.heliyon.2024.e29985\u003c/li\u003e\n\u003cli\u003eSong X, Liu X, Liu F, Wang C. Comparison of machine learning and logistic regression models in predicting acute kidney injury: A systematic review and meta-analysis [J]. Int J Med Inform, 2021, 151: 104484.http://doi.org/10.1016/j.ijmedinf.2021.104484\u003c/li\u003e\n\u003cli\u003eO\u0026apos;Sullivan E D, Hughes J, Ferenbach D A. Renal Aging: Causes and Consequences [J]. J Am Soc Nephrol, 2017, 28(2): 407-20.http://doi.org/10.1681/asn.2015121308\u003c/li\u003e\n\u003cli\u003eShen J, Chu Y, Wang C, Yan S. Risk factors for acute kidney injury after major abdominal surgery in the elderly aged 75 years and above [J]. BMC Nephrol, 2022, 23(1): 224.http://doi.org/10.1186/s12882-022-02822-7\u003c/li\u003e\n\u003cli\u003eWang M, Wang X, Zhu B, Li W, Jiang Q, Zuo Y, et al. The effects of timing onset and progression of AKI on the clinical outcomes in AKI patients with sepsis: a prospective multicenter cohort study [J]. Ren Fail, 2023, 45(1): 1-10.http://doi.org/10.1080/0886022x.2022.2138433\u003c/li\u003e\n\u003cli\u003eWahab A, Smith R J, Lal A, Flurin L, Malinchoc M, Dong Y, Gajic O. CHARACTERISTICS AND PREDICTORS OF PATIENTS WITH SEPSIS WHO ARE CANDIDATES FOR MINIMALLY INVASIVE APPROACH OUTSIDE OF INTENSIVE CARE UNIT [J]. Shock, 2023, 59(5): 702-7.http://doi.org/10.1097/shk.0000000000002112\u003c/li\u003e\n\u003cli\u003eLindner G, Funk G-C. Hypernatremia in critically ill patients [J]. Journal of Critical Care, 2013, 28(2): 216.e11-.e20.http://doi.org/https://doi.org/10.1016/j.jcrc.2012.05.001\u003c/li\u003e\n\u003cli\u003eAtlani M, Kumar A, Pakhare A P, Singhai A, Gadwala R. Potential Association of Hypernatremia With Mortality in Patients With Acute Kidney Injury and COVID-19 [J]. Cureus, 2022, 14(7): e27530.http://doi.org/10.7759/cureus.27530\u003c/li\u003e\n\u003cli\u003eJung S Y, Kim H, Park S, Jhee J H, Yun H R, Kim H, et al. Electrolyte and mineral disturbances in septic acute kidney injury patients undergoing continuous renal replacement therapy [J]. Medicine (Baltimore), 2016, 95(36): e4542.http://doi.org/10.1097/md.0000000000004542\u003c/li\u003e\n\u003cli\u003eGao C, Peng L. Association and prediction of red blood cell distribution width to albumin ratio in all-cause mortality of acute kidney injury in critically ill patients [J]. Front Med (Lausanne), 2023, 10: 1047933.http://doi.org/10.3389/fmed.2023.1047933\u003c/li\u003e\n\u003cli\u003eHong J, Hu X, Liu W, Qian X, Jiang F, Xu Z, et al. Impact of red cell distribution width and red cell distribution width/albumin ratio on all-cause mortality in patients with type 2 diabetes and foot ulcers: a retrospective cohort study [J]. Cardiovasc Diabetol, 2022, 21(1): 91.http://doi.org/10.1186/s12933-022-01534-4\u003c/li\u003e\n\u003cli\u003eRabb H, Griffin M D, McKay D B, Swaminathan S, Pickkers P, Rosner M H, et al. Inflammation in AKI: Current Understanding, Key Questions, and Knowledge Gaps [J]. J Am Soc Nephrol, 2016, 27(2): 371-9.http://doi.org/10.1681/asn.2015030261\u003c/li\u003e\n\u003cli\u003eYu M Y, Lee S W, Baek S H, Na K Y, Chae D W, Chin H J, Kim S. Hypoalbuminemia at admission predicts the development of acute kidney injury in hospitalized patients: A retrospective cohort study [J]. PLoS One, 2017, 12(7): e0180750.http://doi.org/10.1371/journal.pone.0180750\u003c/li\u003e\n\u003cli\u003eThongprayoon C, Cheungpasitporn W, Chewcharat A, Mao M A, Thirunavukkarasu S, Kashani K B. Impacts of admission serum albumin levels on short-term and long-term mortality in hospitalized patients [J]. Qjm, 2020, 113(6): 393-8.http://doi.org/10.1093/qjmed/hcz305\u003c/li\u003e\n\u003cli\u003eLi X, Li X, Zhao W, Wang D. Development and validation of a nomogram for predicting in-hospital death in cirrhotic patients with acute kidney injury [J]. BMC Nephrol, 2024, 25(1): 175.http://doi.org/10.1186/s12882-024-03609-8\u003c/li\u003e\n\u003cli\u003eZhou H, Liu L, Zhao Q, Jin X, Peng Z, Wang W, et al. Machine learning for the prediction of all-cause mortality in patients with sepsis-associated acute kidney injury during hospitalization [J]. Front Immunol, 2023, 14: 1140755.http://doi.org/10.3389/fimmu.2023.1140755\u003c/li\u003e\n\u003cli\u003eFolkestad T, Brurberg K G, Nordhuus K M, Tveiten C K, Guttormsen A B, Os I, Beitland S. Acute kidney injury in burn patients admitted to the intensive care unit: a systematic review and meta-analysis [J]. Crit Care, 2020, 24(1): 2.http://doi.org/10.1186/s13054-019-2710-4\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"rhabdomyolysis, acute kidney injury, in-hospital mortality, machine learning","lastPublishedDoi":"10.21203/rs.3.rs-6824634/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6824634/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground.\u003c/h2\u003e \u003cp\u003eRhabdomyolysis-associated acute kidney injury (RI-AKI) is a serious complication in critically ill patients and is associated with increased in-hospital mortality. However, limited research has focused on predictive modeling of in-hospital mortality among this population.\u003c/p\u003e\u003ch2\u003eObjective.\u003c/h2\u003e \u003cp\u003eTo develop and evaluate machine learning (ML) models for predicting in-hospital mortality in critically ill patients with RI-AKI.\u003c/p\u003e\u003ch2\u003eMethods.\u003c/h2\u003e \u003cp\u003eData were extracted from the MIMIC-IV and eICU Collaborative Research Databases. Patients with RI-AKI were identified, and relevant clinical variables\u0026mdash;including demographics, vital signs, laboratory indicators, comorbidities/complications, and treatments within the first 24 hours of ICU admission\u0026mdash;were collected. The combined dataset was randomly divided into training and testing sets in an 8:2 ratio. Feature selection was performed using the least absolute shrinkage and selection operator (LASSO) and random forest (RF). ML models were constructed using Extreme Gradient Boosting (XGBoost), RF, and logistic regression (LR). Model performance was assessed by area under the receiver operating characteristic curve (AUC), Brier score, sensitivity, specificity, and calibration.\u003c/p\u003e\u003ch2\u003eResults.\u003c/h2\u003e \u003cp\u003eTen key predictors, including age, sodium, phosphorus, and coagulation markers, were identified. In the training set, the XGBoost model achieved the highest AUC (0.889; 95% CI: 0.872\u0026ndash;0.908), outperforming RF (0.797) and LR (0.740). Brier scores were 0.122, 0.185, and 0.203, respectively. Similar results were observed in the testing set.\u003c/p\u003e\u003ch2\u003eConclusions.\u003c/h2\u003e \u003cp\u003eThe XGBoost model demonstrated superior performance in predicting in-hospital mortality among critically ill RI-AKI patients, indicating its potential value in clinical risk stratification. Further external validation is warranted.\u003c/p\u003e","manuscriptTitle":"Machine learning models to predict in-hospital mortality in patients with rhabdomyolysis combined with acute kidney injury","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-24 12:32:51","doi":"10.21203/rs.3.rs-6824634/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8a78faf5-2be9-4877-b00f-6f9ea397a19e","owner":[],"postedDate":"June 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-20T16:23:14+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-24 12:32:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6824634","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6824634","identity":"rs-6824634","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}