{"paper_id":"469b8a6c-3a2c-49e7-8416-74367e850ac9","body_text":"Prediction model of intradialytic hypertension in hemodialysis patients based on machine learning | 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 Prediction model of intradialytic hypertension in hemodialysis patients based on machine learning Hongming Zhou, Qi Guo, Kang wang, Xinzhou Zhang, Yehua Luo, Shaodong Luan, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5355171/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 Objective : The global prevalence of chronic kidney disease (CKD) is escalating, particularly for end-stage renal disease (ESRD), which has led to greater dependence on hemodialysis. This upswing exerts substantial strains on patient families and healthcare systems. A critical concern during hemodialysis is the emergence of Intradialytic Hypertension (IDH), which carries significant health risks. Delayed management of IDH can lead to severe cardiovascular and cerebrovascular complications. The aim of our study was to harness machine learning methodologies to develop a predictive algorithm for IDH, utilizing patient demographic data and dialysis records. Our model equips medical professionals with a robust predictive tool that enhances the detection of patients more susceptible to hypertension during dialysis, thereby advancing the pre-screening for individuals considered at increased risk. Methods : This study developed two predictive models for IDH, named IDH-1 and IDH-2, by employing a suite of machine learning algorithms, namely the Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and TabNet. IDH-1 is specifically engineered to provide immediate predictions of IDH risk prior to a hemodialysis session, utilizing records from the imminent pre-dialysis period combined with historical average dialysis data, whereas IDH-2 employs records from the current dialysis session along with historical average data to forecast the risk of IDH for the next hemodialysis session. The performance evaluation of the models utilized key metrics, including Area Under the Curve (AUC), recall, accuracy, and F1 score, which are crucial in determining the models' precision and reliability. Results : This research analyzed data from 1,405 patients at Shenzhen People's Hospital over 185,125 dialysis sessions and 416 patients at Fuding City Hospital across 71,427 sessions. Data from Shenzhen served as the training set, while Fuding data comprised the test set, supporting the model development and validation process. In the IDH-1 models, the LGBM outperformed SVM and TabNet with an AUC of 0.87. LGBM achieved a recall of 0.73, an accuracy of 0.243, and an F1 score of 0.36. For IDH-2 models, LGBM maintained superior performance, with an AUC of 0.75, a recall of 0.56, an accuracy of 0.17, and an F1 score of 0.26. Predictor importance analysis for the LGBM algorithm identified pre-dialysis diastolic pressures, historical mean arterial pressure, and historical average IDH episodes as significant for the IDH-1 model. For the IDH-2 model, historical average IDH episodes and post-dialysis systolic pressures were most predictive. Conclusions : This study's results highlight the significant potential of machine learning techniques in leveraging demographic and dialysis data to predict IDH in patients undergoing hemodialysis. Chronic Kidney Disease Hemodialysis Machine Learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Chronic kidney disease (CKD) poses a significant challenge to global public health, with epidemiological studies indicating that approximately 13.4% of the global population is affected [ 1 ] . There is considerable variation in CKD prevalence across Asia, with China and India together accounting for 69.1% of the adult CKD cases in the region [ 2 ] . According to the 2015 Global Burden of Disease Study, CKD-related factors were responsible for an estimated 1.2 million deaths [ 3 ] . The prevalence of CKD is on the rise, driven by increasing rates of diabetes and hypertension, as well as an aging population worldwide [ 4 ] . CKD often progresses to end-stage renal disease (ESRD) [ 5 ] , requiring renal replacement therapies such as hemodialysis (HD), peritoneal dialysis, and kidney transplantation. Managing ESRD through hemodialysis is essential [ 6 ] but fraught with multiple potential complications, including cardiovascular disease, infections, vascular access issues, hypotension, electrolyte imbalances, and hypertension. Intradialytic hypertension (IDH) is a common complication observed during dialysis, with its definition evolving over the years [ 7 ] . Initially, Inrig et al. [ 8 ] proposed defining IDH as mean systolic pressures increase exceeding 10 mmHg over three hemodialysis HD sessions. This definition was later expanded by Sebastian et al. [ 9 ] , who required a rise in systolic pressure over 10 mmHg in at least four out of six HD sessions. Shamir et al. [ 10 ] simplified this criterion to a systolic pressure increase of 10 mmHg or more during HD. Raja et al. [ 11 ] adopted a stricter definition, identifying IDH as a 15 mmHg increase in mean arterial pressure (MAP) during or immediately post-dialysis. The KDIGO (kidney disease: Improving Global Outcomes) recently issued a recommendation that a diagnosis of IDH should be considered when there is an increase in systolic blood pressure of more than 10 mmHg before and after dialysis in at least four out of six consecutive HD sessions [ 12 ] . While the definition of IDH has been refined and developed over the past decade, a globally unified clinical standard is yet to be established. In the scientific literature, the prevalence of IDH is reported to range from 17–23% [ 7 , 13 , 14 ] , a variability that may be ascribed to the lack of uniform IDH identification criteria. In patients with ESRD undergoing HD, IDH is implicated in elevating the risk of mortality [ 15 , 16 ] , making it a crucial factor in prognostication [ 17 ] . The pathophysiological underpinnings of IDH are not fully understood, but it is widely recognized to arise from multiple interrelated factors, including overactivation of the renin-angiotensin system, sympathetic nervous system hyperactivity, volume and sodium excess, electrolyte shifts during dialysis, and endothelial dysfunction [ 18 , 19 ] . These elements, individually or synergistically, play central roles in the development of IDH. The absence of a standard definition for IDH hampers the comparison of research outcomes and diminishes their clinical applicability. Nonetheless, the evidence underscores the significance of IDH in clinical settings [ 20 ] , advocating for early interventions that could profoundly improve patient well-being and long-term outcomes. Artificial intelligence refers to the capability of machines to emulate human intelligent behavior [ 21 ] . Machine learning (ML), a crucial subset of AI, enables systems to discern patterns and learn from data input, ultimately endowing them with predictive and decision-making faculties [ 22 ] . The rapid progress in digital technologies has catalyzed significant advancements in ML, with its successful integration into diverse fields such as agriculture [ 23 ] , fisheries [ 24 ] , and healthcare [ 25 ] . ML's deployment spans a wide array of diagnostic and therapeutic domains, including the respiratory [ 26 ] , cardiovascular [ 27 ] , digestive [ 28 ] , urinary [ 29 ] , hematological [ 30 ] , and endocrine systems [ 31 ] . In nephrology, ML techniques have proven particularly effective for the prognostication of renal pathologies, exemplified by the development of predictive models for acute kidney injury [ 32 – 35 ] and chronic renal failure [ 36 – 38 ] . Notable is the model by Chuah et al. [ 39 ] aimed at predicting progression to ESRD, which has demonstrated potential in decelerating the progression of CKD. Recent studies have increasingly focused on the development of machine learning-based predictive models for hypotension during dialysis sessions [ 40 – 42 ] . However, the development of predictive models for hypertension in the same context has been considerably less explored. Addressing this gap, leveraging ML algorithms to analyze both historic and real-time data may unearth patterns integral to IDH, providing clinicians with a tool to foresee and mitigate the risk of IDH in HD patients, thus enhancing clinical outcomes and patient quality of life. To improve the model's generalization capabilities while accounting for differences in interspatial data, we adopted a cross-regional data validation approach. This method entails constructing a predictive model with data from one region and subsequently validating it using data from a different region, thus confirming the model's wide-ranging applicability. The collection of extensive data on dialysis sessions and patient demographics during the study enriched the development of the IDH risk prediction model. We conducted a comparative analysis of various models to identify the most effective decision-support tool, thereby enhancing clinicians' ability to make informed treatment decisions. 2. Methods 2.1 Study population and data collection In this study, demographic data and HD records were collected from patients undergoing HD at the People's Hospital in Shenzhen, China, from January 2017 to October 2022, and at the Hospital of Fuding City, China, from October 2020 to August 2022. Participants were aged between 16 and 100 years with recorded dry body weights. Each patient received dialysis treatments lasting between 3 and 4 hours. Exclusion criteria were applied to patients with any diagnosis of cancer, acute renal failure, acute renal insufficiency, or acute kidney injury documented in their dialysis records, as well as to those with a missing value ratio exceeding 35%. The study received ethical approval from the Ethics Committees of both Shenzhen People's Hospital and Fuding Hospital in China. 2.2 Data Preprocessing In addressing outliers within continuous variables, data points exceeding the 99th percentile were replaced with the 99th percentile value, and those below the 1st percentile were replaced with the 1st percentile value. For categorical features with missing values, decision trees were employed for prediction. Meanwhile, for continuous features, the mean of each feature was calculated and used to impute missing values. To demonstrate our model's external validation, training was conducted using the \"Shenyi\" dataset, followed by performance evaluation on the distinct \"Fuding\" dataset. 2.3 Model Input Features and IDH Definition In this study, two predictive models, IDH-1 and IDH-2, were developed to evaluate the risk of IDH. Model IDH-1 aims for real-time prediction of IDH risk, integrating immediate pre-treatment clinical and historical average dialysis data during the current HD session. This model applies a binary classification system, where a '0' indicates no IDH events and a '1' denotes the occurrence of IDH. Conversely, model IDH-2 predicts IDH risk for the next session (n + 1) by analyzing data from the present HD session (n) alongside historical averages. This prospective strategy allows clinicians to adjust treatment plans in advance, with '0' representing the absence of predicted IDH events and '1' indicating their anticipated occurrence. In our study, we conducted a review of patient demographics and dialysis treatment records to gather a comprehensive dataset. This dataset included patient gender, age, and infectious disease status, along with detailed dialysis treatment parameters such as the duration of treatment (recorded in both days and years), modality used, and the type of anticoagulant administered. Key physiological measurements were also recorded, including systolic and diastolic pressures before and after dialysis sessions, body weight measurements before and after treatment, the initial respiratory rate, estimated dry weight, and various dialysis session parameters like dialysate calcium concentration, conductivity, and temperature, as well as transmembrane pressure. Vascular access characteristics were detailed by recording the type, location, and duration of use of the fistula. Clinical metrics during dialysis, such as intra-dialytic pressures readings, pulse, ultrafiltration rate and volume, venous pressure, MAP, and blood flow rate, were also captured. Additionally, the study involved aggregating certain variables, from which mean values and standard deviations were calculated. Baseline pre-dialysis measurements were acquired 10 to 15 minutes before the start of each hemodialysis treatment. Monitoring throughout the hemodialysis consisted of hourly assessments of blood pressure, pulse, ultrafiltration rate, and fluid volume. IDH was defined as a rise in MAP by more than 15 mmHg above the pre-dialysis level, with the MAP being calculated as (2× diastolic pressure + systolic pressure) / 3 mmHg. Detailed descriptions and classifications of model variables are presented in Table 1. The development of the prediction model utilized three machine learning algorithms, namely Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and TabNet, and was executed in the Python programming language. Details on the hyperparameter optimization for each algorithm are provided in Table S3. 2.4 Statistical analysis In this study, continuous variables conforming to a normal distribution were expressed as mean ± standard deviation (x̄±s), while those not following a normal distribution were described using the median and interquartile range [Q1, Q3]. Categorical variables were reported as counts and percentages [n (%)]. Statistical analyses were conducted using the t-test for normally distributed continuous variables and the Wilcoxon rank-sum test for non-normally distributed continuous variables. Differences in categorical variables between groups were assessed using the Chi-square test. A P-value of less than 0.05 was considered to indicate statistical significance. The predictive capability of the model in this study was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), also known as the c statistic. Additionally, the model's performance was comprehensively assessed by metrics including recall, precision, and the F1 score. Data analysis was performed using Python version 3.7.3. DeLong's test was employed to compare the AUC of different predictive variables. Furthermore, the identification of significant predictors of IDH involved ranking feature importance, which was based on the frequency of feature use for node splitting during the training of decision trees, thus providing insights into the variables most critical for enhancing the model's predictive performance. 3. Results 3.1 Cohort Characteristics This retrospective study involved analyzing HD treatment records from two centers: Shenzhen People's Hospital (March 3, 2017, to October 8, 2022) and Fuding City Hospital (October 31, 2020, to August 13, 2022). At Shenzhen People's Hospital, the initial dataset comprised 222,858 HD sessions for 1,693 patients. Exclusion criteria included diagnoses of acute renal conditions or cancer, resulting in the removal of 22 records, while no patients were excluded. Additionally, 56 patients and 6,232 sessions were excluded due to the absence of systolic, diastolic, or mean arterial pressure readings during dialysis. An additional 195 patients and 11,601 sessions were excluded based on incomplete dry weight records, and 653 sessions were excluded due to age criteria (under 16 or over 100 years). The exclusion of sessions outside the 3- to 4-hour duration or with more than 35% data missingness resulted in the removal of 37 patients and 19,225 records. Due to the criterion of having fewer than two dialysis sessions, 97 patients and 1,405 records were excluded from the IDH-2 model development. The final datasets for the IDH-1 model included 185,125 sessions from 1,405 patients, while the IDH-2 model considered 183,720 sessions from 1,308 patients. At Fuding City Hospital, the dataset of 434 HD patients comprised 77,808 sessions. Application of the same exclusion criteria removed one patient and 1,289 records due to acute illnesses or cancers, three patients and 3,168 records due to lack of blood pressure data during dialysis, and 14 patients and 602 records due to incomplete dry weight data. Although age criteria (under 16 or over 100 years old) were set for exclusion within the screening protocol, they resulted in no patient or record exclusions. Sessions failing to meet the duration criteria or having high data missingness accounted for 1,322 excluded records. For the IDH-2 model, the exclusion of patients with fewer than two sessions reduced the dataset by 416 records. Ultimately, the IDH-1 model analysis for Fuding City Hospital included 416 patients and 71,427 sessions, and the IDH-2 model dataset comprised the same number of patients but with 71,011 sessions. The exclusion process for both centers is detailed in Figs. 1 . For this analysis, the dataset comprised a training set of 185,125 entries from Shenzhen and a test set of 71,427 entries from Fuding. In the training set applicable to both IDH-1 and IDH-2 models, the gender distribution was 34.3% female and 65.7% male, with an observed IDH occurrence rate of 14%. Conversely, the test set showed a slightly higher proportion of females at 39.9% and a male proportion of 60.1%, alongside a lower IDH occurrence rate of 6% for both models. Detailed variable distributions for the data records are presented in Tables 1 to 4 . Tables 1. Characteristics of patients in the IDH-1 model training (Shenyi) cohort. Characteristic all(n = 185125) non- IDH(n = 159290) IDH(n = 25835) P female, n (%) 63561 (34.3) 53545 (33.6) 10016 (38.8) < 0.001 male, n (%) 121564 (65.7) 105745 (66.4) 15819 (61.2) pre-dialysis systolic pressure(mmHg) 142.0 ± 22.3 143.4 ± 22.2 133.8 ± 21.0 < 0.001 pre-dialysis diastolic pressure(mmHg) 83.4 ± 13.4 84.6 ± 13.0 76.2 ± 13.2 < 0.001 dialysis age (days) 1865.4 ± 1581.3 1886.8 ± 1584.0 1733.2 ± 1558.4 < 0.001 age(year), median [Q1,Q3] 56.0 [46.0,67.0] 55.0 [46.0,67.0] 59.0 [48.0,71.0] < 0.001 pre-dialysis weight(kg) 63.4 ± 12.0 63.7 ± 12.0 61.3 ± 11.7 < 0.001 dry weight(kg) 60.4 ± 11.5 60.7 ± 11.5 58.5 ± 11.2 < 0.001 age at first dialysis(year) 51.9 ± 15.5 51.4 ± 15.5 54.5 ± 15.5 < 0.001 historical average pre-dialysis weight(kg) 63.1 ± 12.2 63.4 ± 12.2 61.3 ± 12.0 < 0.001 historical average dry weight(kg) 60.1 ± 11.7 60.4 ± 11.7 58.4 ± 11.5 < 0.001 historical average dialysate calcium concentration(mmol/l) 1.5 ± 0.1 1.5 ± 0.1 1.5 ± 0.1 < 0.001 historical average pre-dialysis systolic pressure(mmHg) 142.0 ± 17.1 141.9 ± 17.1 142.4 ± 17.0 0.001 historical average pre-dialysis diastolic pressure(mmHg) 83.6 ± 10.8 83.8 ± 10.7 82.4 ± 11.3 < 0.001 historical average IDH episodes 0.1 ± 0.1 0.1 ± 0.1 0.2 ± 0.2 < 0.001 historical average intradialytic systolic pressure(mmHg) 139.8 ± 16.1 139.2 ± 15.9 143.2 ± 16.5 < 0.001 historical average intradialytic diastolic pressure(mmHg) 84.1 ± 10.8 84.0 ± 10.6 84.4 ± 11.6 < 0.001 historical average venous pressure(mmHg) 125.6 ± 15.5 126.1 ± 15.3 122.5 ± 16.2 < 0.001 historical average MAP(mmHg) 102.8 ± 10.8 102.6 ± 10.6 104.2 ± 11.4 < 0.001 historical average blood flow rate(ml/min) 239.3 ± 22.0 240.0 ± 22.0 235.0 ± 21.6 < 0.001 historical average dialysate temperature (℃) 36.3 ± 0.8 36.3 ± 0.8 36.3 ± 0.8 1 Tables 2. Characteristics of patients in the IDH-1 model test (Fuding) cohort. Characteristic all(n = 71427) non-IDH(n = 67032) IDH(n = 4395) P female, n (%) 28526 (39.9) 26916 (40.2) 1610 (36.6) < 0.001 male, n (%) 42901 (60.1) 40116 (59.8) 2785 (63.4) pre-dialysis systolic pressure(mmHg) 156.9 ± 22.6 158.0 ± 22.4 141.2 ± 19.7 < 0.001 pre-dialysis diastolic pressure(mmHg) 82.8 ± 14.6 83.7 ± 14.3 69.2 ± 12.7 < 0.001 dialysis age (days) 1579.4 ± 1390.1 1603.5 ± 1402.7 1211.7 ± 1119.5 < 0.001 age(year), median [Q1,Q3] 58.0 [49.0,67.0] 58.0 [49.0,67.0] 59.0 [49.0,67.0] < 0.001 pre-dialysis weight(kg) 59.0 ± 10.4 58.9 ± 10.4 59.3 ± 10.5 0.88 dry weight(kg) 56.0 ± 10.1 56.0 ± 10.1 56.5 ± 10.5 0.045 age at first dialysis(year) 52.6 ± 13.6 52.5 ± 13.6 54.7 ± 13.9 < 0.001 historical average pre-dialysis weight(kg) 59.0 ± 10.7 58.9 ± 10.7 59.7 ± 11.3 < 0.001 historical average dry weight(kg) 56.0 ± 10.4 56.0 ± 10.3 56.5 ± 10.9 0.045 historical average dialysate calcium concentration(mmol/l) 1.5 ± 0.0 1.5 ± 0.0 1.5 ± 0.0 0.914 historical average pre-dialysis systolic pressure(mmHg) 158.2 ± 16.3 158.3 ± 16.3 156.0 ± 16.1 < 0.001 historical average pre-dialysis diastolic pressure(mmHg) 83.2 ± 12.1 83.5 ± 12.0 78.4 ± 12.0 < 0.001 historical average IDH episodes 0.1 ± 0.1 0.1 ± 0.1 0.2 ± 0.1 < 0.001 historical average intradialytic systolic pressure(mmHg) 145.0 ± 14.8 144.6 ± 14.7 151.2 ± 15.2 < 0.001 historical average intradialytic diastolic pressure(mmHg) 77.5 ± 9.2 77.5 ± 9.1 77.7 ± 9.9 1 historical average venous pressure(mmHg) 111.1 ± 22.0 111.4 ± 21.9 106.7 ± 22.9 < 0.001 historical average MAP(mmHg) 100.1 ± 9.3 99.9 ± 9.2 102.2 ± 9.9 < 0.001 historical average blood flow rate(ml/min) 242.3 ± 11.3 242.5 ± 11.2 239.7 ± 13.3 < 0.001 historical average dialysate temperature (℃) 37.0 ± 1.5 37.0 ± 1.5 36.9 ± 1.4 < 0.001 Tables 3. Characteristics of patients in the IDH-2 model training (Shenyi) cohort. Characteristic all(n = 183720) non-IDH(n = 158119) IDH(n = 25601) P female, n (%) 63041 (34.3) 53111 (33.6) 9930 (38.8) < 0.001 male, n (%) 120679 (65.7) 105008 (66.4) 15671 (61.2) pre-dialysis systolic pressure(mmHg) 142.0 ± 22.3 143.4 ± 22.2 133.8 ± 21.0 < 0.001 pre-dialysis diastolic pressure(mmHg) 83.4 ± 13.4 84.6 ± 13.0 76.2 ± 13.2 < 0.001 post-dialysis systolic pressure(mmHg) 139.3 ± 22.6 137.3 ± 21.8 151.6 ± 23.4 < 0.001 post-dialysis diastolic pressure(mmHg) 84.4 ± 13.8 83.5 ± 13.4 89.8 ± 15.0 < 0.001 dialysis age (days) 1873.5 ± 1580.5 1895.0 ± 1583.1 1741.0 ± 1558.0 < 0.001 age(year), median [Q1,Q3] 56.0 [46.0,67.0] 55.0 [46.0,67.0] 59.0 [48.0,71.0] < 0.001 pre-dialysis weight(kg) 63.4 ± 12.0 63.7 ± 12.0 61.3 ± 11.7 < 0.001 dry weight(kg) 60.4 ± 11.5 60.7 ± 11.5 58.5 ± 11.2 < 0.001 post-dialysis weight(kg) 60.6 ± 11.3 60.9 ± 11.3 58.9 ± 11.4 < 0.001 age at first dialysis(year) 51.8 ± 15.5 51.4 ± 15.5 54.5 ± 15.5 < 0.001 historical average intradialytic systolic pressure(mmHg) 140.1 ± 20.2 139.2 ± 19.9 145.7 ± 20.7 < 0.001 standard deviation of historical average intradialytic systolic pressure(mmHg) 8.5 ± 5.4 7.9 ± 5.1 11.8 ± 6.0 < 0.001 historical average intradialytic diastolic pressure(mmHg) 83.8 ± 12.5 83.5 ± 12.2 85.3 ± 13.8 < 0.001 standard deviation ofhistorical average intradialytic diastolic pressure(mmHg) 5.2 ± 4.2 4.7 ± 3.1 8.4 ± 7.2 < 0.001 historical average pre-dialysis weight(kg) 63.1 ± 12.1 63.4 ± 12.1 61.3 ± 12.0 < 0.001 historical average dry weight(kg) 60.2 ± 11.7 60.4 ± 11.7 58.4 ± 11.5 < 0.001 historical average dialysate calcium concentration(mmol/l) 1.5 ± 0.1 1.5 ± 0.1 1.5 ± 0.1 < 0.001 historical average pre-dialysis systolic pressure(mmHg) 142.0 ± 17.1 141.9 ± 17.1 142.4 ± 16.9 0.001 historical average pre-dialysis diastolic pressure(mmHg) 83.6 ± 10.8 83.8 ± 10.7 82.4 ± 11.2 < 0.001 historical average IDH episodes 0.1 ± 0.1 0.1 ± 0.1 0.2 ± 0.2 < 0.001 historical average intradialytic systolic pressure(mmHg) 139.8 ± 16.0 139.2 ± 15.9 143.2 ± 16.4 < 0.001 historical average intradialytic diastolic pressure(mmHg) 84.1 ± 10.7 84.1 ± 10.6 84.4 ± 11.6 < 0.001 historical average venous pressure(mmHg) 125.7 ± 15.3 126.2 ± 15.1 122.5 ± 16.1 < 0.001 historical average MAP(mmHg) 102.8 ± 10.7 102.6 ± 10.6 104.2 ± 11.4 < 0.001 historical average blood flow rate(ml/min) 239.5 ± 21.9 240.2 ± 21.9 235.1 ± 21.5 < 0.001 historical average dialysate temperature (℃) 36.3 ± 0.7 36.3 ± 0.7 36.3 ± 0.8 1 Tables 4. Characteristics of patients in the IDH-2 model test (Fuding) cohort. Characteristic all(n = 71011) non-IDH(n = 66648) IDH(n = 4363) P female, n (%) 28366 (39.9) 26772 (40.2) 1594 (36.5) < 0.001 male, n (%) 42645 (60.1) 39876 (59.8) 2769 (63.5) pre-dialysis systolic pressure(mmHg) 157.0 ± 22.6 158.0 ± 22.4 141.2 ± 19.6 < 0.001 pre-dialysis diastolic pressure(mmHg) 82.9 ± 14.6 83.8 ± 14.3 69.3 ± 12.6 < 0.001 post-dialysis systolic pressure(mmHg) 139.0 ± 21.1 137.9 ± 20.6 156.9 ± 20.1 < 0.001 post-dialysis diastolic pressure(mmHg) 75.1 ± 11.7 74.6 ± 11.5 82.3 ± 12.5 < 0.001 dialysis age (days) 1580.2 ± 1390.1 1604.3 ± 1402.7 1212.4 ± 1119.9 < 0.001 age(year), median [Q1,Q3] 58.0 [49.0,67.0] 58.0 [49.0,66.0] 59.0 [49.0,67.0] < 0.001 pre-dialysis weight(kg) 59.0 ± 10.4 58.9 ± 10.4 59.3 ± 10.5 1 dry weight(kg) 56.0 ± 10.1 56.0 ± 10.1 56.5 ± 10.5 0.08 post-dialysis weight(kg) 56.5 ± 10.1 56.4 ± 10.1 56.9 ± 9.9 0.276 age at first dialysis(year) 52.6 ± 13.6 52.5 ± 13.6 54.7 ± 13.9 < 0.001 historical average intradialytic systolic pressure(mmHg) 144.1 ± 19.0 143.5 ± 18.9 153.5 ± 18.5 < 0.001 standard deviation of historical average intradialytic systolic pressure(mmHg) 10.6 ± 6.2 10.6 ± 6.3 10.9 ± 5.0 0.006 historical average intradialytic diastolic pressure(mmHg) 77.3 ± 11.1 77.1 ± 11.1 79.9 ± 12.0 < 0.001 standard deviation ofhistorical average intradialytic diastolic pressure(mmHg) 6.1 ± 3.6 6.0 ± 3.5 8.2 ± 3.8 < 0.001 historical average pre-dialysis weight(kg) 59.0 ± 10.7 58.9 ± 10.7 59.7 ± 11.3 < 0.001 historical average dry weight(kg) 56.0 ± 10.4 56.0 ± 10.3 56.5 ± 10.9 0.079 historical average dialysate calcium concentration(mmol/l) 1.5 ± 0.0 1.5 ± 0.0 1.5 ± 0.0 1 historical average pre-dialysis systolic pressure(mmHg) 158.2 ± 16.3 158.3 ± 16.3 156.0 ± 16.1 < 0.001 historical average pre-dialysis diastolic pressure(mmHg) 83.2 ± 12.1 83.5 ± 12.0 78.4 ± 12.0 < 0.001 historical average IDH episodes 0.1 ± 0.1 0.1 ± 0.1 0.2 ± 0.1 < 0.001 historical average intradialytic systolic pressure(mmHg) 145.0 ± 14.8 144.6 ± 14.7 151.2 ± 15.2 < 0.001 historical average intradialytic diastolic pressure(mmHg) 77.5 ± 9.2 77.5 ± 9.1 77.8 ± 9.9 1 historical average venous pressure(mmHg) 111.1 ± 22.0 111.4 ± 21.9 106.7 ± 22.9 < 0.001 historical average MAP(mmHg) 100.1 ± 9.3 99.9 ± 9.2 102.3 ± 9.9 < 0.001 historical average blood flow rate(ml/min) 242.3 ± 11.3 242.5 ± 11.2 239.7 ± 13.3 < 0.001 historical average dialysate temperature(℃) 37.0 ± 1.5 37.0 ± 1.5 36.9 ± 1.4 < 0.001 3.2 Model performance comparison In assessing the performance outcomes for the IDH-1 model, it was observed that the LGBM model outperformed its counterparts, achieving the highest AUC of 0.87, as detailed in Table 5 and Fig. 2 . This was closely followed by the SVM model with an AUC of 0.87, and the TabNet model with an AUC of 0.86. Regarding recall rates, the LGBM model also led with a rate of 0.78, surpassing the SVM and TabNet models, which recorded rates of 0.57 and 0.69, respectively. While accuracy rates were similar across the models, the SVM model exhibited a slightly higher accuracy of 0.31, compared to the LGBM and TabNet models at 0.24 and 0.23, respectively. In terms of the F1 score, which measures the model's balance between precision and recall, the SVM model achieved the highest score of 0.401, followed by the LGBM and TabNet models with scores of 0.36 and 0.34, respectively. Statistical analysis of Delong revealed a significant difference between the LGBM model and the TabNet model (p-value = 0), whereas the difference between the LGBM and SVM models was not statistically significant (p-value = 0.12). This suggests that the LGBM model significantly outperforms the TabNet model in predicting IDH events among HD patients and is comparably effective to the SVM model. Further analysis of the IDH-2 models, as shown in Table 6 and Fig. 3 , demonstrates the LGBM algorithm's dominance over the SVM and TabNet models across all evaluated metrics. The LGBM algorithm secured an AUC of 0.75, significantly higher than the SVM and TabNet models, which achieved AUCs of 0.70 and 0.58, respectively. It also led in recall with a score of 0.561, markedly ahead of the SVM and TabNet scores of 0.06 and 0.22, respectively. Although the LGBM model's accuracy was lower than that of the SVM model (0.17 vs. 0.26), it exceeded the TabNet model's accuracy (0.09). Furthermore, the LGBM model excelled in F1 score with 0.26, while the SVM and TabNet models recorded scores of 0.09 and 0.13, respectively. Statistical comparisons underscored the LGBM model's performance superiority with a p-value of 0. The comparative performance of the three models across both IDH-1 and IDH-2 is illustrated in bar graphs (Figs. 4 and 5 ). Table 5 The performance of IDH-1 prediction models in test data. Model AUC Recall Accuracy F1 LGBM 0.87 0.73 0.24 0.36 SVM 0.87 0.57 0.31 0.40 TabNet 0.86 0.69 0.23 0.34 Table 6 The performance of IDH-2 prediction models in test data. Model AUC Recall Accuracy F1 LGBM 0.75 0.56 0.17 0.26 SVM 0.70 0.06 0.26 0.09 TabNet 0.58 0.22 0.09 0.13 3.3 Feature importance analysis Feature importance was determined based on the number of splits in the LGBM model, revealing the top 15 features that substantially contribute to the model. In the IDH-1 model, as shown in Fig. 6 , pre-dialysis diastolic pressure was identified as the most influential feature for predicting IDH incidents, with an importance score of 579. It was closely followed by the historical average MAP and historical average IDH episodes, with scores of 262 and 261, respectively, highlighting the predictive significance of prior blood pressure measurements. Other notable features included pre-dialysis systolic pressure (score: 255), historical average intradialytic diastolic pressure (score: 241), and historical average pre-dialysis diastolic pressure (score: 171). Additional features such as historical average intradialytic systolic pressure (score: 131) and historical average ratio of rise time points range (score: 80) also demonstrated substantial predictive relevance. Contributing to a lesser degree were features like historical average pre-dialysis systolic pressure (score: 51), historical average dialysate temperature (score: 49), historical average difference in rise time points range (score: 45), and patient age (score: 31), among others. In the context of the IDH-2 model, depicted in Fig. 7 , historical average IDH episodes was the most significant feature, with a score of 336, affirming its critical role in predicting IDH during HD. Post-dialysis systolic pressure, with a score of 232, and ratio of rise time points, with a score of 191, were also highly indicative of IDH risk. Further important features were the historical average ratio of rise time points range (score: 131), historical average blood flow rate (score: 118), and historical average venous pressure (score: 103), which mirror the influence of historical hemodynamic patterns and dialysis thermal management on IDH occurrence. Lesser but meaningful contributions to IDH prediction were observed from features such as the difference in rise time points (score: 82), historical average dialysate temperature (score: 58), among others. 4. Discussion The prevalence of IDH has been linked to increased hospitalization and mortality among HD patients [ 43 ] . Currently, IDH detection primarily relies on blood pressure readings exceeding 140/90 mmHg, highlighting the absence of a universal IDH definition. With renal disease prediction improving, machine learning approaches have outperformed traditional statistical methods [ 44 , 45 ] . This underscores the importance of developing predictive models to improve patient outcomes in HD. This study employs three machine learning models including LGBM, SVM, and TabNet to predict IDH in HD patients. These models were selected based on their proven performance on historical datasets and their adeptness at handling various types of data. Specifically, LGBM was chosen due to its efficiency and excellence in processing large datasets, as well as its remarkable performance across several predictive tasks. Such tasks include predicting sepsis-associated acute brain damage [ 46 ] , gestational diabetes [ 47 ] , and acute renal injury following nephrectomy in patients with renal cell carcinoma [ 48 ] . Although the SVM is less computationally efficient, it shows high efficacy in nonlinear problem-solving, demonstrated by its accuracy in predicting breast cancer recurrence [ 49 ] . TabNet, optimized for tabular data, offers excellent interpretability and fast training times. We defined IDH as a MAP increase of 15 mmHg or more during dialysis. Our IDH-1 model, informed by 40 variables including demographic and pre-dialysis data, aimed to predict IDH before HD sessions. In the test dataset, the LGBM-based IDH-1 model surpassed its SVM and TabNet counterparts, achieving an AUC of 0.87. The IDH-2 model, predicting subsequent HD session events, also favored LGBM, with an AUC of 0.88. Although the models reported high recall rates, they had lower precision, possibly due to data recording issues, such as undetected or unrecorded hypertensive events. The challenge of accurately identifying IDH events is compounded by the skewed dataset distribution, the variable prevalence of IDH, and the dynamic nature of blood pressure in HD patients. The study's hourly blood pressure measurement protocol may be affected by these factors, which suggests a need for more rigorous data collection and analysis in future research to improve model accuracy. Our examination of the IDH model underscores the importance of feature importance analysis, illuminating the factors that significantly influence predictions and aid clinicians in identifying and mitigating patient risks. Variables such as blood volume and sodium overloads, heightened activity of the renin-angiotensin-aldosterone system, overactive sympathetic nervous system, and endothelial dysfunction are implicated in the pathogenesis of hypertension during HD.In our study of the IDH-1 model, the top 15 predictors of importance include a variety of historical blood pressure measurements and dialysis conditions. Excessive volume overload is implicated in suboptimal blood pressure control. Elevated pre-dialysis diastolic and systolic pressures suggest that the patient may already be in a state of volume overload prior to HD. Recent studies have underscored the significance of blood pressure variability during HD and its direct correlation with patient outcomes [ 50 ] . Moreover, long-term fluctuations in blood pressure have been closely associated with prognostic outcomes [ 51 ] . Given the low rates of achieving target blood pressure control among HD patients, which range from 25.6–59.0%, the inclusion of historical average blood pressure readings could serve as an effective marker for assessing long-term stability of blood pressure. This may be a potential predictor for the occurrence of IDH. Instances of high blood pressure during past dialysis sessions show significant variation, pointing to higher risks of IDH in future treatments. Timing and interval differences in historical pressure surges may reflect the rate of blood pressure change or the efficacy of managing blood pressure during dialysis, influencing the probability of IDH. Studies have indicated that the likelihood of IDH increases as the temperature of the dialysate rises to 37℃, above the isothermal condition [ 52 ] . Elevated historical dry weight and pre-dialysis weight suggest increased blood volume, which significantly correlates with IDH incidence. Reducing pre-dialysis dry weight has been shown to decrease IDH frequency [ 53 ] , highlighting the impact of careful fluid management. An increased calcium load is associated with a heightened cardiovascular risk and also affects blood pressure [ 54 ] . Research has shown that the use of low-calcium dialysate in HD filtration significantly reduces post-treatment blood pressure levels. Moreover, this study finds that the incidence of IDH is predominantly in older patients, consistent with findings from other scholars [ 55 ] . Additionally, some patients exhibit blood pressure increases during dialysis following ultrafiltration dehydration. This phenomenon may be attributed to decreased volume load, reduced end-diastolic pressure, enhanced myocardial contractility, and increased cardiac output, collectively contributing to IDH. Antihypertensive medication removal during dialysis, particularly without dose adjustment, can lead to a reduction in blood efficacy, potentially raising blood pressure during the treatment. Endothelial dysfunction may lead to IDH by impairing the capacity for vasodilation [ 56 ] . Moreover, post-dialysis hypertension is considered closely associated with cardiovascular and cerebrovascular diseases [ 57 ] . The IDH-2 model incorporates a range of predictive indicators such as historical frequency of IDH incidents, post-dialysis systolic pressure, and various parameters related to pressure increase dynamics and dialysis conditions. Comparing the IDH-1 and IDH-2 models reveals common determinants in predicting IDH. However, the IDH-1 model focuses on rapid pre-dialysis prediction using pre-dialysis blood pressures as prognostic markers, while the IDH-2 model, aimed at post-dialysis forecasting, emphasizes the historical frequency of IDH and post-dialysis systolic pressure. Despite the establishment of effective models for predicting IDH, this study recognizes the presence of certain limitations. The utilization of a retrospective multicenter approach introduces potential biases typically associated with this design, such as the challenges of non-prospective data collection and the risk of selection bias. Furthermore, the reliance on data from Asian populations may impair the generalizability of the models to diverse ethnic groups, thus potentially limiting their global applicability. It is advisable for subsequent research to investigate varied populations to enhance the models' generalizability. 5. Conclusions The present research demonstrated that the LGBM algorithm-based model exhibits superior predictive performance for IDH compared to SVM and TabNet-based models. Two distinct models were developed: IDH-1 for immediate pre-dialysis prediction of IDH, and IDH-2 for forecasting IDH in upcoming dialysis sessions. The study employed patient data from Shenzhen People's Hospital to train the models and validated them using data from Fuding Hospital, thereby evaluating the models' predictive capabilities across different settings. Key indicators, including pre-dialysis diastolic and systolic pressures, historical average IDH episodes, and post-dialysis systolic pressures, were identified as significant for the early detection and management of IDH. Declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of Shenzhen People's Hospital (LL-KY-2021,870), Shenzhen, China and Fujian University of Traditional Chinese Medicine Clinical Special Funding (XB2022051). Author contributions Research idea and study design: L.H.Y, J.J.D, Y.D, H.M.Z, S.L.H, Q.G; data analysis/interpretation: Q.G, H.M.Z, S.L.H. Model development: Q.G. Manuscript drafting: K.W, X.Z.Z, Y.H.L, S.D.L, D.E.T, M.Z.X, J.J.D, H.M.Z, S.L.H. Each author contributed to the manuscript revision. All authors take responsibility for the validity of data in this study. Funding This work was supported by Guangdong Engineering Technology Research Center（507204531040），Guangzhou Development Zone entrepreneurship leading talent project (2017-L153), Guangdong Provincial R&D Program for Key Areas (Grant No. 2023 B0101200010), Shenzhen Longhua District Science and technology innovation special fund project（11501A20220923BE5B6B3；11501A20220923BD5F291） Competing Interest The authors declare that they have no relevant financial interests. Acknowledgments We would like to thank the patients and families who participated in the study. Consent for publication All authors agree to publish Availability of data and material Graphical Abstract References Lv JC, Zhang LX. Prevalence and Disease Burden of Chronic Kidney Disease. Adv Exp Med Biol. 2019. 1165: 3-15. Liyanage T, Toyama T, Hockham C, et al. Prevalence of chronic kidney disease in Asia: a systematic review and analysis. BMJ Glob Health. 2022. 7(1): e007525. Crews DC, Bello AK, Saadi G. Burden, Access, and Disparities in Kidney Disease. Am J Nephrol. 2019. 49(3): 254-262. Ruiz-Ortega M, Rayego-Mateos S, Lamas S, Ortiz A, Rodrigues-Diez RR. Targeting the progression of chronic kidney disease. Nat Rev Nephrol. 2020. 16(5): 269-288. Evans M, Lewis RD, Morgan AR, et al. A Narrative Review of Chronic Kidney Disease in Clinical Practice: Current Challenges and Future Perspectives. Adv Ther. 2022. 39(1): 33-43. Braun MM, Khayat M. Kidney Disease: End-Stage Renal Disease. FP Essent. 2021. 509: 26-32. Ali M, Ejaz A, Iram H, Solangi SA, Junejo AM, Solangi SA. Frequency of Intradialytic Complications in Patients of End-Stage Renal Disease on Maintenance Hemodialysis. Cureus. 2021. 13(1): e12641. Inrig JK, Patel UD, Toto RD, Szczech LA. Association of blood pressure increases during hemodialysis with 2-year mortality in incident hemodialysis patients: a secondary analysis of the Dialysis Morbidity and Mortality Wave 2 Study. Am J Kidney Dis. 2009. 54(5): 881-90. Sebastian S, Filmalter C, Harvey J, Chothia MY. Intradialytic hypertension during chronic haemodialysis and subclinical fluid overload assessed by bioimpedance spectroscopy. Clin Kidney J. 2016. 9(4): 636-43. Shamir AR, Karembelkar A, Yabes J, et al. Association of Intradialytic Hypertension with Left Ventricular Mass in Hypertensive Hemodialysis Patients Enrolled in the Blood Pressure in Dialysis (BID) Study. Kidney Blood Press Res. 2018. 43(3): 882-892. Raja SM, Seyoum Y. Intradialytic complications among patients on twice-weekly maintenance hemodialysis: an experience from a hemodialysis center in Eritrea. BMC Nephrol. 2020. 21(1): 163. Flythe JE, Chang TI, Gallagher MP, et al. Blood pressure and volume management in dialysis: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2020. 97(5): 861-876. Van Buren PN, Kim C, Toto RD, Inrig JK. The prevalence of persistent intradialytic hypertension in a hemodialysis population with extended follow-up. Int J Artif Organs. 2012. 35(12): 1031-8. Losito A, Del Vecchio L, Del Rosso G, Locatelli F. Postdialysis Hypertension: Associated Factors, Patient Profiles, and Cardiovascular Mortality. Am J Hypertens. 2016. 29(6): 684-9. Singh AT, Waikar SS, Mc Causland FR. Association of Different Definitions of Intradialytic Hypertension With Long-Term Mortality in Hemodialysis. Hypertension. 2022. 79(4): 855-862. Zager PG, Nikolic J, Brown RH, et al. \"U\" curve association of blood pressure and mortality in hemodialysis patients. Medical Directors of Dialysis Clinic, Inc. Kidney Int. 1998. 54(2): 561-9. Tsuruya K, Kanda E, Nomura T, Iseki K, Hirakata H. Postdialysis blood pressure is a better predictor of mortality than predialysis blood pressure in Japanese hemodialysis patients: the Japan Dialysis Outcomes and Practice Patterns Study. Hypertens Res. 2020. 43(8): 791-797. Iatridi F, Theodorakopoulou MP, Papagianni A, Sarafidis P. Intradialytic hypertension: epidemiology and pathophysiology of a silent killer. Hypertens Res. 2022. 45(11): 1713-1725. Iatridi F, Theodorakopoulou MP, Papagianni A, Sarafidis P. Management of intradialytic hypertension: current evidence and future perspectives. J Hypertens. 2022. 40(11): 2120-2129. Inrig JK. Intradialytic hypertension: a less-recognized cardiovascular complication of hemodialysis. Am J Kidney Dis. 2010. 55(3): 580-9. Handelman GS, Kok HK, Chandra RV, Razavi AH, Lee MJ, Asadi H. eDoctor: machine learning and the future of medicine. J Intern Med. 2018. 284(6): 603-619. Deo RC. Machine Learning in Medicine. Circulation. 2015. 132(20): 1920-30. Ahmad Loti NN, Mohd Noor MR, Chang SW. Integrated analysis of machine learning and deep learning in chili pest and disease identification. J Sci Food Agric. 2021. 101(9): 3582-3594. Ali A, Fathalla A, Salah A, Bekhit M, Eldesouky E. Marine Data Prediction: An Evaluation of Machine Learning, Deep Learning, and Statistical Predictive Models. Comput Intell Neurosci. 2021. 2021: 8551167. Ting Sim JZ, Fong QW, Huang W, Tan CH. Machine learning in medicine: what clinicians should know. Singapore Med J. 2023. 64(2): 91-97. Fatima FS, Jaiswal A, Sachdeva N. Lung Cancer Detection Using Machine Learning Techniques. Crit Rev Biomed Eng. 2022. 50(6): 45-58. Angraal S, Mortazavi BJ, Gupta A, et al. Machine Learning Prediction of Mortality and Hospitalization in Heart Failure With Preserved Ejection Fraction. JACC Heart Fail. 2020. 8(1): 12-21. Alaimo L, Lima HA, Moazzam Z, et al. Development and Validation of a Machine-Learning Model to Predict Early Recurrence of Intrahepatic Cholangiocarcinoma. Ann Surg Oncol. 2023. 30(9): 5406-5415. Gottlieb ER, Samuel M, Bonventre JV, Celi LA, Mattie H. Machine Learning for Acute Kidney Injury Prediction in the Intensive Care Unit. Adv Chronic Kidney Dis. 2022. 29(5): 431-438. Shouval R, Fein JA, Savani B, Mohty M, Nagler A. Machine learning and artificial intelligence in haematology. Br J Haematol. 2021. 192(2): 239-250. Joshi RD, Dhakal CK. Predicting Type 2 Diabetes Using Logistic Regression and Machine Learning Approaches. Int J Environ Res Public Health. 2021. 18(14): 7346. Dong J, Feng T, Thapa-Chhetry B, et al. Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care. Crit Care. 2021. 25(1): 288. Kate RJ, Pearce N, Mazumdar D, Nilakantan V. A continual prediction model for inpatient acute kidney injury. Comput Biol Med. 2020. 116: 103580. 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. Int J Med Inform. 2021. 151: 104484. Thongprayoon C, Hansrivijit P, Bathini T, et al. Predicting Acute Kidney Injury after Cardiac Surgery by Machine Learning Approaches. J Clin Med. 2020. 9(6): 1767. Su CT, Chang YP, Ku YT, Lin CM. Machine Learning Models for the Prediction of Renal Failure in Chronic Kidney Disease: A Retrospective Cohort Study. Diagnostics (Basel). 2022. 12(10): 2454. Schena FP, Anelli VW, Abbrescia DI, Di Noia T. Prediction of chronic kidney disease and its progression by artificial intelligence algorithms. J Nephrol. 2022. 35(8): 1953-1971. Zacharias HU, Altenbuchinger M, Schultheiss UT, et al. A Predictive Model for Progression of CKD to Kidney Failure Based on Routine Laboratory Tests. Am J Kidney Dis. 2022. 79(2): 217-230.e1. Chuah A, Walters G, Christiadi D, et al. Machine Learning Improves Upon Clinicians' Prediction of End Stage Kidney Disease. Front Med (Lausanne). 2022. 9: 837232. Gómez-Pulido JA, Gómez-Pulido JM, Rodríguez-Puyol D, Polo-Luque ML, Vargas-Lombardo M. Predicting the Appearance of Hypotension During Hemodialysis Sessions Using Machine Learning Classifiers. Int J Environ Res Public Health. 2021. 18(5): 2364. Hong D, Chang H, He X, et al. Construction of an Early Alert System for Intradialytic Hypotension before Initiating Hemodialysis Based on Machine Learning. Kidney Dis (Basel). 2023. 9(5): 433-442. Lee H, Yun D, Yoo J, et al. Deep Learning Model for Real-Time Prediction of Intradialytic Hypotension. Clin J Am Soc Nephrol. 2021. 16(3): 396-406. Kale G, Mali M, Bhangale A, Somani J, Jeloka T. Intradialytic Hypertension Increases Non-access Related Hospitalization and Mortality in Maintenance Hemodialysis Patients. Indian J Nephrol. 2020. 30(2): 85-90. Choi H, Lee JY, Sul Y, et al. Comparing machine learning and logistic regression for acute kidney injury prediction in trauma patients: A retrospective observational study at a single tertiary medical center. Medicine (Baltimore). 2023. 102(33): e34847. Zhao X, Lu Y, Li S, et al. Predicting renal function recovery and short-term reversibility among acute kidney injury patients in the ICU: comparison of machine learning methods and conventional regression. Ren Fail. 2022. 44(1): 1326-1337. Ge C, Deng F, Chen W, et al. Machine learning for early prediction of sepsis-associated acute brain injury. Front Med (Lausanne). 2022. 9: 962027. Kang BS, Lee SU, Hong S, et al. Prediction of gestational diabetes mellitus in Asian women using machine learning algorithms. Sci Rep. 2023. 13(1): 13356. Lee Y, Ryu J, Kang MW, et al. Machine learning-based prediction of acute kidney injury after nephrectomy in patients with renal cell carcinoma. Sci Rep. 2021. 11(1): 15704. Gupta SR. Prediction time of breast cancer tumor recurrence using Machine Learning. Cancer Treat Res Commun. 2022. 32: 100602. Yu J, Chen X, Wang Y, et al. Intradialytic systolic blood pressure variation can predict long-term mortality in patients on maintenance hemodialysis. Int Urol Nephrol. 2021. 53(4): 785-795. Yang J, Huang J, Yu B, et al. Long-term predialysis blood pressure variability and outcomes in hemodialysis patients. J Clin Hypertens (Greenwich). 2022. 24(2): 148-155. Veerappan I, Thiruvenkadam G, Abraham G, Dasari BR, Rajagopal A. Effect of Isothermic Dialysis on Intradialytic Hypertension. Indian J Nephrol. 2019. 29(5): 317-323. Zhang Y, Zhang X, Li J, et al. Dry-weight reduction improves intradialytic hypertension only in patients with high predialytic blood pressure. Blood Press Monit. 2019. 24(4): 185-190. Pirklbauer M, Fuchs L, Heiss R, Ratschiller T, Mayer G. Intradialytic Calcium Kinetics and Cardiovascular Disease in Chronic Hemodialysis Patients. Blood Purif. 2020. 49(6): 723-732. Eftimovska-Otovic N, Grozdanovski R, Taneva B, Stojceva-Taneva O. Clinical Characteristics of Patients with Intradialytic Hypertension. Pril (Makedon Akad Nauk Umet Odd Med Nauki). 2015. 36(2): 187-93. Tawfeek GA, Kora MA, Yassein YS, Baghdadi AM, Elzorkany KM. Association of pre-pro-endothelin gene polymorphism and serum endothelin-1 with intradialytic hypertension in an Egyptian population. Cytokine. 2021. 137: 155293. Cho H, Kwon SK, Lee SW, et al. The Association Among Post-hemodialysis Blood Pressure, Nocturnal Hypertension, and Cardiovascular Risk Factors. Electrolyte Blood Press. 2023. 21(2): 53-60. Additional Declarations No competing interests reported. Supplementary Files GraphicalAbstract.png 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-5355171\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":379950443,\"identity\":\"0dd46416-7c28-4015-a197-d82b05fb80df\",\"order_by\":0,\"name\":\"Hongming Zhou\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Shenzhen Longhua District Central Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Hongming\",\"middleName\":\"\",\"lastName\":\"Zhou\",\"suffix\":\"\"},{\"id\":379950444,\"identity\":\"6fa42523-bd7d-4fd2-b058-6e45acee9261\",\"order_by\":1,\"name\":\"Qi Guo\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Southern University of Science and Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Qi\",\"middleName\":\"\",\"lastName\":\"Guo\",\"suffix\":\"\"},{\"id\":379950445,\"identity\":\"69a8578b-766c-4281-8cdc-e2131513ef0c\",\"order_by\":2,\"name\":\"Kang wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"the Second Affiliated Hospital of Jinan University, Jinan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Kang\",\"middleName\":\"\",\"lastName\":\"wang\",\"suffix\":\"\"},{\"id\":379950446,\"identity\":\"ec1ce3c0-36cd-4d42-9b0e-3273631c5f5a\",\"order_by\":3,\"name\":\"Xinzhou Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"the Second Affiliated Hospital of Jinan University, Jinan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xinzhou\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":379950447,\"identity\":\"8b11864e-a68c-4092-8c1e-01615b800a56\",\"order_by\":4,\"name\":\"Yehua Luo\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Fujian University of Traditional Chinese Medicine\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yehua\",\"middleName\":\"\",\"lastName\":\"Luo\",\"suffix\":\"\"},{\"id\":379950448,\"identity\":\"78f45590-b1ec-44c0-a824-9a5b41f611a3\",\"order_by\":5,\"name\":\"Shaodong Luan\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Shenzhen Longhua District Central Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Shaodong\",\"middleName\":\"\",\"lastName\":\"Luan\",\"suffix\":\"\"},{\"id\":379950449,\"identity\":\"db441cd9-479f-4624-8a3d-97a4091baea4\",\"order_by\":6,\"name\":\"Donge Tang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"the Second Clinical Medical College of Jinan University, Jinan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Donge\",\"middleName\":\"\",\"lastName\":\"Tang\",\"suffix\":\"\"},{\"id\":379950450,\"identity\":\"53e6f028-7cee-40b2-a72a-dabee110d7ea\",\"order_by\":7,\"name\":\"Mingzhi Xu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Zhejiang Cancer Hospital, Chinese Academy of Sciences\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Mingzhi\",\"middleName\":\"\",\"lastName\":\"Xu\",\"suffix\":\"\"},{\"id\":379950451,\"identity\":\"2f1e5404-a724-464b-9dd8-77c7869aa0d3\",\"order_by\":8,\"name\":\"Jingjing Dong\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Zhejiang Cancer Hospital, Chinese Academy of Sciences\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jingjing\",\"middleName\":\"\",\"lastName\":\"Dong\",\"suffix\":\"\"},{\"id\":379950452,\"identity\":\"ce063c5a-38df-49a9-9bd0-4bf1055eb567\",\"order_by\":9,\"name\":\"Lianghong Yin\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"the First Affiliated Hospital of Jinan University, Jinan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Lianghong\",\"middleName\":\"\",\"lastName\":\"Yin\",\"suffix\":\"\"},{\"id\":379950453,\"identity\":\"a44bf90f-ef32-43ac-bb13-32ca530bc1b2\",\"order_by\":10,\"name\":\"Yong Dai\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFklEQVRIie3QMWuDQBTA8Scpulwzn5Q2X8ESsBQHv4q39Jakc4c0XAhcFqFzPkRX6XhwYJYrrjfGJVMHs4fSU9NOGtcO9we5h/rjiQA2239MOKw+7gCclTjfc801GiRTQ1hD0DBpD8J+50Ey3kmOjx+v9H0jzZaTjGNvfdjDS0SY9ym6iK8I97dqN88UYcLhkqQofwhAUcLQc9JFAkH4zTXP55moCZMJwjMX15ZhFHSSomwIDYuy/TA0+TLk+wLRzZZFEmqzBVzppBgZwvqJr8v145aL+0ybLYRTkqqnECc5nXI06yTjgpb6yJeT0Az76hTF3kYecLWIbt881UlMVxhAtuPfH6oHt+d906gCWPY/ttlsNtsPe21rWJdHto4AAAAASUVORK5CYII=\",\"orcid\":\"\",\"institution\":\"Shenzhen People 's Hospital, Second Clinical Medical College of Jinan University\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Yong\",\"middleName\":\"\",\"lastName\":\"Dai\",\"suffix\":\"\"},{\"id\":379950454,\"identity\":\"7d6dde2a-819e-4f79-8a18-4a9e5db1a884\",\"order_by\":11,\"name\":\"shengling huang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"the First Affiliated Hospital of Jinan University, Jinan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"shengling\",\"middleName\":\"\",\"lastName\":\"huang\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-10-29 14:38:11\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-5355171/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-5355171/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":71148379,\"identity\":\"8edaf0d1-2d66-4738-9f8b-036054add417\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:33:44\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":263504,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe study cohort of IDH model. (A)The study cohort of IDH-1 model (Shenzhen People's Hospital). (B)The study cohort of IDH-1 model (Fuding city's Hospital). (C)The study cohort of IDH-2 model (Shenzhen People's Hospital). (D)The study cohort of IDH-2 model (Fuding city's Hospital).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/9fff36e38bcd5249cbb5668f.png\"},{\"id\":71145125,\"identity\":\"883eadbf-c8bb-44f5-a310-0c522b244fb7\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:17:44\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":41224,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe ROC curve for simple IDH-1 model in test data. Receiver operating characteristic curve of IDH-1 predictor.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/127f0a691f4b3a41f4648432.png\"},{\"id\":71148377,\"identity\":\"31d55880-d167-4233-927b-c260118264b5\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:33:44\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":49486,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe ROC curve for simple IDH-2 model in test data. Receiver operating characteristic curve of IDH-2 predictor.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/1af05a0877f3b5a92e1e2104.png\"},{\"id\":71145127,\"identity\":\"ae6598e2-8f40-498f-9df9-e9750728efe9\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:17:44\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":24072,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003ePerformance Metrics for IDH-1 Models. The bar chart compares the performance of three machine learning models: Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and Tabular Neural Network (TabNet), across four metrics: Area Under the Curve (AUC), Recall, Accuracy, and F1 Score. Performance metrics are derived from the validation phase of model training. Bars are color-coded according to the metric for clear differentiation.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/687a9b31922b1c29c6639274.png\"},{\"id\":71145131,\"identity\":\"2555d773-2370-496c-849a-9bcd6702d7a3\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:17:44\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":23499,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003ePerformance Metrics for IDH-2 Models. The bar chart compares the performance of three machine learning models: Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and Tabular Neural Network (TabNet), across four metrics: Area Under the Curve (AUC), Recall, Accuracy, and F1 Score. Performance metrics are derived from the validation phase of model training. Bars are color-coded according to the metric for clear differentiation.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/200a3ff95710c645b90eba3f.png\"},{\"id\":71149543,\"identity\":\"6a5ec4a2-4190-428b-8a8d-aa3cc07a64a6\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:41:44\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":62458,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eRanking of feature importance based on dataset IDH-1. Features are ordered dissentingly by their importance score, indicated by the numerical labels above each bar. The visualization is designed to provide a clear comparison across the different feature importances.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/d76908933a9ed2e4a3124eb7.png\"},{\"id\":71147705,\"identity\":\"58d0186d-adad-44da-b1b7-83faf52cd4cf\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:25:44\",\"extension\":\"png\",\"order_by\":7,\"title\":\"Figure 7\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":56447,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eRanking of feature importance based on dataset IDH-2. Features are ordered dissentingly by their importance score, indicated by the numerical labels above each bar. The visualization is designed to provide a clear comparison across the different feature importances.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"7.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/83a699b44d0ad93ae53d18de.png\"},{\"id\":72452409,\"identity\":\"8b4d9b3e-7e39-4050-8677-dea1aac038b0\",\"added_by\":\"auto\",\"created_at\":\"2024-12-27 08:53:54\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1344881,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/6cd5fb62-ed7c-47fd-ae23-ba05068e1ea8.pdf\"},{\"id\":71145132,\"identity\":\"1b1f13e4-9e72-43ad-bbcb-e8fe6a3f512a\",\"added_by\":\"auto\",\"created_at\":\"2024-12-11 14:17:44\",\"extension\":\"png\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":190554,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"GraphicalAbstract.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5355171/v1/8558c8b62e987e7f57ee4632.png\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Prediction model of intradialytic hypertension in hemodialysis patients based on machine learning\",\"fulltext\":[{\"header\":\"1. Introduction\",\"content\":\"\\u003cp\\u003eChronic kidney disease (CKD) poses a significant challenge to global public health, with epidemiological studies indicating that approximately 13.4% of the global population is affected\\u003csup\\u003e[\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e]\\u003c/sup\\u003e. There is considerable variation in CKD prevalence across Asia, with China and India together accounting for 69.1% of the adult CKD cases in the region\\u003csup\\u003e[\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e]\\u003c/sup\\u003e. According to the 2015 Global Burden of Disease Study, CKD-related factors were responsible for an estimated 1.2\\u0026nbsp;million deaths\\u003csup\\u003e[\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e]\\u003c/sup\\u003e. The prevalence of CKD is on the rise, driven by increasing rates of diabetes and hypertension, as well as an aging population worldwide\\u003csup\\u003e[\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e]\\u003c/sup\\u003e. CKD often progresses to end-stage renal disease (ESRD)\\u003csup\\u003e[\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e]\\u003c/sup\\u003e, requiring renal replacement therapies such as hemodialysis (HD), peritoneal dialysis, and kidney transplantation. Managing ESRD through hemodialysis is essential\\u003csup\\u003e[\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e]\\u003c/sup\\u003e but fraught with multiple potential complications, including cardiovascular disease, infections, vascular access issues, hypotension, electrolyte imbalances, and hypertension.\\u003c/p\\u003e \\u003cp\\u003eIntradialytic hypertension (IDH) is a common complication observed during dialysis, with its definition evolving over the years\\u003csup\\u003e[\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e]\\u003c/sup\\u003e. Initially, Inrig et al.\\u003csup\\u003e[\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e]\\u003c/sup\\u003e proposed defining IDH as mean systolic pressures increase exceeding 10 mmHg over three hemodialysis HD sessions. This definition was later expanded by Sebastian et al.\\u003csup\\u003e[\\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e]\\u003c/sup\\u003e, who required a rise in systolic pressure over 10 mmHg in at least four out of six HD sessions. Shamir et al. \\u003csup\\u003e[\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e]\\u003c/sup\\u003esimplified this criterion to a systolic pressure increase of 10 mmHg or more during HD. Raja et al.\\u003csup\\u003e[\\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e]\\u003c/sup\\u003e adopted a stricter definition, identifying IDH as a 15 mmHg increase in mean arterial pressure (MAP) during or immediately post-dialysis. The KDIGO (kidney disease: Improving Global Outcomes) recently issued a recommendation that a diagnosis of IDH should be considered when there is an increase in systolic blood pressure of more than 10 mmHg before and after dialysis in at least four out of six consecutive HD sessions\\u003csup\\u003e[\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e]\\u003c/sup\\u003e. While the definition of IDH has been refined and developed over the past decade, a globally unified clinical standard is yet to be established.\\u003c/p\\u003e \\u003cp\\u003eIn the scientific literature, the prevalence of IDH is reported to range from 17\\u0026ndash;23%\\u003csup\\u003e[\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e]\\u003c/sup\\u003e, a variability that may be ascribed to the lack of uniform IDH identification criteria. In patients with ESRD undergoing HD, IDH is implicated in elevating the risk of mortality\\u003csup\\u003e[\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e]\\u003c/sup\\u003e, making it a crucial factor in prognostication\\u003csup\\u003e[\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e]\\u003c/sup\\u003e. The pathophysiological underpinnings of IDH are not fully understood, but it is widely recognized to arise from multiple interrelated factors, including overactivation of the renin-angiotensin system, sympathetic nervous system hyperactivity, volume and sodium excess, electrolyte shifts during dialysis, and endothelial dysfunction\\u003csup\\u003e[\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e]\\u003c/sup\\u003e. These elements, individually or synergistically, play central roles in the development of IDH. The absence of a standard definition for IDH hampers the comparison of research outcomes and diminishes their clinical applicability. Nonetheless, the evidence underscores the significance of IDH in clinical settings\\u003csup\\u003e[\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]\\u003c/sup\\u003e, advocating for early interventions that could profoundly improve patient well-being and long-term outcomes.\\u003c/p\\u003e \\u003cp\\u003eArtificial intelligence refers to the capability of machines to emulate human intelligent behavior\\u003csup\\u003e[\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e]\\u003c/sup\\u003e. Machine learning (ML), a crucial subset of AI, enables systems to discern patterns and learn from data input, ultimately endowing them with predictive and decision-making faculties\\u003csup\\u003e[\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e]\\u003c/sup\\u003e. The rapid progress in digital technologies has catalyzed significant advancements in ML, with its successful integration into diverse fields such as agriculture\\u003csup\\u003e[\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e]\\u003c/sup\\u003e, fisheries\\u003csup\\u003e[\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e]\\u003c/sup\\u003e, and healthcare\\u003csup\\u003e[\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e]\\u003c/sup\\u003e. ML's deployment spans a wide array of diagnostic and therapeutic domains, including the respiratory\\u003csup\\u003e[\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e]\\u003c/sup\\u003e, cardiovascular\\u003csup\\u003e[\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e]\\u003c/sup\\u003e, digestive\\u003csup\\u003e[\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e]\\u003c/sup\\u003e, urinary\\u003csup\\u003e[\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e]\\u003c/sup\\u003e, hematological\\u003csup\\u003e[\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e]\\u003c/sup\\u003e, and endocrine systems\\u003csup\\u003e[\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e]\\u003c/sup\\u003e. In nephrology, ML techniques have proven particularly effective for the prognostication of renal pathologies, exemplified by the development of predictive models for acute kidney injury\\u003csup\\u003e[\\u003cspan additionalcitationids=\\\"CR33 CR34\\\" citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e]\\u003c/sup\\u003e and chronic renal failure\\u003csup\\u003e[\\u003cspan additionalcitationids=\\\"CR37\\\" citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e]\\u003c/sup\\u003e. Notable is the model by Chuah et al.\\u003csup\\u003e[\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e]\\u003c/sup\\u003e aimed at predicting progression to ESRD, which has demonstrated potential in decelerating the progression of CKD. Recent studies have increasingly focused on the development of machine learning-based predictive models for hypotension during dialysis sessions\\u003csup\\u003e[\\u003cspan additionalcitationids=\\\"CR41\\\" citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e]\\u003c/sup\\u003e. However, the development of predictive models for hypertension in the same context has been considerably less explored. Addressing this gap, leveraging ML algorithms to analyze both historic and real-time data may unearth patterns integral to IDH, providing clinicians with a tool to foresee and mitigate the risk of IDH in HD patients, thus enhancing clinical outcomes and patient quality of life.\\u003c/p\\u003e \\u003cp\\u003eTo improve the model's generalization capabilities while accounting for differences in interspatial data, we adopted a cross-regional data validation approach. This method entails constructing a predictive model with data from one region and subsequently validating it using data from a different region, thus confirming the model's wide-ranging applicability. The collection of extensive data on dialysis sessions and patient demographics during the study enriched the development of the IDH risk prediction model. We conducted a comparative analysis of various models to identify the most effective decision-support tool, thereby enhancing clinicians' ability to make informed treatment decisions.\\u003c/p\\u003e\"},{\"header\":\"2. Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.1 Study population and data collection\\u003c/h2\\u003e \\u003cp\\u003eIn this study, demographic data and HD records were collected from patients undergoing HD at the People's Hospital in Shenzhen, China, from January 2017 to October 2022, and at the Hospital of Fuding City, China, from October 2020 to August 2022. Participants were aged between 16 and 100 years with recorded dry body weights. Each patient received dialysis treatments lasting between 3 and 4 hours. Exclusion criteria were applied to patients with any diagnosis of cancer, acute renal failure, acute renal insufficiency, or acute kidney injury documented in their dialysis records, as well as to those with a missing value ratio exceeding 35%. The study received ethical approval from the Ethics Committees of both Shenzhen People's Hospital and Fuding Hospital in China.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec4\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.2 Data Preprocessing\\u003c/h2\\u003e \\u003cp\\u003eIn addressing outliers within continuous variables, data points exceeding the 99th percentile were replaced with the 99th percentile value, and those below the 1st percentile were replaced with the 1st percentile value. For categorical features with missing values, decision trees were employed for prediction. Meanwhile, for continuous features, the mean of each feature was calculated and used to impute missing values. To demonstrate our model's external validation, training was conducted using the \\\"Shenyi\\\" dataset, followed by performance evaluation on the distinct \\\"Fuding\\\" dataset.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec5\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.3 Model Input Features and IDH Definition\\u003c/h2\\u003e \\u003cp\\u003eIn this study, two predictive models, IDH-1 and IDH-2, were developed to evaluate the risk of IDH. Model IDH-1 aims for real-time prediction of IDH risk, integrating immediate pre-treatment clinical and historical average dialysis data during the current HD session. This model applies a binary classification system, where a '0' indicates no IDH events and a '1' denotes the occurrence of IDH. Conversely, model IDH-2 predicts IDH risk for the next session (n\\u0026thinsp;+\\u0026thinsp;1) by analyzing data from the present HD session (n) alongside historical averages. This prospective strategy allows clinicians to adjust treatment plans in advance, with '0' representing the absence of predicted IDH events and '1' indicating their anticipated occurrence.\\u003c/p\\u003e \\u003cp\\u003e In our study, we conducted a review of patient demographics and dialysis treatment records to gather a comprehensive dataset. This dataset included patient gender, age, and infectious disease status, along with detailed dialysis treatment parameters such as the duration of treatment (recorded in both days and years), modality used, and the type of anticoagulant administered. Key physiological measurements were also recorded, including systolic and diastolic pressures before and after dialysis sessions, body weight measurements before and after treatment, the initial respiratory rate, estimated dry weight, and various dialysis session parameters like dialysate calcium concentration, conductivity, and temperature, as well as transmembrane pressure. Vascular access characteristics were detailed by recording the type, location, and duration of use of the fistula. Clinical metrics during dialysis, such as intra-dialytic pressures readings, pulse, ultrafiltration rate and volume, venous pressure, MAP, and blood flow rate, were also captured. Additionally, the study involved aggregating certain variables, from which mean values and standard deviations were calculated. Baseline pre-dialysis measurements were acquired 10 to 15 minutes before the start of each hemodialysis treatment. Monitoring throughout the hemodialysis consisted of hourly assessments of blood pressure, pulse, ultrafiltration rate, and fluid volume. IDH was defined as a rise in MAP by more than 15 mmHg above the pre-dialysis level, with the MAP being calculated as (2\\u0026times; diastolic pressure\\u0026thinsp;+\\u0026thinsp;systolic pressure) / 3 mmHg. Detailed descriptions and classifications of model variables are presented in Table\\u0026nbsp;1.\\u003c/p\\u003e \\u003cp\\u003eThe development of the prediction model utilized three machine learning algorithms, namely Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and TabNet, and was executed in the Python programming language. Details on the hyperparameter optimization for each algorithm are provided in Table S3.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec6\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2.4 Statistical analysis\\u003c/h2\\u003e \\u003cp\\u003eIn this study, continuous variables conforming to a normal distribution were expressed as mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;standard deviation (x̄\\u0026plusmn;s), while those not following a normal distribution were described using the median and interquartile range [Q1, Q3]. Categorical variables were reported as counts and percentages [n (%)]. Statistical analyses were conducted using the t-test for normally distributed continuous variables and the Wilcoxon rank-sum test for non-normally distributed continuous variables. Differences in categorical variables between groups were assessed using the Chi-square test. A P-value of less than 0.05 was considered to indicate statistical significance. The predictive capability of the model in this study was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), also known as the c statistic. Additionally, the model's performance was comprehensively assessed by metrics including recall, precision, and the F1 score. Data analysis was performed using Python version 3.7.3. DeLong's test was employed to compare the AUC of different predictive variables. Furthermore, the identification of significant predictors of IDH involved ranking feature importance, which was based on the frequency of feature use for node splitting during the training of decision trees, thus providing insights into the variables most critical for enhancing the model's predictive performance.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"3. Results\",\"content\":\"\\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.1 Cohort Characteristics\\u003c/h2\\u003e \\u003cp\\u003eThis retrospective study involved analyzing HD treatment records from two centers: Shenzhen People's Hospital (March 3, 2017, to October 8, 2022) and Fuding City Hospital (October 31, 2020, to August 13, 2022). At Shenzhen People's Hospital, the initial dataset comprised 222,858 HD sessions for 1,693 patients. Exclusion criteria included diagnoses of acute renal conditions or cancer, resulting in the removal of 22 records, while no patients were excluded. Additionally, 56 patients and 6,232 sessions were excluded due to the absence of systolic, diastolic, or mean arterial pressure readings during dialysis. An additional 195 patients and 11,601 sessions were excluded based on incomplete dry weight records, and 653 sessions were excluded due to age criteria (under 16 or over 100 years). The exclusion of sessions outside the 3- to 4-hour duration or with more than 35% data missingness resulted in the removal of 37 patients and 19,225 records. Due to the criterion of having fewer than two dialysis sessions, 97 patients and 1,405 records were excluded from the IDH-2 model development. The final datasets for the IDH-1 model included 185,125 sessions from 1,405 patients, while the IDH-2 model considered 183,720 sessions from 1,308 patients.\\u003c/p\\u003e \\u003cp\\u003eAt Fuding City Hospital, the dataset of 434 HD patients comprised 77,808 sessions. Application of the same exclusion criteria removed one patient and 1,289 records due to acute illnesses or cancers, three patients and 3,168 records due to lack of blood pressure data during dialysis, and 14 patients and 602 records due to incomplete dry weight data. Although age criteria (under 16 or over 100 years old) were set for exclusion within the screening protocol, they resulted in no patient or record exclusions. Sessions failing to meet the duration criteria or having high data missingness accounted for 1,322 excluded records. For the IDH-2 model, the exclusion of patients with fewer than two sessions reduced the dataset by 416 records. Ultimately, the IDH-1 model analysis for Fuding City Hospital included 416 patients and 71,427 sessions, and the IDH-2 model dataset comprised the same number of patients but with 71,011 sessions. The exclusion process for both centers is detailed in Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eFor this analysis, the dataset comprised a training set of 185,125 entries from Shenzhen and a test set of 71,427 entries from Fuding. In the training set applicable to both IDH-1 and IDH-2 models, the gender distribution was 34.3% female and 65.7% male, with an observed IDH occurrence rate of 14%. Conversely, the test set showed a slightly higher proportion of females at 39.9% and a male proportion of 60.1%, alongside a lower IDH occurrence rate of 6% for both models. Detailed variable distributions for the data records are presented in \\u003cb\\u003eTables\\u0026nbsp;1 to 4\\u003c/b\\u003e.\\u003c/p\\u003e \\u003cp\\u003eTables\\u0026nbsp;1. Characteristics of patients in the IDH-1 model training (Shenyi) cohort.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"No\\\" id=\\\"Taba\\\" border=\\\"1\\\"\\u003e \\u003ccolgroup cols=\\\"5\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCharacteristic\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eall(n\\u0026thinsp;=\\u0026thinsp;185125)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003enon- IDH(n\\u0026thinsp;=\\u0026thinsp;159290)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eIDH(n\\u0026thinsp;=\\u0026thinsp;25835)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e\\u003cem\\u003eP\\u003c/em\\u003e\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003efemale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63561 (34.3)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e53545 (33.6)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e10016 (38.8)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003emale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e121564 (65.7)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e105745 (66.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e15819 (61.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003epre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e142.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e143.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e133.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e84.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e76.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003edialysis age (days)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1865.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1581.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1886.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1584.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1733.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1558.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003eage(year), median [Q1,Q3]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.0 [46.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e55.0 [46.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.0 [48.0,71.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e63.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e61.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003edry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e60.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e60.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e58.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003eage at first dialysis(year)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e51.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e51.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e54.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e63.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e61.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e60.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e60.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e58.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dialysate calcium concentration(mmol/l)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e142.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;17.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e141.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;17.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e142.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;17.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.001\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average pre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e82.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average IDH episodes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e139.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e139.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e143.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e84.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e84.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e84.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average venous pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e125.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e126.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e122.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average MAP(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e102.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e102.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e104.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average blood flow rate(ml/min)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e239.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e240.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e235.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dialysate temperature (℃)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e36.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e36.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e36.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003eTables\\u0026nbsp;2. Characteristics of patients in the IDH-1 model test (Fuding) cohort.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"No\\\" id=\\\"Tabb\\\" border=\\\"1\\\"\\u003e \\u003ccolgroup cols=\\\"5\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCharacteristic\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eall(n\\u0026thinsp;=\\u0026thinsp;71427)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003enon-IDH(n\\u0026thinsp;=\\u0026thinsp;67032)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eIDH(n\\u0026thinsp;=\\u0026thinsp;4395)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eP\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003efemale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e28526 (39.9)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e26916 (40.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1610 (36.6)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003emale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e42901 (60.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e40116 (59.8)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e2785 (63.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003epre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e156.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e158.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e141.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;19.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e82.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e69.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003edialysis age (days)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1579.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1390.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1603.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1402.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1211.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1119.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003eage(year), median [Q1,Q3]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e58.0 [49.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58.0 [49.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.0 [49.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e59.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.88\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003edry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e56.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.045\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eage at first dialysis(year)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e52.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e52.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e54.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e59.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e56.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.045\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average dialysate calcium concentration(mmol/l)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.914\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average pre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e158.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e158.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e156.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e78.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average IDH episodes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e145.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e144.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e151.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e77.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e77.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e77.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average venous pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e111.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e111.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e106.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average MAP(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e100.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e99.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e102.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average blood flow rate(ml/min)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e242.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e242.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e239.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dialysate temperature (℃)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e37.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e37.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e36.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003eTables\\u0026nbsp;3. Characteristics of patients in the IDH-2 model training (Shenyi) cohort.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"No\\\" id=\\\"Tabc\\\" border=\\\"1\\\"\\u003e \\u003ccolgroup cols=\\\"5\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCharacteristic\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eall(n\\u0026thinsp;=\\u0026thinsp;183720)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003enon-IDH(n\\u0026thinsp;=\\u0026thinsp;158119)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eIDH(n\\u0026thinsp;=\\u0026thinsp;25601)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eP\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003efemale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63041 (34.3)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e53111 (33.6)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e9930 (38.8)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003emale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e120679 (65.7)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e105008 (66.4)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e15671 (61.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003epre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e142.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e143.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e133.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e84.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e76.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epost-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e139.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e137.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e151.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;23.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epost-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e84.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e89.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003edialysis age (days)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1873.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1580.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1895.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1583.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1741.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1558.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003eage(year), median [Q1,Q3]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.0 [46.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e55.0 [46.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.0 [48.0,71.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e63.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e61.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003edry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e60.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e60.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e58.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epost-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e60.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e60.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e58.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003eage at first dialysis(year)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e51.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e51.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e54.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e140.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;20.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e139.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;19.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e145.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;20.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003estandard deviation of historical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e8.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;5.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e7.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;5.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e11.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e85.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003estandard deviation ofhistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e5.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;4.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e4.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e8.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;7.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e63.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e63.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e61.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e60.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e60.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e58.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dialysate calcium concentration(mmol/l)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e142.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;17.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e141.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;17.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e142.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.001\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average pre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e82.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average IDH episodes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e139.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e139.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e143.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e84.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e84.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e84.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average venous pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e125.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e126.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e122.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average MAP(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e102.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e102.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e104.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average blood flow rate(ml/min)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e239.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e240.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e235.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dialysate temperature (℃)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e36.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e36.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e36.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003eTables\\u0026nbsp;4. Characteristics of patients in the IDH-2 model test (Fuding) cohort.\\u003c/p\\u003e \\u003cp\\u003e \\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"No\\\" id=\\\"Tabd\\\" border=\\\"1\\\"\\u003e \\u003ccolgroup cols=\\\"5\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eCharacteristic\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eall(n\\u0026thinsp;=\\u0026thinsp;71011)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003enon-IDH(n\\u0026thinsp;=\\u0026thinsp;66648)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eIDH(n\\u0026thinsp;=\\u0026thinsp;4363)\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eP\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003efemale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e28366 (39.9)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e26772 (40.2)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1594 (36.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003emale, n (%)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e42645 (60.1)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e39876 (59.8)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e2769 (63.5)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u0026nbsp;\\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003epre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e157.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e158.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e141.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;19.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e82.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e69.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epost-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e139.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e137.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;20.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e156.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;20.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epost-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e75.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e74.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e82.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003edialysis age (days)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1580.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1390.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1604.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1402.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1212.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1119.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003eage(year), median [Q1,Q3]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e58.0 [49.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58.0 [49.0,66.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.0 [49.0,67.0]\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003epre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e59.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003edry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e56.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.08\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003epost-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e56.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e56.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.276\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eage at first dialysis(year)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e52.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e52.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e54.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e144.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;19.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e143.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;18.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e153.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;18.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003estandard deviation of historical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e10.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e10.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e10.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;5.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.006\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e77.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e77.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e79.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003estandard deviation ofhistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e6.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.6\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e6.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e8.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e59.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e58.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e59.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dry weight(kg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e56.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e56.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;10.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.079\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average dialysate calcium concentration(mmol/l)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e1.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average pre-dialysis systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e158.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e158.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e156.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average pre-dialysis diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e83.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e83.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e78.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;12.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average IDH episodes\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic systolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e145.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.8\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e144.6\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14.7\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e151.2\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;15.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average intradialytic diastolic pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e77.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e77.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.1\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e77.8\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e1\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003ehistorical average venous pressure(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e111.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.0\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e111.4\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;21.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e106.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;22.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average MAP(mmHg)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e100.1\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e99.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e102.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.9\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average blood flow rate(ml/min)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e242.3\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e242.5\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;11.2\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e239.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.3\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\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\\u003ehistorical average dialysate temperature(℃)\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e37.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e37.0\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.5\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e36.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.4\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec9\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.2 Model performance comparison\\u003c/h2\\u003e \\u003cp\\u003eIn assessing the performance outcomes for the IDH-1 model, it was observed that the LGBM model outperformed its counterparts, achieving the highest AUC of 0.87, as detailed in Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e \\u003cb\\u003eand\\u003c/b\\u003e Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e. This was closely followed by the SVM model with an AUC of 0.87, and the TabNet model with an AUC of 0.86. Regarding recall rates, the LGBM model also led with a rate of 0.78, surpassing the SVM and TabNet models, which recorded rates of 0.57 and 0.69, respectively. While accuracy rates were similar across the models, the SVM model exhibited a slightly higher accuracy of 0.31, compared to the LGBM and TabNet models at 0.24 and 0.23, respectively. In terms of the F1 score, which measures the model's balance between precision and recall, the SVM model achieved the highest score of 0.401, followed by the LGBM and TabNet models with scores of 0.36 and 0.34, respectively.\\u003c/p\\u003e \\u003cp\\u003eStatistical analysis of Delong revealed a significant difference between the LGBM model and the TabNet model (p-value\\u0026thinsp;=\\u0026thinsp;0), whereas the difference between the LGBM and SVM models was not statistically significant (p-value\\u0026thinsp;=\\u0026thinsp;0.12). This suggests that the LGBM model significantly outperforms the TabNet model in predicting IDH events among HD patients and is comparably effective to the SVM model.\\u003c/p\\u003e \\u003cp\\u003eFurther analysis of the IDH-2 models, as shown in Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003e \\u003cb\\u003eand\\u003c/b\\u003e Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e, demonstrates the LGBM algorithm's dominance over the SVM and TabNet models across all evaluated metrics. The LGBM algorithm secured an AUC of 0.75, significantly higher than the SVM and TabNet models, which achieved AUCs of 0.70 and 0.58, respectively. It also led in recall with a score of 0.561, markedly ahead of the SVM and TabNet scores of 0.06 and 0.22, respectively. Although the LGBM model's accuracy was lower than that of the SVM model (0.17 vs. 0.26), it exceeded the TabNet model's accuracy (0.09). Furthermore, the LGBM model excelled in F1 score with 0.26, while the SVM and TabNet models recorded scores of 0.09 and 0.13, respectively. Statistical comparisons underscored the LGBM model's performance superiority with a p-value of 0. The comparative performance of the three models across both IDH-1 and IDH-2 is illustrated in bar graphs (Figs.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e and \\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\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 5\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003eThe performance of IDH-1 prediction models in test data.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"5\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eModel\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eAUC\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eRecall\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eAccuracy\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eF1\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLGBM\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.87\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.73\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.24\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.36\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSVM\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.87\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.57\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.31\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.40\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eTabNet\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.86\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.69\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.23\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.34\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\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 6\\u003c/div\\u003e \\u003cdiv class=\\\"CaptionContent\\\"\\u003e \\u003cp\\u003eThe performance of IDH-2 prediction models in test data.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/caption\\u003e \\u003ccolgroup cols=\\\"5\\\"\\u003e \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e \\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e \\u003cthead\\u003e \\u003ctr\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eModel\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003eAUC\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003eRecall\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003eAccuracy\\u003c/p\\u003e \\u003c/th\\u003e \\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003eF1\\u003c/p\\u003e \\u003c/th\\u003e \\u003c/tr\\u003e \\u003c/thead\\u003e \\u003ctbody\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eLGBM\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.75\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.56\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.17\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.26\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eSVM\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.70\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.06\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.26\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.09\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003ctr\\u003e \\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e \\u003cp\\u003eTabNet\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c2\\\"\\u003e \\u003cp\\u003e0.58\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c3\\\"\\u003e \\u003cp\\u003e0.22\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e \\u003cp\\u003e0.09\\u003c/p\\u003e \\u003c/td\\u003e \\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e \\u003cp\\u003e0.13\\u003c/p\\u003e \\u003c/td\\u003e \\u003c/tr\\u003e \\u003c/tbody\\u003e \\u003c/colgroup\\u003e \\u003c/table\\u003e\\u003c/div\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec10\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e3.3 Feature importance analysis\\u003c/h2\\u003e \\u003cp\\u003eFeature importance was determined based on the number of splits in the LGBM model, revealing the top 15 features that substantially contribute to the model. In the IDH-1 model, as shown in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003e, pre-dialysis diastolic pressure was identified as the most influential feature for predicting IDH incidents, with an importance score of 579. It was closely followed by the historical average MAP and historical average IDH episodes, with scores of 262 and 261, respectively, highlighting the predictive significance of prior blood pressure measurements. Other notable features included pre-dialysis systolic pressure (score: 255), historical average intradialytic diastolic pressure (score: 241), and historical average pre-dialysis diastolic pressure (score: 171). Additional features such as historical average intradialytic systolic pressure (score: 131) and historical average ratio of rise time points range (score: 80) also demonstrated substantial predictive relevance. Contributing to a lesser degree were features like historical average pre-dialysis systolic pressure (score: 51), historical average dialysate temperature (score: 49), historical average difference in rise time points range (score: 45), and patient age (score: 31), among others.\\u003c/p\\u003e \\u003cp\\u003eIn the context of the IDH-2 model, depicted in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig7\\\" class=\\\"InternalRef\\\"\\u003e7\\u003c/span\\u003e, historical average IDH episodes was the most significant feature, with a score of 336, affirming its critical role in predicting IDH during HD. Post-dialysis systolic pressure, with a score of 232, and ratio of rise time points, with a score of 191, were also highly indicative of IDH risk. Further important features were the historical average ratio of rise time points range (score: 131), historical average blood flow rate (score: 118), and historical average venous pressure (score: 103), which mirror the influence of historical hemodynamic patterns and dialysis thermal management on IDH occurrence. Lesser but meaningful contributions to IDH prediction were observed from features such as the difference in rise time points (score: 82), historical average dialysate temperature (score: 58), among others.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"4. Discussion\",\"content\":\"\\u003cp\\u003eThe prevalence of IDH has been linked to increased hospitalization and mortality among HD patients\\u003csup\\u003e[\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e]\\u003c/sup\\u003e. Currently, IDH detection primarily relies on blood pressure readings exceeding 140/90 mmHg, highlighting the absence of a universal IDH definition. With renal disease prediction improving, machine learning approaches have outperformed traditional statistical methods\\u003csup\\u003e[\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e]\\u003c/sup\\u003e. This underscores the importance of developing predictive models to improve patient outcomes in HD. This study employs three machine learning models including LGBM, SVM, and TabNet to predict IDH in HD patients. These models were selected based on their proven performance on historical datasets and their adeptness at handling various types of data. Specifically, LGBM was chosen due to its efficiency and excellence in processing large datasets, as well as its remarkable performance across several predictive tasks. Such tasks include predicting sepsis-associated acute brain damage\\u003csup\\u003e[\\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e]\\u003c/sup\\u003e, gestational diabetes\\u003csup\\u003e[\\u003cspan citationid=\\\"CR47\\\" class=\\\"CitationRef\\\"\\u003e47\\u003c/span\\u003e]\\u003c/sup\\u003e, and acute renal injury following nephrectomy in patients with renal cell carcinoma\\u003csup\\u003e[\\u003cspan citationid=\\\"CR48\\\" class=\\\"CitationRef\\\"\\u003e48\\u003c/span\\u003e]\\u003c/sup\\u003e. Although the SVM is less computationally efficient, it shows high efficacy in nonlinear problem-solving, demonstrated by its accuracy in predicting breast cancer recurrence\\u003csup\\u003e[\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e]\\u003c/sup\\u003e. TabNet, optimized for tabular data, offers excellent interpretability and fast training times. We defined IDH as a MAP increase of 15 mmHg or more during dialysis. Our IDH-1 model, informed by 40 variables including demographic and pre-dialysis data, aimed to predict IDH before HD sessions. In the test dataset, the LGBM-based IDH-1 model surpassed its SVM and TabNet counterparts, achieving an AUC of 0.87. The IDH-2 model, predicting subsequent HD session events, also favored LGBM, with an AUC of 0.88. Although the models reported high recall rates, they had lower precision, possibly due to data recording issues, such as undetected or unrecorded hypertensive events. The challenge of accurately identifying IDH events is compounded by the skewed dataset distribution, the variable prevalence of IDH, and the dynamic nature of blood pressure in HD patients. The study's hourly blood pressure measurement protocol may be affected by these factors, which suggests a need for more rigorous data collection and analysis in future research to improve model accuracy.\\u003c/p\\u003e \\u003cp\\u003eOur examination of the IDH model underscores the importance of feature importance analysis, illuminating the factors that significantly influence predictions and aid clinicians in identifying and mitigating patient risks. Variables such as blood volume and sodium overloads, heightened activity of the renin-angiotensin-aldosterone system, overactive sympathetic nervous system, and endothelial dysfunction are implicated in the pathogenesis of hypertension during HD.In our study of the IDH-1 model, the top 15 predictors of importance include a variety of historical blood pressure measurements and dialysis conditions. Excessive volume overload is implicated in suboptimal blood pressure control. Elevated pre-dialysis diastolic and systolic pressures suggest that the patient may already be in a state of volume overload prior to HD. Recent studies have underscored the significance of blood pressure variability during HD and its direct correlation with patient outcomes\\u003csup\\u003e[\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e]\\u003c/sup\\u003e. Moreover, long-term fluctuations in blood pressure have been closely associated with prognostic outcomes\\u003csup\\u003e[\\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e]\\u003c/sup\\u003e. Given the low rates of achieving target blood pressure control among HD patients, which range from 25.6\\u0026ndash;59.0%, the inclusion of historical average blood pressure readings could serve as an effective marker for assessing long-term stability of blood pressure. This may be a potential predictor for the occurrence of IDH. Instances of high blood pressure during past dialysis sessions show significant variation, pointing to higher risks of IDH in future treatments. Timing and interval differences in historical pressure surges may reflect the rate of blood pressure change or the efficacy of managing blood pressure during dialysis, influencing the probability of IDH. Studies have indicated that the likelihood of IDH increases as the temperature of the dialysate rises to 37℃, above the isothermal condition\\u003csup\\u003e[\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e]\\u003c/sup\\u003e. Elevated historical dry weight and pre-dialysis weight suggest increased blood volume, which significantly correlates with IDH incidence. Reducing pre-dialysis dry weight has been shown to decrease IDH frequency\\u003csup\\u003e[\\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e]\\u003c/sup\\u003e, highlighting the impact of careful fluid management. An increased calcium load is associated with a heightened cardiovascular risk and also affects blood pressure\\u003csup\\u003e[\\u003cspan citationid=\\\"CR54\\\" class=\\\"CitationRef\\\"\\u003e54\\u003c/span\\u003e]\\u003c/sup\\u003e. Research has shown that the use of low-calcium dialysate in HD filtration significantly reduces post-treatment blood pressure levels. Moreover, this study finds that the incidence of IDH is predominantly in older patients, consistent with findings from other scholars\\u003csup\\u003e[\\u003cspan citationid=\\\"CR55\\\" class=\\\"CitationRef\\\"\\u003e55\\u003c/span\\u003e]\\u003c/sup\\u003e. Additionally, some patients exhibit blood pressure increases during dialysis following ultrafiltration dehydration. This phenomenon may be attributed to decreased volume load, reduced end-diastolic pressure, enhanced myocardial contractility, and increased cardiac output, collectively contributing to IDH. Antihypertensive medication removal during dialysis, particularly without dose adjustment, can lead to a reduction in blood efficacy, potentially raising blood pressure during the treatment. Endothelial dysfunction may lead to IDH by impairing the capacity for vasodilation\\u003csup\\u003e[\\u003cspan citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e]\\u003c/sup\\u003e. Moreover, post-dialysis hypertension is considered closely associated with cardiovascular and cerebrovascular diseases\\u003csup\\u003e[\\u003cspan citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e]\\u003c/sup\\u003e. The IDH-2 model incorporates a range of predictive indicators such as historical frequency of IDH incidents, post-dialysis systolic pressure, and various parameters related to pressure increase dynamics and dialysis conditions. Comparing the IDH-1 and IDH-2 models reveals common determinants in predicting IDH. However, the IDH-1 model focuses on rapid pre-dialysis prediction using pre-dialysis blood pressures as prognostic markers, while the IDH-2 model, aimed at post-dialysis forecasting, emphasizes the historical frequency of IDH and post-dialysis systolic pressure.\\u003c/p\\u003e \\u003cp\\u003eDespite the establishment of effective models for predicting IDH, this study recognizes the presence of certain limitations. The utilization of a retrospective multicenter approach introduces potential biases typically associated with this design, such as the challenges of non-prospective data collection and the risk of selection bias. Furthermore, the reliance on data from Asian populations may impair the generalizability of the models to diverse ethnic groups, thus potentially limiting their global applicability. It is advisable for subsequent research to investigate varied populations to enhance the models' generalizability.\\u003c/p\\u003e\"},{\"header\":\"5. Conclusions\",\"content\":\"\\u003cp\\u003eThe present research demonstrated that the LGBM algorithm-based model exhibits superior predictive performance for IDH compared to SVM and TabNet-based models. Two distinct models were developed: IDH-1 for immediate pre-dialysis prediction of IDH, and IDH-2 for forecasting IDH in upcoming dialysis sessions. The study employed patient data from Shenzhen People's Hospital to train the models and validated them using data from Fuding Hospital, thereby evaluating the models' predictive capabilities across different settings. Key indicators, including pre-dialysis diastolic and systolic pressures, historical average IDH episodes, and post-dialysis systolic pressures, were identified as significant for the early detection and management of IDH.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003eEthics approval and consent to participate\\u003c/p\\u003e\\n\\u003cp\\u003eThis study was approved by the Ethics Committee of Shenzhen People's Hospital (LL-KY-2021,870), Shenzhen, China and Fujian University of Traditional Chinese Medicine Clinical Special Funding (XB2022051).\\u003c/p\\u003e\\n\\u003cp\\u003eAuthor contributions\\u003c/p\\u003e\\n\\u003cp\\u003eResearch idea and study design: L.H.Y, J.J.D, Y.D, H.M.Z, S.L.H, Q.G; data analysis/interpretation: Q.G, H.M.Z, S.L.H. Model development: Q.G. Manuscript drafting: K.W, X.Z.Z, Y.H.L, S.D.L, D.E.T, M.Z.X, J.J.D, H.M.Z, S.L.H. Each author contributed to the manuscript revision. All authors take responsibility for the validity of data in this study.\\u003c/p\\u003e\\n\\u003cp\\u003eFunding\\u003c/p\\u003e\\n\\u003cp\\u003eThis work was supported by Guangdong Engineering Technology Research Center（507204531040），Guangzhou Development Zone entrepreneurship leading talent project (2017-L153), Guangdong Provincial R\\u0026amp;D Program for Key Areas (Grant No. 2023 B0101200010), Shenzhen Longhua District Science and technology innovation special fund project（11501A20220923BE5B6B3；11501A20220923BD5F291）\\u003c/p\\u003e\\n\\u003cp\\u003eCompeting Interest\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare that they have no relevant financial interests.\\u003c/p\\u003e\\n\\u003cp\\u003eAcknowledgments\\u003c/p\\u003e\\n\\u003cp\\u003eWe would like to thank the patients and families who participated in the study.\\u003c/p\\u003e\\n\\u003cp\\u003eConsent for publication\\u003c/p\\u003e\\n\\u003cp\\u003eAll authors agree to publish\\u003c/p\\u003e\\n\\u003cp\\u003eAvailability of data and material\\u003c/p\\u003e\\n\\u003cp\\u003eGraphical Abstract\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n \\u003cli\\u003eLv JC, Zhang LX. Prevalence and Disease Burden of Chronic Kidney Disease. Adv Exp Med Biol. 2019. 1165: 3-15.\\u003c/li\\u003e\\n \\u003cli\\u003eLiyanage T, Toyama T, Hockham C, et al. Prevalence of chronic kidney disease in Asia: a systematic review and analysis. BMJ Glob Health. 2022. 7(1): e007525.\\u003c/li\\u003e\\n \\u003cli\\u003eCrews DC, Bello AK, Saadi G. Burden, Access, and Disparities in Kidney Disease. Am J Nephrol. 2019. 49(3): 254-262.\\u003c/li\\u003e\\n \\u003cli\\u003eRuiz-Ortega M, Rayego-Mateos S, Lamas S, Ortiz A, Rodrigues-Diez RR. Targeting the progression of chronic kidney disease. Nat Rev Nephrol. 2020. 16(5): 269-288.\\u003c/li\\u003e\\n \\u003cli\\u003eEvans M, Lewis RD, Morgan AR, et al. A Narrative Review of Chronic Kidney Disease in Clinical Practice: Current Challenges and Future Perspectives. Adv Ther. 2022. 39(1): 33-43.\\u003c/li\\u003e\\n \\u003cli\\u003eBraun MM, Khayat M. Kidney Disease: End-Stage Renal Disease. FP Essent. 2021. 509: 26-32.\\u003c/li\\u003e\\n \\u003cli\\u003eAli M, Ejaz A, Iram H, Solangi SA, Junejo AM, Solangi SA. Frequency of Intradialytic Complications in Patients of End-Stage Renal Disease on Maintenance Hemodialysis. Cureus. 2021. 13(1): e12641.\\u003c/li\\u003e\\n \\u003cli\\u003eInrig JK, Patel UD, Toto RD, Szczech LA. Association of blood pressure increases during hemodialysis with 2-year mortality in incident hemodialysis patients: a secondary analysis of the Dialysis Morbidity and Mortality Wave 2 Study. Am J Kidney Dis. 2009. 54(5): 881-90.\\u003c/li\\u003e\\n \\u003cli\\u003eSebastian S, Filmalter C, Harvey J, Chothia MY. Intradialytic hypertension during chronic haemodialysis and subclinical fluid overload assessed by bioimpedance spectroscopy. Clin Kidney J. 2016. 9(4): 636-43.\\u003c/li\\u003e\\n \\u003cli\\u003eShamir AR, Karembelkar A, Yabes J, et al. Association of Intradialytic Hypertension with Left Ventricular Mass in Hypertensive Hemodialysis Patients Enrolled in the Blood Pressure in Dialysis (BID) Study. Kidney Blood Press Res. 2018. 43(3): 882-892.\\u003c/li\\u003e\\n \\u003cli\\u003eRaja SM, Seyoum Y. Intradialytic complications among patients on twice-weekly maintenance hemodialysis: an experience from a hemodialysis center in Eritrea. BMC Nephrol. 2020. 21(1): 163.\\u003c/li\\u003e\\n \\u003cli\\u003eFlythe JE, Chang TI, Gallagher MP, et al. Blood pressure and volume management in dialysis: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2020. 97(5): 861-876.\\u003c/li\\u003e\\n \\u003cli\\u003eVan Buren PN, Kim C, Toto RD, Inrig JK. The prevalence of persistent intradialytic hypertension in a hemodialysis population with extended follow-up. Int J Artif Organs. 2012. 35(12): 1031-8.\\u003c/li\\u003e\\n \\u003cli\\u003eLosito A, Del Vecchio L, Del Rosso G, Locatelli F. Postdialysis Hypertension: Associated Factors, Patient Profiles, and Cardiovascular Mortality. Am J Hypertens. 2016. 29(6): 684-9.\\u003c/li\\u003e\\n \\u003cli\\u003eSingh AT, Waikar SS, Mc Causland FR. Association of Different Definitions of Intradialytic Hypertension With Long-Term Mortality in Hemodialysis. Hypertension. 2022. 79(4): 855-862.\\u003c/li\\u003e\\n \\u003cli\\u003eZager PG, Nikolic J, Brown RH, et al. \\u0026quot;U\\u0026quot; curve association of blood pressure and mortality in hemodialysis patients. Medical Directors of Dialysis Clinic, Inc. Kidney Int. 1998. 54(2): 561-9.\\u003c/li\\u003e\\n \\u003cli\\u003eTsuruya K, Kanda E, Nomura T, Iseki K, Hirakata H. Postdialysis blood pressure is a better predictor of mortality than predialysis blood pressure in Japanese hemodialysis patients: the Japan Dialysis Outcomes and Practice Patterns Study. Hypertens Res. 2020. 43(8): 791-797.\\u003c/li\\u003e\\n \\u003cli\\u003eIatridi F, Theodorakopoulou MP, Papagianni A, Sarafidis P. Intradialytic hypertension: epidemiology and pathophysiology of a silent killer. Hypertens Res. 2022. 45(11): 1713-1725.\\u003c/li\\u003e\\n \\u003cli\\u003eIatridi F, Theodorakopoulou MP, Papagianni A, Sarafidis P. Management of intradialytic hypertension: current evidence and future perspectives. J Hypertens. 2022. 40(11): 2120-2129.\\u003c/li\\u003e\\n \\u003cli\\u003eInrig JK. Intradialytic hypertension: a less-recognized cardiovascular complication of hemodialysis. Am J Kidney Dis. 2010. 55(3): 580-9.\\u003c/li\\u003e\\n \\u003cli\\u003eHandelman GS, Kok HK, Chandra RV, Razavi AH, Lee MJ, Asadi H. eDoctor: machine learning and the future of medicine. J Intern Med. 2018. 284(6): 603-619.\\u003c/li\\u003e\\n \\u003cli\\u003eDeo RC. Machine Learning in Medicine. Circulation. 2015. 132(20): 1920-30.\\u003c/li\\u003e\\n \\u003cli\\u003eAhmad Loti NN, Mohd Noor MR, Chang SW. Integrated analysis of machine learning and deep learning in chili pest and disease identification. J Sci Food Agric. 2021. 101(9): 3582-3594.\\u003c/li\\u003e\\n \\u003cli\\u003eAli A, Fathalla A, Salah A, Bekhit M, Eldesouky E. Marine Data Prediction: An Evaluation of Machine Learning, Deep Learning, and Statistical Predictive Models. Comput Intell Neurosci. 2021. 2021: 8551167.\\u003c/li\\u003e\\n \\u003cli\\u003eTing Sim JZ, Fong QW, Huang W, Tan CH. Machine learning in medicine: what clinicians should know. Singapore Med J. 2023. 64(2): 91-97.\\u003c/li\\u003e\\n \\u003cli\\u003eFatima FS, Jaiswal A, Sachdeva N. Lung Cancer Detection Using Machine Learning Techniques. Crit Rev Biomed Eng. 2022. 50(6): 45-58.\\u003c/li\\u003e\\n \\u003cli\\u003eAngraal S, Mortazavi BJ, Gupta A, et al. Machine Learning Prediction of Mortality and Hospitalization in Heart Failure With Preserved Ejection Fraction. JACC Heart Fail. 2020. 8(1): 12-21.\\u003c/li\\u003e\\n \\u003cli\\u003eAlaimo L, Lima HA, Moazzam Z, et al. Development and Validation of a Machine-Learning Model to Predict Early Recurrence of Intrahepatic Cholangiocarcinoma. Ann Surg Oncol. 2023. 30(9): 5406-5415.\\u003c/li\\u003e\\n \\u003cli\\u003eGottlieb ER, Samuel M, Bonventre JV, Celi LA, Mattie H. Machine Learning for Acute Kidney Injury Prediction in the Intensive Care Unit. Adv Chronic Kidney Dis. 2022. 29(5): 431-438.\\u003c/li\\u003e\\n \\u003cli\\u003eShouval R, Fein JA, Savani B, Mohty M, Nagler A. Machine learning and artificial intelligence in haematology. Br J Haematol. 2021. 192(2): 239-250.\\u003c/li\\u003e\\n \\u003cli\\u003eJoshi RD, Dhakal CK. Predicting Type 2 Diabetes Using Logistic Regression and Machine Learning Approaches. Int J Environ Res Public Health. 2021. 18(14): 7346.\\u003c/li\\u003e\\n \\u003cli\\u003eDong J, Feng T, Thapa-Chhetry B, et al. Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care. Crit Care. 2021. 25(1): 288.\\u003c/li\\u003e\\n \\u003cli\\u003eKate RJ, Pearce N, Mazumdar D, Nilakantan V. A continual prediction model for inpatient acute kidney injury. Comput Biol Med. 2020. 116: 103580.\\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. Int J Med Inform. 2021. 151: 104484.\\u003c/li\\u003e\\n \\u003cli\\u003eThongprayoon C, Hansrivijit P, Bathini T, et al. Predicting Acute Kidney Injury after Cardiac Surgery by Machine Learning Approaches. J Clin Med. 2020. 9(6): 1767.\\u003c/li\\u003e\\n \\u003cli\\u003eSu CT, Chang YP, Ku YT, Lin CM. Machine Learning Models for the Prediction of Renal Failure in Chronic Kidney Disease: A Retrospective Cohort Study. Diagnostics (Basel). 2022. 12(10): 2454.\\u003c/li\\u003e\\n \\u003cli\\u003eSchena FP, Anelli VW, Abbrescia DI, Di Noia T. Prediction of chronic kidney disease and its progression by artificial intelligence algorithms. J Nephrol. 2022. 35(8): 1953-1971.\\u003c/li\\u003e\\n \\u003cli\\u003eZacharias HU, Altenbuchinger M, Schultheiss UT, et al. A Predictive Model for Progression of CKD to Kidney Failure Based on Routine Laboratory Tests. Am J Kidney Dis. 2022. 79(2): 217-230.e1.\\u003c/li\\u003e\\n \\u003cli\\u003eChuah A, Walters G, Christiadi D, et al. Machine Learning Improves Upon Clinicians\\u0026apos; Prediction of End Stage Kidney Disease. Front Med (Lausanne). 2022. 9: 837232.\\u003c/li\\u003e\\n \\u003cli\\u003eG\\u0026oacute;mez-Pulido JA, G\\u0026oacute;mez-Pulido JM, Rodr\\u0026iacute;guez-Puyol D, Polo-Luque ML, Vargas-Lombardo M. Predicting the Appearance of Hypotension During Hemodialysis Sessions Using Machine Learning Classifiers. Int J Environ Res Public Health. 2021. 18(5): 2364.\\u003c/li\\u003e\\n \\u003cli\\u003eHong D, Chang H, He X, et al. Construction of an Early Alert System for Intradialytic Hypotension before Initiating Hemodialysis Based on Machine Learning. Kidney Dis (Basel). 2023. 9(5): 433-442.\\u003c/li\\u003e\\n \\u003cli\\u003eLee H, Yun D, Yoo J, et al. Deep Learning Model for Real-Time Prediction of Intradialytic Hypotension. Clin J Am Soc Nephrol. 2021. 16(3): 396-406.\\u003c/li\\u003e\\n \\u003cli\\u003eKale G, Mali M, Bhangale A, Somani J, Jeloka T. Intradialytic Hypertension Increases Non-access Related Hospitalization and Mortality in Maintenance Hemodialysis Patients. Indian J Nephrol. 2020. 30(2): 85-90.\\u003c/li\\u003e\\n \\u003cli\\u003eChoi H, Lee JY, Sul Y, et al. Comparing machine learning and logistic regression for acute kidney injury prediction in trauma patients: A retrospective observational study at a single tertiary medical center. Medicine (Baltimore). 2023. 102(33): e34847.\\u003c/li\\u003e\\n \\u003cli\\u003eZhao X, Lu Y, Li S, et al. Predicting renal function recovery and short-term reversibility among acute kidney injury patients in the ICU: comparison of machine learning methods and conventional regression. Ren Fail. 2022. 44(1): 1326-1337.\\u003c/li\\u003e\\n \\u003cli\\u003eGe C, Deng F, Chen W, et al. Machine learning for early prediction of sepsis-associated acute brain injury. Front Med (Lausanne). 2022. 9: 962027.\\u003c/li\\u003e\\n \\u003cli\\u003eKang BS, Lee SU, Hong S, et al. Prediction of gestational diabetes mellitus in Asian women using machine learning algorithms. Sci Rep. 2023. 13(1): 13356.\\u003c/li\\u003e\\n \\u003cli\\u003eLee Y, Ryu J, Kang MW, et al. Machine learning-based prediction of acute kidney injury after nephrectomy in patients with renal cell carcinoma. Sci Rep. 2021. 11(1): 15704.\\u003c/li\\u003e\\n \\u003cli\\u003eGupta SR. Prediction time of breast cancer tumor recurrence using Machine Learning. Cancer Treat Res Commun. 2022. 32: 100602.\\u003c/li\\u003e\\n \\u003cli\\u003eYu J, Chen X, Wang Y, et al. Intradialytic systolic blood pressure variation can predict long-term mortality in patients on maintenance hemodialysis. Int Urol Nephrol. 2021. 53(4): 785-795.\\u003c/li\\u003e\\n \\u003cli\\u003eYang J, Huang J, Yu B, et al. Long-term predialysis blood pressure variability and outcomes in hemodialysis patients. J Clin Hypertens (Greenwich). 2022. 24(2): 148-155.\\u003c/li\\u003e\\n \\u003cli\\u003eVeerappan I, Thiruvenkadam G, Abraham G, Dasari BR, Rajagopal A. Effect of Isothermic Dialysis on Intradialytic Hypertension. Indian J Nephrol. 2019. 29(5): 317-323.\\u003c/li\\u003e\\n \\u003cli\\u003eZhang Y, Zhang X, Li J, et al. Dry-weight reduction improves intradialytic hypertension only in patients with high predialytic blood pressure. Blood Press Monit. 2019. 24(4): 185-190.\\u003c/li\\u003e\\n \\u003cli\\u003ePirklbauer M, Fuchs L, Heiss R, Ratschiller T, Mayer G. Intradialytic Calcium Kinetics and Cardiovascular Disease in Chronic Hemodialysis Patients. Blood Purif. 2020. 49(6): 723-732.\\u003c/li\\u003e\\n \\u003cli\\u003eEftimovska-Otovic N, Grozdanovski R, Taneva B, Stojceva-Taneva O. Clinical Characteristics of Patients with Intradialytic Hypertension. Pril (Makedon Akad Nauk Umet Odd Med Nauki). 2015. 36(2): 187-93.\\u003c/li\\u003e\\n \\u003cli\\u003eTawfeek GA, Kora MA, Yassein YS, Baghdadi AM, Elzorkany KM. Association of pre-pro-endothelin gene polymorphism and serum endothelin-1 with intradialytic hypertension in an Egyptian population. Cytokine. 2021. 137: 155293.\\u003c/li\\u003e\\n \\u003cli\\u003eCho H, Kwon SK, Lee SW, et al. The Association Among Post-hemodialysis Blood Pressure, Nocturnal Hypertension, and Cardiovascular Risk Factors. Electrolyte Blood Press. 2023. 21(2): 53-60.\\u003c/li\\u003e\\n\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"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\":\"info@researchsquare.com\",\"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\":\"Chronic Kidney Disease, Hemodialysis, Machine Learning\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-5355171/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-5355171/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e\\u003cstrong\\u003eObjective\\u003c/strong\\u003e: The global prevalence of chronic kidney disease (CKD) is escalating, particularly for end-stage renal disease (ESRD), which has led to greater dependence on hemodialysis. This upswing exerts substantial strains on patient families and healthcare systems. A critical concern during hemodialysis is the emergence of Intradialytic Hypertension (IDH), which carries significant health risks. Delayed management of IDH can lead to severe cardiovascular and cerebrovascular complications. The aim of our study was to harness machine learning methodologies to develop a predictive algorithm for IDH, utilizing patient demographic data and dialysis records. Our model equips medical professionals with a robust predictive tool that enhances the detection of patients more susceptible to hypertension during dialysis, thereby advancing the pre-screening for individuals considered at increased risk.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eMethods\\u003c/strong\\u003e: This study developed two predictive models for IDH, named IDH-1 and IDH-2, by employing a suite of machine learning algorithms, namely the Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and TabNet. IDH-1 is specifically engineered to provide immediate predictions of IDH risk prior to a hemodialysis session, utilizing records from the imminent pre-dialysis period combined with historical average dialysis data, whereas IDH-2 employs records from the current dialysis session along with historical average data to forecast the risk of IDH for the next hemodialysis session. The performance evaluation of the models utilized key metrics, including Area Under the Curve (AUC), recall, accuracy, and F1 score, which are crucial in determining the models' precision and reliability.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eResults\\u003c/strong\\u003e: This research analyzed data from 1,405 patients at Shenzhen People's Hospital over 185,125 dialysis sessions and 416 patients at Fuding City Hospital across 71,427 sessions. Data from Shenzhen served as the training set, while Fuding data comprised the test set, supporting the model development and validation process. In the IDH-1 models, the LGBM outperformed SVM and TabNet with an AUC of 0.87. LGBM achieved a recall of 0.73, an accuracy of 0.243, and an F1 score of 0.36. For IDH-2 models, LGBM maintained superior performance, with an AUC of 0.75, a recall of 0.56, an accuracy of 0.17, and an F1 score of 0.26. Predictor importance analysis for the LGBM algorithm identified pre-dialysis diastolic pressures, historical mean arterial pressure, and historical average IDH episodes as significant for the IDH-1 model. For the IDH-2 model, historical average IDH episodes and post-dialysis systolic pressures were most predictive.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConclusions\\u003c/strong\\u003e: This study's results highlight the significant potential of machine learning techniques in leveraging demographic and dialysis data to predict IDH in patients undergoing hemodialysis.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Prediction model of intradialytic hypertension in hemodialysis patients based on machine learning\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-12-11 14:17:39\",\"doi\":\"10.21203/rs.3.rs-5355171/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"9df6f1c5-3515-4005-b493-408a6e0512d3\",\"owner\":[],\"postedDate\":\"December 11th, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2024-12-27T08:53:23+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2024-12-11 14:17:39\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-5355171\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-5355171\",\"identity\":\"rs-5355171\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}