Hemodialysis Population Research Articles

Echocardiographic Changes in Prevalent Hemodialysis Population Based on Cardiac Symptomatology

BACKGROUND: Despite the prevalence of cardiovascular complications in hemodialysis (HD) patients, routine screening for asymptomatic individuals remains underexplored in clinical practice, particularly beyond those assessed for kidney transplantation. This study aimed to investigate echocardiographic changes in prevalent HD patients, irrespective of symptomatic presentation. SUBJECTS AND METHODS: A prospective, single-center study spanning 36 months included 79 HD patients. Grouping was based on cardiac symptomatology assessed with the New York Heart Association (NYHA) classification: group 1 comprised 18 asymptomatic patients (NYHA class I), while group 2 included 61 patients with moderate cardiac symptoms (NYHA classes II and III). Routine laboratory analyses, miRNA 133, hydration, and nutritional parameters were assessed, with echocardiography performed a day post-HD. RESULTS: Demographic data, HD quality, blood pressure, therapy, and most echocardiographic indicators showed no significant differences. Asymptomatic patients exhibited noteworthy cardiac structural and functional abnormalities: 55.5% had left ventricular hypertrophy (LVH), and 72.2% had increased left atrial volume index (LAVi). Concentric hypertrophy was observed in 7/18 asymptomatic patients, while 15/61 symptomatic patients demonstrated concentric left ventricle remodeling. In comparison to group 2, group 1 displayed higher hemoglobin, uric acid, and miRNA 133 concentrations, along with better hydration control and higher lean tissue index. All these dependent variables show statistically significant relation to the NYHA: approximately 13% of the variance in Hb levels and 18% of the variance in miRNA can be explained by NYHA classification. CONCLUSION: This study advocates for routine echocardiographic and cardiac examinations for all HD patients from the treatment's onset, irrespective of symptoms. Maintaining optimal volume, nutrition, urate concentrations, and hemoglobin is crucial for comprehensive cardiac care in this population. Additionally, exploring whether interventions targeting these non-cardiac factors could alleviate symptoms and improve quality of life in symptomatic HD patients would be worthwhile.

The prevalence and risk of mortality associated with intradialytic hypertension among patients with end-stage kidney disease on haemodialysis: A systematic review and meta-analysis.

Intradialytic hypertension (IDHTN) is a common but less frequently recognised complication of haemodialysis. However, it is associated with increased overall mortality in patients on haemodialysis. This systematic review and meta-analysis aimed to determine the prevalence of IDHTN and associated mortality risk in the global haemodialysis population. A systematic search of PubMed and EMBASE was undertaken to identify articles with relevant data published between 1990 and 2023. The pooled prevalence of IDHTN in the global haemodialysis population was determined using the DerSimonian-Laird random-effects meta-analysis. The pooled hazards ratio for mortality in patients with IDHTN was also computed from the studies that reported mortality among haemodialysis patients with IDHTN. The study protocol was registered with PROSPERO (CRD42023388278). Thirty-two articles from 17 countries were included, with a pooled population of 127,080 hemodialysis patients (median age 55.1 years, 38.2% females). Most studies had medium methodological quality (53.1%, n = 17). The overall pooled prevalence of IDHTN was 26.6% [(95% CI 20.2-33.4%), n = 27 studies, I2 = 99.3%, p<0.001 for heterogeneity], with significant differences depending on the definition used. The pooled proportion of haemodialysis sessions with IDHTN was 19.9% [(95% 12.5-28.6%, n = 8 studies, I2 = 99.3%, p<0.001 for heterogeneity)] with significant differences across the different definition criteria. The p-value for the Begg test was 0.85. The median pre-dialysis blood pressure was not significantly associated with IDHTN. The pooled hazard ratio for mortality was 1.37 (95% CI 1.09-1.65), n = 5 studies, I2 = 13.7%, and p-value for heterogeneity = 0.33. The prevalence of IDHTN is high and varies widely according to the definition used. A consensus definition of IDHTN is needed to promote uniformity in research and management. The increased mortality risk forecasted by IDHTN highlights the need for optimal blood pressure control in patients on hemodialysis.

Disturbance in the potential cardiovascular-bone-skeletal muscle axis and morbidity and mortality in patients undergoing haemodialysis: the Q-Cohort Study.

Disturbances in the cardiovascular system, bone and skeletal muscle are independent risk factors for death among patients receiving haemodialysis (HD). However, the combined impact of disorders of these three organs on morbidity and mortality is unclear in the HD population. A total of 3031 Japanese patients on maintenance HD were prospectively followed. The outcomes were all-cause mortality, major adverse cardiovascular events (MACE) and bone fracture. Patients were divided into four groups (G1-G4) according to the baseline number of diseased organs represented as histories of cardiovascular disease and bone fractures and the presence of low skeletal muscle mass as follows: G1, no organ; G2, one organ; G3, two organs; G4, three organs. Multivariable-adjusted survival models were used to analyse associations between the number of diseased organs and outcomes. During a 4-year follow-up, 499 deaths, 540 MACE and 140 bone fractures occurred. In the Cox proportional hazards model, the risk for all-cause mortality was significantly higher in G2, G3 and G4 than in G1 as the reference {hazard ratio: G2, 2.16 [95% confidence interval (CI) 1.65-2.84], G3, 3.10 [95% CI 2.27-4.23] and G4, 3.11 [95% CI 1.89-5.14]}. Similarly, the risks for developing MACE and bone fractures were significantly elevated as the number of organ disorders increased. Multiple disorders of the cardiovascular-bone-skeletal muscle axis are strong predictors of morbidity and mortality in patients undergoing HD.

Preventive dental care reduces risk of cardiovascular disease and pneumonia in hemodialysis population: a nationwide claims database analysis

This study aims to investigate the impact of dental care utilization status on the occurrence of fatal complications such as cerebral/cardiovascular disease (CVD) and infectious diseases in patients with end-stage renal disease (ESRD) undergoing hemodialysis. This retrospective cohort study was performed using the Japanese claims database and included patients who first underwent hemodialysis between April 2014 and September 2020. The exposure variable of interest was the pattern of dental utilization, which was categorized into three groups, “dental treatment group”, “preventive dental care group”, and “no-dental visit group”. The primary outcomes were the time interval until a composite end point of first major cardiovascular event (acute myocardial infarction, heart failure, or cerebral infarction) of infectious disease (pneumonia and sepsis). The secondary outcomes were the time interval until the incidence of each component of primary outcomes. Survival analyses, including log-rank tests and Cox proportional hazards regression analyses, were performed. Among the 10,873 patients who underwent the first dialysis treatment, 6152 were assigned to the no-dental visit group, 2221 to the dental treatment group, and 2500 to the preventive dental care group. The preventive dental care group had significantly lower hazard ratios (HRs) of the incidence of CVD (adjusted hazard ratio [aHR]: 0.86, 95% confidence interval [CI]: 0.77–0.96) and infectious diseases (aHR: 0.86, 95% CI: 0.76–0.97). As for pneumonia, preventive dental care and dental treatment groups had significantly lower HRs (aHR: 0.74 and 0.80, 95% CI: 0.61–0.88, 0.66–0.96) than the no-dental visit group. This study demonstrated that dental visits for preventive dental care were associated with a significant risk reduction in CVD and infectious complications in patients with ESRD undergoing hemodialysis.

#2990 Prevalence and risk factors of altered control of BP during the night in a large hemodialysis population investigated by 48 hours ABPM

Abstract Background and Aims Altered BP control during the night (non-dipping, reverse dipping and nocturnal hypertension) is recognized as a powerful risk factor for death and CV events in the hemodialysis (HD) population. However, these alterations have been described in relatively small studies and the epidemiology and risk factors for subtypes non-dipping and nocturnal hypertension are still poorly characterized in the current literature. We investigated the problem in a large Registry (The EURECAm Registry) that is the sole that adopted state of the art 48 h ABPM recording in the HD population. Method We included in this analysis 534 HD patients, enrolled in 7 centers, led by members of the EURECA-m working group, in 3 European countries. 48 h ABPM was measured by using well validated instruments (AAMI/ESH/ISO). As recommended by the European Society of Hypertension guidelines, recordings were made at 15-minute intervals during the day and 30 minutes during the night. Hypertension was defined as 48 h blood pressure (BP) &amp;gt;130/80, nocturnal hypertension as a nighttime BP &amp;gt;120/70 mmHg, non-dipping as a night/day systolic BP fall &amp;lt;10% and reverse dipping as a night/day ratio &amp;gt;1. Results Among 534 HD patients, 317 were hypertensive and. 217 normotensive. As many as 274 patients were non-dippers (51%), 32% were reverse dippers, and 381 (71%) had nocturnal hypertension. Of note, forty-one per cent of non-dippers, 37% of reverse dippers and 19% of patients with nocturnal hypertension were normotensives. As expected, the majority of reverse dippers (87%) had frank nocturnal hypertension. In a combined analysis of non-dippers and reverse dippers, as compared to dippers, these patients more frequently exhibited atrial fibrillation and had lower Heart Rate both pre-dialysis (72± 11 vs 77±13 bpm, P = 0.004) and post-dialysis (74± 13 vs 78±14 bpm, P = 0.03) and lower serum albumin (3.9±0.4 vs 4.0±0.3 g/dL, P = 0.02) but did not differ for age and gender. Conclusion In a large cohort of HD patients investigated by the state-of-the-art technique, 48 h ABPM alterations in the nocturnal BP profile were almost universal, and nocturnal hypertension was the more prevalent alteration, followed by non-dipping and reverse dipping. These alterations were independent of age and gender. Long-term, granular follow-up analyses are warranted to identify the prognostic value of these alterations.

#2654 Erythropoietin hypo-responsiveness in haemodialysis patients, contributing factors and outcome

Abstract Background and Aims Erythropoietin (ESA) resistance or hypo-responsiveness (HR) is a condition when patients do not achieve the desired haemoglobin concentration or require incremental ESA doses to maintain stable Hb concentration. Contributing factors include iron deficiency, inflammation, infection, mineral bone disease (CKD-MBD), insufficient dialysis, malnutrition, and hematologic abnormalities. This study aimed to determine the prevalence of HR in maintenance haemodialysis (HD) population and its contributing factors. The outcome in terms of all-cause mortality and need for hospitalization over 6 months follow-up was studied. Method We prospectively studied 100 maintenance HD patients above 18 years age over one and a half years. Those with a known haematological disorder, malignancy or active infection were excluded. Enrolled patients were classified into 2 groups: 1) Initial ESA Group: Initiated on ESA therapy and had Hb less than 10 gm/dl. These were also incident HD patients. Initial ESA HR for the study was defined as any Hb increase less than 0.8 gm/dl over 1 month after initiating ESA. The ESA dose and modification remained as per current existing practice in the dialysis unit. 2) Subsequent ESA Group: Already receiving ESA therapy and on HD for at least 3 months. Patients on previously stable doses of ESA who required two increments in ESA doses up to 50% beyond the previous dose to maintain Hb were taken as HR. Patients who required Epoetin or equivalent ESA dose of more than 300 IU/kg/week were also included in HR. All underwent anaemia workup at start and monthly haemoglobin estimation. They were also tested for CRP and CKD-MBD. All were followed for 6 months; hospitalization (excluding vascular access) and mortality was noted. The Charlson comorbidity index was used to ensure equitability. Results Seventeen cases were in Initial ESA group and 83 cases were in subsequent ESA group. ESA type was not associated with responsiveness in either group. In the ‘initial’ group, 8/17 were HR. Iron deficiency was noted in 75 and 33 percent of HR vs R patients (p =0.15). Median CRP in HR group was significantly higher at 10 mg/dL (7.4-20.4) as compared to R 2.8mg/dl (1.5-4), p=0.034. Distribution of other bio-chemical parameters was comparable. HR patients had more hospitalizations (62.5% vs 22.2%; p =0.153). In subsequent ESA group 10/83 (12%) cases were HR. Mean average weekly ESA dose (IU/Kg/week) in HR (286 ± 28) was significantly higher than R group (137 ± 44, p &amp;lt;0.0001). Median CRP (mg/dL) in HR was 20.2 (8.8-27.6) and was significantly higher than R 2.6 (1.2-5.4, p=0.0001). Median iPTH (pg/mL) in HR group was 899 (801-1053) significantly higher than R 418 (242-632; p =&amp;lt;0.0001). Mean serum phosphorus (mg/dL) in HR was 6.4 ± 1.7 which was significantly higher as compared to R (5.3 ± 1.6, p value=0.038). Median ALP of 366 (276-408) IU/L in HR was significantly higher than in R 126 (93-196, p=0.0007). Both groups were equitable in adequacy. Hospitalization rate was not (60 vs 30%, p=0.09) different. Mortality rate was significantly higher in HR as compared to R (40% vs 1.4%, p=0.0005). Conclusion The study revealed a substantial occurrence of ESA hypo-responsiveness in HD population. High CRP, signifying inflammation and CKD-MBD turned out to be significant contributors to development of hypo-responsiveness. ESA hypo-responsiveness is associated with higher mortality over short term.

#620 Remaining life expectancy of Korean hemodialysis patients: how much longer can they live?

Abstract Background and Aims Hemodialysis (HD) patients have a higher mortality rate compared to general population. However, no study has investigated life expectancy in Korean HD patients so far. Therefore, this study aimed to calculate the remaining life expectancy among Korean maintenance HD patients and compare it to those of general population as well as HD patients from other countries. Method Baseline data were retrieved from HD quality assessment data from October to December 2015. Among the total number of patients who were alive at the beginning of 2016 (20,537 men and 14,404 women), a total of 22,439 (12,848 men and 9,591 women) were still alive at the end of 2021 while 12,502 (7,689 men and 4,813 women) were deceased during 6 years of follow-up. We used the life table method to calculate the expected remaining years of life in two-year increments. Results The mean age of the patients was 60.5 ± 12.4 years, and 58.7% were male. The average duration of HD was 5.63 ± 5.11 years. The most common cause of end-stage kidney disease was diabetes mellitus (42.0%), followed by hypertension (27.1%) and glomerulonephritis (10.9%). The remaining life expectancy for 60-year-old patients were 11.64 years for men and 14.64 years for women. The average remaining lifetimes of the HD population were only about a half of the general population (Fig.). Diabetic patients demonstrated shorter life expectancy compared to the patients with hypertension or glomerulonephritis. The remaining life expectancy of Korean HD patients was similar to that of Japanese and was almost double that of HD patients in Western countries such as Europe and United States. Conclusion HD population showed shorter life expectancy compared to the general population. Asian HD population seemed to demonstrate better patient survival compared to other countries. Longitudinal analysis should be warranted to analyze the effect of advanced dialysis technology upon improved survival rate among HD population.

#1080 Low-T3 syndrome: “endemic” in hemodialysis patients?

Abstract Background and Aims In hemodialysis patients, low-T3 syndrome is more prevalent than in the general population and was found to be associated with malnutrition and inflammation. Although evidence suggest that low free T3 (fT3) levels predict cardiovascular and all-cause mortality, low-T3 syndrome is usually not treated. The aim of this single-center retrospective observational study was to determine the prevalence of thyroid axis derangements and examine the relationship between thyroid hormones and biochemical and clinical variables in our hemodialysis (HD) patients cohort. Method Patients on thrice-weekly maintenance HD for at least 6 months were enrolled. Patients taking amiodarone, levothyroxine or methimazole were excluded. Data about age, sex, ethnicity, hemodialysis vintage, type of dialysis (bicarbonate dialysis/online hemodiafiltration/acetate-free-biofiltration/hemofiltration with endogenous reinfusion/expanded hemodialysis) were obtained from medical records and values of fT3, fT4, TSH, albumin, LDL, hemoglobin—routinely checked for every patient during hemodialysis sessions—were acquired. Continuous variables are expressed as median and IQR or mean ± SD and categorical variables as absolute value or percentage. The normality of data distribution was tested by Sapiro-Wilk test. Correlations were studied with Pearson and Spearman. T-test and Mann-Whitney U test were performed to compare continuous and categorical variables. Statistical significance was reached with p &amp;lt; 0.05. Results A total of 57 patients on thrice-weekly maintenance HD were enrolled. Most of them were male (70.2%), Caucasians (89.5%), with a mean age of 69 ± 14 years and with a median dialytic vintage of 63 [29; 113] months. Hypertension was the most frequent comorbidity and more than half of the patients was either affected by ischemic, hypertensive or dilatative cardiomyopathy. Most of the patients underwent bicarbonate hemodialysis (42.1%) and overall 57.9% of the study population was treated by convective techniques. Hemoglobin, albumin, LDL cholesterol, fT3 and fT4 values was normally distributed, while those of TSH and PCR were not. A very high prevalence of low-T3 syndrome (75%) was found, 5 patients showed subclinical hypothyroidism and only in a minority of patients thyroid function tests were normal (12%) (Fig. 1). A significant negative correlation between fT3 and age (p = 0.041) and a positive one between TSH and dialytic vintage (p = 0.017) were found (Table 1). In addition, a significant positive correlation between albumin and fT3 was detected, supporting the known association between low T3 and malnutrition. Interestingly, PCR negatively correlated with fT3 and positively with fT4. Patients treated by diffusive techniques showed no statistically significant difference of thyroid hormone levels compared to convective therapies (Table 1). Conclusion Low-T3 syndrome affects the great majority of patients in our hemodialysis population, with a prevalence even higher than that provided by literature. Our investigations confirm that older patients seem to show lower fT3 levels and a longer hemodialysis vintage may correlate with higher TSH levels. The association with inflammation was also confirmed. Considering the known impact on cardiovascular and all-cause mortality already demonstrated, it is important to routinely check thyroid hormones in HD. In addition, further studies should be performed in order to establish the possible impact of the therapeutic correction of this hormonal derangement on hemodialysis patients’ mortality and cardiovascular outcomes.

#91 Ultrafiltration rate induced cardiac strain (ULRICA) – study. Preliminary results of a prospective multicentre study in Sweden

Abstract Background and Aims Cardiovascular disease is the principal cause of mortality in the haemodialysis population, attributable not only to traditional risk factors but also to those specific to dialysis treatment such as fluid overload and ultrafiltration. These are modifiable risk factors. The cardiac biomarkers N-terminal pro-B-type natriuretic peptide (proBNP) and hs-Troponin T (TnT) increase during dialysis (Laveborn et al. 2012). Previous studies (Goto et al. 2023) hypothesize that this increase was related to ultrafiltration rate. The aim of the study was to determine if modifying ultrafiltration rate could reduce the elevation of these cardiac biomarkers. Method ULRICA is a multicentre study registered at ClinicalTrials.gov (NCT06153888), conducted in six centres across Sweden. It investigated the effect of ultrafiltration rate in adult, stable haemodialysis patients with a) interdialytic weight gain (IDWG) of more than 2.5% of dry body weight and b) an ultrafiltration rate of over 0.72 IDWG/h (standard UF rate). In a crossover setting each patient underwent one treatment according to standard prescription and, after one week, a second treatment with prolonged UF time to achieve an ultrafiltration rate of less than 0.6 IDWG/h or a reduction of ultrafiltration rate of at least 20% compared to standard treatment. All treatments were performed using low flux dialysers. Blood samples were collected before, after 180 min and after treatment. Levels of proBNP and TnT were then adjusted for haemoconcentration after the formula of Schneditz et al. (2012). Echocardiographic measurements were derived from echocardiographs performed within one year from study start under stable conditions. Due to the small sample size and non-normally distributed values, data were analysed using Wilcoxon signed-rank test for comparisons between treatments and Spearman's rank correlation coefficient for correlations. Results This preliminary one-centre report included 12 patients (four women, eight men) with a median age of 68.5 (range 32-86 years). There was a significant increase in TnT (p=0.008) and proBNP (p= 0.002) during dialysis with standard UF rate. The median levels of cardiac markers before treatment did not differ between the two treatments, TnT 61.5 ng/l (IQR 33.3-87.2) resp. 59.5 ng/l (IQR 35.5-90.0) and proBNP 12177.5 ng/l (IQR 3112.5-34773.0) resp. 9105 ng/l (IQR 2555.3-36803.0); (p=0.906 for TnT and p=0.638 for proBNP). The median increase of TnT was 4.1 ng/l (IQR 1.2-8.4) during standard treatment and 5.8 ng/l (IQR 0.1-10.3) during the prolonged treatment (p= 0.158). The median increase in proBNP was 3257.5 ng/l (IQR 855.0- 15074.7) during standard treatment and 2996.9 ng/l (IQR 724.1-13338.4) during the prolonged treatment (p=0.814). The changes at 180 min were also non-significant (p=0.182 for TnT and p=0.875 for proBNP). There was a positive correlation between the change in proBNP during treatment and age (R=0.85, p&amp;lt;0.001) and left ventricular mass index (LVMI) (R=0.89, p=0.019). There was also a correlation between age and LVMI (R=0.83, p=0.042). Conclusion Preliminary data of a small group of patients confirm an increase of TnT and proBNP during dialysis. Age and LVMI are strongly associated with increase of proBNP. To show whether a reduction of ultrafiltration rate per se might reduce the release of cardiac biomarkers during haemodialysis treatment, further inclusion of patients is in progress.

#861 Intradialytic oxygen saturation in chronic hemodialysis patients—results from a large U.S. population

Abstract Background and Aims The study of intradialytic oxygen saturation (SO2) in hemodialysis (HD) patients is a significant yet under-explored area. This study's objective was to evaluate the epidemiological aspects of SO2 during HD. Method SO2 levels during HD were measured using the Crit-Line Monitor (CLM; Fresenius Medical Care North America, Waltham, MA), a device certified by the US Food and Drug Administration (FDA) for the measurement of hematocrit and SO2 in the extracorporeal circuit. CLM measurements were recorded at 10-second intervals and transmitted into Amazon Web Services (AWS) via Apache KAFKA, a real time streaming software. Apache KAFKA was rolled out in phased approach, started early 2021. We extracted CLM data between January 14, 2021, and July 30, 2023. CLM values with the following characteristics were deemed implausible and hence, excluded: SO2 ≤ 0 or ≥99.9; hematocrit &amp;lt; 15% or &amp;gt;60%. We computed basic statistics of SO2 for every HD session. We then excluded the sessions with the standard deviation of SO2 &amp;lt; 0.1%, an interquartile range of zero, and sessions with less than 360 SO2 measurements (i.e., less than 60 minutes of recording). Results We collected data from over 5.6 million HD sessions. After data cleaning, we retained 4.9 million HD sessions from 72,656 patients treated in 727 dialysis clinics across the US. The average duration of a session was 204.5 (±42.8) minutes. Per HD session, on average 1,227 (±257) SO2 measurements were recorded. The distribution of the number of SO2 measurements per HD session is shown in Fig. 1A. The mean of the session-level average SO2 was 88% (±15%). The distribution of session-level average SO2 is shown in Fig. 1B. The bimodal distribution of SO2 is attributed to the use of arterio-venous and central venous vascular accesses, respectively. Conclusion We collected and analysed a substantial dataset of intradialytic SO2 in a large US HD population. These data provide a rich source for future research into SO2 and its clinical correlates. Exploring intradialytic SO2 in depth will offer valuable insights to improve patient care and optimize treatment procedures.

#1085 Prediction models for hospitalization in patients with advanced chronic kidney disease: systematic review, external validation and development study

Abstract Background and Aims Hospitalization is a common complication in patients with advanced chronic kidney disease (CKD). Predicting hospitalization would be beneficial for clinicians and patients. Clinicians may be assisted in adequate healthcare planning by being able to anticipate hospitalizations. Additionally, patients have indicated wanting more information about their risk of hospitalization. By providing patients a personal indication of their risk of hospitalization, they better know what to expect of the future which can foster a feeling of control. Especially for incident dialysis patients this might help, as they enter a new and to them still largely unknown stage of their disease. Moreover, if a patient knows they have a high risk of hospitalization, they may better cope with the hospitalization if it occurs, as it is less unexpected. Therefore, we aimed to 1) identify and give an overview of current prediction models for hospitalization in CKD patients, 2) externally validate these models, and 3) develop a new model if no well-performing models were identified. Method We performed a systematic review in PubMed to identify prediction models for hospitalization, length of stay (LOS), and readmission. Identified models were appraised using the Prediction model Risk of Bias Assessment Tool. We targeted three external validations in hemodialysis (HD), peritoneal dialysis (PD), and pre-dialysis patients in a Dutch dialysis and European pre-dialysis cohort. Prediction models were externally validated on discrimination (C-statistic) and calibration (calibration plot, slope, and calibration-in-the-large). We developed a Fine-Gray model for hospitalization within one year in patients initiating HD, taking into account the competing risk of death. Informed by model performance, we also investigated Cox and logistic regression as modeling strategies. We performed a sensitivity analysis, redefining the outcome to be predicted as the 1-year risk of hospitalization of 3 days or longer. Results From the systematic search, we identified 17 models in 6 studies. All models were developed in the USA or in China. Two models predicted hospitalization rate in incident PD patients and 3 models predicted 30-day readmission in maintenance HD patients and in CKD stage 3-5 patients not on dialysis discharged after a heart failure hospitalization. The risk of 1 and 6 hospitalizations within one year in prevalent HD patients were predicted by one model each. Prolonged LOS in PD patients was predicted by ten models. All models were of low quality with a high risk of bias. The most common shortcomings were using future information to make predictions, inadequate handling of missing data, and model overfitting. Due to underreporting and unavailability of predictors, we could only externally validate 3 models. These models were poorly calibrated and had poor discrimination. We developed a new model in 1,020 incident HD patients predicting 1-year risk of hospitalization. We used age, cardiovascular disease, diabetes mellitus, malignancy, estimated glomerular filtration rate, and body mass index as predictors, selected based on literature and clinical expertise. The model discriminated poorly (C-statistic, 95% confidence interval [CI]; 0.55, 0.53-0.58), also in targeted external validations in 620 PD patients (0.55, 0.52-0.58) and 1,718 pre-dialysis patients (0.52, 0.52-0.54). Calibration was poor for all models. Using different modelling strategies, model performance did not improve. In the sensitivity analysis, discrimination improved but was still low (HD population: 0.58, 0.55-0.62; PD population: 0.59, 0.54-0.63; pre-dialysis: 0.55, 0.52-0.57). Conclusion Hospitalizations are important but difficult to predict for patients with advanced CKD. Current models are insufficient to predict hospitalizations. We offer two main suggestions to improve future predictions of hospitalizations. First, the outcome of hospitalization should be more specific (e.g., cardiovascular hospitalizations) to reduce outcome heterogeneity. Second, the selection of predictors should be extended with more patient-centered factors, such as patient-reported outcomes.

#2067 Prediction of gastrointestinal bleeding hospitalization risk in hemodialysis using machine learning techniques

Abstract Background and Aims INitiativeS on advancing Patients’ outcomes In REnal disease (INSPIRE) is an academia and industry collaboration set forth to identify critical investigations needed to advance the practice of medicine in nephrology. Gastrointestinal bleeding (GIB) is one of the most common types of bleeding events in the kidney dialysis population [1]. We aimed to develop a model to predict GIB hospitalization risk in a hemodialysis (HD) patient within 180 days. We evaluated advanced machine learning algorithm (XGBoost) and compared it to traditional machine learning algorithm (logistic regression) modeling techniques. Method We used data from a kidney care network from 2017 through 2020. We included data from adult dialysis patients (age ≥18 years). GIB related hospitalization was defined based on international classification of diseases (ICD) diagnosis codes recorded as the primary, secondary, or tertiary discharge reason for hospitalization. Two distinct models were created using XGBoost and logistic regression algorithms and using the same dataset. Both models were evaluated using metrics such as area under the receiver operating curve (AUROC), accuracy, sensitivity, and specificity. Missing data was imputed using mean for quantitative data and mode for qualitative data. The dataset then was randomly divided into 60% training, 20% validation and 20% test dataset. The test data, comprising unseen patients and data was used to evaluate the performance of the model. This means that the model had never encountered the test data during its learning phase, which included training the model and then validating the model. Results The incidence of 180-day GIB hospitalization was 1.12% in the HD population (n = 5 116 with GIB hospitalization/ n = 451 653 without GIB hospitalization), and consistent in the unseen test dataset (n = 465 with GIB hospitalization/ n = 38586 without GIB hospitalization). The XGBoost model showed higher predictive performance compared to logistic regression (Table). The AUROC was 0.72 (95% confidence interval (CI) 0.69, 0.74) for the XGBoost model. In comparison, the AUROC was 0.615 (95% CI 0.60, 0.64) for the logistic regression model. Specificity was 67% versus 58% for the XGBoost model versus logistic regression model respectively. However, sensitivity for both the models was 65%. The top predictors for major GIB were consistent in both models for many factors, yet some factors showed different importance on outcome prediction (Fig. 1). Higher risk of GIB hospitalization was associated with older age, lower ferritin levels, and recent all-cause hospitalizations in both models. The XGBoost showed high importance on outcome prediction for lower hemoglobin and higher serum 25 hydroxy (25OH) vitamin D values, and the logistic regression model showed high importance for higher amounts of saline delivered during an HD treatment and lower intact parathyroid hormone (iPTH) levels. Conclusion We found advanced machine learning prediction modeling (XGBoost) appears suitable for identifying a HD patient at risk for a GIB hospitalization in the next 180 days, and outperforms traditional machine learning (logistic regression) modeling techniques. Although both models showed identical sensitivity, the XGBoost model had higher specificity. Prospective testing is needed to confirm the model's performance. The association between bone mineral metabolism markers and GIB hospitalization risk is unexpected and warrants further investigation.

#892 Hemodialysis close to home: lessons from 9 years dialysis hubs

Abstract Background and Aims In the Netherlands, more than 18.000 patients are treated with renal replacement therapy (RRT) (end of 2022), of which 5.216 patients with hemodialysis (HD). Only 5% (255 patients) are treated with HD outside a dialysis center, at home. Home HD can be either performed by the patient alone, or with assistance from spouse or other informal caregiver or from a health care professional. When applicable, patients and there (dialysis)partners are trained to perform their treatment independently. Home HD can give more flexibility and independence which could improve quality of life. To make HD close to home available for patients even if their own house is not suitable and to deal with limited resources, such as nursing support, we have started in 2015 to set up small dialysis facilities in collaboration with other healthcare providers, such as nursing homes or general practitioners, and have called them Hubs. By setting up Hubs, we ensure that dialysis remains available close to home for patients who cannot carry out the treatment completely independently or with their own partner, but need help from a care worker. Method We provide an overview of the logistical and technical matters involved in setting up a successful Hub and will discuss the experience so far with 11 Dianet Hubs in the Netherlands. Results Preparatory phase: In the run-up to a new dialysis hub a standard check list is used. The Quality and Safety department supervises the process and translates control measures into cooperation agreements. For a hub to be successful, partnership with local dialysis centers and hospitals is key. Together the best location can be pinpointed. Annual audits are carried out. Patient characteristics and medical back up: Because a Hub is a home HD setting, good stable vascular access is mandatory, together with hemodynamic stability during sessions. However, because the HD population has multiple comorbidities, in case of instability, acute consultation with the referral center is necessary. Protocols guide the health care professional at the Hub in those circumstances. Furthermore, every hub presents new challenges that are embedded in the work processes. Current situation: Eleven dialysis hubs, each accommodating up to 8 patients are in operation in collaboration with a partnering dialysis center. Since 2015, processes have been refined, risks identified and hubs almost continuously occupied. Due to patient factors, sometimes incenter dialysis is necessary, for example in case of access problems. We learned during the pandemic, that accessibility of the dialysis hubs located in (primary) care homes is an important factor to consider when choosing a Hub location. Both patients and partnering dialysis centers feel that our Hubs meet a need for close to home assisted HD. Conclusion With 11 dialysis hubs we have built up a wealth of experience. From employer perspective, deployment of dialysis nurses in hubs is profitable. HD close to home offers patients an alternative for home HD. There is great support within healthcare facilities; they experience the impact of dialysis on patients’ lives. Annual audits guarantee quality and safety.

#2470 Isolated diastolic hypertension in hemodialysis patients

Abstract Background and Aims Isolated diastolic hypertension (IDH) is relatively rare in the general population (prevalence about 6.5%, Front Cardiovasc Med. 2021;8:810105) and of uncertain etiologic implications. The prevalence of this alteration in the hemodialysis (HD) population and the etiologic implications of IDH have not been studied. Method We investigated the prevalence and the risk for all-cause death and cardiovascular events of isolated diastolic hypertension in a population of 534 HD patients enrolled in 7 centres in 3 European countries (the EURECA-m working group). BP was measured by a golden standard method in this population (48 h ABPM) using well-validated instruments (AAMI/ESH/ISO). Along with recommendations by the European Society of Hypertension guidelines, recordings were made at 15-minute intervals during the day and 30 minutes at night. According to a consensus document by the EURECAm working group (Nephrol Dial Transplant. 2019;34:1542-1548), isolated diastolic hypertension was defined as 48 h diastolic BP &amp;gt; 80 mmHg with a systolic BP &amp;lt;130. Results Among 534 HD patients, thirty-one (5.8%) had isolated diastolic HTN, which is very close to the prevalence estimated in the general population (6.5%). During a mean follow-up of 2.6 years, 138 patients died, 188 had at least one CV event, and 249 experienced the combined event. In univariate analyses, diastolic hypertension was associated with death (HR 0.21; 95% CI: 0.05-0.86, P = 0.03) and the combined outcome of death and CV events (HR 0.42, 95% CI: 0.20-0.88, P = 0.02). However, these associations were no longer significant after adjustment for age and sex (all P &amp;gt; 0.45). Additional sensitivity analyses substantially confirmed these findings. No effect modification by age or gender was observed in these associations. Conclusion In a large cohort of HD patients studied by state-of-the-art 48-hour BP monitoring, the prevalence of isolated diastolic hypertension in HD patients (5.8%) is close to that registered in the general population (6.5%). This alteration predicts the risk for clinical outcomes in unadjusted analyses but not after adjustment for age and sex. Isolated diastolic hypertension is as rare as it is in the general population and unlikely to contribute to the high risk of adverse clinical outcomes in the dialysis population.

#1384 Obstructive sleep apnea and fluid management in the hemodialysis: are clinicians putting existing scientific evidence into practice?

Abstract Background and Aims Obstructive Sleep Apnea (OSA) prevalence was found to be higher in an intermittent hemodialysis (iHD) population of the French-speaking Switzerland than in the general population of the same geographical area. The high prevalence of OSA in iHD patients was linked to fluid overload. Interdialytic fluid management targeting a low dry weight showed to reduce OSA's severity, offering a novel therapeutic option in this population. The aims of this study were: 1) to assess the prevalence of confirmed OSA in an end-stage renal disease (ESRD) population undergoing iHD in the Italian-speaking area of Switzerland; 2) to assess the awareness of the treating nephrologists about the OSA's diagnosis of their patients; and 3) to evaluate whether nephrologists integrate the current knowledge in their fluid management strategy of iHD patients with OSA. Method We performed a multicenter, cross-sectional study between July 2022 and July 2023, screening all patients attending the four iHD units of the Ente Ospedaliero Cantonale (EOC), the public hospital network of Canton Ticino (Switzerland), and including those with confirmed OSA. For these patients, we collected anthropometric parameters, iHD protocol characteristics and fluid status from electronic patient dossiers. Fluid overload was measured by multifrequency bioelectrical impedance (Body Composition Monitor®, BCM®; Fresenius Medical Care). Nephrologists of the four iHD units were asked to identify patients with known OSA diagnosis, without consulting the medical dossiers, and were interviewed about their specific iHD strategy in patients with OSA. We compared the fluid management of patients identified as “OSA positive” and those with OSA diagnosis, but misclassified as “OSA negative” by the treating nephrologist. Results Out of 193 patients treated by iHD in the four HD units, 45 patients had a confirmed diagnosis of OSA and were included in this study. Mean age was 76.0 ± 7.5 y, 82.2% were men, mean BMI was 29.1 ± 4.5 kg/m2. They were treated 3 times per week, in mean for 3.76 ± 0.37 hours; 22 were treated by hemodialysis and 23 by hemodiafiltration. The 23% prevalence of diagnosed OSA observed in the study population was significantly lower than the 56% prevalence found in the French Switzerland iHD cohort (p &amp;lt; 0.001). Out of the 45 patients with diagnosed OSA, only 27 (60.0%) were correctly identified by the treating nephrologist. We found no difference in the dry weight target defined by the treating nephrologist between the patients correctly identified as “OSA positive” and those misclassified as “OSA negative”, both being close to the dry weight calculated with BCM® (mean difference 0.7 ± 2.3 vs 0.3 ± 1.4 kg; p = 0.564). The pre-dialysis fluid overload of the whole population was 2.4 ± 1.7 kg, being higher in the “OSA positive” group (2.9 ± 2.0 kg) than in the “OSA negative” group (1.8 ± 0.7 kg; p = 0.0172). However, we found no difference in post-dialysis achievement of dry weight between the groups (residual overweight 0.2 ± 1.0 and 0.1 ± 0.7 kg respectively; p = 0.672). In the interviews, only half of the nephrologists (4/8) were aware of the correlation between fluid overload and OSA, and two of them reported to integrate this knowledge in their fluid management strategy, even if not systematically. Conclusion Nephrologists' attention to sleep apnea in iHD patients seems to be low, despite its high prevalence in this population and scientific evidence supporting that targeting fluid overload is an effective option to improve OSA severity. However, nephrologists tried to achieve the dry weight very strictly in all patients, independently of OSA. More intensive training of nephrologists to the clinical and diagnostic peculiarities of OSA in patients with ESKD is required, in order to improve OSA diagnosis and targeted therapeutic care.

#336 Early detection of vascular access complication in pediatric CKD hemodialysis patients

Abstract Background and Aims Vascular access (VA) dysfunction is a major clinical complication in the hemodialysis (HD) population and has a direct effect on dialysis outcome. Where in patients undergoing HD, neointimal hyperplasia results in vascular stenosis and subsequent thrombosis, which lead to vascular access failure. Interleukin-10 (IL-10), hs CRP, and tumor necrosis factor- (TNF–α) are inflammatory markers which are proved to be involved in the process neointimal hyperplasia &amp; subsequent complications. We aimed to evaluate the value of these inflammatory markers in early detection of arterio-venous fistula (AVF) failure, in pediatric patients on regular HD, together with the clinical &amp; radiological diagnoses, for proper intervention for correcting AVF failure according to guidelines. Method It was an prospective cohort study, where we included 59 pediatric chronic kidney disease patients’ stage 5, on regular HD (CKD5d), who were following up regularly with us for 1 year. Patients with nonfunctioning AVFs due to complications as thrombus occluding fistula, infections, or large aneurysm, were excluded from the beginning of the study. Monitoring of VA by pre-cannulation physical examination was performed to detect the physical signs of dysfunction, at least once per month according to KDOQI guidelines. Venous pressure (VP) and trans-membrane pressure (TMP) were recorded at each HD session, to detect any abnormalities and to record the pressure changes with the appearance of any AVF complications. Doppler ultrasound (DUS) was performed routinely every 3 months for all HD patients, &amp; at the time of development AVF complications. Blood samples for measuring IL10, TNF, hs CRP were collected from the patients every 2 weeks for 12 months, prior to HD sessions, where we included the most recent samples prior to, and at the time of development complications, which were clinically and radiologically confirmed. Results The mean (±SD) age of our patients was 13.97 (±2.65) years, where we had 39 (56.5%) males, and 30 (43.5%) females, with a median (IQR) HD duration of 4.0 (5.0) years. On assessing the AVF complications, 33 patients (55.9%) had experienced complications, where thrombosis was the most common reported one (17 patients, 28.8%), followed by aneurysm formation in 14 patients (23.7%), while 6 patients (10%) had stenosed AVF, early VA failure in 2 patients (3.4%),1 patient (1.7%) had pseudoaneurysm, and another patient (1.7%) had perivascular hematoma. Mean serum levels of Hs-CRP, TNF-α, IL-10 were elevated at timing of early development of complication, in comparison to their baseline levels, meanwhile serum IL10, and TNF alpha were significantly higher among patients who developed AVF stenosis. The baseline hs-CRP had related positively with AVF diameter at the time of the development of the complications. Conclusion The inflammatory markers have no clinically applied role in early detection of AVF abnormalities, therefore, physical examination and DUS are deemed sufficient.

#445 Octogenarians in hemodialysis: not everyone ages the same

Abstract Background and Aims The average life expectancy in Western Europe in 2022 was 79 years for males and 84 years for females. Over the past years, the number of ageing patients in hemodialysis (HD) has increased significantly in all countries. There are several considerations for dialysis maintenance in older patients with an increased burden of age-related problems. It is important to identify the specific characteristics of this elderly population to develop appropriate care strategies and to estimate future needs for renal support services. The aim was to describe the clinical characteristics of our prevalent dialysis population by age, with particular reference to octogenarians. Method In this retrospective observational study between 2008 and 2022, the medical charts of 253 prevalent HD patients in a referral university hospital were analyzed. Patients were divided into four groups according to their age (four age groups: ≤60, 61–70, 71–80, 80&amp;gt; years old). Patient demographics and comorbidities were compared between age groups. Percentage and type of AVF (arteriovenous fistula), primary failures, and AVF reparations were recorded at onset of HD and at follow-up. Interdialytic weigh gain (average for pre-dialysis measurement of three hemodialysis sessions per week for two weeks) was also informed. Results 57 patients older than 81 years underwent dialysis (mean age 85.2±3.3 years), 31 (54%) older than 84 years (11 women, 20 men). 48% of patients over the age of 80 had to start HD urgently. In octogenarian, length on time in hemodialysis was 4.3 ± 6.46 years, they were no differences based on age; vintage HD in women and men older than 84 years was 5.2 ± 3.4 years and 5.3 ± 3.7 years, respectively. Females accounted for 40.4%, higher than other age groups, but not statistically significant. None of the over-80s self-reported active smoking, with differences between groups (p=0,000). Diabetes was less common in the youngest and oldest patients (23.9%vs 49.1%vs 56.8% vs 43.9% for patients ≤60, 61-70, 71-80 and 80&amp;gt;age, respectively, p=0.001). Similar trends were observed for ischemic heart disease and peripheral vascular disease (3%vs23,6%vs25,7% vs15,7% and 9%vs27,3%vs24,3vs15,8%; p=0,000 and p=0,035, respectively). Among octogenarians, 36.8% had experienced heart failure, the highest proportion of any age group (14,9%vs23,6%vs25,7% vs36,8% for patients ≤60, 61-70, 71-80 and 80&amp;gt;age, p=0.046), but interdialytic gains were not greater in this group than in others (70%vs53,7%vs45,9%vs47,4% for patients ≤60, 61-70, 71-80 and 80&amp;gt;age, p=0.019). Chronic lung disease occurred significantly less often in octogenarians compared to younger patients (17,9%vs30,9%vs36,5%vs15,2% in the ≤60, 61-70, 71-80 and 80&amp;gt; age groups respectively (p=0.026). Cerebrovascular disease and active neoplasm did not differ based on age. Underlying chronic kidney disease was unknown in 28.1% of patients over 80 years of age. Diabetic nephropathy was the main known condition of nephropathy (16,4%vs27,3%vs23%vs15,8% for patients ≤60, 61-70, 71-80 and 80&amp;gt;age, p=0.052). (see Table 1). In all age groups, AVF was the most common VA in incident and in prevalent HD patients (52,2%vs61,8%vs62,2%vs 77,2% and 82,1%vs80%vs86,5%vs 78,9 % in the ≤60, 61-70, 71-80 and 80&amp;gt; age groups respectively (p=0,086, p=0,640)). Radio-cephalic made up the largest proportion of AVFs in the total population. VA primary failure is more common in octogenarians (1,5%vs16,4%vs10,8%vs19,3% for patients ≤60, 61-70, 71-80 and 80&amp;gt; age, p=0.010). There were no differences in the need for AVF repair between groups (see Table 2). Conclusion Our octogenarian HD population is a selected population with a lower-than-expected comorbidity burden compared to younger age groups. Very elderly patients are more compliant and survive more than five years after HD initiation. Although vascular access may be more difficult to achieve initially, the outcome is comparable to that of younger patients in our study. Strategies to best manage this vulnerable hemodialysis population remain to be explored.

#3075 Itch and ache on dialysis: the impact on the quality of life in Tunisian hemodialysis patients

Abstract Background and Aims Pruritus associated with chronic kidney disease (CKD-aP) is a common condition. However, although pruritus seriously affects the physical and mental health of patients, clinicians tend to underestimate this condition. The aim of this study is to determine the prevalence of Pa-MRC and its impact on the quality of life (QOL) in the Tunisian hemodialysis population. Method This is a prospective observational study conducted at the hemodialysis unit at Taher Sfar University Hospital (February - May 2022). The presence and severity of Pa-MRC were documented based on a 5-D pruritus questionnaire, demographic and clinico-biological data were extracted from the dialysis unit register. Results A total of 63 chronic hemodialysis patients (CHD) were recruited in this study, 68.3% of whom presented pruritus, most often generalized (2/3 cases). More than half of the patients reported pruritus throughout the day. Thirty patients reported moderate to severe Pa-CKD. Patients were not treated for pruritus in 72% of cases, but 92% of those treated were satisfied with their treatment. Patients reporting a more severe Pa-CKD had a shorter duration of dialysis, higher phosphate and parathormone levels and more frequent diabetes (p = 0.03). The QOL deteriorated when the severity of pruritus increased (p = 0.01). Patients reported having difficulty falling asleep (52%), irritability (61.4%), concentration disorders (31.8%). In addition, pruritus limits the daily life of patients in 59.1% of cases with an impact on nutritional status in 13.6%. Conclusion This study reveals that Pa-CKD is a distressing condition with a huge impact on the QOL of CHD patients. However, there is an unmet need for treatment, hence the need for new therapeutic approaches.

#1098 Utilizing a novel individual patient performance score for mortality prediction in hemodialysis patients

Abstract Background and Aims Chronic kidney disease (CKD) management has advanced significantly over recent years, yet mortality rates among hemodialysis (HD) patients remain high. This study aims to assess the impact of changes in an individualized patient performance score (IPPS) on survival in a large cohort of HD patients. Method This multicentre, observational, prospective cohort study included HD patients across 19 countries, starting from January 1st, 2022. We followed these patients for one year, categorizing them based on their survival status at year's end. We calculated the mean IPPS monthly, from June to December 2021, and divided the patients into tertiles based on their December 2021 scores. Subsequently, we formed five groups based on their IPPS dynamics over six months. We used Kaplan-Meier and Cox regression analyses to evaluate survival, presenting the results as hazard ratios (HR) and 95% confidence intervals (95% CI). A p-value below 0.05 was considered statistically significant. The IPPS, ranging from 0 to 100, was calculated based on eight key performance areas, with higher scores indicating better medical performance: Vascular Access (20 points): type of vascular access, episodes of thrombosis and infection); HD adequacy (20 points): eKt/V, blood flow and treatment time; Anemia (20 points): hemoglobin concentration, transferrin saturation, ferritin; Arterial hypertension (10 points): mean arterial pressure; Mineral Bone disease (15 points): iPTH, plasmatic phosphorus and calcium; Fluid status (5 points): percentage of interdialytic fluid gain; Nutrition (5 points): plasmatic albumin and phosphorus and Others (5 points): influenzae vaccination, transplantation status and hospitalization events. The final score is obtained by the weighted sum of positive clinical factors, subtracted from the weighted sum of negative clinical factors and is reported as a number ranging from 0 to 100, with a higher score representing a better medical performance. Results The study included 10 458 HD patients. In the survival analysis, a significant increase in mortality odds was observed in patients with a stable IPPS of 72 or lower, compared to those with a score consistently above 86. Patients transitioning from the lowest to the highest tertile in six months had a mortality HR similar to those consistently in the highest tertile. On the other hand, patients showing a decrease in IPPS score from the highest to the lowest tertile in the same period had a mortality HR similar to the intermediate group (Table 1). Conclusion A lower total IPPS score (&amp;lt;72) was associated with a threefold increase in mortality risk compared to scores above 86. An increase in IPPS from the lowest to the highest tertile correlated with a significant reduction in mortality, equating to the survival rates of patients consistently scoring high. These findings underscore the potential of IPPS as a valuable tool for guiding clinical interventions in the HD population to enhance patient survival.

#2560 Intradialytic serum phosphate variations are associated with low PTH levels

Abstract Background and Aims Secondary hyperparathyroidism (SHPT) is one of the most common complications in patients with chronic kidney disease (CKD). The prevalence of SHPT progressively increases with the severity of CKD, affecting approximately 50-70% of patients on hemodialysis (HD). SHPT significantly impacts morbidity, mortality, hospitalization risks, healthcare costs, and the quality of life of CKD patients, making its management of paramount importance in this population. A positive serum phosphorus (P) balance is one of the critical determinants of SHPT, and the management of P levels in HD patients revolves around nutritional therapy, medications, and optimal kidney replacement therapy (KRT). Numerous studies have explored the role of KRT in P control among prevalent HD patients. However, whether the reduction of P achieved during KRT affects parathyroid hormone (PTH) levels is still a matter of debate. Currently, a direct evaluation of the relationship between intra-HD P reduction and PTH levels is lacking. Method We conducted a retrospective observational study on the prevalent HD population at the Division of Nephrology, University Hospital of Verona, from January to December 2022. We Included clinically stable adult patients undergoing HD for over 6 months, with multiple recorded visits during the follow-up. Demographic, clinical, laboratory, and medication data were collected. Time-varying variables were updated at each study visit. Food intake was assessed through a 3-day food record (HD day, non-HD day, and weekend day). Patients were instructed to maintain their usual ideally low-phosphate diet and to meticulously record and weigh all foods and preparations consumed. This assessment was conducted every 4 months during the observation period. From the 3-day food records, the dietary phosphorus intake was calculated using a national database and reported as mg of P per day. A dietitian evaluated dietary questionnaires. The primary outcome of interest was PTH levels. The absolute intra-HD change in P (intra-HD ∆P), defined as the difference between pre- and post-HD P levels, served as the main exposure. Multivariable adjusted linear mixed models were used to investigate the relationship between intra-HD ∆P and PTH levels. Potential confounders were included in the following three models of hierarchical adjustment: (i) Model 1, unadjusted; (ii) Model 2, adjusted for age and sex; (iii) Model 3, adjusted for all covariates in Model 2 plus URR, HD modality, active vitamin D, phosphate binders, calcimimetics, comorbidities. Since the relationship between intra-HD ∆P and PTH is influenced by the pre-HD P, all models were adjusted for pre-HD P. The association between PTH and intra-HD ∆P was also assessed in a subgroup of patients who provided a valid dietary diary. A multivariable linear mixed model adjusted for all covariates in Model 3 plus dietary phosphorus intake was used to assess this association. Results A total of 211 patients contributed to 904 study visits. A significant and positive relationship was observed between intra-HD ∆P and pre-HD P (β = 0.76, 95% CI 0.75, 0.78, p &amp;lt; 0.001) and urea reduction ratio (β = 0.38, 95% CI 0.35, 0.41; p &amp;lt; 0.001) (Figure). An increase in intra-HD ∆P was significantly and independently associated with low PTH levels (β = −0.17, 95% CI −0.30, −0.03; p = 0.016; Table). Overall, 86 out of 211 patients provided valid dietary data. The relationship between intra-HD ∆P and PTH levels was confirmed after additional adjustment for dietary P intake (p = 0.041). Conclusion The extent of intra-HD P reduction significantly correlates with low PTH levels. Strategies focused on optimizing or enhancing depurative efficiency in KRT can exert a substantial impact on managing positive phosphorus balance and secondary hyperparathyroidism (SHPT). The assessment of intra-HD P reduction may play a pivotal role in the management and follow-up of SHPT in HD patients.