Kidney failure has been recognized as one of the most costly chronic conditions among United States Veterans. Approximately 13,000 Veterans develop kidney failure each year, and most require hemodialysis initiation. Hemodialysis patients suffer significantly increased risk of death and hospitalizations, and excessive body fluid is a major cause of this. While empiric aggressive fluid removal during dialysis is one approach to limit fluid overload, this can cause dangerous decreases in blood pressure during dialysis that independently contribute to the high death rate. In this study, I aim to test a new strategy that prescribes fluid removal based on a patient's recent blood pressure patterns during dialysis. This clinical trial will compare my strategy to standard care and assess the outcomes of overall blood pressure change between dialysis treatments in addition to the number of times the blood pressure becomes dangerously low during dialysis. Another aim is to determine how differences in the structure and function of the heart influence blood pressure during dialysis.

United States Veterans have disproportionately higher risk for end stage renal disease (ESRD) compared to the general population. Veterans with ESRD on maintenance hemodialysis (HD) suffer from alarmingly high mortality rates and hospitalizations mainly related to cardiovascular disease. Extracellular volume (ECV) excess is a primary contributing factor to cardiovascular disease and the heightened mortality rate in HD patients. Extracellular volume excess remains difficult to identify in clinical practice, and the standard approach to fluid management in the clinical setting involves arbitrary trial and error attempts to remove fluid without invoking hemodynamic instability such as intradialytic hypotension. Bioimpedance spectroscopy (BIS) is a useful research tool for assessing ECV; however, it is not feasible in routine practice, and there is little data on how its use affects intermediate and hard clinical outcomes. There is an unmet need for an approach to guide ultrafiltration in clinical practice that addresses both reduction of ECV and other mortality outcomes as well as minimization of intradialytic hypotension. The long-term goal of this study is to develop a more precise, patient-specific fluid management approach to be tested in a large clinical trial aimed at reducing mortality in Veterans on HD. The overall objective of this project is to utilize the investigators' novel, patient-specific ultrafiltration algorithm as an intervention in a clinical trial using mortality risk factors as the primary outcomes. The central hypothesis is that prescribing ultrafiltration prospectively based on an individual patient's intradialytic blood pressure slopes (IBPS) from recent treatments is superior to standard care at reducing ambulatory blood pressure and ECV without increasing risk for intradialytic hypotension. Aim 1 will use an un-blinded, controlled randomized clinical trial to demonstrate the effects of an IBPS-based ultrafiltration prescription compared to standard clinical practice. Each month, updated ultrafiltration prescriptions for the IBPS group will be determined based on the most recent treatment data. The primary outcome will be change in mean systolic 44-hour ambulatory blood pressure after 4 months. Other outcomes will include 1) change in post-HD ECV/body weight using multifrequency bioimpedance spectroscopy, 2) change in post-HD total peripheral resistance index using a non-invasive cardiac output monitor, and 3) between-group comparison of the frequency of intradialytic hypotension and intradialytic symptoms. Aim 2 will involve a cross sectional analysis of baseline data of subjects from Aim 1 in addition to consecutive enrollment of additional hypertensive HD patients. In addition to Aim 1 measurements, subjects will undergo transthoracic echocardiograms on a non-HD day to obtain measurements of left ventricular ejection fraction as a metric of systolic function, mitral inflow and mitral annulus velocities as a metric of diastolic function, and left ventricular mass index. Mixed linear models will be used to determine how these metrics independently influence the association between ECV/body weight and IBPS. The strength of the association between IBPS and ECV/body weight will then be determined within each tertile of the distributions of systolic and diastolic dysfunction. Finally, there will be an assessment of how the echocardiogram based metrics influence the likelihood of intradialytic hypotension with prospective follow up while controlling for ECV/body weight. If successful, this study will provide nephrologists with an easily-implemented, individualized approach to fluid management in Veterans on HD that safely reduces ECV excess and related mortality risk factors. Because no approach currently exits, this could immediately change clinical practice of managing Veterans with ESRD. The long term impact will be the opportunity to utilize the data generated to design a large, multi-center trial directly evaluating this intervention's effect on mortality in Veterans with ESRD on HD. The observed reduction in blood pressure can be used to determine expected mortality, while the echocardiogram data can be used to determine phenotypes of patients that may need to be considered for inclusion/exclusion criteria.

Inclusion Criteria:

• Hemodialysis patient
• Pre-dialysis systolic blood pressure greater than 140 mmHg averaged over 2 weeks OR post-dialysis systolic blood pressure greater than 130 mmHg averaged over 2 weeks

Exclusion Criteria:

• HemoDialysis Vintage Less than 2 months
• Pregnancy
• Mean systolic blood pressure nadir <95 mmHg in 2 weeks screening
• Mean pre- or post-dialysis systolic blood pressure >180 mmHg
• Mean pre to post-HD decrease in blood pressure >60 mmHg
• Routine intradialytic clonidine use
• Routine intradialytic midodrine use
• Documented antihypertensive medication non-adherence
• Mean ultrafiltration rate >13 mL/kg/hr during 2 week screening
• For bioimpedance measurements only: amputation of a major extremity, presence of cardiac defibrillator or pacemaker, presence of a metallic implant (prosthetic joint)