Ultrafiltration Profiling and Outcomes Among Individuals on Maintenance Hemodialysis

• Sponsor: University of North Carolina, Chapel Hill (Other)
• Overall Status: Completed
• CT.gov ID: NCT03301740
• Collaborator: National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (NIH)
• Enrollment: 34
• Locations: 2
• Arms: 2
• Actual Duration (Months): 9.2
• Patients Per Site: 17
• Patients Per Site Per Month: 1.8

Study Details

Study Description

Brief Summary

The rate of fluid removal (ultrafiltration, UF) during hemodialysis (HD) may contribute to cardiovascular morbidity and mortality among individuals receiving maintenance HD. More rapid UF rates are associated with higher morbidity and mortality. Ultrafiltration profiling, the practice of varying UF rates to maximize fluid removal during periods of greatest hydration and plasma oncotic pressures, is one treatment modification that may reduce UF-related harm without necessitating reduction in interdialytic fluid intake or longer HD treatments. To date, UF profiling has not been adequately studied independent of sodium profiling.

This study investigates the comparative effect of UF profiling versus non-profiled conventional HD on select cardiovascular and patient-reported outcomes. Participants will complete two phases of UF profiling and two phases of conventional HD and will act as their own controls.

Condition or Disease	Intervention/Treatment	Phase
End Stage Renal Disease	Other: UF profiling during HD; Other: Conventional HD	N/A

Study Design

• Study Type: Interventional
• Actual Enrollment: 34 participants
• Allocation: Randomized
• Intervention Model: Crossover Assignment
• Intervention Model Description: The study is a 4-phase crossover trial in which participants are successively alternated between conventional HD and conventional HD + linear UF profiling across 4 phases with intervening wash-out periods. The study compares conventional HD (the participant's standard HD prescription with no UF profiling) to conventional HD + linear UF profiling. Participants will be randomly allocated to conventional HD or conventional HD + linear UF profiling for phase 1. Participants will undergo 9 treatments during each phase for a total of 18 conventional HD treatments and 18 UF profiled treatments. Participants will undergo 3 wash-out HD treatments between phases.The study is a 4-phase crossover trial in which participants are successively alternated between conventional HD and conventional HD + linear UF profiling across 4 phases with intervening wash-out periods. The study compares conventional HD (the participant's standard HD prescription with no UF profiling) to conventional HD + linear UF profiling. Participants will be randomly allocated to conventional HD or conventional HD + linear UF profiling for phase 1. Participants will undergo 9 treatments during each phase for a total of 18 conventional HD treatments and 18 UF profiled treatments. Participants will undergo 3 wash-out HD treatments between phases.
• Masking: Triple (Participant, Investigator, Outcomes Assessor)
• Masking Description: The study will be double-blinded in that participants will not be informed of the treatment paradigm. Investigators, including those performing transthoracic echocardiography (TTE) interpretation, will be blinded to treatment paradigm. Ultrasonographers performing intra-HD TTEs will be blinded to treatment paradigm.
• Primary Purpose: Treatment
• Official Title: Ultrafiltration Profiling and Outcomes Among Individuals on Maintenance Hemodialysis
• Actual Study Start Date: Mar 12, 2018
• Actual Primary Completion Date: Dec 15, 2018
• Actual Study Completion Date: Dec 18, 2018

Arms and Interventions

Arm	Intervention/Treatment
Experimental: UF Profiling Phase First First treatment phase begins with linear UF profiling during HD. Participants randomized to starting with the experimental UF profiling phase will receive 9 HD treatments with UF profiling (1st experimental phase). Following a 3 conventional HD treatment wash-out period, participants will then cross over to 9 conventional HD treatments (1st control phase). Following a 2nd 3 conventional HD treatment wash-out period, participants will cross over to 9 HD treatments with UF profiling (2nd experimental phase). Following a 3rd conventional HD treatment wash-out period, participants will cross over to 9 conventional HD treatments (2nd control phase).	Other: UF profiling during HD; Experimental arm: Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; pre-programmed "profile 2" on a Fresenius 2008K machine, the machine used in all participating clinics). Ultrafiltration profiling will be performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.; Other: Conventional HD; Control arm: Conventional HD (routine care) is the participant's standard HD prescription without UF profiling.
Experimental: Conventional HD Phase First First treatment phase begins with conventional HD. Participants randomized to starting with the control conventional HD phase will receive 9 conventional HD treatments (1st control phase). Following a 3 conventional HD treatment wash-out period, participants will then cross over to 9 HD treatments with UF profiling (1st experimental phase). Following a 2nd 3 conventional HD treatment wash-out period, participants will cross over to 9 conventional HD treatments (2nd control phase). Following a 3rd conventional HD treatment wash-out period, participants will cross over to 9 HD treatments with UF profiling (2nd experimental phase).	Other: UF profiling during HD; Experimental arm: Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; pre-programmed "profile 2" on a Fresenius 2008K machine, the machine used in all participating clinics). Ultrafiltration profiling will be performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.; Other: Conventional HD; Control arm: Conventional HD (routine care) is the participant's standard HD prescription without UF profiling.

Outcome Measures

Primary Outcome Measures

1. Occurrence of Intradialytic Hypotension (Intradialytic Hypotension Defined as Nadir Systolic BP <90 mmHg) [Every study hemodialysis treatment, up to 36 treatments per participant over 15 weeks]

   Intradialytic blood pressure (BP) was measured with an upper extremity cuff in seated position at 15-minute intervals during each hemodialysis treatment per standard dialysis clinic protocols. Intradialytic hypotension was defined as the presence of a nadir systolic BP <90 mmHg. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

2. Pre- to Post-hemodialysis Treatment Change in Troponin T Level in ng/mL at Weeks 3, 7, 11, and 15, Using Mixed Model Analysis [Weeks 3, 7, 11, and 15]

   Troponin T blood samples were collected at 4 study visits. Specifically, at the 7th hemodialysis treatment in each respective study phase (i.e., at week 3, 7, 11, and 15 study visits). Pre- to post-hemodialysis troponin T change was calculated as: post-dialysis troponin T - pre-dialysis troponin T (ng/mL). A lower change value reflects less cardiac strain. Based on the pre-specified protocol, the reported values represent change in troponin T between pre- and post-hemodialysis using mixed model (repeated measures logistic regression) analysis that considered all specified time-points (i.e., weeks 3, 7, 11, and 15).

3. Occurrence of a ≥10% Troponin T Percentage Rise From Pre- to Post-hemodialysis Treatment [Weeks 3, 7, 11, and 15]

   Troponin T blood samples were collected before and after each participant's 7th hemodialysis treatment of each study phase (4 times during the study). Troponin T percentage change was calculated as [(Post-HD troponin T - pre-HD troponin T) / pre-HD troponin T] x100. Troponin T percentage rise was defined as a troponin T percentage change ≥10%. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

4. Change From Baseline in Percent Left Ventricular Global Longitudinal Strain (GLS) [Weeks 3 and 7]

   Left ventricular GLS was measured with transthoracic echocardiography at baseline and at 30 minutes before HD treatment end during the 7th treatment in the first phase of each arm. Left ventricular GLS change was calculated as peak intradialytic stress GLS - baseline GLS (%). A lower change value reflects lesser cardiac strain. Median differences were estimated using Wilcoxon (Mann-Whitney) tests.

Secondary Outcome Measures

1. Nadir Systolic Blood Pressure During Hemodialysis in mmHg [Every study hemodialysis treatment, up to 36 treatments per participant over 15 weeks]

   Intradialytic BP was measured with an upper extremity cuff in seated position at 15-minute intervals during each hemodialysis treatment per standard dialysis clinic protocols. Nadir systolic BP was defined as the lowest intradialytic systolic BP measurement during each hemodialysis treatment. Lower values reflect greater cardiac strain. Beta-coefficients were estimated using a mixed model (repeated measures linear regression model).

2. Occurrence of Failed Target Weight Achievement (Failed Target Weight Achievement Defined as a Difference in Prescribed Target Weight and Post-dialysis Weight That is >1 kg or <-1 kg) [Every study hemodialysis treatment, up to 36 treatments per participant over 15 weeks]

   The treating nephrologist prescribed the target weight per routine clinical care. Post-dialysis weight was measured after each hemodialysis treatment in the standing position, per dialysis clinic protocol. Failed target weight achievement was defined as a difference in prescribed target weight and post-dialysis weight that was >1 kg or <-1 kg. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

3. Occurrence of Patient-reported Clinically Important Cramping During Dialysis (Clinically Important Cramping Defined as Moderate, Severe, or Very Severe Cramping) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. cramping) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important cramping was defined as cramping ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

4. Occurrence of Patient-reported Clinically Important Nausea or Upset Stomach During Dialysis (Clinically Important Nausea or Upset Stomach Defined as Moderate, Severe, or Very Severe Nausea or Upset Stomach) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. nausea or upset stomach) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important nausea/upset stomach was defined as nausea/upset stomach ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

5. Occurrence of Patient-reported Clinically Important Vomiting or Throwing up During Dialysis (Clinically Important Vomiting or Throwing up Defined as Moderate, Severe, or Very Severe Vomiting or Throwing up) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. vomiting or throwing up) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important vomiting/throwing up was defined as vomiting/throwing up ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

6. Occurrence of Patient-reported Clinically Important Dizziness or Lightheadedness During Dialysis (Clinically Important Dizziness or Lightheadedness Defined as Moderate, Severe, or Very Severe Dizziness or Lightheadedness) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. dizziness or lightheadedness) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important dizziness/lightheadedness was defined as dizziness/lightheadedness ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

7. Occurrence of Patient-reported Clinically Important Racing Heart or Heart Palpitations During Dialysis (Clinically Important Racing Heart or Heart Palpitations Defined as Moderate, Severe, or Very Severe Racing Heart or Heart Palpitations) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. racing heart or heart palpitations) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important racing heart/heart palpitations was defined as racing heart/heart palpitations ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

8. Occurrence of Patient-reported Clinically Important Chest Pain During Dialysis (Clinically Important Chest Pain Defined as Moderate, Severe, or Very Severe Chest Pain) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. chest pain) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important chest pain was defined as chest pain ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

9. Occurrence of Patient-reported Clinically Important Shortness of Breath During Dialysis (Clinically Important Shortness of Breath Defined as Moderate, Severe, or Very Severe Shortness of Breath) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

   Participants' dialysis-related symptoms (e.g. shortness of breath) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important shortness of breath was defined as shortness of breath ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

10. Occurrence of Patient-reported Clinically Important Thirst or Dry Mouth During Dialysis (Clinically Important Thirst or Dry Mouth Defined as Moderate, Severe, or Very Severe Thirst or Dry Mouth) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

    Participants' dialysis-related symptoms (e.g. thirst or dry mouth) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important thirst/dry mouth was defined as thirst/dry mouth ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

11. Occurrence of Patient-reported Clinically Important Headache During Dialysis (Clinically Important Headache Defined as Moderate, Severe, or Very Severe Headache) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

    Participants' dialysis-related symptoms (e.g. headache) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important headache was defined as headache ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

12. Occurrence of Patient-reported Clinically Important Itching During Dialysis (Clinically Important Itching Defined as Moderate, Severe, or Very Severe Itching) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

    Participants' dialysis-related symptoms (e.g. itching) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important itching was defined as itching ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

13. Occurrence of Patient-reported Clinically Important Restless Legs During Dialysis (Clinically Important Restless Legs Defined as Moderate, Severe, or Very Severe Restless Legs) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

    Participants' dialysis-related symptoms (e.g. restless legs) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important restless legs was defined as restless legs ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

14. Occurrence of Patient-reported Clinically Important Tingling or Feeling of Pins and Needles During Dialysis (Clinically Important Tingling or Feeling of Pins and Needles Defined as Moderate, Severe, or Very Severe Tingling or Feeling of Pins and Needles) [Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15]

    Participants' dialysis-related symptoms (e.g. tingling or feeling of pins and needles during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important tingling/feeling of pins and needles was defined as tingling/feeling of pins and needles ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).

Eligibility Criteria

Criteria

• Ages Eligible for Study: 18 Years to 85 Years
• Sexes Eligible for Study: All
• Accepts Healthy Volunteers: No

Inclusion Criteria:

• UF rate >10 mL/h/kg in >30% of treatments in a 30-day screening period (require ≥6 outpatient HD treatments in this period)

• Age 18-85 years

• Ability to converse comfortably in English or Spanish

• Receipt of in-center maintenance HD at Carolina Dialysis clinics in Carrboro or Siler City, North Carolina

• ≥90 days on HD

• Free of bloodstream infection during screening period

• Willingness to undergo all study testing

• Evidence of a signed and dated informed consent document

Exclusion Criteria:

• Systolic BP unable to be measured by arm cuff

• 1 hospitalization during screening period

• Unstable angina per treating nephrologist

• End-stage cirrhosis per treating nephrologist

• New York Heart Association class IV heart failure per treating nephrologist

• Pregnant

• More than 4 times per week HD

• Incarcerated

• Anticipated kidney transplant within 6 months per treating nephrologist

• Non-adherence to HD prescription (>2 unexplained absences during screening period)

• Sodium profiling or UF profiling in standard HD prescription

• Decisionally challenged, unable to provide informed consent

Contacts and Locations

Locations

	Site	City	State	Country	Postal Code
1	Carolina Dialysis - Carrboro	Carrboro	North Carolina	United States	27510
2	Carolina Dialysis - Siler City	Siler City	North Carolina	United States	27344

Sponsors and Collaborators

• University of North Carolina, Chapel Hill
• National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)

Investigators

• Principal Investigator: Jennifer Flythe, MD, MPH, University of North Carolina, Chapel Hill

Study Documents (Full-Text)

• Study Protocol and Statistical Analysis Plan - Apr 4, 2018

More Information

Publications

None provided.

• Responsible Party: University of North Carolina, Chapel Hill
• ClinicalTrials.gov Identifier: NCT03301740
• Other Study ID Numbers: 17-1057; 1K23DK109401
• First Posted: Oct 4, 2017
• Last Update Posted: May 15, 2020
• Last Verified: Jan 1, 2019

• Individual Participant Data (IPD) Sharing Statement: No
• Plan to Share IPD: No
• Studies a U.S. FDA-regulated Drug Product: No
• Studies a U.S. FDA-regulated Device Product: No
• Keywords provided by University of North Carolina, Chapel Hill: Ultrafiltration profiling; Hemodialysis; Dialysis; Cardiovascular outcomes; Patient-reported outcomes; Crossover study design; End Stage Renal Disease; Ultrafiltration; Echocardiogram; Blood pressure; Hypotension; Symptoms
• Additional relevant MeSH terms: Kidney Diseases; Kidney Failure, Chronic

Study Results

Participant Flow

Arm/Group Title	UF Profiling Phase First	Conventional HD Phase First
Arm/Group Description	First treatment phase begins with linear UF profiling during HD. Participants randomized to starting with the experimental UF profiling phase will receive 9 HD treatments with UF profiling (1st experimental phase). Following a 3 conventional HD treatment wash-out period, participants will then cross over to 9 conventional HD treatments (1st control phase). Following a 2nd 3 conventional HD treatment wash-out period, participants will cross over to 9 HD treatments with UF profiling (2nd experimental phase). Following a 3rd conventional HD treatment wash-out period, participants will cross over to 9 conventional HD treatments (2nd control phase).	First treatment phase begins with conventional HD. Participants randomized to starting with the control conventional HD phase will receive 9 conventional HD treatments (1st control phase). Following a 3 conventional HD treatment wash-out period, participants will then cross over to 9 HD treatments with UF profiling (1st experimental phase). Following a 2nd 3 conventional HD treatment wash-out period, participants will cross over to 9 conventional HD treatments (2nd control phase). Following a 3rd conventional HD treatment wash-out period, participants will cross over to 9 HD treatments with UF profiling (2nd experimental phase).
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	18	16
COMPLETED	18	16
NOT COMPLETED	0	0
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	18	16
COMPLETED	18	16
NOT COMPLETED	0	0
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	18	16
COMPLETED	17	16
NOT COMPLETED	1	0
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	17	16
COMPLETED	17	16
NOT COMPLETED	0	0
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	17	16
COMPLETED	17	16
NOT COMPLETED	0	0
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	17	16
COMPLETED	17	16
NOT COMPLETED	0	0
Period Title: Phase 1 Intervention (3 Weeks)
STARTED	17	16
COMPLETED	16	16
NOT COMPLETED	1	0

Baseline Characteristics

Arm/Group Title	UF Profiling Phase First	Conventional HD Phase First	Total
Arm/Group Description	First treatment phase begins with linear UF profiling during HD. Participants randomized to starting with the experimental UF profiling phase will receive 9 HD treatments with UF profiling (1st experimental phase). Following a 3 conventional HD treatment wash-out period, participants will then cross over to 9 conventional HD treatments (1st control phase). Following a 2nd 3 conventional HD treatment wash-out period, participants will cross over to 9 HD treatments with UF profiling (2nd experimental phase). Following a 3rd conventional HD treatment wash-out period, participants will cross over to 9 conventional HD treatments (2nd control phase).	First treatment phase begins with conventional HD. Participants randomized to starting with the control conventional HD phase will receive 9 conventional HD treatments (1st control phase). Following a 3 conventional HD treatment wash-out period, participants will then cross over to 9 HD treatments with UF profiling (1st experimental phase). Following a 2nd 3 conventional HD treatment wash-out period, participants will cross over to 9 conventional HD treatments (2nd control phase). Following a 3rd conventional HD treatment wash-out period, participants will cross over to 9 HD treatments with UF profiling (2nd experimental phase).	Total of all reporting groups
Overall Participants	18	16	34
Age (Years) [Mean (Standard Deviation) ]
Mean (Standard Deviation) [Years]	55 (15)	57 (15)	56 (15)
Sex: Female, Male (Count of Participants)
Female	5 27.8%	3 18.8%	8 23.5%
Male	13 72.2%	13 81.3%	26 76.5%
Ethnicity (NIH/OMB) (Count of Participants)
Hispanic or Latino	7 38.9%	4 25%	11 32.4%
Not Hispanic or Latino	11 61.1%	12 75%	23 67.6%
Unknown or Not Reported	0 0%	0 0%	0 0%
Race (NIH/OMB) (Count of Participants)
American Indian or Alaska Native	0 0%	0 0%	0 0%
Asian	0 0%	0 0%	0 0%
Native Hawaiian or Other Pacific Islander	0 0%	0 0%	0 0%
Black or African American	12 66.7%	10 62.5%	22 64.7%
White	6 33.3%	6 37.5%	12 35.3%
More than one race	0 0%	0 0%	0 0%
Unknown or Not Reported	0 0%	0 0%	0 0%
Region of Enrollment (Count of Participants)
United States	18 100%	16 100%	34 100%

Outcome Measures

1. Primary Outcome

Title	Occurrence of Intradialytic Hypotension (Intradialytic Hypotension Defined as Nadir Systolic BP <90 mmHg)
Description	Intradialytic blood pressure (BP) was measured with an upper extremity cuff in seated position at 15-minute intervals during each hemodialysis treatment per standard dialysis clinic protocols. Intradialytic hypotension was defined as the presence of a nadir systolic BP <90 mmHg. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Every study hemodialysis treatment, up to 36 treatments per participant over 15 weeks

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	578	571

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in intradialytic hypotension between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.5
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.2
	Confidence Interval	(2-Sided) 95%; 0.8 to 1.7
	Parameter Dispersion	Type: Value:
	Estimation Comments

2. Primary Outcome

Title	Pre- to Post-hemodialysis Treatment Change in Troponin T Level in ng/mL at Weeks 3, 7, 11, and 15, Using Mixed Model Analysis
Description	Troponin T blood samples were collected at 4 study visits. Specifically, at the 7th hemodialysis treatment in each respective study phase (i.e., at week 3, 7, 11, and 15 study visits). Pre- to post-hemodialysis troponin T change was calculated as: post-dialysis troponin T - pre-dialysis troponin T (ng/mL). A lower change value reflects less cardiac strain. Based on the pre-specified protocol, the reported values represent change in troponin T between pre- and post-hemodialysis using mixed model (repeated measures logistic regression) analysis that considered all specified time-points (i.e., weeks 3, 7, 11, and 15).
Time Frame	Weeks 3, 7, 11, and 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Mean (95% Confidence Interval) [ng/mL]	12.3	11.3

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in troponin T change between UF profiling and conventional HD. Repeated measure linear regression was performed, giving each subject a random intercept term. The coefficient is the difference in the outcome between UF profiling and conventional HD.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.2
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Beta coefficient
	Estimated Value	-2.8
	Confidence Interval	(2-Sided) 95%; -6.9 to 1.4
	Parameter Dispersion	Type: Value:
	Estimation Comments

3. Primary Outcome

Title	Occurrence of a ≥10% Troponin T Percentage Rise From Pre- to Post-hemodialysis Treatment
Description	Troponin T blood samples were collected before and after each participant's 7th hemodialysis treatment of each study phase (4 times during the study). Troponin T percentage change was calculated as [(Post-HD troponin T - pre-HD troponin T) / pre-HD troponin T] x100. Troponin T percentage rise was defined as a troponin T percentage change ≥10%. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 3, 7, 11, and 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	64	63

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in troponin T rise between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.1
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	0.5
	Confidence Interval	(2-Sided) 95%; 0.2 to 1.3
	Parameter Dispersion	Type: Value:
	Estimation Comments

4. Primary Outcome

Title	Change From Baseline in Percent Left Ventricular Global Longitudinal Strain (GLS)
Description	Left ventricular GLS was measured with transthoracic echocardiography at baseline and at 30 minutes before HD treatment end during the 7th treatment in the first phase of each arm. Left ventricular GLS change was calculated as peak intradialytic stress GLS - baseline GLS (%). A lower change value reflects lesser cardiac strain. Median differences were estimated using Wilcoxon (Mann-Whitney) tests.
Time Frame	Weeks 3 and 7

Outcome Measure Data

Analysis Population Description

All participants who had baseline GLS measurement and had one GLS measurement in each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	29	27
Median (95% Confidence Interval) [GLS change (%)]	0.9	1.3

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in left ventricular GLS change between UF profiling and conventional HD. Wilcoxon (Mann-Whitney) tests were performed assessing the difference of the endpoint between the 2 treatment groups.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.2
	Comments	Statistically significant p-value is <0.05
	Method	Wilcoxon (Mann-Whitney)
	Comments
Method of Estimation	Estimation Parameter	Mean Difference (Final Values)
	Estimated Value	-0.8
	Confidence Interval	(2-Sided) 95%; -2.0 to 0.5
	Parameter Dispersion	Type: Value:
	Estimation Comments

5. Secondary Outcome

Title	Nadir Systolic Blood Pressure During Hemodialysis in mmHg
Description	Intradialytic BP was measured with an upper extremity cuff in seated position at 15-minute intervals during each hemodialysis treatment per standard dialysis clinic protocols. Nadir systolic BP was defined as the lowest intradialytic systolic BP measurement during each hemodialysis treatment. Lower values reflect greater cardiac strain. Beta-coefficients were estimated using a mixed model (repeated measures linear regression model).
Time Frame	Every study hemodialysis treatment, up to 36 treatments per participant over 15 weeks

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Mean (95% Confidence Interval) [mmHg]	104.9	102.1

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in nadir systolic BP between UF profiling and conventional HD. Repeated measure linear regression was performed, giving each subject a random intercept term. The coefficient is the difference in the outcome between UF profiling and conventional HD.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.9
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Beta coefficient
	Estimated Value	-0.2
	Confidence Interval	(2-Sided) 95%; -2.0 to 1.7
	Parameter Dispersion	Type: Value:
	Estimation Comments

6. Secondary Outcome

Title	Occurrence of Failed Target Weight Achievement (Failed Target Weight Achievement Defined as a Difference in Prescribed Target Weight and Post-dialysis Weight That is >1 kg or <-1 kg)
Description	The treating nephrologist prescribed the target weight per routine clinical care. Post-dialysis weight was measured after each hemodialysis treatment in the standing position, per dialysis clinic protocol. Failed target weight achievement was defined as a difference in prescribed target weight and post-dialysis weight that was >1 kg or <-1 kg. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Every study hemodialysis treatment, up to 36 treatments per participant over 15 weeks

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	578	571

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in failed target weight achievement between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.8
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.0
	Confidence Interval	(2-Sided) 95%; 0.7 to 1.3
	Parameter Dispersion	Type: Value:
	Estimation Comments

7. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Cramping During Dialysis (Clinically Important Cramping Defined as Moderate, Severe, or Very Severe Cramping)
Description	Participants' dialysis-related symptoms (e.g. cramping) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important cramping was defined as cramping ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important cramping between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.9
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.0
	Confidence Interval	(2-Sided) 95%; 0.4 to 2.1
	Parameter Dispersion	Type: Value:
	Estimation Comments

8. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Nausea or Upset Stomach During Dialysis (Clinically Important Nausea or Upset Stomach Defined as Moderate, Severe, or Very Severe Nausea or Upset Stomach)
Description	Participants' dialysis-related symptoms (e.g. nausea or upset stomach) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important nausea/upset stomach was defined as nausea/upset stomach ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important nausea/upset stomach between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.5
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	0.7
	Confidence Interval	(2-Sided) 95%; 0.2 to 2.2
	Parameter Dispersion	Type: Value:
	Estimation Comments

9. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Vomiting or Throwing up During Dialysis (Clinically Important Vomiting or Throwing up Defined as Moderate, Severe, or Very Severe Vomiting or Throwing up)
Description	Participants' dialysis-related symptoms (e.g. vomiting or throwing up) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important vomiting/throwing up was defined as vomiting/throwing up ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important Vomiting/throwing up score between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	1.0
	Comments
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.0
	Confidence Interval	(2-Sided) 95%; 0.1 to 16.0
	Parameter Dispersion	Type: Value:
	Estimation Comments

10. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Dizziness or Lightheadedness During Dialysis (Clinically Important Dizziness or Lightheadedness Defined as Moderate, Severe, or Very Severe Dizziness or Lightheadedness)
Description	Participants' dialysis-related symptoms (e.g. dizziness or lightheadedness) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important dizziness/lightheadedness was defined as dizziness/lightheadedness ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important dizziness/lightheadedness between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.04
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	0.2
	Confidence Interval	(2-Sided) 95%; 0.1 to 0.9
	Parameter Dispersion	Type: Value:
	Estimation Comments

11. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Racing Heart or Heart Palpitations During Dialysis (Clinically Important Racing Heart or Heart Palpitations Defined as Moderate, Severe, or Very Severe Racing Heart or Heart Palpitations)
Description	Participants' dialysis-related symptoms (e.g. racing heart or heart palpitations) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important racing heart/heart palpitations was defined as racing heart/heart palpitations ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important racing heart/heart palpitations between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.3
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	3.1
	Confidence Interval	(2-Sided) 95%; 0.3 to 32.4
	Parameter Dispersion	Type: Value:
	Estimation Comments

12. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Chest Pain During Dialysis (Clinically Important Chest Pain Defined as Moderate, Severe, or Very Severe Chest Pain)
Description	Participants' dialysis-related symptoms (e.g. chest pain) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important chest pain was defined as chest pain ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important chest pain between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	1.0
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.0
	Confidence Interval	(2-Sided) 95%; 0.2 to 6.0
	Parameter Dispersion	Type: Value:
	Estimation Comments

13. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Shortness of Breath During Dialysis (Clinically Important Shortness of Breath Defined as Moderate, Severe, or Very Severe Shortness of Breath)
Description	Participants' dialysis-related symptoms (e.g. shortness of breath) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important shortness of breath was defined as shortness of breath ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important shortness of breath between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.7
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.3
	Confidence Interval	(2-Sided) 95%; 0.3 to 6.5
	Parameter Dispersion	Type: Value:
	Estimation Comments

14. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Thirst or Dry Mouth During Dialysis (Clinically Important Thirst or Dry Mouth Defined as Moderate, Severe, or Very Severe Thirst or Dry Mouth)
Description	Participants' dialysis-related symptoms (e.g. thirst or dry mouth) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important thirst/dry mouth was defined as thirst/dry mouth ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important thirst/dry mouth between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.2
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	0.6
	Confidence Interval	(2-Sided) 95%; 0.3 to 1.2
	Parameter Dispersion	Type: Value:
	Estimation Comments

15. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Headache During Dialysis (Clinically Important Headache Defined as Moderate, Severe, or Very Severe Headache)
Description	Participants' dialysis-related symptoms (e.g. headache) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important headache was defined as headache ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important headache between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.8
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	0.9
	Confidence Interval	(2-Sided) 95%; 0.4 to 2.3
	Parameter Dispersion	Type: Value:
	Estimation Comments

16. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Itching During Dialysis (Clinically Important Itching Defined as Moderate, Severe, or Very Severe Itching)
Description	Participants' dialysis-related symptoms (e.g. itching) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important itching was defined as itching ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important itching between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.02
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	2.8
	Confidence Interval	(2-Sided) 95%; 1.2 to 6.6
	Parameter Dispersion	Type: Value:
	Estimation Comments

17. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Restless Legs During Dialysis (Clinically Important Restless Legs Defined as Moderate, Severe, or Very Severe Restless Legs)
Description	Participants' dialysis-related symptoms (e.g. restless legs) during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important restless legs was defined as restless legs ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important restless legs/difficulty keeping legs still between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.7
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	1.2
	Confidence Interval	(2-Sided) 95%; 0.5 to 2.6
	Parameter Dispersion	Type: Value:
	Estimation Comments

18. Secondary Outcome

Title	Occurrence of Patient-reported Clinically Important Tingling or Feeling of Pins and Needles During Dialysis (Clinically Important Tingling or Feeling of Pins and Needles Defined as Moderate, Severe, or Very Severe Tingling or Feeling of Pins and Needles)
Description	Participants' dialysis-related symptoms (e.g. tingling or feeling of pins and needles during the last week were assessed using an investigator-developed 12-question symptom questionnaire administered once weekly throughout the study (6 times per study arm). Each symptom was graded using a 5-point symptom severity Likert scale (response options: none, mild, moderate, severe, very severe). Clinically important tingling/feeling of pins and needles was defined as tingling/feeling of pins and needles ranked as moderate, severe or very severe. Odds ratios were estimated using a mixed model (repeated measures logistic regression model).
Time Frame	Weeks 1, 2, 3, 5, 6, 7, 9, 10, 11, 13, 14, 15

Outcome Measure Data

Analysis Population Description

All participants who completed at least one phase of each study arm.

Arm/Group Title	UF Profiling	Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.	Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
Measure Participants	33	33
Number [Number of occurrences]	245	242

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	UF Profiling, Conventional HD
	Comments	Null hypothesis = no difference in clinically important tingling/feeling of pins and needles between UF profiling and conventional HD. Likelihood ratio tests were performed using repeated measure logistic regression with the binary treatment indicator as a fixed effect predictor and assigning each subject a random intercept term. A likelihood ratio test assessed the significance of whether the coefficient corresponding to the binary treatment indicator equals to zero.
	Type of Statistical Test	Equivalence
	Comments	The equivalence margin is the standard error of the estimated differences between UF and standard treatment from the repeated measure mixed model.
Statistical Test of Hypothesis	p-Value	0.5
	Comments	Statistically significant p-value is <0.05
	Method	Mixed Models Analysis
	Comments
Method of Estimation	Estimation Parameter	Odds Ratio (OR)
	Estimated Value	0.6
	Confidence Interval	(2-Sided) 95%; 0.2 to 2.4
	Parameter Dispersion	Type: Value:
	Estimation Comments

Adverse Events

Time Frame	From the time informed consent is obtained to study participation end, up to 21 weeks.
Arm/Group Title	UF Profiling		Conventional HD
Arm/Group Description	Linear UF profiling (linearly decreasing UF rate with a UF rate starting at 1.33 times the rate that would be needed at a constant UF rate to achieve the desired post-weight; preprogrammed "profile 2" on a Fresenius 2008K machine). Ultrafiltration profiling performed using Fresenius 2008K dialysis machines in accordance with manufacturer instructions.		Conventional HD (routine care) is the participant's standard HD prescription without UF profiling
All Cause Mortality
	UF Profiling		Conventional HD
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	0/34 (0%)		0/34 (0%)
Serious Adverse Events
	UF Profiling		Conventional HD
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	0/34 (0%)		0/34 (0%)
Other (Not Including Serious) Adverse Events
	UF Profiling		Conventional HD
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	2/34 (5.9%)		5/34 (14.7%)
Blood and lymphatic system disorders
Epistaxis	1/34 (2.9%)	1	1/34 (2.9%)	1
Cardiac disorders
Hypervolemia	0/34 (0%)	0	1/34 (2.9%)	2
General disorders
Non-cardiac chest pain	0/34 (0%)	0	1/34 (2.9%)	1
Infections and infestations
Clostridium difficile diarrhea	1/34 (2.9%)	1	0/34 (0%)	0
Musculoskeletal and connective tissue disorders
Bone fracture	0/34 (0%)	0	1/34 (2.9%)	8
Surgical and medical procedures
Arteriovenous vascular access surgery	0/34 (0%)	0	1/34 (2.9%)	3

Limitations/Caveats

[Not Specified]

More Information

Certain Agreements

All Principal Investigators ARE employed by the organization sponsoring the study.

There is NOT an agreement between Principal Investigators and the Sponsor (or its agents) that restricts the PI's rights to discuss or publish trial results after the trial is completed.