Advertisement

Twice-Weekly Hemodialysis With Adjuvant Pharmacotherapy and Transition to Thrice-Weekly Hemodialysis: A Pilot Study

Published:December 18, 2021DOI:https://doi.org/10.1053/j.ajkd.2021.12.001

      Abstract

      Rationale & Objective

      Thrice-weekly hemodialysis (HD) is the most common treatment modality for kidney dysfunction requiring dialysis (KDRD) in the US. We conducted a pilot study to assess the feasibility and safety of incremental-start HD in patients with incident KDRD.

      Study Design

      Pilot study.

      Setting and Participants

      Adults with eGFR ≥5 mL/min/1.73 m2 and urine volume ≥500 mL/24 h beginning maintenance HD at 14 outpatient dialysis units.

      Intervention

      Randomized allocation (1:1 ratio) to twice-weekly HD and adjuvant pharmacologic therapy for 6 weeks followed by thrice-weekly HD (incremental HD group) or thrice-weekly HD (conventional HD group).

      Outcomes

      The primary outcome was feasibility. Secondary outcomes included changes in urine volume and solute clearance.

      Results

      Of 77 patients invited to participate, 51 consented to do so, representing 66% of eligible patients. We randomized 23 patients to the incremental HD group and 25 patients to the conventional HD group. Protocol-based loop diuretics, sodium bicarbonate and patiromer were prescribed to 100%, 39% and 17% of patients on twice-weekly HD, respectively. At a mean follow-up of 281.9 days, participant adherence was 96% to the HD schedule (22/23 and 24/25 in the incremental and conventional groups, respectively) and 100% in both groups to serial timed urine collection. The Incidence Rate Ratio for all-cause hospitalization was 0.31 (95% CI 0.08, 1.17); and 7 deaths were recorded (1 in the incremental and 6 in the conventional group). At week 24, the incremental HD group had 51.0 percent lower decline in urine volume (95% CI, -0.7, 102.8) and 57.9 percent lower decline in the averaged urea and creatinine clearances (95% CI, -22.6, 138.4).

      Limitations

      Small sample size, time-limited twice-weekly HD.

      Conclusions

      It is feasible to enroll patients with incident KDRD into a randomized study of incremental-start HD with adjuvant pharmacotherapy who adhere to the study protocol during follow-up. Larger multicenter clinical trials are indicated to determine the efficacy and safety of incremental HD with longer twice-weekly HD periods.

      Keywords

      Introduction

      Almost all Americans with kidney dysfunction requiring dialysis (KDRD) initiated on treatment with maintenance hemodialysis (HD) are prescribed standard dialytic therapy of fixed frequency (thrice-weekly HD) and dose (dialysis single-pool Kt/V urea [spKt/V] ≥ 1.2, corresponding to standard Kt/V urea [stdKt/V] ≥ 2.1)1.This standard HD therapy disregards individual levels of residual kidney function
      • Murea M.
      • Deira J.
      • Kalantar-Zadeh K.
      • Casino F.G.
      • Basile C.
      The spectrum of kidney dysfunction requiring chronic dialysis therapy: Implications for clinical practice and future clinical trials.
      ,
      • Murea M.
      Precision medicine approach to dialysis including incremental and decremental dialysis regimens.
      . The HD dose, validated in clinical trials that involved solely patients on HD for >2yearsand with no residual kidney function, was generalized as ‘optimal’ dialysis dose to all patients prescribed maintenance HD
      • Lowrie E.G.
      • Laird N.M.
      • Parker T.F.
      • Sargent J.A.
      Effect of the hemodialysis prescription of patient morbidity: report from the National Cooperative Dialysis Study.
      ,
      • Eknoyan G.
      • Beck G.J.
      • Cheung A.K.
      • et al.
      Effect of dialysis dose and membrane flux in maintenance hemodialysis.
      . However, more than 90% of incident dialysis patients have an estimated glomerular filtration rate (eGFR) ≥5 mL/min/1.73 m2 at dialysis initiation
      • Vilar E.
      • Wellsted D.
      • Chandna S.M.
      • Greenwood R.N.
      • Farrington K.
      Residual renal function improves outcome in incremental haemodialysis despite reduced dialysis dose.
      and those with substantial levels of residual kidney function could likely be treated safely and effectively with less frequent dialysis therapy until residual kidney function was diminished
      • Kalantar-Zadeh K.
      • Unruh M.
      • Zager P.G.
      • et al.
      Twice-weekly and incremental hemodialysis treatment for initiation of kidney replacement therapy.
      ,
      • Murea M.
      • Moossavi S.
      • Garneata L.
      • Kalantar-Zadeh K.
      Narrative Review of Incremental Hemodialysis.
      .
      In retrospective studies, incremental HD, i.e., twice-weekly HD at initiation of dialysis therapy, later switched to thrice-weekly HD according to changes in residual kidney function and/or other clinical indications, yielded adequate symptom control
      • Ghahremani-Ghajar M.
      • Rojas-Bautista V.
      • Lau W.L.
      • et al.
      Incremental Hemodialysis: The University of California Irvine Experience.
      ,
      • Bowline I.G.
      • Russell G.B.
      • Bagwell B.
      • Crossley B.
      • Fletcher A.J.
      • Murea M.
      Temporal trends in fluid management with incremental hemodialysis.
      . Compared to outright thrice-weekly HD, incremental HD conferred similar or better patient survival
      • Obi Y.
      • Rhee C.M.
      • Mathew A.T.
      • et al.
      Residual Kidney Function Decline and Mortality in Incident Hemodialysis Patients.
      ,
      • Obi Y.
      • Streja E.
      • Rhee C.M.
      • et al.
      Incremental Hemodialysis, Residual Kidney Function, and Mortality Risk in Incident Dialysis Patients: A Cohort Study.
      ; similar or better quality of life
      • Park J.I.
      • Park J.T.
      • Kim Y.L.
      • et al.
      Comparison of outcomes between the incremental and thrice-weekly initiation of hemodialysis: a propensity-matched study of a prospective cohort in Korea.
      ,
      • Dai L.
      • Lu C.
      • Liu J.
      • et al.
      Impact of twice- or three-times-weekly maintenance hemodialysis on patient outcomes: A multicenter randomized trial.
      ; and longer preservation of residual kidney function
      • Zhang M.
      • Wang M.
      • Li H.
      • et al.
      Association of initial twice-weekly hemodialysis treatment with preservation of residual kidney function in ESRD patients.
      . Since residual renal function might afford excretion of toxins not removed by dialysis
      • Marquez I.O.
      • Tambra S.
      • Luo F.Y.
      • et al.
      Contribution of residual function to removal of protein-bound solutes in hemodialysis.
      , reduce systemic inflammation
      • Marquez I.O.
      • Tambra S.
      • Luo F.Y.
      • et al.
      Contribution of residual function to removal of protein-bound solutes in hemodialysis.
      ,
      • Shafi T.
      • Jaar B.G.
      • Plantinga L.C.
      • et al.
      Association of residual urine output with mortality, quality of life, and inflammation in incident hemodialysis patients: the Choices for Healthy Outcomes in Caring for End-Stage Renal Disease (CHOICE) Study.
      , and prolong patient survival
      • Shafi T.
      • Jaar B.G.
      • Plantinga L.C.
      • et al.
      Association of residual urine output with mortality, quality of life, and inflammation in incident hemodialysis patients: the Choices for Healthy Outcomes in Caring for End-Stage Renal Disease (CHOICE) Study.
      • Kalantar-Zadeh K.
      • Casino F.G.
      Let us give twice-weekly hemodialysis a chance: revisiting the taboo.
      • Mathew A.T.
      • Fishbane S.
      • Obi Y.
      • Kalantar-Zadeh K.
      Preservation of residual kidney function in hemodialysis patients: reviving an old concept.
      • Obi Y.
      • Chou J.
      • Kalantar-Zadeh K.
      Introduction to the Critical Balance - Residual Kidney Function and Incremental Transition to Dialysis.
      , there is biologic plausibility for potential benefits of incremental HD. Notwithstanding these data, prospective studies with randomized HD frequency assignments are needed to test the emerging concept that incremental HD is effective and safe in defined clinical scenarios
      • Murea M.F.J.
      • Anjay R.
      • Emaad A.R.
      • Gupta N.
      • Kovach C.
      • Vachharajani T.J.
      • Kalantar-Zadeh K.
      • Casino F.G.
      • Basile C.
      Kidney dysfunction requiring dialysis is a heterogeneous syndrome; se should treat it like one.
      . Herein, we report the results of the TWOPLUS Pilot Study (NCT03740048) designed to assess the feasibility of individual randomization to incremental-start vs conventional HD
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      .

      Methods

       Study Design

      A detailed study protocol has been published elsewhere
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      . In brief, the TWOPLUS Pilot Study was a prospective, individually-randomized, unblinded, parallel-group controlled trial with the primary objective of assessing the feasibility of time-limited twice-weekly HD with protocol-based pharmacotherapy in patients with incident KDRD and residual kidney function. The research team considered that, on account of resource limitation (i.e., budget, personnel, and stakeholder negotiation) the model of time-limited twice-weekly HD permits evaluation of the achievability of key processes of care that pertain to treatment of patients with and implementation of incremental HD at outpatient dialysis units. The study design was reviewed by dialysis administration leadership before implementation and the protocol was approved by the Institutional Review Board of Wake Forest School of Medicine in North Carolina, USA. All participants provided written informed consent, and study conduct adhered to the Declaration of Helsinki. Preparatory processes prior to recruitment commencement consisted of: 1) establishing schedules of outpatient twice-weekly HD at the participating dialysis units; 2) training of research coordinators; 3) provision of instructional guidance to dialysis personnel regarding study-related procedures; 4) procurement, storing and dispensing of patiromer potassium binder from the sponsoring agency; and 5) establishing procedures required to deliver study biosamples to a centralized laboratory. These processes are detailed in the Item S1.

       Study Setting

      The study was conducted at 14 outpatient and 1 inpatient dialysis units affiliated with a large healthcare system in North Carolina, USA.

       Participant Recruitment and Randomization

      Eligibility criteria are listed in Table S1. Recruitment began on 14 June 2019, was paused between 13 March 2020 and 31 May 2020 due to COVID-19 pandemic, resumed on 01 June 2020 and ended on 10 December 2020with the last randomization (Figure S1 and Item S1)
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      ,
      • Levey A.S.
      • Stevens L.A.
      • Schmid C.H.
      • et al.
      A new equation to estimate glomerular filtration rate.
      . Randomization was determined by a computer algorithm, in random blocks of 2 or4 size and 1:1 allocation, stratified by type of vascular access used at HD initiation (catheter or arteriovenous access).

       Intervention and Control

      The incremental HD group received the experimental intervention, which consisted of twice-weekly HD for 6 weeks coupled with adjuvant pharmacologic therapy (loop diuretic, sodium bicarbonate, potassium binder patiromer required according to the protocol) followed by transition to thrice-weekly HD at week 7. The conventional HD group served as the control group and continued to receive thrice-weekly HD (Table S2). Dialysis prescription (e.g., duration, blood flow rate, dialysate flow rate) was adjusted by the treating team in both treatment groups to achieve dialysis single-pool Kt/Vurea (spKdt/Vurea) of ≥1.2 and urea reduction ration (URR) of ≥65%
      KDOQI Clinical Practice Guideline for Hemodialysis Adequacy
      2015 update.
      .

       Study Visits and Data Collection

      All study-specific assessments occurred during the patients’ regularly scheduled hemodialysis session. Baseline timed 24-hour urine collections were obtained before randomization. Follow up inter-dialytic urine collections were obtained during the longest inter-dialytic interval in week 6, week 12 and week 24, with a timeframe window of 2 weeks for each assessment time point. Blood samples to calculate residual renal clearances were collected at the end of the HD treatment preceding timed urine collection and the beginning of the successive HD treatment ending the period of timed urine collection
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      . Renal stdKt/V, dialysis stdKt/V and total stdKt/V were calculated at each assessment time point
      • Daugirdas J.T.
      • Depner T.A.
      • Greene T.
      • Levin N.W.
      • Chertow G.M.
      • Rocco M.V.
      Standard Kt/Vurea: a method of calculation that includes effects of fluid removal and residual kidney clearance.
      . All tests were performed at one LabCorp laboratory in North Carolina.

       Outcomes

      The primary outcome was feasibility, assessed as (1) ≥70% of eligible patients are recruited, (2) ≥95% of participants randomized in the intervention group will adhere to the HD regimen, (3) ≥80% patients adhere to study-specific timed urine collection and (4) ≤5% of participants randomized in the control group will cross over to a regimen of less frequent HD. Feasibility metrics were selected based on a consensus opinion among investigators regarding medically acceptable rates of adherence to the tested intervention
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      . Adherence to protocol operationalization at outpatient dialysis facilities was monitored. Serious adverse events related to hospitalization and death; and events of additional outpatient HD treatments for volume overload or metabolic imbalances were recorded (Item S1). Secondary outcomes included changes in residual kidney function parameters (urine volume per 24 hours, renal urea clearance, and renal creatinine clearance) from baseline to weeks 6, 12 and 24; dialysis and total (dialysis + renal) urea solute clearance; and volume management parameters.

       Sample size

      This pilot study was designed to evaluate study feasibility (rate of enrollment and adherence) of a future multicenter controlled trial of incremental HD; and to get an estimate of the standard deviation on outcomes related to clinical events and changes of residual kidney function parameters. A total sample of 50 would be able to estimate feasibility with a 95% confidence interval width at most ±14.5%. Within each group, feasibility could be estimated with a confidence interval width no greater than ±20.5%
      • Lewis M.
      • Bromley K.
      • Sutton C.J.
      • McCray G.
      • Myers H.L.
      • Lancaster G.A.
      Determining sample size for progression criteria for pragmatic pilot RCTs: the hypothesis test strikes back.
      .

       Statistical Analyses

      Continuous variables are summarized with means (standard deviation [SD]), medians (1st quartile, 3rd quartile), or mean (95% confidence interval [CI]) and categorical variables are given as proportion per participant or per visit as appropriate. Standardized differences for baseline characteristics were calculated
      • Austin P.C.
      Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.
      . Descriptive statistics were used to report feasibility outcomes. The results are reported based on intention-to-treat analysis. Participants were censored at death, HD withdrawal, or conversion to peritoneal dialysis. Participants who withdrew consent, were withdrawn from the study because of non-compliance with HD treatments, or were lost to follow-up, were not included in data analysis.
      To analyze the relationship between cumulative all-cause hospitalization and treatment assignment, a Cox proportional hazards model utilizing the counting process for recurrent events was used; this model produces estimated of the hazard distribution for the two treatment groups
      • Andersen P.K.
      • Gill R.D.
      Cox's Regression Model for Counting Processes: A Large Sample Study.
      . All-cause hospitalization rates (number of days total, taking into account time on study) were analyzed using a negative binomial regression model with an offset of time at risk and Incidence Rate Ratio (IRR) were reported. Cox proportional hazards regression was used to assess the association between group assignment and overall survival. Fisher’s Exact Test was used to estimate the 95% CI around the proportion estimates

      Rothman KJ, Boice JD. Epidemiologic Analysis with a Programmable Calculator. National Institutes of Health. Pub No 79-1649. 1979;31-32.

      .
      Changes in least squares mean level of biochemical parameters, tested at successive time points, were analyzed across individuals with data available at both analysis time points using pairwise comparisons within a repeated measures mixed effect regression model. Estimate percent change from baseline to week 6, week 12 and week 24 in residual kidney function parameters and stdKt/V were analyzed using mixed effect linear models. Each set of data was viewed as a separate experiment; as such, for consistency, analyses were not adjusted for multiple comparisons or co-variates. Statistical analyses were performed using SAS, version 9.4, Cary, NC, USA.

      Results

       Primary Outcome: Feasibility

       Recruitment

      Over a 15.5-month period, 185 adult patients diagnosed with KDRD were started on thrice-weekly maintenance HD. After prescreening, 77 (41%) patients were approached for study participation. Of those, 51 (66%) consented to study participation, resulting in 48 patients randomized to treatment with twice-weekly HD and adjuvant pharmacotherapy for 6 weeks followed by thrice-weekly HD (incremental HD group, n=23) or continued usual care treatment with thrice-weekly HD (conventional HD group, n=25) (Figure 1 and Figure S1). Participant characteristics at the time of enrollment are summarized in Table 1. Patients randomized to conventional HD had higher urine volume and eGFR and higher prevalence of diabetes, congestive heart failure, cerebrovascular disease and malignancy.
      Figure thumbnail gr1
      Figure 1Study Flow Diagram. Eligibility, best case scenario if patients who refused to participate (n=26) were to be eligible: (B-D)/A; worst case scenario if patients who declined study participation were to not be eligible: (B-D-26)/A. Consent rate, C:B; Number randomized, E:A; Drop out, (Fi+Fc)/E.
      Table 1Participant sociodemographic and clinical characteristics, overall and according to treatment allocation
      VariableOverall (n=48)Incremental HD (n=23)Conventional HD (n=25)Standardized Difference#
      Sociodemographic characteristics at enrollment
      Age at enrollment (years), mean (SD)61.3 (14.0)59.1 (15.0)63.3 (12.8)0.31
      Female, n (%)21 (44%)10 (44%)11 (44%)0.02
      Race, n (%)
      White19 (40%)11 (48%)8 (32%)0.08
      Black27 (56%)11 (48%)16 (64%)0.09
      Hispanic2 (4%)1 (4%)1 (4%)0.00
      Body-mass index, mean (SD)32.1 (8.8)32.5 (9.4)31.8 (8.4)0.08
      Body surface area (m2), mean (SD)2.06 (0.3)2.1 (0.2)2.0 (0.3)0.34
      Clinical characteristics at enrollment
      Received HD* prior to randomization, n (%)35 (73%)17 (74%)18 (72%)0.01
      HD before randomization (n), mean (SD)3.0 (1.7)3.0 (1.7)3.0 (1.9)0.00
      KDRD etiology, n (%)
      Diabetes mellitus21 (44%)8 (35%)13 (52%)0.12
      Glomerulonephritis/Vasculitis1 (2%)1 (4%)0 (0%)0.52
      Other26 (54%)14 (61%)12 (48%)0.04
      Urine volume (mL/24 h)1215 (820)914 (522)1424 (955)0.66
      Vascular Access, n (%)
      Arteriovenous fistula10 (21%)5 (22%)5 (20%)0.00
      Central venous catheter38 (79%)18 (78%)20 (80%)0.03
      Comorbidities, n (%)
      Diabetes mellitus31 (65%)12 (52%)19 (76%)0.11
      Coronary artery disease8 (17%)4 (17%)4 (16%)0.00
      Congestive heart failure12 (25%)4 (17%)8 (32%)0.16
      Peripheral arterial disease5 (10%)2 (9%)3 (12%)0.10
      Cerebrovascular disease8 (17%)3 (13%)5 (20%)0.12
      COPD/Asthma8 (17%)4 (17%)4 (16%)0.00
      Malignancy9 (19%)3 (13%)6 (24%)0.16
      HIV2 (4%)2 (9%)0 (0%)0.52
      Dementia0 (0%)0 (0%)0 (0%)0.00
      Medications, n (%)
      Renin-angiotensin-aldosterone inhibitor12 (29%)6 (26%)6 (24%)0.00
      Statin31 (65%)14 (61%)17 (68%)0.05
      Beta blocker33 (69%)18 (78%)15 (60%)0.04
      Antiplatelet agent4 (8%)0 (0%)4 (16%)0.52
      Anticoagulant8 (17%)3 (13%)5 (20%)0.12
      Laboratory data prior to first HD treatment
      Blood urea nitrogen (mg/dL)54.2 (24.4)51.6 (23.2)58.3 (25.7)0.27
      Serum creatinine (mg/dL)5.9 (3.2)5.5 (2.8)6.4 (3.6)0.28
      eGFR (mL/min/1.73 m2)9.9 (5.1)9.4 (3.3)10.3 (6.4)0.17
      Serum sodium (mEq/L)137.9 (2.8)138.3 (2.9)137.5 (2.7)0.28
      Serum potassium (mEq/L)4.2 (0.6)4.1 (0.6)4.2 (0.6)0.17
      Serum bicarbonate (mEq/L)22.9 (3.4)22.5 (3.7)23.3 (3.1)0.24
      Hemoglobin (g/dL)9.5 (1.6)9.4 (1.4)9.5 (1.8)0.06
      Ferritin (μg/L)248.9 (243.5)279.7 (188.2)220.6 (294.3)0.24
      Transferrin saturation (%)37.8 (45.7)43.4 (48.4)30.9 (43.2)0.27
      Serum calcium (mg/dL)8.4 (0.8)8.3 (1.0)8.4 (0.7)0.12
      Serum phosphorus (mg/dL)5.3 (1.9)5.0 (1.7)5.6 (2.0)0.32
      Parathyroid hormone, intact (μg/L)400.9 (274.9)388.6 (225.0)412.3 (320.8)0.08
      Albumin (g/dL)3.4 (0.5)3.4 (0.4)3.3 (0.6)0.19
      Hemodialysis prescription prior to randomization
      Treatment time (minutes), mean (SD)201.9 (35.5)204.1 (39.6)199.8 (32.0)0.12
      Blood flow (mL/min), mean (SD)304.2 (60.9)295.7 (63.8)312.0 (58.2)0.27
      Dialysate flow (mL/min), mean (SD)539.6 (84.4)530.4 (75.5)548.0 (91.8)0.22
      Potassium bath (mEq/L), mean (SD)2.5 (0.5)2.5 (0.5)2.4 (0.5)0.20
      Calcium bath (mEq/L), mean (SD)2.6 (0.2)2.6 (0.2)2.6 (0.3)0.00
      *Subjects were excluded if received > 6 HD sessions by the date all eligibility criteria were met and randomization was enacted. Vascular access used at dialysis initiation.
      HD prescription parameters, prescribed thrice-weekly by treating provider prior to study enrollment and randomization.
      #The standardized mean difference is reported in absolute values.
      Percentages may not total 100 because of rounding. To convert values for serum creatinine to micromoles per liter, multiply by 88.4. Body-mass index is the weight in kilograms divided by the square of height in meters.
      eGFR denotes estimated glomerular filtration rate (as assessed by Chronic Kidney Disease Epidemiology equation)
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      ,
      • Levey A.S.
      • Stevens L.A.
      • Schmid C.H.
      • et al.
      A new equation to estimate glomerular filtration rate.
      ; HD, hemodialysis; KDRD, kidney dysfunction requiring dialysis.

       Adherence to Study Protocol

      At participant level. At mean follow-up of 288.9 days in incremental HD group and 275.3 days in conventional HD group, adherence to prescribed HD regimen was 96% in each group (22/23 in incremental HD group and 24/25 in conventional HD group). The drop-out rate was 13% (3/23) in incremental HD group and 12% (3/25) in conventional HD group, corresponding to a drop-out rate of 0.76 (95% CI of 0.25, 1.78) per 1,000 participant-days in each group. All participants who completed 6 and 12 months of follow-up, complied with inter-dialytic timed urine collection. Two patients crossed over from the twice-weekly to thrice-weekly HD regimen in the first 6 weeks (see below). There was no cross-over from the thrice-weekly to twice-weekly HD regimen.
      At facility level. We noted incidents of missed laboratory data; these were due to missed collection of blood samples before or after HD surrounding the timeframe of urine collection, or mishandling of urine specimen by the laboratory personnel. Overall, these events occurred at higher rate in the first 2 months of the study (up to 36%) and declined (<10%) at later time points during the study.

       Pharmacologic Therapy

      Adjuvant pharmacologic therapy in the Incremental HD group. According to the protocol, all patients randomized to incremental HD group received diuretics, with a mean (SD) 120 (40) furosemide-equivalent milligrams/day. Thirty-nice percent of participants (9/23) in the incremental HD group received sodium bicarbonate, with a mean dose of 1300 (650) mg/day. Seventeen percent of participants (4/23) in the incremental HD group received patiromer at a dose of 8.4g/day on non-HD days. Sodium bicarbonate and patiromer were discontinued after transition from twice-weekly to thrice-weekly HD.
      Other medications prescribed during the study by the treating team, according to standard care, are listed in Table S3.

       Serious Adverse Events

      During the first 6 weeks of the study, 2/23 (9%) patients in the incremental HD group were electively switched, by the treating provider, from twice to thrice-weekly HD at day 21 and day 30 for anemia and volume management, respectively. No events of unscheduled outpatient HD treatments or hospitalizations due to volume overload, metabolic imbalances or uremia occurred in the incremental HD group.
      Table 2 summarizes serious adverse events in each treatment group. A total of 52 hospitalizations occurred, 19 in the incremental and 33 in the conventional HD group (Item S1). The cumulative hazard function for hospitalization in each treatment group is shown in Figure 2. The annual cumulative hospitalization rate was 1.06 (95% CI, 0.64 to 1.66) in incremental HD group and 1.84 (95% CI, 1.31 to 2.59) in conventional HD group, with an IRR for all-cause hospitalization of 0.31 (95% CI, 0.08-1.17).
      Table 2Serious adverse events according to trial group
      EventIncremental HD groupConventional HD group
      Number of patients in analysis2222
      Days in study, per patient, mean (SD)288.9 (80.3)275.3 (98.0)
      Number of patients hospitalized1112
      Total number of hospitalizations1933
      Total days hospitalized71172
      Hospitalization rate, per 1000 person-days11.0 (8.6 - 13.8)26.3 (22.5 - 30.5)
      Cumulative hospitalization rate, per person-year (95% CI)1.06 (0.64 to 1.66)1.84 (1.31 to 2.59)
      Time to first hospitalization, days, median (1st - 3rd quartile)30.0 (17.0 - 94.5)58.0 (17.0 - 110.0)
      Proportion of hospitalizations in the first 90 days42% (8 of 19)33% (11 of 33)
      Length of hospital stay, days, median (1st - 3rd quartile)

      (per hospitalization, per person)
      2.0 (2.0 - 5.0)5.0 (2.5 - 6.5)
      Patients with ≥1 hospitalization, n (% of all patients in analysis)3 (14%)7 (32%)
      Hospitalization cause, n (%) of all hospitalizations
      Cardiovascular1 (5%)4 (12%)
      Cerebrovascular4 (21%)1 (3%)
      Fluid management-6 (18%)
      Hyperkalemia1 (5%)*-
      Encephalopathy-5 (15%)
      Vascular access infection2 (11%)-
      Vascular access complication, non-infectious-1 (33%)
      Infection, not related to vascular access4 (21%)9 (27%)
      Other6 (32%)7 (21%)
      All-cause death rate, per 1000 person-days0.15 (0.004 - 0.85)0.92 (0.34 - 1.99)
      Participants who withdrew consent, were lost to follow-up, or were withdrawn from the study were excluded from serious adverse events analysis. Events were calculated from date of randomization.
      Hospitalization rate, per 1000 person-days, takes into account the number of hospital days per 1000 days.
      The cumulative hospitalization rate takes into account the total number of hospitalizations per each individual, independent of hospital length of stay.
      *Event occurred with thrice-weekly HD prescription, 299 days after transition from twice-weekly HD.
      Figure thumbnail gr2
      Figure 2Cumulative Hazard of Hospitalization
      Seven deaths were recorded (1 in the incremental HD group and 6 in conventional HD group; hazard ratio for death, incremental HD group vs conventional HD group, 0.18; 95% CI 0.02 to 1.47), with an estimated first-year survival of 95.0% (standard error, 4.9%) in the incremental HD group and 63.2% (standard error, 13.1%) in conventional HD group (Table S4).

       Secondary Outcomes

       Residual Kidney Function and Urea Solute Clearance

      Estimated means (95% CI) and percent changes in residual kidney function parameters from baseline to scheduled time points are shown in Table 3 and Table 4, respectively.
      Table 3Parameters of residual kidney function and stdKt/V urea solute clearance
      ParameterStudy Time Point
      BaselineWeek 6Week 12Week 24
      Incremental HDConventional HDIncremental HDConventional HDIncremental HDConventional HDIncremental HDConventional HD
      NMean (95% CI)NMean (95% CI)NMean (95% CI)NMean (95% CI)NMean (95% CI)NMean (95% CI)NMean (95% CI)NMean (95% CI)
      Residual kidney function by treatment group
      Urine volume, mL/24 h23914 (654 –1174)251424 (976 – 1872)19504 (335 – 673)17553 (384 – 723)19479 (288 - 670)19484 (292 - 676)17862 (624 - 1099)15876 (638 - 1114)
      Renal urea clearance, mL/min/1.73 m2183.3 (2.0 – 4.5)174.1 (2.9 – 5.3)192.3 (1.6 – 2.9)172.3 (1.6 – 3.0)192.1 (1.4 – 2.8)192.0 (1.4 – 2.7)171.6 (0.9 – 2.2)152.2 (1.6 – 2.8)
      Renal creatinine clearance, mL/min/1.73 m2188.4 (5.6 – 11.3)1810.8 (8.1 – 13.5)155.5 (2.9 – 8.1)196.8 (4.7 – 9.0)155.3 (3.2 – 7.4)175.8 (4.1 – 7.5)165.6 (3.9 – 7.3)155.4 (3.7 – 7.2)
      Averaged renal urea and creatinine clearances, mL/min/1.73 m2186.1 (4.1 – 8.1)177.2 (5.3 – 9.1)114.0 (2.8 – 5.3)133.9 (2.7 – 5.0)154.4 (2.9 – 5.8)174.3 (3.0 – 5.6)163.6 (2.6 – 4.6)153.8 (2.7 – 4.8)
      StdKt/V urea solute clearance by treatment group
      Dialysis spKt/Vurea221.43 (1.32 - 1.53)211.30 (1.20 - 1.40)221.42 (1.32 - 1.51)181.40 (1.30 - 1.50)211.42 (1.34 - 1.50)181.36 (1.29 - 1.45)181.45 (1.29 - 1.61)171.53 (1.37 - 1.69)
      eKt/Vurea221.23 (1.14 - 1.32)211.10 (1.01 - 1.19)221.22 (1.4 - 1.31)181.19 (1.1. - 1.28)211.22 (1.15 - 1.29)181.16 (1.09 - 1.23)181.24 (1.11 - 1.38)171.30 (1.16 - 1.43)
      Dialysis stdKt/Vurea221.47 (1.40 - 1.54)212.10 (2.03 - 2.17)221.48 (1.42 - 1.55)182.19 (2.12 - 2.26)212.25 (2.18 - 2.33)182.17 (2.09 - 2.25)182.28 (2.19 - 2.37)172.28 (2.19 - 2.37)
      Renal stdKt/Vurea180.99 (0.63 - 1.36)171.24 (0.88 - 1.61)190.67 (0.42 - 0.92)170.66 (0.37 – 0.95)190.49 (0.32 - 0.66)170.59 (0.41 - 0.77)170.41 (0.27 - 0.55)150.54 (0.39 - 0.68)
      Total stdKt/Vurea (dialysis + renal stdKt/V)182.46 (2.09 - 2.84)173.37 (3.00 - 3.74)192.15 (1.90 - 2.39)172.90 (2.61 - 3.18)192.76 (2.59 - 2.94)172.77 (2.59 – 2.96)172.71 (2.55 - 2.87)152.82 (2.66 – 2.99)
      Dialysis spKt/Vurea221.43 (1.32 - 1.53)211.30 (1.20 - 1.40)221.42 (1.32 - 1.51)181.40 (1.30 - 1.50)211.42 (1.34 - 1.50)181.36 (1.29 - 1.45)181.45 (1.29 - 1.61)171.53 (1.37 - 1.69)
      Data are presented as least squares mean (95% confidence intervals [CI]). Residual renal clearance was calculated based on 24-hour urine collection performed at baseline at enrollment; and inter-dialytic urine collection performed during week 6, week 12 and week 24. Interdialytic urine collection started at the end of first HD session and ended at the beginning for the second HD session for both treatment groups. Duration of interdialytic urine collection (minute) was calculated as 72 x 60 – HD treatment time, if on twice-weekly HD; and as 48 x 60 – HD treatment time, if on thrice-weekly HD. To calculate renal clearances, blood samples were collected at the end of the HD treatment preceding timed urine collection and the beginning of the successive HD treatment ending the period of timed urine collection
      • Murea M.
      • Moossavi S.
      • Fletcher A.J.
      • et al.
      Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
      ,
      • Levey A.S.
      • Stevens L.A.
      • Schmid C.H.
      • et al.
      A new equation to estimate glomerular filtration rate.
      . Dialysis spKt/Vurea was obtained at the outpatient dialysis facility according to standard of care, during the first week of the month, before the second HD treatment of the week. Dialysis stdKt/V was calculated using an equation that includes effects of fluid removal. Urea distribution volume (V) was normalized using the formula 3.271×V(2/3), where V was determined based on Watson formula
      • Daugirdas J.T.
      • Depner T.A.
      • Greene T.
      • Levin N.W.
      • Chertow G.M.
      • Rocco M.V.
      Standard Kt/Vurea: a method of calculation that includes effects of fluid removal and residual kidney clearance.
      . Renal stdKt/Vurea was calculated based on renal urea clearance (mL/min) obtained at baseline, week 6, week 12 and week 24. Total stdKt/Vurea represents the sum of dialysis stdKt/V and renal stdKt/V at each respective time point. N denotes number of participants with data at the respective time point; eKt/V denotes equilibrated Kt/V;HD, hemodialysis, spKt/V, single pool Kt/V; stdKt/V, standard Kt/V. Approximate study time point, +/- 2 weeks.
      Table 4Percent change in parameters of residual kidney function and stdKt/V urea solute clearance over time
      OutcomeBaseline to Week 6Week 6 to Week 12Baseline to Week 12Baseline to Week 24
      NIncremental HDNConventional HDDifference (95% CI)NIncremental HDNConventional HDDifference (95% CI)NIncremental HDNConventional HDDifference (95% CI)NIncremental HDNConventional HDDifference (95% CI)
      Residual kidney function by treatment group
      Urine volume, mL/24 h18-35.3 (-51.0, -19.5)17-57.0 (-73.2, -40.8)21.7 (-0.9, 44.3)186.0 (-25.5, 37.6)173.8 (-27.8, 35.4)2.2 (-42.4, 46.8)17-36.3 (-53.4, 19.2)17-60.8 (-78.1, -43.4)24.5 (0.1, 48.8)1422.8 (-13.7, 59.2)14-28.2 (--65.0, 8.5)51.0 (-0.7, 102.8)
      Urea clearance, mL/min/1.73 m215-1.2 (-41.7, 39.2)15-3.7 (43.9, 36.6)2.4 (-54.6, 59.5)180.8 (-32.6, 34.2)167.8 (-27.2, 42.8)-7.0 (-55.3, 41.4)16-4.2 (-43.0, 34.6)16-24.3 (-62.9, 14.3)20.1 (-34.7, 74.8)142.8 (-50.0, 55.7)13-4.8 (-58.4, 48.8)7.6 (-67.6, 82.9)
      Creatinine clearance, mL/min/1.73 m211-31.3 (-70.2, 7.6)17-17.7 (-49.2, 13.9)-13.6 (-63.7, 36.5)1411.5 (-27.8, 50.9)18-2.4 (-33.4, 28.5)14.0 (-36.1, 64.1)12-30.1 (-63.8, 3.6)18-22.5 (-49.8, 4.8)-7.6 (-59.9, 35.8)1317.9 (-53.7, 89.5)13-20.8 (-91.8, 50.3)38.6 (-62.3, 139.5)
      Averaged renal urea and creatinine clearances, mL/min/1.73 m29-27.9 (-55.8, -0.1)11-31.2 (-56.0, -6.4)3.2 (-34.1, 40.5)922.5 (-11.2, 56.2)1123.0 (-7.7, 53.8)-0.5 (-46.1, 45.1)9-17.7 (-51.8, 16.5)11-14.9 (-45.3, 15.5)-2.7 (-48.4, 43.0)1312.4 (-45.3, 70.0)13-45.5 (-101.7, 10.7)57.9 (-22.6, 138.4)
      StdKt/V urea solute clearance by treatment group
      Dialysis stdKt/Vurea201.2 (-3.2, 5.7)164.9 (0.1, 9.8)-3.7 (-10.3, 3.0)1952.1 (47.7, 56.4)15-0.7 (-5.5, 4.2)52.7 (46.2, 59.3)1953.5 (47.2, 59.7)163.9 (-2.7,10.4)49.6 (40.5, 58.6)1755.3 (47.5, 63.0)198.6 (1.1, 16.1)46.6 (35.8, 57.4)
      Renal stdKt/Vurea17-22.7 (-46.7, 1.3)11-11.7 (-40.6, 17.2)-11.0 (-48.5, 26.6)16-0.9 (-39.4, 37.6)1115.6 (-29.3, 60.4)-16.4 (-75.5, 42.7)17-24.5 (-62.2, 13.2)14-22.4 ( -63.7, 18.9)-2.1 (-58.1, 53.8)17-17.8 (-55.9, 20.2)15-32.2 (-72.8, 8.5)14.3 (-41.4, 70.0)
      Total stdKt/Vurea (dialysis + renal stdKt/V)17-10.6 (-21.3, 0.2)11-6.7 (-19.4, 5.9)-3.8 (-20.4, 12.8)1634.8 (21.7, 47.3)11-1.1 (-16.2, 14.0)35.6 (15.8, 55.3)1718.6 (4.6, 32.6)14-9.7 (-24.8, 5.3)28.3 (7.7, 48.9)1718.9 (4.5, 33.3)15-9.8 (-25.2, 5.5)28.7 (7.7, 49.8)
      Data are presented as least squares mean (95% confidence interval). Clearance was normalized to body surface area. Percent change was calculated based on least squares mean values at different assessment time points as [(value time point b – value time point a)/value time point a]*100; where time point a and b represent the earlier and the later time points, respectively, in the interval calculated. Difference in estimated percent change in residual kidney function parameters between groups was calculated as mean % change in incremental HD group – mean % change in conventional HD group.
      Urine volume: At the end of week 6, the urine volume decreased by 35.3% (95% CI, -51.0 to -19.5) in the incremental HD group and by 57.0% in the conventional HD group (95% CI, -73.2 to -40.8), for an estimated difference of 21.7 percentage points (95% CI, -0.9 to 44.3) lower decline in urine volume in the incremental HD group compared to conventional HD group (Table 4). Both groups experienced a rise in urine output between week 6 and 24. By the end of week 24, the urine volume increased by 22.8% (95% CI, -13.7 to 59.2) in the incremental HD group and decreased by 28.2% (95% CI, -65.0 to 8.5) in the conventional HD group, corresponding to an estimated difference of 51.0 percentage points (95% CI, -0.7 to 102.8) lower decline in urine volume in the incremental HD group compared to conventional HD group.
      Renal urea and creatinine clearance (mL/min/1.73 m2): At baseline, the proportion of patients with renal urea clearance ≥2 mL/min was 86% the incremental HD group and 88% in the conventional HD group. From baseline to week 24, the averaged renal urea and creatinine clearances increased by 12.4% (95% CI, -45.3 to 70.0) in incremental HD group and decreased by 45.5% (95% CI, -101.7 to 10.7) in the conventional HD group, largely due to more pronounced difference in percent change in renal creatinine clearance as compared to renal urea clearance between the two groups (Table 4). The proportion of patients in the incremental HD group with different cut-offs of residual kidney function at the time of transition from twice-weekly to thrice-weekly HD is shown in Figure S2.
      Dialysis Adequacy Measures: According to study design, a significant increase in dialysis stdKt/V was present in the incremental HD group at week 12 (Table 3). Predialysis biochemical parameters were largely similar between the groups (Table S5).

       Volume Management

      Parameters related to volume management with HD are shown in Table 5.
      Table 5Parameters related to volume management
      ParameterStudy Time Point
      BaselineWeek 6Week 12Week 24
      Incremental HDConventional HDIncremental HDConventional HDIncremental HDConventional HDIncremental HDConventional HD
      NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)NMean (SD) or Median (Q1, Q3)
      Dry weight (kg)2392.0 (20.1)2587.5 (21.7)2191.8 (20.7)2188.4 (21.2)2090.9 (21.1)2087.6 (20.6)1992.1 (22.7)1987.5 (21.0)
      HD treatment time (minutes)23232.6 (24.8)25209.2 (21.9)21234.8 (27.3)21214.8 (34.3)20230.6 (23.5)20211.4 (29.8)19224.7 (28.0)19218.1 (32.2)
      Inter-dialytic weight gain*, % of dry weight231.3 (0.4, 2.5)251.2 (0.3, 2.4)211.5 (0.0, 2.2)210.9 (-0.4, 2.1)201.9 (0.4, 2.9)202.2 (1.0, 3.6)192.5 (1.0, 3.8)191.8 (0.6, 2.5)
      Residual weight*, % of dry weight23-0.4 (-0.9, 0.2)25-0.2 (-0.5, 04)21-0.6 (-1.1, -0.2)21-0.4 (-0.8, 0.2)20-0.6 (-1.0, 0.0)200.0 (-0.3, 0.6)190.1 (-0.5, 1.3)19-0.1 (-0.6, 0.5)
      Ultrafiltration rate*, mL/kg/hour234.7 (3.3)255.0 (3.6)214.7 (3.7)214.4 (3.3)205.9 (3.3)206.4 (4.0)196.4 (3.7)194.8 (3.7)
      Drop in SBP#, % of pre-dialysis SBP2314.0 (7.2, 20.6)2511.8 (5.1, 15.3)2116.6 (10.0, 23.8)2112.6 (9.4, 25.3)2013.8 (5.8, 31.1)2018.7 (8.5, 28.9)1916.4 (5.1, 23.4)1914.2 (4.4, 23.3)
      *#Calculated based on one-week average *values and #nadir SBP, per each participant, observed during the week laboratory data were obtained at outpatient dialysis unit. These include baseline at enrollment, week 6, week 12 and week 24. Pre- and post-dialysis weights were recorded in kg. Dry weights were established by the treating providers. Inter-dialytic weight gain, as percentage of dry weight, was calculated as [(predialysis weight – dry weight)/dry weight] × 100. Residual weight, as percentage of dry weight, was calculated as [(postdialysis weight – dry weight)/dry weight] × 100. Ultrafiltration rate was calculated as {[(predialysis weight – postdialysis weight) × 1000]/dry weight]}: dialysis treatment time (in hours). Drop in SBP, as percentage of pre-dialysis SBP, was calculated as [(predialysis SBP – nadir SBP)/predialysis SBP] × 100. HD denotes hemodialysis; N, number of participants with data at the respective time point; Q, quartile; SBP, systolic blood pressure. Approximate study time point, +/- 2 weeks.

      Discussion

      This pilot study showed that random allocation to incremental-start HD versus conventional HD in eligible patients with incident KDRD is feasible. The study met 3 of 4 a-priori defined feasibility criteria (Table 6). Several important lessons were learned.
      Table 6Lessons Learned from TWOPLUS Pilot Feasibility Trial
      FeasibilityLessons Learned
      ExpectedObserved
      • ≥70% of eligible patients are recruited
      • 41% (77/185) of patients with incident KDRD met preliminary criteria for study participation
      • 66% (51/77) of study candidates consented
      • 26% (48/185) of adults with incident KDRD met all eligibility criteria and were enrolled in the study
      • Partnership between Investigators and dialysis organizations is necessary to efficiently implement the study across a broader dialysis population and varied dialysis practice climates
      • ≥95% of participants randomized will adhere to the HD regimen
      • 96% (22/23 and 24/25) adhered to the assigned HD protocol
      • Changes in HD prescription from twice-weekly to thrice-weekly HD were required in 9% (2/23) prior to end of week 6, based on patient’s clinical status
      • Provider engagement is critical to assure safety of study participants
      • ≥80% patients adhere to study-specific assessments
      • 100% adhered to timed urine collection
      • Frequent communications between research team and dialysis personnel and laboratory personnel were necessary
      • Optimize stakeholder engagement, communication with dialysis personnel regarding urine and blood sample collection
      • Use Fidelity Checklist and Protocol Implementation Log with real-time observation and feedback to dialysis units in a larger multicenter clinical trial
      • ≤5% of participants randomized in the conventional HD group cross over to less-frequent HD group
      • 0% drop-in rate
      • Engaging “on-the-ground” clinicians is a requisite for the trial to become routinely incorporated into the delivery of clinical care
      First, the study demonstrates that multi-stakeholder collaboration between patients, investigators, dialysis administration, dialysis personnel and treating providers is required to proceed with a larger multi-center clinical trial. We note a lower than expected consent rate, which can negatively impact enrollment efficiency and generalizability of results. Thus, for a larger clinical trial, partnership between investigators and large dialysis organizations is necessary to a) broaden the pool of eligible candidates, and b) test implementation acceptability and sustainability across varied dialysis establishments and patient populations. Indeed, multi-stakeholder engagement has been recognized as a requisite to facilitate the conduct of clinical trials in nephrology
      • Dember L.M.
      • Archdeacon P.
      • Krishnan M.
      • et al.
      Pragmatic Trials in Maintenance Dialysis: Perspectives from the Kidney Health Initiative.
      .
      Second, this study uncovered the need for refining eligibility for future studies. Based on residual kidney function eligibility criteria chosen for this pilot, 13% of participants had renal urea clearance <2.0 mL/min/1.73 m2, compromising the achievement of total (dialysis + renal) stdKt/V ≥2.10 with twice-weekly HD at spKt/V ≥1.20 in those with low residual renal urea clearance
      KDOQI Clinical Practice Guideline for Hemodialysis Adequacy
      2015 update.
      ,
      • Daugirdas J.T.
      • Depner T.A.
      • Greene T.
      • Levin N.W.
      • Chertow G.M.
      • Rocco M.V.
      Standard Kt/Vurea: a method of calculation that includes effects of fluid removal and residual kidney clearance.
      . For implementation purposes and intervention scalability, a chief practical consideration is that patients in the incremental HD group receive HD treatments targeted for same metrics (i.e., urea reduction ratio ≥65% and skKt/V ≥1.20) as patients in conventional HD group, the only difference being the frequency of HD. Urea kinetic models identified thresholds of renal urea clearance that confer validity to this practical approach
      • Daugirdas J.T.
      • Depner T.A.
      • Greene T.
      • Levin N.W.
      • Chertow G.M.
      • Rocco M.V.
      Standard Kt/Vurea: a method of calculation that includes effects of fluid removal and residual kidney clearance.
      . Thus, renal urea clearance of ≥2.0 mL/min/1.73 m2 should be among residual kidney function eligibility criteria in future clinical trials, particularly with protocols consisting of longer, individualized periods of twice-weekly HD.
      Our preliminary results suggest incremental-start HD may confer better preservation of residual kidney function than conventional HD, a finding previously reported in observational studies
      • Obi Y.
      • Streja E.
      • Rhee C.M.
      • et al.
      Incremental Hemodialysis, Residual Kidney Function, and Mortality Risk in Incident Dialysis Patients: A Cohort Study.
      ,
      • Zhang M.
      • Wang M.
      • Li H.
      • et al.
      Association of initial twice-weekly hemodialysis treatment with preservation of residual kidney function in ESRD patients.
      ,
      • Wang A.Y.
      Preserving Residual Kidney Function in Hemodialysis Patients-Back in the Spotlight.
      • Tattersall J.
      Residual renal function in incremental dialysis.
      • Basile C.
      • Casino F.G.
      Incremental haemodialysis and residual kidney function: more and more observations but no trials.
      • Bolasco P.
      • Casula L.
      • Contu R.
      • Cadeddu M.
      • Murtas S.
      Evaluation of Residual Kidney Function during Once-Weekly Incremental Hemodialysis.
      . Both groups had significant decline in urine volume at week 6 and week 12, followed by a rebound in urine output by week 24, possibly related to dialysis-induced kidney injury followed by recovery
      • Benichou N.
      • Gaudry S.
      • Dreyfuss D.
      The artificial kidney induces acute kidney injury: yes.
      . The estimated treatment effect on urine volume at week 24 seemed to favor incremental HD, suggesting dialysis-induced kidney injury may had been less pronounced with twice-weekly HD. Although we did not monitor participants’ adherence to pharmacotherapy, we speculate that diuretic prescription, which was enforced at randomization in the incremental group, may have contributed to the observed between-group differences in urine output. Additionally, we identified solute-specific differences in residual kidney function between the groups, with larger decline in renal creatinine clearance in conventional group and similar decline in renal urea clearance. This may suggest tubular secretory function could be better preserved with incremental HD which, in turn, may be protective against adverse clinical outcomes
      • Sirich T.L.
      • Aronov P.A.
      • Plummer N.S.
      • Hostetter T.H.
      • Meyer T.W.
      Numerous protein-bound solutes are cleared by the kidney with high efficiency.
      ,
      • Suchy-Dicey A.M.
      • Laha T.
      • Hoofnagle A.
      • et al.
      Tubular Secretion in CKD.
      . A growing body of literature indicates several manifestations of KDRD (e.g., heart failure, arrhythmias, and sudden death)are directly associated with accumulation of protein-bound uremic solutes that are not readily dialyzable, but rather are actively secreted by transporters in the proximal renal tubule
      • Lowenstein J.
      • Grantham J.J.
      The rebirth of interest in renal tubular function.
      .
      Results should be interpreted in the context of several limitations. Given a) imbalances in baseline characteristics between the two groups, b) small sample size, and c) time-delineated twice-weekly HD, the effect estimates are solely exploratory. Out testing of time-delineated prescription of twice-weekly HD for 6 weeks may leave lingering concerns as to whether incremental HD with longer periods of twice-weekly HD can be effectively and safely implemented. We theorize patients’ conformity to increasing HD frequency in the absence of clinical complications and when more than half had continued levels of substantial residual kidney function (Figure S2) lends assurance responsible levels of adherence can be anticipated in a clinical trial of individualized incremental HD that will employ ascending HD frequency with erosion of residual kidney function or clinical requirements.
      This study showed core components of incremental HD can be achieved: intervention eligibility and enrollment among patients with incident KDRD and residual kidney function are favorable; intervention implementation at outpatient HD units is attainable; and patient adherence to recommended changes in HD prescription and to serial timed urine collection are good. We believe these results, coupled with the pilot study conducted in Europe
      • Vilar E.
      • Kaja Kamal R.M.
      • Fotheringham J.
      • et al.
      A multicenter feasibility randomized controlled trial to assess the impact of incremental versus conventional initiation of hemodialysis on residual kidney function.
      , provide the necessary data to advance the research in individualized HD to a larger multicenter clinical trial. We propose individualized HD in the incremental study group will have individualized periods of twice-weekly HD with spKt/V urea ≥1.20 and adjuvant pharmacotherapy; with progress from twice-weekly to thrice-weekly HD being driven by changes in residual kidney function and/or clinical status. We emphasize implementation of practice-embedded incremental HD clinical trial at a larger scale will require diligent planning consisting of systematic education of dialysis staff and nephrology providers about practical periodic patient assessments
      • Baigent C.
      • Herrington W.G.
      • Coresh J.
      • et al.
      Challenges in conducting clinical trials in nephrology: conclusions from a Kidney Disease-Improving Global Outcomes (KDIGO) Controversies Conference.
      .
      With this framework, a multicenter controlled trial with the primary objective to establish noninferiority for safety of incremental HD is justified. The proposed primary outcome is incidence rate of all-cause mortality, all-cause ED visits not leading to a hospitalization, and all-cause hospitalizations at 24 months. This rate would be calculated as the total number of safety events divided by the total number of person-years observed in the study. A participant could have recurring ED visits, hospitalizations, and/or die during the study and all events would be included in the primary outcome analysis. A sample size of 350 total (175 per treatment group) would allow rejection of a noninferiority hazard ratio (HR) limit of 1.20 with ≥85% power at one-sided level of significance equal to 0.025 under assumptions detailed in Item S1.
      In conclusion, this pilot trial showed time-delineated treatment with twice-weekly HD and adjuvant pharmacologic therapy followed by conversion to thrice-weekly HD, along with serial timed urine collection is feasible. These findings support indication to progress to a larger multicenter clinical trial with modified eligibility criteria and individualized periods of twice-weekly HD to conclusively investigate clinical effectiveness and safety of individualized HD.

      Supplementary data

      REFERENCES

        • KDOQI Clinical Practice Guideline for Hemodialysis Adequacy
        2015 update.
        Am J Kidney Dis. 11/2015 2015; 66: 884-930https://doi.org/10.1053/j.ajkd.2015.07.015
        • Murea M.
        • Deira J.
        • Kalantar-Zadeh K.
        • Casino F.G.
        • Basile C.
        The spectrum of kidney dysfunction requiring chronic dialysis therapy: Implications for clinical practice and future clinical trials.
        Semin Dial. Oct 12 2021; https://doi.org/10.1111/sdi.13027
        • Murea M.
        Precision medicine approach to dialysis including incremental and decremental dialysis regimens.
        Curr Opin Nephrol Hypertens. Jan 2021; 30: 85-92https://doi.org/10.1097/mnh.0000000000000667
        • Lowrie E.G.
        • Laird N.M.
        • Parker T.F.
        • Sargent J.A.
        Effect of the hemodialysis prescription of patient morbidity: report from the National Cooperative Dialysis Study.
        N Engl J Med. 11/12/1981 1981; 305: 1176-1181
        • Eknoyan G.
        • Beck G.J.
        • Cheung A.K.
        • et al.
        Effect of dialysis dose and membrane flux in maintenance hemodialysis.
        N Engl J Med. 12/19/2002 2002; 347: 2010-2019
        • Vilar E.
        • Wellsted D.
        • Chandna S.M.
        • Greenwood R.N.
        • Farrington K.
        Residual renal function improves outcome in incremental haemodialysis despite reduced dialysis dose.
        Nephrol Dial Transplant. 8/2009 2009; 24: 2502-2510
        • Kalantar-Zadeh K.
        • Unruh M.
        • Zager P.G.
        • et al.
        Twice-weekly and incremental hemodialysis treatment for initiation of kidney replacement therapy.
        Am J Kidney Dis. Aug 2014; 64: 181-186https://doi.org/10.1053/j.ajkd.2014.04.019
        • Murea M.
        • Moossavi S.
        • Garneata L.
        • Kalantar-Zadeh K.
        Narrative Review of Incremental Hemodialysis.
        Kidney Int Rep. Feb 2020; 5: 135-148https://doi.org/10.1016/j.ekir.2019.11.014
        • Ghahremani-Ghajar M.
        • Rojas-Bautista V.
        • Lau W.L.
        • et al.
        Incremental Hemodialysis: The University of California Irvine Experience.
        Semin Dial. May 2017; 30: 262-269https://doi.org/10.1111/sdi.12591
        • Bowline I.G.
        • Russell G.B.
        • Bagwell B.
        • Crossley B.
        • Fletcher A.J.
        • Murea M.
        Temporal trends in fluid management with incremental hemodialysis.
        Clin Nephrol. Oct 2019; 92: 165-173https://doi.org/10.5414/cn109660
        • Obi Y.
        • Rhee C.M.
        • Mathew A.T.
        • et al.
        Residual Kidney Function Decline and Mortality in Incident Hemodialysis Patients.
        J Am Soc Nephrol. 12/2016 2016; 27: 3758-3768
        • Obi Y.
        • Streja E.
        • Rhee C.M.
        • et al.
        Incremental Hemodialysis, Residual Kidney Function, and Mortality Risk in Incident Dialysis Patients: A Cohort Study.
        Am J Kidney Dis. 8/2016 2016; 68: 256-265
        • Park J.I.
        • Park J.T.
        • Kim Y.L.
        • et al.
        Comparison of outcomes between the incremental and thrice-weekly initiation of hemodialysis: a propensity-matched study of a prospective cohort in Korea.
        Nephrol Dial Transplant. Feb 1 2017; 32: 355-363https://doi.org/10.1093/ndt/gfw332
        • Dai L.
        • Lu C.
        • Liu J.
        • et al.
        Impact of twice- or three-times-weekly maintenance hemodialysis on patient outcomes: A multicenter randomized trial.
        Medicine. 2020; 99
        • Zhang M.
        • Wang M.
        • Li H.
        • et al.
        Association of initial twice-weekly hemodialysis treatment with preservation of residual kidney function in ESRD patients.
        Am J Nephrol. 2014; 40: 140-150https://doi.org/10.1159/000365819
        • Marquez I.O.
        • Tambra S.
        • Luo F.Y.
        • et al.
        Contribution of residual function to removal of protein-bound solutes in hemodialysis.
        Clin J Am Soc Nephrol. Feb 2011; 6: 290-296https://doi.org/10.2215/cjn.06100710
        • Shafi T.
        • Jaar B.G.
        • Plantinga L.C.
        • et al.
        Association of residual urine output with mortality, quality of life, and inflammation in incident hemodialysis patients: the Choices for Healthy Outcomes in Caring for End-Stage Renal Disease (CHOICE) Study.
        Am J Kidney Dis. Aug 2010; 56: 348-358https://doi.org/10.1053/j.ajkd.2010.03.020
        • Kalantar-Zadeh K.
        • Casino F.G.
        Let us give twice-weekly hemodialysis a chance: revisiting the taboo.
        Nephrol Dial Transplant. Sep 2014; 29: 1618-1620https://doi.org/10.1093/ndt/gfu096
        • Mathew A.T.
        • Fishbane S.
        • Obi Y.
        • Kalantar-Zadeh K.
        Preservation of residual kidney function in hemodialysis patients: reviving an old concept.
        Kidney Int. 8/2016 2016; 90 (S0085-2538(16)30054-0 [pii];10.1016/j.kint.2016.02.037): 262-271
        • Obi Y.
        • Chou J.
        • Kalantar-Zadeh K.
        Introduction to the Critical Balance - Residual Kidney Function and Incremental Transition to Dialysis.
        Semin Dial. May 2017; 30: 232-234https://doi.org/10.1111/sdi.12600
        • Murea M.F.J.
        • Anjay R.
        • Emaad A.R.
        • Gupta N.
        • Kovach C.
        • Vachharajani T.J.
        • Kalantar-Zadeh K.
        • Casino F.G.
        • Basile C.
        Kidney dysfunction requiring dialysis is a heterogeneous syndrome; se should treat it like one.
        Curr Opin Nephrol Hypertens. 11/3/21 2021; https://doi.org/10.1097/MNH.0000000000000754
        • Murea M.
        • Moossavi S.
        • Fletcher A.J.
        • et al.
        Renal replacement treatment initiation with twice-weekly versus thrice-weekly haemodialysis in patients with incident dialysis-dependent kidney disease: rationale and design of the TWOPLUS pilot clinical trial.
        BMJ Open. May 24 2021; 11e047596https://doi.org/10.1136/bmjopen-2020-047596
        • Levey A.S.
        • Stevens L.A.
        • Schmid C.H.
        • et al.
        A new equation to estimate glomerular filtration rate.
        Ann Intern Med. 5/5/2009 2009; 150: 604-612
        • Daugirdas J.T.
        • Depner T.A.
        • Greene T.
        • Levin N.W.
        • Chertow G.M.
        • Rocco M.V.
        Standard Kt/Vurea: a method of calculation that includes effects of fluid removal and residual kidney clearance.
        Kidney Int. 4/2010 2010; 77: 637-644https://doi.org/10.1038/ki.2009.525
        • Lewis M.
        • Bromley K.
        • Sutton C.J.
        • McCray G.
        • Myers H.L.
        • Lancaster G.A.
        Determining sample size for progression criteria for pragmatic pilot RCTs: the hypothesis test strikes back.
        Pilot Feasibility Stud. Feb 3 2021; 7: 40https://doi.org/10.1186/s40814-021-00770-x
        • Austin P.C.
        Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples.
        Stat Med. Nov 10 2009; 28: 3083-3107https://doi.org/10.1002/sim.3697
        • Andersen P.K.
        • Gill R.D.
        Cox's Regression Model for Counting Processes: A Large Sample Study.
        The Annals of Statistics. 1982; 10 (21): 1100-1120https://doi.org/10.1214/aos/1176345976
      1. Rothman KJ, Boice JD. Epidemiologic Analysis with a Programmable Calculator. National Institutes of Health. Pub No 79-1649. 1979;31-32.

        • Dember L.M.
        • Archdeacon P.
        • Krishnan M.
        • et al.
        Pragmatic Trials in Maintenance Dialysis: Perspectives from the Kidney Health Initiative.
        J Am Soc Nephrol. Oct 2016; 27: 2955-2963https://doi.org/10.1681/asn.2016030340
        • Wang A.Y.
        Preserving Residual Kidney Function in Hemodialysis Patients-Back in the Spotlight.
        J Am Soc Nephrol. Dec 2016; 27: 3504-3507https://doi.org/10.1681/asn.2016060693
        • Tattersall J.
        Residual renal function in incremental dialysis.
        Clin Kidney J. Dec 2018; 11: 853-856https://doi.org/10.1093/ckj/sfy082
        • Basile C.
        • Casino F.G.
        Incremental haemodialysis and residual kidney function: more and more observations but no trials.
        Nephrol Dial Transplant. Nov 1 2019; 34: 1806-1811https://doi.org/10.1093/ndt/gfz035
        • Bolasco P.
        • Casula L.
        • Contu R.
        • Cadeddu M.
        • Murtas S.
        Evaluation of Residual Kidney Function during Once-Weekly Incremental Hemodialysis.
        Blood Purif. 2021; 50: 246-253https://doi.org/10.1159/000509790
        • Benichou N.
        • Gaudry S.
        • Dreyfuss D.
        The artificial kidney induces acute kidney injury: yes.
        Intensive Care Med. Mar 2020; 46: 513-515https://doi.org/10.1007/s00134-019-05891-9
        • Sirich T.L.
        • Aronov P.A.
        • Plummer N.S.
        • Hostetter T.H.
        • Meyer T.W.
        Numerous protein-bound solutes are cleared by the kidney with high efficiency.
        Kidney Int. Sep 2013; 84: 585-590https://doi.org/10.1038/ki.2013.154
        • Suchy-Dicey A.M.
        • Laha T.
        • Hoofnagle A.
        • et al.
        Tubular Secretion in CKD.
        J Am Soc Nephrol. Jul 2016; 27: 2148-2155https://doi.org/10.1681/ASN.2014121193
        • Lowenstein J.
        • Grantham J.J.
        The rebirth of interest in renal tubular function.
        Am J Physiol Renal Physiol. Jun 1 2016; 310: F1351-F1355https://doi.org/10.1152/ajprenal.00055.2016
        • Vilar E.
        • Kaja Kamal R.M.
        • Fotheringham J.
        • et al.
        A multicenter feasibility randomized controlled trial to assess the impact of incremental versus conventional initiation of hemodialysis on residual kidney function.
        Kidney Int. Aug 18 2021; https://doi.org/10.1016/j.kint.2021.07.025
        • Baigent C.
        • Herrington W.G.
        • Coresh J.
        • et al.
        Challenges in conducting clinical trials in nephrology: conclusions from a Kidney Disease-Improving Global Outcomes (KDIGO) Controversies Conference.
        Kidney Int. Aug 2017; 92: 297-305https://doi.org/10.1016/j.kint.2017.04.019