Bio-impedance spectroscopy added to a fluid management protocol does not improve preservation of residual kidney function in incident hemodialysis patients in a randomized controlled trial

Abstract

Avoiding excessive dialysis-associated volume depletion may help preserve residual kidney function (RKF). To establish whether knowledge of the estimated normally hydrated weight from bioimpedance measurements (BI-NHW) when setting the post-hemodialysis target weight (TW) might mitigate rate of loss of RKF, we undertook an open label, randomized controlled trial in incident patients receiving HD, with clinicians and patients blinded to bioimpedance readings in controls. A total of 439 patients with over 500 ml urine/day or residual GFR exceeding 3 ml/min/1.73m2 were recruited from 34 United Kingdom centers and randomized 1:1, stratified by center. Fluid assessments were made for up to 24 months using a standardized proforma in both groups, supplemented by availability of BI-NHW in the intervention group. Primary outcome was time to anuria, analyzed using competing-risk survival models adjusted for baseline characteristics, by intention to treat. Secondary outcomes included rate of RKF decline (mean urea and creatinine clearance), blood pressure and patient-reported outcomes. There were no group differences in cause-specific hazard rates of anuria (0.751; 95% confidence interval (0.459, 1.229)) or sub-distribution hazard rates (0.742 (0.453, 1.215)). RKF decline was markedly slower than anticipated, pooled linear rates in year 1: –0.178 (–0.196, –0.159)), year 2: –0.061 (–0.086, –0.036)) ml/min/1.73m2/month. Blood pressure and patient-reported outcomes did not differ by group. The mean difference agreement between TW and BI-NHW was similar for both groups, Bioimpedance: –0.04 kg; Control: –0.25 kg. Thus, use of a standardized clinical protocol for fluid assessment when setting TW is associated with excellent preservation of RKF. Hence, bioimpedance measurements are not necessary to achieve this. Avoiding excessive dialysis-associated volume depletion may help preserve residual kidney function (RKF). To establish whether knowledge of the estimated normally hydrated weight from bioimpedance measurements (BI-NHW) when setting the post-hemodialysis target weight (TW) might mitigate rate of loss of RKF, we undertook an open label, randomized controlled trial in incident patients receiving HD, with clinicians and patients blinded to bioimpedance readings in controls. A total of 439 patients with over 500 ml urine/day or residual GFR exceeding 3 ml/min/1.73m2 were recruited from 34 United Kingdom centers and randomized 1:1, stratified by center. Fluid assessments were made for up to 24 months using a standardized proforma in both groups, supplemented by availability of BI-NHW in the intervention group. Primary outcome was time to anuria, analyzed using competing-risk survival models adjusted for baseline characteristics, by intention to treat. Secondary outcomes included rate of RKF decline (mean urea and creatinine clearance), blood pressure and patient-reported outcomes. There were no group differences in cause-specific hazard rates of anuria (0.751; 95% confidence interval (0.459, 1.229)) or sub-distribution hazard rates (0.742 (0.453, 1.215)). RKF decline was markedly slower than anticipated, pooled linear rates in year 1: –0.178 (–0.196, –0.159)), year 2: –0.061 (–0.086, –0.036)) ml/min/1.73m2/month. Blood pressure and patient-reported outcomes did not differ by group. The mean difference agreement between TW and BI-NHW was similar for both groups, Bioimpedance: –0.04 kg; Control: –0.25 kg. Thus, use of a standardized clinical protocol for fluid assessment when setting TW is associated with excellent preservation of RKF. Hence, bioimpedance measurements are not necessary to achieve this. Lay SummaryPatients on hemodialysis benefit from keeping some of their own kidney function for as long as possible after starting dialysis. Removing too much fluid during dialysis, by setting a low target weight for the end of a dialysis session, could accelerate its rate of loss. We wanted to see how this could be affected by developing a standardized protocol for assessing fluid status in people new to dialysis and whether device called bioimpedance, which calculates the target weight independently, was better at guiding clinicians in avoiding setting target weights too low. A total of 437 people from 34 dialysis centers across the UK took part in the randomized trial for up to 2 years. Using bioimpedance did not result in better outcomes as clinicians were just as good in setting the target weight whether or not they used the device. We expected that approximately 25% would lose their own kidney function after 1 year. We found that this was much lower in both groups, such that <25% lost their kidney function by 2 years. There was good evidence that clinical staff engaged with patients’ views when deciding whether to change the target weight. Safety, transplantation rates, and numbers of deaths were not affected. Bioimpedance does not improve on setting the target weight in the context of a standardized approach to fluid management. Applying a strategy that avoids excessive fluid removal is associated with better-than-expected preservation of kidney function. Patients on hemodialysis benefit from keeping some of their own kidney function for as long as possible after starting dialysis. Removing too much fluid during dialysis, by setting a low target weight for the end of a dialysis session, could accelerate its rate of loss. We wanted to see how this could be affected by developing a standardized protocol for assessing fluid status in people new to dialysis and whether device called bioimpedance, which calculates the target weight independently, was better at guiding clinicians in avoiding setting target weights too low. A total of 437 people from 34 dialysis centers across the UK took part in the randomized trial for up to 2 years. Using bioimpedance did not result in better outcomes as clinicians were just as good in setting the target weight whether or not they used the device. We expected that approximately 25% would lose their own kidney function after 1 year. We found that this was much lower in both groups, such that <25% lost their kidney function by 2 years. There was good evidence that clinical staff engaged with patients’ views when deciding whether to change the target weight. Safety, transplantation rates, and numbers of deaths were not affected. Bioimpedance does not improve on setting the target weight in the context of a standardized approach to fluid management. Applying a strategy that avoids excessive fluid removal is associated with better-than-expected preservation of kidney function. Most people starting dialysis have significant residual kidney function (RKF), and observational studies have consistently found that if this is preserved, it is associated with better survival and improved quality of life.1Termorshuizen F. Dekker F.W. van Manen J.G. et al.Relative contribution of residual renal function and different measures of adequacy to survival in hemodialysis patients: an analysis of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD)-2.J Am Soc Nephrol. 2004; 15: 1061-1070Google Scholar, 2Obi Y. Rhee C.M. Mathew A.T. et al.Residual kidney function decline and mortality in incident hemodialysis patients.J Am Soc Nephrol. 2016; 27: 3758-3768Google Scholar, 3Merkus M.P. Jager K.J. Dekker F.W. et al.Quality of life in patients on chronic dialysis: self-assessment 3 months after the start of treatment. The Necosad Study Group.Am J Kidney Dis. 1997; 29: 584-592Google Scholar Despite this, there are few trials of interventions that might improve the preservation of RKF in hemodialysis (HD) patients, and where these have been undertaken, they are typically of fewer than 50 participants.4Schiffl H. Lang S.M. Fischer R. Ultrapure dialysis fluid slows loss of residual renal function in new dialysis patients.Nephrol Dial Transplant. 2002; 17: 1814-1818Google Scholar, 5Lu W. Ren C. Han X. et al.The protective effect of different dialysis types on residual renal function in patients with maintenance hemodialysis: a systematic review and meta-analysis.Medicine (Baltimore). 2018; 97e12325Google Scholar, 6Lang S.M. Bergner A. Töpfer M. Schiffl H. Preservation of residual renal function in dialysis patients: effects of dialysis-technique-related factors.Perit Dial Int. 2001; 21: 52-57Google Scholar There is also plenty of evidence of inconsistency in the design and application of dialysis unit protocols to guide fluid management. This inconsistency was evident in a UK-wide survey of practices undertaken in preparation for the design of this study, where 50% of units claimed to use volume control to reduce dependence on antihypertensive medication.7Dasgupta I. Farrington K. Davies S.J. et al.UK national survey of practice patterns of fluid volume management in haemodialysis patients: a need for evidence.Blood Purif. 2016; 41: 324-331Google Scholar The Dialysis Outcomes and Practice Patterns Study also found considerable variation in practices related to fluid management and that a protocol specifying the frequency of assessment was associated with better outcomes.8Dasgupta I. Thomas G.N. Clarke J. et al.Associations between hemodialysis facility practices to manage fluid volume and intradialytic hypotension and patient outcomes.Clin J Am Soc Nephrol. 2019; 14: 385-393Google Scholar Bioimpedance (BI) is frequently used in HD units to monitor fluid status and body composition. There is evidence that overhydration and loss of lean tissue mass, measured using BI, are associated with shorter survival9Tabinor M. Elphick E. Dudson M. et al.Bioimpedance-defined overhydration predicts survival in end stage kidney failure (ESKF): systematic review and subgroup meta-analysis.Sci Rep. 2018; 8: 4441Google Scholar, 10Dekker M.J.E. Marcelli et the between and in hemodialysis patients: from the J 2016; Scholar, M. et fluid and lower weight are associated with mortality in a hemodialysis Dial Transplant. 2018; Scholar, C. C. et fluid and mortality in Am Soc Nephrol. Scholar over and such as and blood is is whether a to in guiding the of the target weight (TW) when fluid status in overhydration might help in blood pressure may volume RKF at M. et trial of analysis clinical for guiding in hemodialysis patients: effects on blood and Nephrol. Scholar for the use of BI, the for and in the UK that there was evidence to its for and frequency bioimpedance to guide fluid management in people with chronic kidney Scholar This the of the UK of and to a to the use of in guiding fluid its after the outcome of was for no Bioimpedance fluid management in dialysis Scholar of the of loss of RKF in HD is the removal of which the to volume kidney and kidney R. E. et during hemodialysis: intradialytic blood decline and effects of Am Soc Nephrol. 2019; Scholar to the we that avoiding setting the TW the estimated normally hydrated weight from (BI-NHW) where possible might the by excessive volume to preserve RKF. To establish whether this use of the BI-NHW was associated with a lower rate of loss of RKF, we a randomized controlled trial in which clinicians setting the TW were blinded to the in the control group. We that the of a protocol that the of fluid assessments and the of RKF and the use of a proforma to the assessment in is to that dialysis fluid assessments are an of that patient this was to be we on the of the an analysis of the of the The trial protocol was the of S.J. et and design of a randomized controlled trial to whether bioimpedance fluid management residual kidney function in incident haemodialysis Nephrol. Scholar this was an randomized UK-wide trial of incident HD patients the of and were HD patients 3 months of HD, using and for were evidence of RKF, as ml of volume or a measured rate were the to or for RKF and of or expected transplantation were the trial and to the for The trial was by and Study included and and a receiving The UK and their or was to fluid status were in the use of the fluid assessment proforma 1 in the and S.J. et and design of a randomized controlled trial to whether bioimpedance fluid management residual kidney function in incident haemodialysis Nephrol. and to the TW as to excessive volume where the patients randomized to the intervention the BI-NHW could be used in to clinical patients in the control the TW was using clinical To achieve measurements were by in both groups, the BI-NHW measurements were to the proforma for patients in the intervention the were fluid assessments were made for 3 months and for up to the of 2 with for assessments if The trial an open approach to of the device. an by Kidney of that were to their patient evidence to the and for The was the centers were in the use of the if necessary to the change in 2 in the The the BI-NHW by the weight would be if the and were normally et to fluid from the of body J Scholar were fluid assessments by an were unit and the and undertook blinded quality control assessments of The outcome of was RKF, 2 months and measured both as time to anuria the outcome and as or ml of volume in the with a at 2 and as the rate of decline in measured The was from an and and blood as the mean of the urea and creatinine adjusted for body using a 3 in the and of this which when some of the blood were have been E. J. et residual kidney function in haemodialysis patients using of the use of estimated blood to Scholar Secondary outcomes included and blood pressure at the time of fluid assessment and patient-reported of which a good is to M. C. A. et and of the new of Scholar and et outcome measures in the of hemodialysis patients.J Scholar 3 This was on the time to We from S.M. et of loss of residual renal function new dialysis patients.J Am Soc Nephrol. Scholar, et of the rate of decline of residual renal function in incident dialysis Int. 2002; Scholar, W. S.M. et decline of residual renal function in hemodialysis and Int. 2002; Scholar, et of residual renal function in patients 14: Scholar, M. A. et residual renal function in patients on using a dialysis Scholar and from UK dialysis that of incident patients would be by months and that a significant benefit would reduce this to and transplantation from the UK an decline in RKF, and a total of were to have with to a hazard of This in patients to be randomized 1:1, for a loss to We that patients would have months of with approximately 50% up to a of 2 years. to for was to 2 for of was 1:1, stratified by using of from 2 to was a by during The trial analysis was were on an models were used to cause-specific hazard and and survival analysis to hazard of anuria, and The of were the cause-specific of anuria and the of anuria the hazard rates of anuria between Patients change or of kidney function were at the of The analysis was adjusted for baseline RKF, that or using a use of and The difference between in the rate of decline in RKF was analyzed using a to the rate of change in GFR in for 1 and with for baseline characteristics, as for the outcome Blood pressure and patient-reported outcomes are using mean and or and for and and for To the of the the difference between the TW and the BI-NHW was for fluid assessment using a with measurements from the 34 dialysis included and The is in the 439 patients were recruited from 34 an the from to months after with the This at the and not to To for was to 2 for of and a months after to the trial was by the at which time clinical in the UK was for 3 this fluid assessments were not and measurements of RKF were to to baseline and characteristics, as in of the bioimpedance and control at of the patients in the control did not have baseline mean mean mean hydrated mean blood mean blood mean on on on mean HD, of the patients in the control did not have baseline in a new HD, The hazard rate of for start and baseline use of and not differ between the group and the control group with a cause-specific of confidence interval and of see of the are in The hazard rates also did not differ with a cause-specific of and a of The for the were and for and and for Patients from the to the the adjusted 3 in the control group was not used as this was to the and and cause-specific for anuria, adjusted for baseline patient in the (0.459, (0.453, or use hazard in a new hazard The rate of decline in RKF as is in 3 and The baseline GFR was a in the group than in the control group the of the whether adjusted or did not differ between The linear rate of decline in RKF in the group for 1 and 2 was and The control group of and for 1 and the were –0.178 and –0.061 The linear to GFR and a this is there were no patterns of by that this was the total of was the used measurements of decline of residual kidney function with linear (–0.196, (–0.086, confidence in a new confidence blood pressure and after dialysis at baseline and at time of the did not differ between There was a in the at this was during dialysis low blood pressure and of were and did not differ between The dialysis time was and to as the trial did not differ between and these time or these these these these of these blood standardized of quality of the for the good is are for time these in a new time by blood standardized of quality of the for the good is are for time was to how clinicians were in setting the TW in both the group and the control group. the total fluid assessments undertaken, were in in the group than in measurements as in the The mean difference between the TW and the BI-NHW was –0.25 in both where were made to or the patients were typically or their TW in the group and or TW in the control group There were no differences in the TW the TW and the or the intradialytic fluid removal between and see A analysis of the difference between target and BI-NHW found no significant difference between the with for and just of the total residual included and the proforma of patients’ when deciding on the and it is that there was a of agreement between clinicians and patients to associated with the difference between the target weight and the normally hydrated weight for fluid volume confidence in a new between patient and the to change or or not change the TW at the fluid change in TW better at TW and to at TW not to better at lower TW to not better at TW and to at TW not to better at lower TW to not target weight. in a new confidence target weight. there were during the of the in the and in the control were by to with in both and deaths in the group in the control group deaths in There were 34 of see trial found that the BI-NHW to a standardized fluid management in which clinicians were to a TW to volume did not result in better preservation of RKF. The for this may be of are that RKF was much better than studies would and that the of the BI-NHW by clinicians was just as good in the control group as in the group. the good preservation of it might be that Bioimpedance To observational evidence that the of excessive volume depletion is a approach to fluid management in patients with RKF. Despite the of RKF to people on dialysis, is M.P. et pressure and volume management in dialysis: from a Kidney Outcomes Int. Scholar and there are few clinical trials of interventions that might to its better when the we the rate of RKF loss the that observational studies and trials both and would The Dialysis and which RKF between and months of dialysis that of HD patients S.M. et of loss of residual renal function new dialysis patients.J Am Soc Nephrol. Scholar The Netherlands Cooperative Study on the Adequacy of Dialysis found in the incident HD GFR from to 2 by that with a pooled of in the year of a than we over the 24 et of the rate of decline of residual renal function in incident dialysis Int. 2002; Scholar The also much in volume and measured the RKF at and by starting at a lower measured GFR of on dialysis were in the group. the Dialysis 50% of with RKF in the group were at et of hemodialysis on residual kidney Int. Scholar the Dialysis Outcomes and Practice Patterns Study volume from HD the starting dialysis with a volume of ml in by months and by 24 M. et volume in hemodialysis patients: and mortality outcomes in the Dialysis Outcomes and Practice Patterns Study J Kidney Dis. 2019; Scholar included in a of the effects of on the rate of RKF rates of decline than we did not time to H. Lang S.M. Fischer R. Ultrapure dialysis fluid slows loss of residual renal function in new dialysis patients.Nephrol Dial Transplant. 2002; 17: 1814-1818Google W. Ren C. Han X. et al.The protective effect of different dialysis types on residual renal function in patients with maintenance hemodialysis: a systematic review and meta-analysis.Medicine (Baltimore). 2018; 97e12325Google Scholar trial the effects of on RKF in HD patients, GFR by over 1 with the loss of in et and loss of kidney function in hemodialysis patients: a randomized controlled J Kidney Dis. Scholar for time to anuria over the rate of RKF decline as outcome was the this on patients, by patient and its with outcomes in We were also by the of the which the effects of a dialysis fluid on RKF in dialysis Clarke M. et of fluid on dialysis Am Soc Nephrol. Scholar The trial found a significant effect on time to anuria that did not a significant difference in the rate of RKF with a rate of RKF decline in the months with the with mean of and in the control group and and in the intervention are not slower than the rates we in year 1: and year 2: over 2 in with in not of the rate of loss of RKF on HD and dialysis have that this is for that this be possible of the low rate of decline in RKF is the use in both of a protocol that included fluid measurements of RKF, and a standardized proforma that clinicians to a systematic approach when setting the were to use this, and the agreement between BI-NHW and TW in both groups, with the analysis no variation in its that this was There was also excellent agreement in both between the TW and the a of when with S.M. of weights and target weight with and J Am Soc Nephrol. Scholar in the use of a approach by a is to Using and in clinical to improve J 2004; Scholar and it is that a fluid assessment protocol in was associated with better outcomes in the Dialysis Outcomes and Practice Patterns Study I. Thomas G.N. Clarke J. et al.Associations between hemodialysis facility practices to manage fluid volume and intradialytic hypotension and patient outcomes.Clin J Am Soc Nephrol. 2019; 14: 385-393Google Scholar is that this approach be of that setting the TW is a good of fluid assessment proforma that patients have a as to the setting of their which a using a J. et for Scholar clinicians to this which was in of the fluid assessments undertaken, it was that in the of there was good agreement between patient and clinical Most observational evidence that application of a strategy that avoids excessive volume depletion is associated with RKF and its for low fluid and fewer for rates, which in are both associated with M. et fluid and lower weight are associated with mortality in a hemodialysis Dial Transplant. 2018; S.M. of weights and target weight with and J Am Soc Nephrol. S.M. the the of and weight J Am Soc Nephrol. 8: Scholar also a change in the strategy in survey of fluid management practices in UK dialysis I. Farrington K. Davies S.J. et al.UK national survey of practice patterns of fluid volume management in haemodialysis patients: a need for evidence.Blood Purif. 2016; 41: 324-331Google Scholar for the low rate of RKF loss could be the effect of of clinical We to the and as to as as recruited from of UK when with new incident patients dialysis in to the UK UK to Scholar they were a with a similar of and a of with and similar of the of effect on RKF, there were no differences by group in and blood pressure readings or patient-reported outcomes. is that associated with volume status were low when with and this might the good preservation of et quality of and mortality in hemodialysis patients: the Dialysis Outcomes and Practice Patterns Study J Kidney Dis. Scholar fluid and need for fluid removal during dialysis were A analysis of quality of life is as part of a assessment that be trial We were not to to the target the and a and to for this by the this did not up the of anuria it is also that the was in numbers than the decline in RKF. a trial would be to a benefit from using BI-NHW the difference in the setting of the TW between the This a of trials that have been to the benefit of for and frequency bioimpedance to guide fluid management in people with chronic kidney A. et of bioimpedance analysis estimated in maintenance dialysis patients: a systematic review and Nephrol. M. Davies S.J. The use of bioimpedance to guide fluid management in patients receiving 2018; 27: Scholar and C. C. F. et randomized trial on a strategy in patients on chronic hemodialysis with Int. Scholar in the management of fluid This is the in with W. et bioimpedance to fluid management of dialysis Int. 2016; Scholar, et fluid management benefit to dialysis from clinical Dial Int. 2018; Scholar, S.J. The of bioimpedance in fluid management of patients J Am Soc Nephrol. 15: Scholar trial was by which to some loss of and fluid assessments is a to the of trial and management that the trial did not have to be and the outcome were it be that there was a significant rate in the does not to fluid assessments to preserve RKF in incident HD patients, and TW using a standardized clinical approach to setting the TW with the of the This would that a standardized approach to clinical fluid assessment be the of evidence that preservation of RKF be over a significant of time in incident HD patients, and we have a of this from blood and E. J. et residual kidney function in haemodialysis patients using of the use of estimated blood to Scholar studies of interventions to preserve RKF are and that their the no The trial are to the of a This and be made to the in The of this trial were at the in at the trials The was by the for and The views and are of the and not of the of and or the of is by a from

Topics & Concepts