Haemodialysis dose and serum bicarbonate (Chapter 7)
1UK Renal Registry, Bristol and 2Department of Renal Medicine, University Hospital of Wales
Correspondence and offprint requests to: Charlie Tomson, UK Renal Registry, Bristol. Email: charlie.tomson{at}nbt.nhs.uk
 |
Abstract |
|---|
 Top Abstract IntroductionCompleteness of dataDialysis doseSerum bicarbonateReferences |
|---|
Data from 21 renal units was insufficient to allow analyses of the dose of dialysis in those units. Amongst the remainder, there is evidence of a progressive increase in the proportion of patients meeting the Renal Association audit standard for urea reduction ratio (URR).
In the UK as a whole, 81% of prevalent haemodialysis patients met the standard for URR in 2005. Greater achievement of the standard in a given unit is associated with a higher median URR in that unit, although there is some evidence that some units have been able to narrow the distribution of achieved URR values.
Achievement of the standard remains, as in previous years’ Reports, less common amongst patients recently established on haemodialysis compared with those established on haemodialysis for longer.
Correction of acidosis, as measured by serum bicarbonate concentration remains highly variable, although there is continued uncertainty about the interpretation of routine measurements of venous serum bicarbonate concentration in haemodialysis patients.
Overall, 
64% of UK haemodialysis patients, and 50% of peritoneal dialysis patients met the Renal Association standard for serum bicarbonate in 2005.
|
Introduction |
|---|
|
TopAbstractIntroductionCompleteness of dataDialysis doseSerum bicarbonateReferences |
|---|
Dialysis dose is an important predictor of outcome amongst patients receiving conventional thrice weekly dialysis and is highly susceptible to clinical intervention. Serum bicarbonate in contrast, bears an uncertain relationship to outcome, is highly influenced by non-patient-related factors such as delay in analysis after venepuncture, and it is less clear how clinicians can improve achievement of the desired bicarbonate concentration.
|
Completeness of data |
|---|
|
TopAbstractIntroductionCompleteness of dataDialysis doseSerum bicarbonateReferences |
|---|
No data on urea reduction ratio (URR) were received from Barts, Brighton, Hammersmith/Charing Cross, Royal Free, Newcastle or Wirral. Both Brighton and Newcastle are running CCL Clinicalvision, which currently does not support calculation of URRs. Most remaining centres returned data on >90% of patients, the exceptions being Belfast (89%), Cambridge (56%), Carshalton (64%), Chelmsford (80%), Clwyd (88%), Dudley (71%), Dundee (2%), Guys (81%), Kings (79%), Manchester West (52%), Oxford (66%), Preston (76%), Swansea (69%), Wolverhampton (79%) and Wrexham (69%) (Table 7.1).
|
The Scottish Renal Registry does not currently report serum bicarbonate data from Scottish Renal Units to the UK Renal Registry.
The completeness is recorded as within the last 6 months for England, Wales and Northern Ireland centres and within the last year for Scotland.
Centres reporting data on <20 patients or <50% of prevalent patients were not included in the centre level analyses. The number preceding the centre name in each figure indicates the percentage of missing data for that centre.
|
Dialysis dose |
|---|
|
TopAbstractIntroductionCompleteness of dataDialysis doseSerum bicarbonateReferences |
|---|
Introduction
The Renal Association guidelines offer both Kt/V and URR as markers of haemodialysis dose. The relevant audit standards agreed by the Renal Association [1] are as follows:
HD should take place at least three times per week in nearly all patients. Reduction of dialysis frequency to twice per week because of insufficient dialysis facilities is unacceptable. (Good practice).
Every patient receiving thrice weekly HD should show: either urea reduction ratio (URR) consistently >65% or equilibrated Kt/V of >1.2 (calculated from pre- and post-dialysis urea values, duration of dialysis and weight loss during dialysis) (B).
Patients receiving twice weekly dialysis for reasons of geography should receive a higher sessional dose of dialysis, with a total Kt/V urea (combined residual renal and HD) of >1.8. If this cannot be achieved, then it should be recognized that there is a compromise between the practicalities of dialysis and the patient's long-term health. (Good practice).
Measurement of the ‘dose’ or ‘adequacy’ of HD should be performed monthly in all patients. All dialysis units should collect and report to the Registry, data on pre- and post-dialysis urea values, duration of dialysis and weight loss during dialysis. (Good practice).
Post-dialysis blood samples should be collected either by the slow-flow method, the simplified stop-flow method, or the stop-dialysate-flow method (Appendix 2). The method used should remain consistent within renal units and should be reported to the Registry. (B)
For pragmatic reasons (because most centres do not report duration of dialysis or weight loss during dialysis) the Registry has chosen URR for comparative audit. Data on post-dialysis sampling methods were last collected by telephone survey in 2002 [2]. No reliable data is held on whether the important variations in post-dialysis sampling methodology identified at that time still persist.
As in all other analyses, data are taken from the last quarter of the year (unless otherwise stated); if that data point is missing, data from the 3rd quarter are taken. Data on frequency of dialysis are not routinely reported by all centres and were last collected systematically as part of the 2002 National Renal Survey [3]. For the purposes of the analyses reported subsequently, data from patients known to be receiving twice weekly dialysis are omitted. However, not all centres report frequency of dialysis, so it is possible that some data from a very small number of patients receiving twice weekly dialysis are included in the analyses, but this would not have a large influence on the overall centre mean.
HD session length has been shown to predict outcome independently of URR [4]. The Registry is able to collect data on recorded session time but a few centres report prescribed session time. No data are currently collected on dialyser characteristics (e.g. surface area, clearance, flux, membrane type).
Several centres in the UK now use on-line measurement of ionic dialysance to measure small molecular clearance during haemodialysis, relying on small studies that have demonstrated a close linear relationship between this measure and conventional measures of urea clearance [4]. However, the Registry strongly encourages these centres to continue to perform and report conventional pre- and post-dialysis measurements of blood urea concentration at least on a 3-monthly basis, to allow continued comparative audit.
No consensus has yet been reached on a ‘common currency’ by which to define the dose of peritoneal dialysis and so no attempt has been made to report comparative audits of peritoneal dialysis dose. Consensus is required on whether the Registry should collect ‘raw’ data from 24 h urine and dialysate collections or calculated weekly Kt/Vurea and creatinine clearance; if the latter, a uniform methodology for derivation of these values will be required.
Achieved URR
Median URR achieved in each renal unit is shown in Figure 7.1. The percentage of reported patients meeting the Renal Association audit standard of a URR of 
65% is shown in Figure 7.2. Figure 7.3 demonstrates that the two are closely related; however, the dispersion of values on this plot above a URR of 68% suggests that some higher performing units are achieving the standard in a high proportion of patients by narrowing the distribution rather than simply shifting the distribution upwards [5].
|
|
|
Changes in URR over time
Figure 7.4 shows the change in median URR between 1998 and 2005 in each renal unit. Figure 7.5 shows the change in percentage of reported dialysis patients with a URR
65% in each unit over 1998–2005. Figure 7.6 shows summary data for England and Wales over the same time period. Although the median URR has remained at 71% over the last 3 years, the percentage of patients achieving a URR > 65% has risen from 77% to 81%.
|
|
|
Variation of achieved URR with time on dialysis
As in previous analyses, the percentage of patients with URR
65% is higher amongst patients who have been on RRT for longer than in those who recently started (Figure 7.7). However, the latter group has improved from 48% in 1999 to 68% in 2005. Figure 7.8 shows the percentage of patients with URR 65% during the first quarter of treatment.
|
|
Commentary
There has been a progressive increase over time in the proportion of UK haemodialysis patients meeting the Renal Association audit standards for URR. However, although an increased dialysis dose is being achieved in patients just starting RRT, there is evidence that these standards are less frequently met in patients starting dialysis than in ‘well-established’ patients. This is possibly due to difficulties relating to vascular access in the first few months of dialysis. Previous reports [3] analysed whether this was partly due to selective drop-out (to death or other modalities) of those not initially achieving the audit standard and it was shown that this was not the case, with lower URRs achieved throughout the first year even in those patients that survived at least 2 years.
|
Serum bicarbonate |
|---|
|
TopAbstractIntroductionCompleteness of dataDialysis doseSerum bicarbonateReferences |
|---|
Introduction
The relevant audit standard agreed by the Renal Association [1] is as follows:
Serum bicarbonate, before a haemodialysis (HD) session, measured with minimal delay after venepuncture should be between 20 and 26 mmol/l. (C)
For continuous ambulatory peritoneal dialysis (CAPD) patients serum bicarbonate, measured with minimal delay after venepuncture, should be between 25 and 29 mmol/l. (B)
Haemodialysis
Median pre-dialysis serum bicarbonate amongst prevalent haemodialysis patients in each renal unit is given in Figure 7.9; the percentage of patients in each unit meeting the Renal Association standards is given in Figure 7.10. Figure 7.11 presents the same data as in Figure 7.10 as a funnel plot and Table 7.2 can be used to look up the data for individual centres.
|
|
|
|
Peritoneal dialysis
Median serum bicarbonate amongst prevalent peritoneal dialysis patients in each renal unit is given in Figure 7.12; the percentage of patients in each unit meeting the Renal Association standards is shown in Figure 7.13. Figure 7.14 presents the same data as in Figure 7.13 as a funnel plot and Table 7.3 can be used to look up the data for individual centres.
|
|
|
|
Transplant
Median serum bicarbonate amongst prevalent transplant patients in each renal unit is given in Figure 7.15. Mean serum creatinine and eGFR for the same populations are given in Table 7.4.
|
|
Commentary
An in-depth survey of the causes of variations between renal units in performance against the audit standard for serum bicarbonate concentration was reported in the 2004 Report [6]. Few of these causes of variation have been eliminated and the analyses reported here should, therefore, be interpreted with caution. However, more renal units than expected fall outside three standard deviations from the mean, suggesting that real differences in unit performance are present; it is recommended that those units whose data fall below the 3SD line review their practices relating to measurement of serum bicarbonate and to the correction of acidosis.
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionCompleteness of dataDialysis doseSerum bicarbonateReferences |
|---|
- Renal Association. Treatment of Adults and Children with renal failure. Standards and audit measures (2002) 3rd. Royal College of Physicians of London.
- Ansell D, Feest TG, eds. UK Renal Registry 5th Annual Report, 2002. 85–100. In Chapter 7: Adequacy of haemodialysis (urea reduction ratio).
- Ansell D. UK Renal Registry 6th Annual Report, 2003. 81–94. In Chapter 6: Adequacy of haemodialysis (urea reduction ratio).
- Saran R, Bragg-Gresham JL, Levin NW, et al. Longer treatment time and slower ultrafiltration in hemodialysis: associations with reduced mortality in the DOPPS. Kidney Int (2006) 69:1222–1228.[CrossRef][ISI][Medline]
- Di Filippo S, Andrulli S, Manzoni C, Corti M, Locatelli F. On-line assessment of dialysis dose. Kidney Int (1998) 54:263–267.[CrossRef][ISI][Medline]
- Ansell D, Feest TG, eds. UK Renal Registry 7th Annual Report, 2004. 59–68. In Chapter 6: Adequacy of haemodialysis and serum bicarbonate.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||