Chapter 7: Haemoglobin. Summary. Introduction. Inclusion criteria

Transcription

1 Chapter 7: Haemoglobin Summary Improvement in haemoglobin concentrations of patients receiving dialysis treatment continued in % of haemodialysis patients and 88% of peritoneal dialysis patients had a haemoglobin concentration above the Renal Association target of 10g/dl. Haemoglobin in the first quarter of dialysis treatment has also risen but 43% of individuals new to dialysis still had a haemoglobin <10g/dl in 2002 (45% in 2001). 63% of haemodialysis patients and 73% of peritoneal dialysis patients achieve a haemoglobin above the European guidelines of 11 g/dl. There is still wide variation between dialysis centres in proportion of new and prevalent patients who are anaemic. The variation in haemoglobin from year to year in small dialysis centres reflects the fluctuations in haemoglobin concentration of individuals receiving dialysis treatment. It is not possible to make judgements about quality of treatment in a centre on the basis of a single set of data. Introduction This chapter describes data reported to the Renal Registry relating to management of renal anaemia at the end of Correction of anaemia in individuals with chronic renal failure has been shown to improve quality of life and is likely to increase length of life. The importance of this aspect of patient care is reflected in the fact that both national and international standards have been set for management of renal anaemia. US and European guidelines set a target for an individual s haemoglobin of 11g/dl. The Renal Association have set a target haemoglobin of 10g/dl and this was confirmed in the latest addition of the Standards document published in August However, the third edition of the Renal Association Standards document has made changes to the standards relating to anaemia treatment. Whilst the target haemoglobin has not been altered the standard of 85% of patients on dialysis having a haemoglobin 10g/dl has been removed. The standard now states that : Individuals with CRF should achieve a haemoglobin of 10g/dl within 6 months of being seen by a nephrologist unless there is a specific reason why it could not be achieved. There is no longer a fixed benchmark against which renal centres can be compared. Instead, comparison with other centres through data submitted to the Renal Registry provides the guide to performance. The Renal Registry records data on patients receiving renal replacement therapy (RRT) and the date of start of RRT is recorded. At present the Registry cannot provide information on haemoglobin levels 6 months after seeing a nephrologist. Therefore, whilst the Registry data give information on the performance of renal centres with regard to prevalent dialysis patients and patients close to the start of dialysis, there is no information on whether centres are reaching the target within 6 months of first seeing a patient. Inclusion criteria Patients treated with dialysis during the last quarter of 2002 were included in the analy- 95

2 The UK Renal Registry The Sixth Annual Report sis if they had been on the same modality of dialysis in the same centre for 3 months. The latest available haemoglobin reading from each patient in the last quarter of 2002 was used in the analysis. Haemoglobin achievement by dialysis units The data describing the haemoglobin distribution in each centre are presented in Table 7.1 for haemodialysis (HD) and Table 7.2 for peritoneal dialysis (PD) and in Figures 7.1 and 7.2. The percentage with haemoglobin 11g/dl is included as this is the European standard and may be used by some centres. The percentage achieving the Renal Association target for haemodialysis patients was 82% in England and 84% in Wales, whilst for patients on PD this was 88% and 89% for England and Wales respectively. As described in previous reports there was considerable variation in performance. Two centres in 2002 had 27% of prevalent HD patients with a haemoglobin <10g/dl whilst Bradford achieved a haemoglobin of 10g/dl in 97% of haemodialysis patients and 11g/dl in 86%. Two centres (Sunderland and Clwyd) reported haemoglobin 10g/dl in % of PD patients whilst 2 centres appeared to be outliers with a low percentage of PD patients reaching the target (72% and 77% 10g/dl) Caution needs to be applied when interpreting data from smaller centres. High levels of achievement of the target were reported by a number of centres in last year s report that this year have considerably reduced. Southend reduced from 94% 10g/dl for HD in 2001 to 84% in 2002, Truro reduced from 91% in 2001 to 78% in 2002 whilst Sunderland increased from 67% in 2001 to % in It is unlikely that all of these variations reflect real changes in practice and they are more likely to be due to the inherent variability of haemoglobin in individuals with chronic renal failure. Such variation is hidden in the data from large centres but not when the number of data points is small. Individual centres will wish to investigate apparently significant changes in data year on year but it was not possible to make qualitative judgments about the standard of management on the basis of a single set of data. Overall the spread of data does not appear to be different this year compared to previous years. The % ranges for England and Wales have not changed. There is no evidence that centres are in general becoming more successful at targeting haemoglobin, which is perhaps not surprising given the observed variability in individual haemoglobin concentration. As in previous years a close relationship between median haemoglobin and percentage with haemoglobin greater than 10g/dl or 11g/dl is demonstrated for haemodialysis patients. Whilst less tight, the relationship has also held true for peritoneal dialysis patients in previous years, but in this years data the relationship is less clear. There is a suggestion that the percentage over target may reach a plateau around 95% 10g/dl. This could be an indication of the proportion of individuals on dialysis whose anaemia cannot be corrected, even with best management, together with those who become anaemic because of illness regardless of their pre-illness haemoglobin level. Haemoglobin concentrations of patients recently started RRT Haemoglobin concentrations in the first 3 months of starting dialysis have been analysed and the data are shown in Table 7.3 and Figures 7.11 to The haemoglobin data were extracted locally as the latest value in that quarter. The large range of percentage 10g/dl between centres shown in previous years has been maintained. In one centre as few as 33% of patients had haemoglobin 10g/dl in the first quarter whilst in other centres this was as high as %. 96

3 Chapter 7 Haemoglobin % data return Table 7.1. Haemoglobin data for patients on haemodialysis Median Hb g/dl % range Quartile range Mean Hb g/dl Standard deviation % with Hb 10 % with Hb 11 Bangr Bradf Bristl Camb Carls Carsh Clwyd Covnt Crdff Extr Glouc Guys H&C Heart Hull Ipswi Kings Leic LGI Livrpl Middlbr Newc Notts Oxfrd Plym Ports Prstn Redng Sheff Stevn Sthend StJms Sund Swnse Truro Wirrl Wolve Words Wrex York Eng Wls E&W

4 The UK Renal Registry The Sixth Annual Report % data return Table 7.2. Haemoglobin data for patients on peritoneal dialysis Median Hb g/dl % range Quartile Range Mean Hb g/dl Standard deviation % with Hb 10 % with Hb 11 Bangr Bradf Bristl Camb Carls Carsh Clwyd Covnt Crdff Extr Glouc Guys H&C Heart Hull Ipswi Kings Leic LGI Livrpl Middlbr Newc Notts Oxfrd Plym Ports Prstn Redng Sheff Stevn Sthend StJms Sund Swnse Truro Wirrl Wolve Words Wrex York Eng Wls E&W

5 Chapter 7 Haemoglobin 6 E&W 14 Wls 6 Eng 15 Clw yd 0 Bangr 5 Hull 4 Middlbr 10 Words 12 Guys 1 Glouc 32 Sw nse 2 Kings 2 Truro 6 Prstn 9 Heart 3 New c 6 Sund 14 Ports 1 Oxfrd 3 Redng 2 Leic 2 Extr 25 Camb 2 Wolve 6 Notts 14 Carsh 3 Covnt 0 Ipsw i 2 Sthend 6 Sheff 3 Livrpl 6 Stevn 8 Carls 3 LGI 14 Wrex 0 StJms 15 Plym 21 Wirrl 10 York 7 Crdff 0 Bradf < % 10% 20% 30% % % % % % % % Percentage of patients Figure 7.1. Distribution of haemoglobin in patients on HD 99

6 The UK Renal Registry The Sixth Annual Report 3 E&W 7 W ls 3 Eng 6 Hull 20 Swnse 14 Ports 0 E xtr 1 Kings 0 Leic 0 W ords 0 Newc 3 Glouc 2 Sheff 3 Covnt 0 Stevn 0 O xfrd 43 Sthend 4 Carls 8 Livrpl 3 N o tts 0 StJm s 0 Prstn 0 Redng 2 Carsh 0 Middlbr 3 Cam b 0 W olve 0 Bangr 0 Heart 1 Guys 0 York 2 Crdff 0 B ris tl 3 LGI 0 Bradf 6 W rex < >= T ruro 8 Plym 0 Ip s w i 12 Sund 9 Clwyd 0% 10% 20% 30% % % % % % % % Percentage of patients Figure 7.2. Distribution of haemoglobin in patients on PD 14.0 Haemoglobin g/dl Haemoglobin g/dl Upper quartile Median Lower quartile Bradf 21 Wirrl 7 Crdff 0 StJms 3 LGI 3 Livrpl 14 Wrex 15 Plym 2 Sthend 14 Carsh 10 York 3 Redng 6 Stevn 6 Notts 2 Wolve 6 Sheff 3 Covnt 2 Leic 1 Oxfrd 0 Ipswi 3 Newc 14 Ports 25 Camb 6 Prstn 2 Kings 4 Middlbr 1 Glouc 2 Extr 8 Carls 6 Sund 32 Swnse 0 Bangr 9 Heart 12 Guys 2 Truro 10 Words 5 Hull 15 Clwyd 6 Eng 14 Wls 6 E&W Figure 7.3. Haemoglobin and quartile ranges for HD patients

7 Chapter 7 Haemoglobin New patients HD at start of 2001 % of patients Bradf Crdff York Wirrl Plym StJms Wrex LGI Carls Stevn Livrpl Sheff Sthend Bristl H&C Ipswi Covnt Carsh Notts Wolve Camb Extr Leic Redng Oxfrd Ports Sund Newc Heart Prstn Truro Kings Swnse Glouc Guys Words Middlbr Hull Bangr Clwyd Eng Wls E&W Figure 7.4. Percentage of HD patients, by centre, achieving the Renal Association target Hb Haemoglobin g/dl g/dl Bradf 0 Middlbr 3 LGI 0 York 6 Wrex 0 Wolve 2 Carsh 4 Carls 0 Bangr 0 StJms 0 Stevn 43 Sthend 2 Crdff 8 Livrpl 0 Ipswi 1 Kings 8 Plym 3 Glouc 9 Clwyd 3 Camb 1 Guys 0 Oxfrd 0 Newc 0 Prstn 0 Words 3 Covnt 0 Leic 0 Redng 0 Truro 12 Sund 2 Sheff 3 Notts 0 Heart 14 Ports 20 Swnse 0 Extr 6 Hull 3 Eng 7 Wls 3 E&W Upper quartile Median Lower quartile Figure 7.5. Haemoglobin and quartile ranges for PD patients 95 % of patients Upper 95% Cl % with hb >=10 Lower 95% Cl 55 9 Clwyd 12 Sund 0 Ipswi 8 Plym 0 Truro 6 Wrex 0 Bradf 3 LGI 2 Crdff 0 York 1 Guys 0 Heart 0 Bangr 0 Wolve 3 Camb 0 Middlbr 2 Carsh 0 Redng 0 Prstn 0 StJms 3 Notts 8 Livrpl 43 Sthend 4 Carls 0 Oxfrd 0 Stevn 3 Covnt 2 Sheff 3 Glouc 0 Newc 0 Words 0 Leic 1 Kings 0 Extr 14 Ports 20 Swnse 6 Hull 3 Eng 7 Wls 3 E&W Figure 7.6. Percentage of HD patients, by centre, achieving the Renal Association target Hb 101

8 The UK Renal Registry The Sixth Annual Report % Hb > 10g/dL Median Hb % Hb > 11g/dL Median Hb Figure 7.7. Percentage patients with Hb >10g/dl and HB >11g/dl plotted against median Hb for HD patients % Hb > 10g/dL Percentage patients with Hb > 10 g/dl on peritoneal dialysis Median Hb % Hb > 11g/dL Percentage patients with Hb > 11 g/dl on peritoneal dialysis Median Hb Figure 7.8. Percentage patients with Hb >10g/dl and HB >11g/dl plotted against median Hb for PD patients % % of of patients Percentage haemoglobin > 11 g/dl : haemodialysis 0 Bradf 7 Crdff 15 Plym 21 Wirrl 0 StJms 3 LGI 2 Sthend 10 York 3 Livrpl 14 Carsh 3 Redng 6 Notts 6 Stevn 2 Wolve 14 Wrex 3 Covnt 0 Ipswi 8 Carls 6 Sheff 14 Ports 1 Oxfrd 6 Prstn 3 Newc 2 Extr 2 Leic 4 Middlbr 25 Camb 2 Kings 32 Swnse 1 Glouc 6 Sund 0 Bangr 9 Heart 12 Guys 5 Hull 2 Truro 10 Words 15 Clwyd 6 Eng 14 Wls 6 E&W Upper 95% Cl % with hb>=11 Lower 95% Cl Figure 7.9. Percentage of HD patients with haemoglobin >11g/dl 102

9 Chapter 7 Haemoglobin % of patients % of patients Upper 95% Cl 35 % with hb>=11 30 Lower 95% Cl 25 6 Wrex 0 Middlbr 12 Sund 3 LGI 0 Bradf 0 Ipswi 0 Redng 0 StJms 0 Wolve 2 Carsh 0 Bangr 8 Livrpl 8 Plym 0 York 0 Truro 2 Crdff 3 Glouc 0 Stevn 1 Guys 3 Camb 0 Newc 1 Kings 0 Oxfrd 0 Prstn 43 Sthend 0 Words 0 Heart 0 Leic 3 Covnt 4 Carls 3 Notts 2 Sheff 9 Clwyd 20 Swnse 14 Ports 0 Extr 6 Hull 3 Eng 7 Wls 3 E&W Figure Percentage of PD patients with haemoglobin >11g/dl 7 E&W 12 Wls 6 Eng 17 Plym 17 Wrex 10 Guys 2 Crdff 0 Redng 5 Carsh 22 Camb 7 Covnt 2 York 0 Oxfrd 12 Sund 3 Wolve 0 Ipsw i 2 Leic 8 Clw yd 7 Prstn 2 Truro 2 StJms 2 Glouc 4 Bangr 0 Carls 5 New c 10 Bradf 0 Notts 15 LGI 4 Words 3 Livrpl 3 Sthend 7 Stevn 4 Extr 8 Sheff 4 Kings 16 Ports 28 2 Heart 2 Middlbr 3 Hull 0% 10% 20% 30% % % % % % % % < > = 12.0 % patients Figure Distribution of haemoglobin for new patients 103

10 The UK Renal Registry The Sixth Annual Report Table 7.3. Haemoglobin levels for new patients starting dialysis % data return Median HB g/dl % range Quartile range % HB >10g/dl Bangr Bradf Bristl Camb Carls Carsh Clwyd Covnt Crdff Extr Glouc Guys H&C Heart Hull Ipswi Kings Leic LGI Livrpl Middlbr Newc Notts Oxfrd Plym Ports Prstn Redng Sheff Stevn Sthend StJms Sund Swnse Truro Wolve Words Wrex York Eng Wls E&W

11 Chapter 7 Haemoglobin 13.0 Haemoglobin g/dl UpperQuartile Median haemoglobin Lower Quartile 17 Wrex 2 Crdff 5 Carsh 2 York 17 Plym 10 Guys 8 Clwyd 3 Wolve 0 Ipswi 0 Oxfrd 15 LGI 22 Camb 7 Covnt 0 Carls 7 Prstn 0 Redng 2 Leic 2 Glouc 5 Newc 10 Bradf 2 Truro 4 Words 3 Livrpl 4 Bangr 2 StJms 0 Notts 12 Sund 3 Sthend 7 Stevn 4 Extr 8 Sheff 16 Ports 28 Swnse 2 Heart 3 Hull 4 Kings 2 Middlbr 6 Eng 12 Wls 7 E&W Figure Haemoglobin median and quartile range for new patients % Hb > 10g/dL Upper 95% CI % with hb >= 10 Lower 95% CI 0 17 Plym 17 Wrex 10 Guys 2 Crdff 0 Redng 5 Carsh 22 Camb 7 Covnt 2 York 0 Oxfrd 3 Wolve 12 Sund 0 Ipswi 2 Leic 8 Clwyd 2 Truro 7 Prstn 2 StJms 2 Glouc 4 Bangr 0 Carls 5 Newc 10 Bradf 0 Notts 4 Words 15 LGI 3 Livrpl 3 Sthend 7 Stevn 4 Extr 8 Sheff 4 Kings 16 Ports 28 Swnse 2 Heart 2 Middlbr 3 Hull 6 Eng 12 Wls 7 E&W Figure Percentage of new patients, by centre, achieving the Renal Association target % of patients Upper 95% Cl % with hb>=10 Lower 95% Cl 0 Bradf 7 Crdff 10 York 21 Wirrl 15 Plym 0 StJms 14 Wrex 3 LGI 8 Carls 6 Stevn 3 Livrpl 6 Sheff 2 Sthend 0 Ipswi 3 Covnt 14 Carsh 6 Notts 2 Wolve 25 Camb 2 Extr 2 Leic 3 Redng 1 Oxfrd 14 Ports 6 Sund 3 Newc 9 Heart 6 Prstn 2 Truro 2 Kings 32 Swnse 1 Glouc 12 Guys 10 Words 4 Middlbr 5 Hull 0 Bangr 15 Clwyd 6 Eng 14 Wls 6 E&W Figure Percentage of patients with haemoglobin >10g/dl: new and prevalent patients 105

12 The UK Renal Registry The Sixth Annual Report Figure 7.14 compares the first quarter haemoglobin value of new patients with the haemoglobin of prevalent patients. In some centres the percentage with target haemoglobin in the first quarter of dialysis is very close to that percentage in prevalent patients. In other centres there is a large gap. This variation is likely to reflect differences in pre-dialysis anaemia management but will also be influenced by proportions of patients referred late for treatment. The change in the Renal Association standard to require a haemoglobin 10g/dl in all patients whether on dialysis or pre-dialysis, within 6 months of being seen by a nephrologist may impact upon these variations in the future. Changes in anaemia management over time Every year that the Registry has reported, there has been an increase in median haemoglobin and an increase in percentage reaching the target haemoglobin, although the percentage increase has slowed (Figure 7.15). In haemodialysis patients 82.2% have a haemoglobin >10 g/dl compared with 81.4% at the end of For peritoneal dialysis patients this has increased from 86.5% >10 g/dl at the end of 2001, to 88.1% >10 g/dl at the end of In 1998 only % of patients starting RRT had a haemoglobin >10 g/dl (Figure 7.16) % > Hb > 10g/dL Haemodialysis Upper 95% CI % with hb >= 10 Lower 95% CI Peritoneal dialysis Figure Change in percentage of prevalent patients with Hb >10g/dl in E&W Haemoglobin > 10 g/dl New patients England & Wales Upper 95% CI % with Hb >= 10 Lower 95% CI % > Hb > 10 10g/dL New patients Prevalent dialysis patients Figure Change in percentage of patients starting RRT with Hb >10g/dl in E&W

13 Chapter 7 Haemoglobin compared with 57% in The years showed a dramatic increase, although the rate of this increase is now slowing. In this year s report, Chapter 16 analyses data on the late referral (seen by a nephrologist <3 months before starting renal replacement therapy) of patients. Late referral occurs in 30% of patients starting renal replacement therapy and analysis indicates this rate has remained unchanged over Further large improvements in haemoglobin of patients starting renal replacement therapy may rely on targeting late referral. Analysing these data by a cross-sectional basis on the 31st December each year (Figure 7.17), the time taken to increase haemoglobin can be seen. It is still taking 6 12 months for patients on haemodialysis to achieve maximum haemoglobin level. The Renal Standards document recommends these targets should be achieved within 6 months of seeing a nephrologist. Temporal changes in haemoglobin in different renal units Serial data are shown for those centres that have submitted data to the Registry since the first quarter As has previously been noted there is great variation in haemoglobin levels and proportion of patients achieving the target in small centres from one quarter to the next. This variation is much less obvious in larger centres HD 1999 PD 1999 HD 2000 PD 2000 HD 2001 PD 2001 HD 2002 PD 2002 Hb g/dl <=6 m onths 6 m - 1yr 1-2 yrs >2 years M edian all Years Figure Change in median Hb by length of time on RRT 107

14 The UK Renal Registry The Sixth Annual Report st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Lower quartile Median haemoglobin Upper quartile 7 Bradf Bristl Camb Carls Carsh Covnt Crdff Extr Glouc Guys Heart Hull Leic LGI Livrpl Middlbr st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Lower quartile Median haemoglobin Upper quartile Notts Oxfrd Plym Ports Prstn Redng Sheff Stevn Sthend StJms Sund Swnse Truro W olve W ords W rex York Eng&W Eng W ls Figure Median haemoglobin and quartile range January 2000 December 2002 for patients receiving haemodialysis Haemoglobin g/dl Haemoglobin g/dl 108

15 Chapter 7 Haemoglobin 1st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Upper 95% CI % with Hb >= 1 0 Lower 95% CI Bradf Bristl Camb Carls Carsh Covnt Crdff Extr Glouc Guys Heart Hull Leic LGI Livrpl Middlbr 1st point - Jan nd point - Jan rd point - Jan th point - D ec 2002 Upper 95% CI % with Hb >= 1 0 Lower 95% CI Notts Oxfrd Plym Ports Prstn Redng Sheff Stevn Sthend StJms Sund Sw nse Truro W olve W ords W rex Y ork Eng&W Eng W ls Figure Percentage haemoglobin >10g/dl January 2000 December 2002 for patients receiving haemodialysis % > Hb 10 g/dl % > Hb 10 g/dl 109

16 The UK Renal Registry The Sixth Annual Report st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Upper quartile Median haemoglobin Lower quartile Bradf Bristl Camb Carls Carsh Clwyd Covnt Crdff Extr Glouc Guys Heart Hull Leic LGI Livrpl Middlbr 1st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Upper quartile Median haem oglobin Lower quartile Notts Notts Oxfrd Plym Ports Prstn Prstn Redng Sheff Stevn Sthend Sthend StJms Sund Sw nse Truro Truro Wolve Words Wrex Y ork Y ork Eng&W Eng Wls Figure Median haemoglobin and quartile range January 2000 December 2002 for patients receiving peritoneal dialysis Haemoglobin g/dl Haemoglobin g/dl 110

17 Chapter 7 Haemoglobin 1st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Upper 95% CI % with Hb >= 10 Lower 95% CI Bradf Bristl Camb Carls Carsh Covnt Crdff Extr Glouc Guys Heart Hull Leic LGI Livrpl Middlbr 1st point - Jan nd point - Jan rd point - Jan th point - Dec 2002 Upper 95% CI % with Hb >= 10 Lower 95% CI Notts Oxfrd Plym Ports Prstn Redng Sheff Stevn Sthend StJms Sund Swnse Truro W olve W ords W rex York Eng&W Eng W ls Figure Percentage haemoglobin >10g/dl January 2000 December 2002 for patients receiving peritoneal dialysis % > Hb 10 g/dl % > Hb 10 g/dl 111

18 The UK Renal Registry The Sixth Annual Report Conclusion There has been a continued rise in the haemoglobin concentrations of dialysis patients and the proportion reaching the Renal Association target. There is some evidence that this rise may be reaching a plateau in peritoneal dialysis patients. There continues to be marked difference in haemoglobin concentrations between recently started patients and prevalent dialysis patients. Anaemia is unavoidable when patients present as uraemic emergencies but is also the result of both late referral to nephrologists and variations in predialysis anaemia management that could be improved. There is evidence of variation of haemoglobin levels in centres over time. This reflects the effect of fluctuating haemoglobin concentrations in individuals receiving dialysis treatment, which has also been identified in previous Registry reports. It is therefore not possible to make judgements about quality of treatment within a centre on the basis of a single set of data especially if it has relatively few dialysis patients. 112