Anaemia management and cardiomyopathy in renal failure

Rosemary L. Donne and Robert N. Foley

Department of Nephrology, Salford Royal Hospitals NHS Trust, Stott Lane, Salford M6 8HD, UK
Abstract
Patients with renal failure are at great cardiovascular
risk, with attributable death rates 1020 times those of
an age-matched population. Most patients develop
cardiomyopathy, with a continuum of left ventricu-
lar dilation (LV), hypertrophy and systolic dysfunc-
tion. Untreated, these conditions predispose to cardiac
failure, a dominant and highly lethal cardiovascular
syndrome in this population. Several prospective
observational studies have demonstrated anaemia to
be an independent risk factor for each step in the
process: haemodynamic overload, maladaptive LV
enlargement, LV burn-out and death. Recent evidence
suggests that physiological haemoglobin targets (e.g.
)12 gudl) may be optimal for maintaining cardiac
health and quality of life, especially in patients without
pre-existing clinical cardiac disease. Ongoing studies
should determine whether a physiologically targeted
approach to anaemia management reduces the burden
of cardiomyopathy in renal failure.
Keywords: anaemia; cardiac disease; chronic kidney
disease; epoetin; haemoglobin
Introduction
Two prototypes of LV enlargement occur in response
to chronic haemodynamic stress. In pressure overload,
e.g. as in essential hypertension or aortic stenosis,
thickening of the LV wall allows generation of greater
intraventricular pressure, necessary to overcome the
impediment to outow. This concentric left ventricular
hypertrophy (LVH) is characterized by normal or
decreased chamber volume. In volume overload, e.g. as
in anaemia or aortic regurgitation, an eccentric LVH
develops with lengthening of cardiac myobrils and
LV dilation, leading to augmented stroke volume,
according to the Starling mechanism. Both types of LV
enlargement are common in end-stage renal disease
(ESRD). Both LV mass and volume are predictors of
later mortality in ESRD: mass is prognostically import-
ant in concentric LVH, while volume is important in
LV dilation w1x.
Cardiovascular disease is a major determinant of
morbidity and mortality in patients with chronic renal
failure. After adjustment for age, gender and diabetes,
cardiovascular mortality in dialysis patients has been
estimated to be 1020 times higher than in the general
population. Cardiac failure has been shown repeatedly
to be of prime importance in these patients, on the basis
of incidence rates and adverse impact on survival w2x.
The contribution of anaemia as a haemodynamic risk
factor to the development of accelerated cardiomy-
opathy is likely to be substantial w2x. It has been known
for several years that most patients have echocardio-
graphic abnormalities at the initiation of dialysis,
including LVH, LV dilation and systolic dysfunction.
More recent evidence suggests that these abnormalities
develop more rapidly as renal function declines, typic-
ally years before ESRD occurs. This has been linked
most clearly to minor falls in haemoglobin, even within
ranges close to physiological for gender w3x. If these
relationships are causal, then avoidance of anaemia
would improve clinical outcomes in chronic kidney
disease. This hypothesis, as such, has not yet been
addressed in the literature. However, several studies
have focused on the change in clinical outcome, par-
ticularly cardiovascular risk and quality of life, associ-
ated with adjusting haemoglobin to more physiological
levels. In the remainder of this article, we will examine
these recently published treatment studies.
The United States Normal Hematocrit Trial
The rst and largest trial, which examined the con-
sequences of physiological vs subphysiological haemo-
globin levels, was the United States Normal Hematocrit
Trial w4x. This study involved 1233 haemodialysis
patients, in whom clinical evidence of congestive
heart failure or ischaemic heart disease were absolute
inclusion criteria. Patients were assigned randomly
Correspondence and offprint requests to: Dr R. N. Foley, Hope
Hospital, Stott Lane, Salford M6 8HD, UK. Tel: q44 161 787
5710, fax: q44 161 787 5713, e-mail: rfoley@hope.srht.nwest.nhs.uk
Nephrol Dial Transplant (2002) 17 [Suppl 1]: 3740
#
2002 European Renal AssociationEuropean Dialysis and Transplant Association
to treatment either with increasing doses of recom-
binant human erythropoietin (r-HuEPO, epoetin),
to achieve and maintain a haematocrit of 42%,
or with doses of epoetin sufcient to maintain a
haematocrit of 30% throughout the study. The primary
end-point was a composite of death or rst non-fatal
myocardial infarction. After 29 months, at the time of
the second interim analysis, 183 deaths and 19 rst
non-fatal myocardial infarctions were observed in the
normal target haematocrit group compared with 150
and 14, respectively, in the low target haematocrit
group. The study was halted at the third interim an-
alysis as it was felt unlikely that the primary end-point
would ever favour the higher target, given that at this
time an almost statistically signicant difference in
primary end-point was observed favouring the lower
target. The reasons for the observed discrepancy
between the two haematocrit groups was not clear,
but the higher haematocrit values themselves did not
appear to account for the disparate outcomes. Higher
rates of vascular access loss were clearly present in the
higher target haemoglobin group.
The Canadian Normalization of Hemoglobin Trial
The Canadian Normalization of Hemoglobin Trial
focused on haemodialysis patients with evidence of
concentric or eccentric LVH on echocardiography, but
without associated symptoms. A total of 146 anaemic
haemodialysis patients were assigned randomly to
haemoglobin target concentrations of either 10 or
13.5 gudl for a period of 48 weeks. In patients with
concentric LVH, the changes in LV mass index were
similar in the normal and low target haemoglobin
groups, but progressive LV dilation could be prevented
with a higher target (Figures 1 and 2). In the LV dila-
tion group, the changes in cavity volume and mass
were similar in both targets. Differences in quality of
life were observed, with normalization of haemoglobin
leading to improvements in fatigue, depression and
relationships w5x (Figure 3).
Cross-over trial comparing full and partial
anaemia correction in haemodialysis patients
This Australian study included 14 haemodialysis
patients who were treated with epoetin to increase
haemoglobin from a baseline concentration of 8.5 gudl
to target concentrations of either 10 or 14 gudl.
Patients were assigned randomly to one of the target
levels for 6 weeks before being crossed over to the
alternative target. Compared with the lower haemo-
globin target, the higher target led to a signicant
reduction in cardiac output and in LV end-diastolic
diameter (reductions of 21 and 8%, respectively). The
observed decrease of 5% in the LV mass index (LVMI)
was not statistically signicant. It remains a matter of
speculation that failure to see a difference in LVMI
reects the relatively short duration of the study, given
that other studies have suggested that it may take years
for benecial interventions to lead to LVH regression
in haemodialysis patients. Finally, quality of life, as
assessed by the Sickness Impact Prole, was found to
be enhanced with a higher haemoglobin level w6x.
Treating anaemia in severe, resistant congestive
heart failure
This was an uncontrolled study, novel by virtue of the
fact that it included a group of patients in whom
cardiac failure, and not renal failure, was the primary
medical problem. In the retrospective arm of the study,
anaemia (dened as a haemoglobin concentration of
-12 gudl) was a common nding in patients with
congestive heart failure. The prevalence of anaemia
Fig. 1. Changes in LVMI from baseline to 48 weeks in patients with concentric LVH. Hb, haemoglobin. (Reproduced with permission
from w5x.)
38 R. L. Donne and R. N. Foley
increased with the severity of cardiac failure, such that
the prevalence of anaemia reached 79% in those in
New York Heart Association Class IV. In the pros-
pective intervention study, 26 anaemic patients were
selected who had severe symptoms of cardiac failure,
despite maximal conventional therapy. These patients
were treated with subcutaneous epoetin and intrave-
nous iron to haemoglobin levels )12 gudl for a mean
period of 7.2 months. LV ejection fraction increased,
hospitalization rate fell by 91.9%, New York Heart
Association Class fell signicantly, as did the doses of
diuretics and the rate of decline of renal function w7x.
Conclusions
Oxygen delivery is a fundamental physiological need.
Anaemia, cardiac dysfunction and renal dysfunction
are intricately linked. Most patients have echocardio-
graphic abnormalities at the initiation of dialysis,
which primarily include LVH and dilation. More
recent evidence suggests that these abnormalities
develop more rapidly as renal function declines, and
have been linked to minor falls in haemoglobin, even
within ranges close to physiological w3x. The aim of this
review was to assess whether there are indications in
Fig. 2. Changes in LV cavity index from baseline to 48 weeks in patients with concentric LVH. Hb, haemoglobin. (Reproduced with
permission from w5x.)
Fig. 3. Quality of life in normal (target: 13.5 gudl) vs low (target: 10 gudl) haemoglobin (Hb) concentration groups. Fatigue was scored on a
Likert scale between 7 (no problem) and 1 (a severe problem). (Adapted from w5x.)
39 Anaemia, cardiomyopathy and renal failure
recently published treatment studies that avoidance of
anaemia is associated with improved clinical outcome,
particularly cardiomyopathy and quality of life.
None of the studies discussed above examined the
impact of the avoidance of anaemia, as they all inter-
vened long after anaemia developed. Nevertheless, of
the four studies discussed, three suggest that clinical
outcome is indeed improved when patients achieve
physiological haemoglobin targets. One study (the US
Normal Hematocrit Trial w4x) suggested an adverse
effect on clinical outcome with a normal haematocrit
compared with a low haematocrit. However, the dif-
ference in primary outcome between the two haem-
atocrit groups did not reach statistical signicance,
and the observed discrepancy was not clearly related
to haematocrit levels.
Overall, these ndings suggest that correction of
anaemia is benecial, particularly with respect to
cardiomyopathy. Currently ongoing controlled clinical
trials should conrm whether avoidance of anaemia,
i.e. early intervention and a physiologically targeted
approach, can indeed improve clinical outcome in
patients with chronic kidney disease.
References
1. Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray DC,
Barre PE. The prognostic importance of left ventricular geometry
in uremic cardiomyopathy. J Am Soc Nephrol 1995; 5: 20242031
2. Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of
cardiovascular disease in chronic renal disease. Am J Kidney Dis
1998; 32 wSuppl 3x: S112S119
3. Levin A, Thompson CR, Ethier J et al. Left ventricular mass
index increase in early renal disease: impact of decline in
hemoglobin. Am J Kidney Dis 1999; 34: 125134
4. Besarab A, Bolton WK, Browne JK et al. The effects of normal as
compared with low hematocrit values in patients with cardiac
disease who are receiving hemodialysis and epoetin. N Engl J Med
1998; 339: 584590
5. Foley RN, Parfrey PS, Morgan J et al. Effect of hemoglobin levels
in hemodialysis patients with asymptomatic cardiomyopathy.
Kidney Int 2000; 58: 13251335
6. McMahon LP, Mason K, Skinner SL, Burge CM, Grigg LE,
Becker GJ. Effects of haemoglobin normalization on quality of
life and cardiovascular parameters in end-stage renal failure.
Nephrol Dial Transplant 2000; 15: 14251430
7. Silverberg DS, Wexler D, Blum M et al. The use of subcutaneous
erythropoietin and intravenous iron for the treatment of the
anemia of severe, resistant congestive heart failure improves
cardiac and renal function and functional cardiac class and
markedly reduces hospitalizations. J Am Coll Cardiol 2000; 35:
17371744
40 R. L. Donne and R. N. Foley