RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
Chapter 4
Impaired Renal Function in Patients
with Ischemic and Non-Ischemic CHF;
Association with Neurohormonal Activation and Survival
Tom D.J. Smilde, Hans L. Hillege, Gerjan Navis,
Frans Boomsma, Dick de Zeeuw and Dirk J. van Veldhuisen
American Heart Journal 2004; 148: 165-172
37
CHAPTER 4
ABSTRACT
Background Renal dysfunction is a strong predictor of mortality in chronic heart failure
(CHF). Most CHF patients have atherosclerotic vascular disease, and several authors have
suggested that impaired renal function is only a marker of advanced atherosclerosis. We
compared renal function in patients with ischemic and non-ischemic CHF, and examined
associations with prognosis and extent of neurohormonal activation.
Methods In a large survival study (1906 patients), patients with documented coronary artery
disease (CAD, n=995), were compared to those with idiopathic dilated cardiomyopathy
(IDC, n=429). In a smaller substudy, plasma neurohormones were determined in 270 and
37 patients (CAD and IDC, respectively). All patients had advanced CHF (New York Heart
Association [NYHA] functional class III-IV). At baseline, mean age was 64±10 years, and
mean left ventricular ejection fraction (LVEF) 0.26±0.08. Baseline glomerular filtration
rate was calculated using the Cockcroft-Gault equation (GFRc).
Results GFRc was a strong predictor for mortality in both groups on multivariate analysis.
The relative risk was for IDC patients 3.04; P < 0.01(for the lowest quartile <53 mL/min)
and for CAD patients 1.81; P = 0.01(for the lowest quartile <42 mL/min). Plasma
neurohormones showed a relation with GFRc in both groups.
Conclusions GFRc is related to survival and plasma neurohormones in both patients groups.
In IDC patients this association appears to be at least as strong as in CAD patients.
38
RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
INTRODUCTION
Chronic heart failure (CHF) is characterised by a reduction in cardiac output, which leads
to an inability to maintain adequate tissue perfusion. This reduction in cardiac output leads
to a disproportionate decrease in kidney perfusion1 and might contribute to renal dysfunction.
Renal dysfunction is frequently observed in patients with CHF and is strongly correlated
with the severity and prognosis of this condition.2-4 It is thus tempting to speculate that the
severity of hemodynamic impairment is directly related to the degree of renal dysfunction
in CHF. However, it has also been suggested that renal insufficiency is merely a marker for
the severity of other risk markers, in particular the presence of generalized cardiovascular
disease.5 Indeed, it has been shown that CHF patients with coronary artery disease (CAD),
generally referred to as “ischemic” CHF, have a high incidence of renovascular disease.6
In order to examine this issue we evaluated, in a large study population7, patients with
CHF due to documented idiopathic dilated cardiomyopathy (IDC) and patients with CHF
due to documented CAD. We investigated the prognostic value of an estimated renal function
relative to traditional prognostic markers in both groups. Several studies have observed an
interplay between neurohormones, the severity of the disease, hemodynamic indexes, volume status, renal function and its prognostic properties in patients with CHF.8-11 Therefore,
we explored (in a subset of subjects) the association between renal function and a number
of neurohormonal parameters.
METHODS
Design and study population
All patients were included in the second Prospective Randomised study of Ibopamine on
Mortality and Efficacy (PRIME-II). Details of this study have been described elsewhere.7 In
summary, the study was performed to investigate the effect of the oral dopamine agonist
ibopamine on mortality in advanced CHF. Patient enrolment began in September 1992,
but the study was prematurely discontinued in August 1995 when a significant higher
fatality rate was observed in the ibopamine group than in the placebo group.7 The study
was conducted in 13 European countries, and 1,906 patients were included. Patients aged
from 18 - 80 years, with CHF, characterised by clinical signs and symptoms (New York
Heart Association functional [NYHA] class III to IV), and a radionuclide left ventricular
ejection fraction (LVEF) < 0.35, were eligible for the study. In a predefined sub-study,
which was performed in the Netherlands, blood for plasma neurohormone detection was
collected from 372 patients.12 Methods for neurohormones have been described in detail
previously.2;12 N-terminal brain natriuretic peptide (N-terminal BNP) was measured using
a radioimmunoassay with reagents including antibody, standards, and radiolabel. The assay
uses 50 µl of unextracted plasma and has a standard range of 60/1000 pmol.l-1. All samples
giving results of >900 pmol.l-1 were re-analysed in appropriate dilutions with physiological
salt. In 12 consecutive assays, variability was 14, 11, 4 and 4 % at concentrations of 131,
199, 293 and 901 pmol.l-1, respectively. Brain natriuretic peptide (BNP) was determined
by an immunoradiometric assay (Shionoria, Osaka, Japan).
39
CHAPTER 4
Definition of non-ischemic and ischemic CHF
In the original PRIME-II population (n=1906), aetiology was documented and categorised
in all patients.7 All original patients with “ischemic heart disease” (n=1025), were classified
as “ischemic” CHF in the present study only, if they had a documented history of myocardial
infarction (MI), as was reported before13, leaving a study population 995 patients. In this
CAD group 51% of the patients had angina pectoris, 3% had hypertension, 22% had
diabetes mellitus and 9% also had valvular heart disease. “Non-ischemic” CHF was defined
as follows: all original PRIME-II patients with “cardiomyopathy” (n=600) were taken, and
patients were excluded, in whom angina pectoris, hypertension, diabetes mellitus, and/ or
valvular disease was reported, leaving a population of 429 patients.
Renal function
In this study the glomerular filtration rate (GFR) was used as an indicator for the renal
function. As in the previous study performed from the PRIME-II2, we used the CockcroftGault equation (GFRc)14 as a reflection for the GFR. The GFRc, under steady-state
conditions, is estimated in this formula from the serum creatinine, and from the influence
of age and body weight on the production of serum creatinine. GFRc = [(140-age in years)
x (body weight in Kg)] / (72x serum creatinine in mg/dl). In women, the value of the GFRc
is multiplied by 0.85.
Statistical methods
We performed a retrospective, post-hoc analysis of the two selected patient populations
(IDC and CAD). Kaplan Meier methods and Cox regression analyses were used to study
the influence of baseline GFRc on survival in the study population. In order to determine
whether the effects of the GFRc, on the mortality, were independent, the statistical analysis
included adjustments for several possible risk factors, including sex, blood pressure, heart
rate, rhythm and concomitant medication. The effect of ibopamine on the survival and
other baseline characteristics that were prognostically relevant were also used in this analysis.
Continuous variables were modelled with indicator variables into quartiles, and relative
risks were calculated for the second, third and fourth quartile compared with the lowest
(reference) risk quartile. Test for trends are presented. The variables with P < 0.10 in the
univariate Cox regression were used in the multiple Cox regression analysis. Cumulative
relative risks were calculated within the sub-groups defined by GFRc strata with degree of
LVEF and NYHA class. Interaction terms were used to examine effect modification. To
reduce the risk of bias by the empirical use of arbitrary values for missing items of data, we
excluded observations with missing values for contributing variables in the multivariate
model.
Pearson and Spearman correlation coefficients were calculated to determine which plasma
neurohormone had an univariate correlation with the GFRc and LVEF. All reported
probability values were 2-tailed and a P < 0.05 was considered statistically significant. For
all statistical analysis SPSS version 11.0 was used.
40
RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
RESULTS
The baseline characteristics of the two study populations are depicted in Table 1. The IDC
group (n=429) was younger (60 vs. 66 years; P < 0.001) and tended to have a better renal
function (GFRc 74 vs. 58 mL/min; P < 0.001) than the patients with CAD (n=995). Atrial
fibrillation was more often present in IDC patients (29% vs. 22%; P = 0.008). There were
also slight differences in use of medication, in particular for digitalis, nitrates and calcium
channel blockers. Plasma neurohormones were generally similar between the two groups,
with the exception of norepinephrine, which was lower in IDC patients (414 vs. 575 pg/
mL; P = 0.009) than in CAD patients.
Renal function (GFRc) and overall mortality
During the follow up of 277 days (range, 0 to 1091 days) 77 of the 429 (18%) patients
with IDC and 248 of the 995 (25%) of patients with CAD died. In both groups, baseline
GFRc was strongly related to all cause mortality in an unadjusted univariate model.
In IDC patients, a marked stepwise increase in the cumulative incidence of mortality
for the lower quartiles of GFRc in IDC was observed (Figure 1). Table 2 shows the
multivariate Cox proportional hazards regression analysis of the predictors that were
significantly associated with mortality in IDC patients. In this group NYHA class was the
most powerful prognostic risk marker for all cause mortality, as expressed by the Wald
statistics. This was followed by GFRc, with a relative risk of 3.04 (P = 0.001) in the lowest
quartile (<53 mL/min) when compared with the group of patients in the highest quartile
(>90 mL/min). The LVEF and use of ACE-inhibitors were also significant predictors but less
powerful than the GFRc or NYHA class. In the multivariate model additional adjustments
for univariate prognostic variables for mortality, such as age, body weight, heart rate, serum
urea, intraventricular conduction disorders and furosemide dose were conducted, but were
nonsignificant and therefore not included.
Table 1. Baseline characteristics of the IDC and CAD populations.
Age (years)
IDC (n=429)
CAD (n=995)
P-value
59.6 ± 11.1
66.3 ± 8.3
<0.001
Death, all cause
77 (18%)
248 (25%)
0.001
Sex (male)
80 (19%)
158 (16%)
NS
Rhythm (Atrial fibrillation)
123 (29%)
160 (16%)
<0.001
NYHA class
III
III/IV
IV
270 (63%)
124 (29%)
35 (8%)
592 (69%)
320 (32%)
83 (8%)
NS
NS
NS
Evidence of heart failure
LVEF
LVEDD (mm)
0.25 ± 0.08
7.1 ± 0.95
0.26 ± 0.09
6.8 ± 0.96
NS
<0.001
41
CHAPTER 4
IDC (n=429)
Physical examination
Heart rate (bpm)
Systolic BP (mmhg)
Diastolic BP (mmhg)
Body weight (kg)
CAD (n=995)
P-value
83 ± 16
120 ± 18
76 ± 11
77 ± 15
79 ± 14
121 ± 19
74 ± 11
75 ± 12
<0.001
NS
<0.001
0.015
74 ± 30
58 ± 8
<0.001
139 ± 4
4±0
15 ± 28
111 ± 45
138 ± 4
4±1
14 ± 16
126 ± 46
0.042
NS
NS
<0.001
Medication
Digitalis
Diuretics
<80mg
>80mg
ACE inhibitor
Low dose
Moderate dose
High dose
Nitrates
Beta-blockers
Calcium Channel blockers
Antiarrhythmics
Anti-platelet agents
Ibopamine
330 (77%)
423 (99%)
285 (67%)
143 (33%)
404 (94%)
62 (15%)
223 (55%)
119 (29%)
115 (27%)
26 (6%)
13 (3%)
119 (28%)
52 (12%)
204 (48%)
546 (55%)
985 (99%)
588 (60%)
399 (40%)
907 (91%)
172 (17%)
492 (50%)
243 (25%)
571 (57%)
74 (7%)
104 (11%)
241 (24%)
397 (40%)
507 (51%)
<0.001
NS
Plasma neurohormones
Norepinephrine (pg/mL)
Epinephrine (pg/mL)
Dopamine (pg/mL)
Endotheline (pg/mL)
ANP (pmol/L)
N-terminal ANP (pmol/L)
BNP (pmol/L)
N-terminal BNP(pmol/L)
Aldosterone (pg/mL)
Renin (pmol/L)
414 (311-581)
42 (30-80)
16 (11-26)
5 (3-20)
88 (46-160)
946 (438-1542)
42 (11-105)
486 (151-779)
122 (64-236)
106 (22-256)
GFRc (mL/min)
Serum
Sodium (mmol/L)
Potassium (mmol/L)
Urea (mmol/L)
Creatinine (mmol/L)
575 (381-713)
40 (27-68)
20 (13-32)
5.3 (3.5-10.7)
103 (65-177)
1090 (630-1798)
57 (23-118)
594 (298-1110)
108 (65-219)
77 (37-209)
NS
<0.001
NS
<0.001
NS
<0.001
NS
0.009
0.040
NS
NS
NS
NS
NS
NS
NS
NS
Continuous variables are presented as mean ± SD. Plasma neurohormones are expressed as median (25 th –75 th
percentiles), and other variables are n (%).
ANP = atrial natriuretic peptide. BP = blood pressure. BNP = brain natriuretic peptide. CAD = coronary artery
disease. GFRc = glomerular filtration rate (Cockcroft-Gault equation). IDC = idiopathic dilated cardiomyopathy.
LVEDD = left ventricular end-diastolic diameter. LVEF = left ventricular ejection fraction. NS= not significant.
NYHA = New York Heart Association.
42
RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
In patients with CAD the GFRc was a less powerful prognostic risk marker for all cause
mortality, with a relative risk of 1.81 (P = 0.012) in the lowest quartile (<42mL/min) when
compared with the highest quartile (>73mL/min). In Figure 2 a marked stepwise increase
in the cumulative incidence of mortality is shown only in the lowest quartile of GFRc in
CAD. In Table 3 the multivariate Cox proportional hazards regression analysis of the
predictors that were significantly associated with mortality in CAD patients are shown.
Other variables in the multivariable model, which had an elevated risk of the lowest quartile
compared to the highest quartile, were urea, serum sodium, serum potassium and LVEF.
Also, use of digitalis, NYHA class, use bèta-blockers and use of ACE inhibitors were variables
included in the multivariable model. In this multivariate model also additional adjustments
for univariate prognostic variables for mortality, such as age, heart rate, body weight, systolic
blood pressure, diastolic blood pressure, diabetes mellitus, use of diuretics, furosemide dose
and intraventricular conduction disorders were conducted, but were nonsignificant and
therefore not included. All other medications, mentioned in Table 1, were in univariate
analysis not significantly related to mortality.
Figure 1. The proportional relationship of
GFRc in IDC with mortality in a multivariate
Cox-adjusted survival analysis.
Figure 2. The proportional relationship of
GFRc in CAD with mortality in a multivariate
Cox-adjusted survival analysis.
43
CHAPTER 4
Table 2. Stepwise Cox proportional multivariate hazard analysis of risk factors at baseline for overall mortality
in IDC patients.
IDC (n=429)
LVEF
GFRc (mL/min)
ACE-inhibitors
NYHA (class)
Range
RR-univariate
(95% CI)
RR-multivariate
(95% CI)
>0.30
0.25-0.29
0.21-0.24
<0.20
1.03(0.48-2.22)
2.49(1.23-5.06)
2.06(1.06-4.01)
1.20(0.55-2.65)
3.53(1.70-7.32)
2.50(1.26-4.97)
>90
69-90
53-69
<53
1.28(0.57-2.85)
1.76(0.83-3.72)
3.82(1.92-7.59)
0.87(0.36-2.10)
1.83(0.83-4.06)
3.04(1.43-6.44)
Yes
No
2.19(1.05-4.56)
2.16(1.01-4.66)
III
III/IV
IV
3.19(1.94-5.06)
2.86(1.36-6.02)
3.02(1.76-5.18)
4.36(1.97-9.61)
P-value
Wald
0.001
15.7
0.001
16.3
0.049
3.88
0.000
21.3
Continuous variables are presented in quartiles, with the range of each quartile.
CI = confidence interval. GFRc = glomerular filtration rate (Cockcroft-Gault equation). IDC = idiopathic
dilated cardiomyopathy. LVEF = left ventricular ejection fraction. NYHA = New York Heart Association. RR
= relative risk.
In Figure 3, the influence of affected GFRc on the distribution of relative risk for
mortality in both aetiologies of CHF are shown together. This Figure illustrates that in
IDC an increase of relative risk is present in a less impaired renal function in comparison to
CAD. According to National Kidney Foundation in chronic kidney disease the renal function
should be divided in a GFR above and below 60 mL/min.16 In both groups we divided the
GFRc according to these recommendations, and performed the same survival analysis. In
both groups a GFRc < 60 mL/min resulted in an increased risk for mortality (IDC RR;
2.77, P < 0.001 and in CAD RR; 1.41, P = 0.041).
In a secondary analysis, no interaction term was statistically significant in a multivariate
analysis, including ibopamine treatment and GFRc in patients with IDC (P = 0.343) and
CAD (P = 0.422).
There was no interaction observed between the GFRc, NYHA class and LVEF in either
IDC or CAD. In IDC an inverse correlation was observed between baseline GFRc and
NYHA class (r = -0.154, P = 0.002). No correlation was observed between baseline LVEF
and GFRc and between baseline LVEF and NYHA class (r = -0.012, P = 0.813 and r =
0.015, P = 0.773, respectively). In CAD an inverse correlation was found between NYHA
class and baseline GFRc or LVEF (r = -0.143, P < 0.001 and r = -0.099, P = 0.003). No
correlation was observed between baseline GFRc and LVEF (r = 0.011, P = 0.734).
Relationship of plasma neurohormones with GFRc
In a small, randomly selected, subset of patients with IDC (n=37) and with CAD (n=270),
44
RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
Table 3. Stepwise Cox proportional multivariate hazard analysis of risk factors at baseline for overall mortality
in CAD patients.
CAD (n=995)
LVEF
NYHA (class)
GFR c (mL/min)
Sodium (mmol/L)
Potassium (mmol/L)
Urea (mmol/L)
Digitalis
ACE-inhibitors
Beta-Blockers
Range
RR-univariate
(95% CI)
RR-multivariate
(95% CI)
>0.31
0.26-0.30
0.21-0.25
<0.20
0.87(0.58-1.31)
0.96(0.65-1.40)
1.50(1.07-2.11)
0.84(0.53-1.31)
0.95(0.63-1.43)
1.56(1.07-2.26)
III
III/IV
IV
1.50(1.14-1.96)
2.66(1.82-3.88)
1.20(0.87-1.65)
2.05(1.32-3.18)
>73
55-72
42-55
<42
1.16(0.77-1.74)
1.41(0.96-2.08)
2.59(1.81-3.72)
1.02(0.63-1.63)
1.07(0.68-1.71)
1.81(1.14-2.87)
>141
139-141
137-139
<136
1.25(0.84-1.87)
1.74(1.14-2.66)
2.55(1.73-3.76)
1.08(0.69-1.67)
1.43(0.89-2.27)
1.88(1.22-2.92)
>4.6
4.3-4.6
4.0-4.3
<4.0
0.70(0.49-1.01)
0.71(0.48-1.04)
1.27(0.91-1.77)
1.02(0.66-1.56)
1.21(0.78-1.87)
1.73(1.15-2.59)
<7.1
7.2-10.5
10.6-16.1
>16.1
1.32(0.84-2.06)
2.36(1.56-3.56)
2.89(1.94-4.32)
1.29(0.78-2.13)
2.14(1.33-3.44)
1.71(1.03-2.86)
1.81(1.39-2.36)
1.69(1.24-2.29)
Yes
No
Yes
No
2.14(1.51-3.02)
1.53(1.01-2.30)
Yes
No
3.34(1.38-8.11)
2.85(1.04-7.65)
P-value
Wald
0.01
11.40
0.01
10.23
<0.01
10.96
0.01
11.63
<0.05
10.14
<0.01
11.81
0.001
11.14
<0.05
4.02
<0.05
4.18
Continuous variables are presented in quartiles, with the range of each quartile.
CI = confidence interval. GFRc = glomerular filtration rate (Cockcroft-Gault equation). IDC = idiopathic dilated
cardiomyopathy. LVEF = left ventricular ejection fraction. NYHA = New York Heart Association. RR = relative
an extensive set of neurohormonal parameters was measured2. In comparison with normal
values most neurohormones were elevated, but epinephrine, dopamine and aldosterone
were within the normal range. Correlation coefficient’s for the neurohormones and GFRc
are presented in Table 4. In both groups, the strongest correlations were observed with the
four natriuretic peptides, followed by plasma norepinephrine. In CAD patients, but not in
IDC patients, statistically significant correlation’s were also observed for most neurohormones,
which was related to the larger study population.
45
CHAPTER 4
Figure 3. Relation, in quartiles, of GFRc from IDC and CAD to the risk of mortality using the multivariate
proportional hazards regression model.
Table 4. Relationship between neurohormones and GFRc in IDC and CAD patients.
GFRc (mL/min)
IDC (n=37)
ANP (pmol/L)
N-terminal ANP (pmol/L)
BNP (pmol/L)
N-terminal BNP (pmol/L)
Norepinephrine (pg/mL)
Epinephrine (pg/mL)
Endothelin (pg/mL)
Dopamine (pg/mL)
Aldosterone (pg/mL)
Renin (mU/mL)
CAD (n=270)
r
P-value
r
P-value
-0.520
-0.518
-0.678
-0.719
-0.359
-0.108
-0.030
-0.298
0.083
0.156
<0.01
<0.01
<0.01
<0.01
0.03
NS
NS
NS
NS
NS
-0.374
-0.489
-0.511
-0.595
-0.223
-0.004
-0.208
-0.236
-0.254
-0.241
<0.01
<0.01
<0.01
<0.01
<0.01
NS
<0.01
<0.01
<0.01
<0.01
ANP = atrial natriuretic peptide. BNP = brain natriuretic peptide. CAD = coronary artery disease. GFRc =
glomerular filtration rate (Cockcroft-Gault equation). IDC = idiopathic dilated cardiomyopathy. NS= not significant. r = correlation coefficient.
46
RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
DISCUSSION
The main finding of the present post-hoc analysis is, that renal function (as reflected by
GFRc) is associated with prognosis and with plasma neurohormonal activation (in particular
the natriuretic peptides), in both patients with ischemic and non-ischemic CHF.
Over the last couple of years many studies have been reported in which renal dysfunction
was associated with an increased mortality rate in CHF.2;3;15 The precise mechanism by
which an impaired renal function exerts an adverse prognostic effect is unclear but several
explanations have been suggested. In general, it is assumed that renal function impairment
reflects the impact of pathophysiological mechanisms that are relevant to prognosis in CHF
patients.
First, renal function impairment may reflect the severity of the disturbance of systemic
hemodynamics in CHF. The reduction in cardiac output leads to a reduction in renal
perfusion, resulting in renal dysfunction.1 Within this context our data on natriuretic peptides
are of interest, as they are closely related to the volume status.16;17 In response to the reduced
renal perfusion the kidney retains water and salt resulting in an elevated extracellular fluid
volume. The ensuing increase in ventricular volume and wall tension leads to secretion of
natriuretic peptides.18 Whereas it should be mentioned that our study did not directly assess
hemodynamics, the strong association of natriuretic peptides with GFRc in both populations
is consistent with a contribution of hemodynamic status and renal perfusion impairment in
renal dysfunction.
Second, renal dysfunction may reflect the extent of neurohumoral activation, which can
adversely affect prognosis in CHF as well as affect renal function. Both the vasoconstrictive
properties of norepinephrine and angiotensin II, and their pro-fibrotic properties can be
involved in the occurrence of renal dysfunction.19;20 In this study norepinephrine was
associated with GFRc in both groups. Regretfully, angiotensin II was not measured in this
study. Third, renal insufficiency might be merely a marker for the severity of the risk markers
and for the presence of cardiovascular and renovascular disease.21 In the majority of the
reported patient populations the cause of CHF was underlying atherosclerotic disease22;23,
therefore structural changes such as progressive nephrosclerosis may have contributed to
the lower GFR in CHF.24 This suggests that renal insufficiency is a marker for the severity
of these traditional markers for CAD, such as diabetes, hypertension or underlying
atherosclerosis.25;26 Our present data show that both in patients with high and patients with
low risk of atherosclerosis renal dysfunction was associated with an increased mortality.
This suggest that renal dysfunction is a risk marker in CHF that occurs independently of
the presence of atherosclerosis. Fourth, as possible causes for renal function impairment in
CHF the use of medication such as diuretics and ACE-inhibitors should be mentioned, in
particular as regards patients in whom renal perfusion is impaired already. Drug use, however,
was not an independent predictor of mortality in our study, so drug use does not seem to
explain the link between renal function and prognosis. Finally, it should be mentioned that
we cannot exclude the presence of sporadic cases of intrinsic renal parenchymal disease in
our population – as we have no data on urinalysis – but is unlikely that these possible
sporadic cases explain the prognostic value of renal function in our data set.
47
CHAPTER 4
Alternatively to renal function being a reflection of relevant aspects of the severity of
CHF, renal function impairment itself may induce unfavourable cardiovascular effects, such
as disturbed calcium and phosphate homeostasis with possible adverse cardiovascular effects27
and, moreover, renal function impairment itself may trigger neurohumoral activation.28
Clearly, our data prompt for further investigation of the mechanisms underlying the
prognostic value of renal function.
Renal function and mortality in IDC vs. CAD
In the IDC group, renal function was less impaired in comparison with the CAD group,
but IDC patients were also somewhat younger. Although the study was not designed to
answer this question the associated risk of mortality tended to be more equally distributed
over the quartiles in IDC patients than in CAD patients (Figure 1 and 2). In other words,
this might suggest, although differences were not statistically significant, that with a similar
degree of renal dysfunction, IDC patients had a more elevated risk as CAD patients (Figure
3). Therefore, it could be argued that patients with a non-ischemic cause of CHF are more
vulnerable to (early) alterations in volume status and/or increased neurohormonal activation.
The hypothesis is consistent with the strong correlation in IDC between GFRc and plasma
neurohormones, especially BNP and N-terminal BNP. Although such a correlation was
also present in CAD patients, it was less pronounced, but together it illustrates that impaired
renal function is closely related with volume status in patients of both aetiologies of CHF.
This hypothesis is supported by data from Volpe et al29, who found that patients with IDC
were unable to normally adjust their cardiac performance in response to volume loading.
Limitations of the study
A number of potential limitations may be identified in the present study. First, in our study
GFR was estimated by the Cockcroft-Gault (C-G) equation- that is, an indirect, creatininebased assessment of renal function. This and other equations were mainly validated in renal
populations instead of CHF populations. Thus, the possibility exists that some bias is present in the renal function estimate, because in CHF fluid retention and muscle wasting
affects the anthropometric assumptions underlying the C-G equation. However, none of
the other available renal function equations has been validated in CHF either. Whereas
accuracy of C-G is limited in the absence of clear-cut elevations in serum creatinine most
other available equations performed even worse.30 Second, this study provides only crosssectional observational data and therefore can only be used to generate new hypotheses. In
this respect, it must be noted that half of the patients were treated with ibopamine, which
showed an increased risk for mortality in the original study. The two patient groups were
not randomised for to this analysis and differences between the groups are present, such as
age, heart rate and medication use. These differences were corrected, but their true influence
may not have been adequately represented by the multivariate analysis. Finally, patients
were described as having non-ischemic CHF, but no cardiac catheterization was performed
to determine presence of coronary atherosclerosis. It is assumed that ischemia was not present, if the symptoms for ischemia or known risk markers for atherosclerosis were excluded.
Additionally, the possibility exist that renal atherosclerosis is present without the presence
of coronary atherosclerosis.
48
RENAL FUNCTION
IN
ISCHEMIC
AND
NON-ISCHEMIC ADVANCED HEART FAILURE
CONCLUSIONS
Renal function is related to survival and plasma neurohormones, especially natriuretic
peptides, in both patients with ischemic and non-ischemic CHF. In non-ischemic patients
these associations appear to be at least as strong as in ischemic patients.
Reference List
1.
Ljungman S, Lausch RN, Cody R. Role of the kidney in congestive heart failure. Relationship of cardiac
index to kidney function. Drugs. 1990;39:discussion 22-24.
2.
Hillege HL, Girbes AR, de Kam PJ, Boomsma F, de Zeeuw D, Charlesworth A, Hampton JR, van
Veldhuisen DJ. Renal function, neurohormonal activation, and survival in patients with chronic heart
failure. Circulation. 2000;102:203-210.
3.
Dries DL, Exner DV, Domanski MJ, Greenberg B, Stevenson LW. The prognostic implications of renal
insufficiency in asymptomatic and symptomatic patients with left ventricular systolic dysfunction. J Am
Coll Cardiol. 2000;35:681-689.
4.
McCullough PA, Soman SS, Shah SS, Smith ST, Marks KR, Yee J, Borzak S. Risks associated with renal
dysfunction in patients in the coronary care unit. J Am Coll Cardiol. 2000;36:679-684.
5.
McCullough PA. Why is chronic kidney disease the “spoiler” for cardiovascular outcomes? J Am Coll
Cardiol. 2003;41:725-728.
6.
MacDowall P, Kalra PA, O’Donoghue DJ, Waldek S, Mamtora H, Brown K. Risk of morbidity from
renovascular disease in elderly patients with congestive cardiac failure. Lancet. 1998;352:13-16.
7.
Hampton JR, van Veldhuisen DJ, Kleber FX, Cowley AJ, Ardia A, Block P, Cortina A, Cserhalmi L,
Follath F, Jensen G, Kayanakis J, Lie KI, Mancia G, Skene AM. Randomised study of effect of ibopamine
on survival in patients with advanced severe heart failure. Second Prospective Randomised Study of
Ibopamine on Mortality and Efficacy (PRIME II) Investigators. Lancet. 1997;349:971-977.
8.
Lee SC, Stevens TL, Sandberg SM, Heublein DM, Nelson SM, Jougasaki M, Redfield MM, Burnett JC,
Jr. The potential of brain natriuretic peptide as a biomarker for New York Heart Association class during
the outpatient treatment of heart failure. J Card Fail. 2002;8:149-154.
9.
Anand IS, Ferrari R, Kalra GS, Wahi PL, Poole-Wilson PA, Harris PC. Edema of cardiac origin. Studies
of body water and sodium, renal function, hemodynamic indexes, and plasma hormones in untreated
congestive cardiac failure. Circulation. 1989;80:299-305.
10. Kruger S, Graf J, Kunz D, Stickel T, Hanrath P, Janssens U. brain natriuretic peptide levels predict
functional capacity in patients with chronic heart failure. J Am Coll Cardiol. 2002;40:718-722.
11. Park MH, Scott RL, Uber PA, Harris BC, Chambers R, Mehra MR. Usefulness of B-type natriuretic
peptide levels in predicting hemodynamic perturbations after heart transplantation despite preserved left
ventricular systolic function. Am J Cardiol. 2002;90:1326-1329.
12. van Veldhuisen DJ, Boomsma F, de Kam PJ, Man i, V, Crijns HJ, Hampton JR, Lie KI. Influence of age
on neurohormonal activation and prognosis in patients with chronic heart failure. Eur Heart J. 1998;19:753760.
13. Felker GM, Shaw LK, O’connor CM. A standardized definition of ischemic cardiomyopathy for use in
clinical research. J Am Coll Cardiol. 2002;39:210-218.
14. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:3141.
15. Gottlieb SS, Abraham W, Butler J, Forman DE, Loh E, Massie BM, O’connor CM, Rich MW, Stevenson
LW, Young J, Krumholz HM. The prognostic importance of different definitions of worsening renal
function in congestive heart failure. J Card Fail. 2002;8:136-141.
49
CHAPTER 4
16. Finn WL, Gordon RD, Tunny TJ, Klemm SA, Hamlet SM. Effects of volume expansion and contraction
on plasma levels of atrial natriuretic peptide in man. Clin Exp Pharmacol Physiol. 1988;15:311-315.
17. Hara H, Ogihara T, Shima J, Saito H, Rakugi H, Iinuma K, Kumahara Y, Minamino T. Plasma atrial
natriuretic peptide level as an index for the severity of congestive heart failure. Clin Cardiol. 1987;10:437442.
18. Maeda K, Tsutamoto T, Wada A, Hisanaga T, Kinoshita M. Plasma brain natriuretic peptide as a biochemical
marker of high left ventricular end-diastolic pressure in patients with symptomatic left ventricular
dysfunction. Am Heart J. 1998;135:825-832.
19. Magri P, Rao MA, Cangianiello S, Bellizzi V, Russo R, Mele AF, Andreucci M, Memoli B, De Nicola L,
Volpe M. Early impairment of renal hemodynamic reserve in patients with asymptomatic heart failure is
restored by angiotensin II antagonism. Circulation. 1998;98:2849-2854.
20. Koobi P, Kalliovalkama J, Jolma P, Rysa J, Ruskoaho H, Vuolteenaho O, Kahonen M, Tikkanen I, Fan M,
Ylitalo P, Porsti I. AT1 receptor blockade improves vasorelaxation in experimental renal failure. Hypertension.
2003;41:1364-1371.
21. Tracy RE, Malcom GT, Oalmann MC, Newman WP, III, Guzman MA. Nephrosclerosis, glycohemoglobin,
cholesterol, and smoking in subjects dying of coronary heart disease. Mod Pathol. 1994;7:301-309.
22. McDonagh TA, Morrison CE, Lawrence A, Ford I, Tunstall-Pedoe H, McMurray JJ, Dargie HJ.
Symptomatic and asymptomatic left-ventricular systolic dysfunction in an urban population. Lancet.
1997;350:829-833.
23. Packer M, O’connor CM, Ghali JK, Pressler ML, Carson PE, Belkin RN, Miller AB, Neuberg GW, Frid
D, Wertheimer JH, Cropp AB, DeMets DL. Effect of amlodipine on morbidity and mortality in severe
chronic heart failure. Prospective Randomized Amlodipine Survival Evaluation Study Group. N Engl J
Med. 1996;335:1107-1114.
24. Ruilope LM. Kidney dysfunction: a sensitive predictor of cardiovascular risk. Am J Hypertens. 2001;14:213S217S.
25. Garg AX, Clark WF, Haynes RB, House AA. Moderate renal insufficiency and the risk of cardiovascular
mortality: results from the NHANES I. Kidney Int. 2002;61:1486-1494.
26. Culleton BF, Larson MG, Wilson PW, Evans JC, Parfrey PS, Levy D. Cardiovascular disease and mortality
in a community-based cohort with mild renal insufficiency. Kidney Int. 1999;56:2214-2219.
27. Locatelli F, Cannata-Andia JB, Drueke TB, Horl WH, Fouque D, Heimburger O, Ritz E. Management
of disturbances of calcium and phosphate metabolism in chronic renal insufficiency, with emphasis on the
control of hyperphosphataemia. Nephrol Dial Transplant. 2002;17:723-731.
28. Ligtenberg G, Blankestijn PJ, Oey PL, Klein IH, Dijkhorst-Oei LT, Boomsma F, Wieneke GH, van
Huffelen AC, Koomans HA. Reduction of sympathetic hyperactivity by enalapril in patients with chronic
renal failure. N Engl J Med. 1999;340:1321-1328.
29. Volpe M, Tritto C, DeLuca N, Rubattu S, Rao MA, Lamenza F, Mirante A, Enea I, Rendina V, Mele AF.
Abnormalities of sodium handling and of cardiovascular adaptations during high salt diet in patients with
mild heart failure. Circulation. 1993;88:1620-1627.
30. Bostom AG, Kronenberg F, Ritz E. Predictive performance of renal function equations for patients with
chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol. 2002;13:2140-2144.
50