Context Hypertension is a leading cause of end-stage renal disease (ESRD) in
the United States, with no known treatment to prevent progressive declines
leading to ESRD.
Objective To compare the effects of 2 levels of blood pressure (BP) control and
3 antihypertensive drug classes on glomerular filtration rate (GFR) decline
Design Randomized 3 × 2 factorial trial with enrollment from February
1995 to September 1998.
Setting and Participants A total of 1094 African Americans aged 18 to 70 years with hypertensive
renal disease (GFR, 20-65 mL/min per 1.73 m2) were recruited from
21 clinical centers throughout the United States and followed up for 3 to
Interventions Participants were randomly assigned to 1 of 2 mean arterial pressure
goals, 102 to 107 mm Hg (usual; n = 554) or 92 mm Hg or less (lower; n = 540),
and to initial treatment with either a β-blocker (metoprolol 50-200 mg/d;
n = 441), an angiotensin-converting enzyme inhibitor (ramipril 2.5-10 mg/d;
n = 436) or a dihydropyridine calcium channel blocker, (amlodipine 5-10 mg/d;
n = 217). Open-label agents were added to achieve the assigned BP goals.
Main Outcome Measures Rate of change in GFR (GFR slope); clinical composite outcome of reduction
in GFR by 50% or more (or ≥25 mL/min per 1.73 m2) from baseline,
ESRD, or death. Three primary treatment comparisons were specified: lower
vs usual BP goal; ramipril vs metoprolol; and amlodipine vs metoprolol.
Results Achieved BP averaged (SD) 128/78 (12/8) mm Hg in the lower BP group
and 141/85 (12/7) mm Hg in the usual BP group. The mean (SE) GFR slope from
baseline through 4 years did not differ significantly between the lower BP
group (−2.21 [0.17] mL/min per 1.73 m2 per year) and the
usual BP group (−1.95 [0.17] mL/min per 1.73 m2 per year; P = .24), and the lower BP goal did not significantly reduce
the rate of the clinical composite outcome (risk reduction for lower BP group
= 2%; 95% confidence interval [CI], −22% to 21%; P = .85). None of the drug group comparisons showed consistent significant
differences in the GFR slope. However, compared with the metoprolol and amlodipine
groups, the ramipril group manifested risk reductions in the clinical composite
outcome of 22% (95% CI, 1%-38%; P = .04) and 38%
(95% CI, 14%-56%; P = .004), respectively. There
was no significant difference in the clinical composite outcome between the
amlodipine and metoprolol groups.
Conclusions No additional benefit of slowing progression of hypertensive nephrosclerosis
was observed with the lower BP goal. Angiotensin-converting enzyme inhibitors
appear to be more effective than β-blockers or dihydropyridine calcium
channel blockers in slowing GFR decline.
Hypertension is reported to be the second leading cause of end-stage
renal disease (ESRD).1 African Americans are
6 times more likely to develop ESRD from hypertension than whites.2 Observational studies show a direct relationship between
the level of blood pressure (BP) and renal disease progression.3,4 Post
hoc analyses of clinical trials also suggest that lowering BP may retard progression
of renal disease and reduce cardiovascular risk.5-10 African
Americans, however, were not well represented in the aforementioned studies.5-7,9,11
Several studies document that African Americans with chronic kidney
disease have faster declines in renal function compared with whites with similar
BPs.12-14 In the
first trial to randomize patients to different BP levels and examine the outcome
on kidney disease progression, the Modification of Diet in Renal Disease trial,
a benefit of the lower BP goal (≤92 mm Hg) was suggested in the small subgroup
of 53 African Americans.15 However, whether
a lower BP goal actually retards progression of renal disease in African Americans
In trials that enrolled individuals with renal disease from diabetes
and other etiologies, angiotensin-converting enzyme inhibitors significantly
reduce progression of kidney disease. However, few African Americans were
included in such trials.21-24
Angiotensin-converting enzyme inhibitor use is lower in African Americans
with hypertension and chronic kidney disease compared with whites. This is
a consequence of many factors including a lack of clinical end point and safety
data and lower antihypertensive potency when they are used as monotherapy
compared with other classes of antihypertensive agents.
The African American Study of Kidney Disease and Hypertension (AASK)
prospectively addressed 2 questions in patients with hypertensive nephrosclerosis.25 First, does very aggressive lowering of BP result
in slower declines in kidney function? Second, does the type of antihypertensive
agent used to initiate BP lowering matter with regard to kidney disease outcomes?
The study design has been previously described.25,26 Briefly,
participants were self-identified African Americans with hypertension (n =
1094) who were aged 18 to 70 years with a glomerular filtration rate (GFR)
between 20 and 65 mL/min per 1.73 m2 and no other identified causes
of renal insufficiency. Exclusion criteria included diastolic BP of less than
95 mm Hg, known history of diabetes mellitus (fasting glucose, ≥140 mg/dL
or random glucose, >200 mg/dL), urinary protein to creatinine ratio of more
than 2.5, accelerated or malignant hypertension within 6 months, secondary
hypertension, evidence of non–BP-related causes of chronic kidney disease,
serious systemic disease, clinical congestive heart failure, or specific indication
for or contraindication to a study drug or study procedure. The protocol and
procedures were approved by the institutional review board at each center,
and all participants gave written informed consent. An independent data and
safety monitoring board was also established by the National Institute of
Diabetes and Digestive and Kidney Diseases.
Participant enrollment began in February 1995 and ended in September
1998. Figure 1 summarizes the numbers
of participants recruited, randomized, and followed up. Planned follow-up
to the end of the study in September 2001 was 3 to 6.4 years. On the recommendation
of the data and safety monitoring board, the amlodipine arm was halted in
September 2000,25 at which point patients randomized
to amlodipine were switched to open-label medication. The study's visit schedule,
including GFR measurements, was continued and patients in all 3 drug groups
remained on their randomly assigned BP goals through the end of the trial.
Based on a 3 × 2 factorial design, participants were randomized
equally to a usual mean arterial pressure goal of 102 to 107 mm Hg or to a
lower mean arterial pressure goal of 92 mm Hg or lower, and to treatment with
1 of 3 antihypertensive drugs (a sustained-release β-blocker, metoprolol,
50 to 200 mg/d; an angiotensin-converting enzyme inhibitor, ramipril, 2.5
to 10 mg/d; or a dihydropyridine calcium channel blocker, amlodipine, 5 to
10 mg/d). If the BP goal could not be achieved by the randomized drug, additional
open-labeled antihypertensives (furosemide, doxazosin, clonidine, and hydralazine
or minoxidil) were added sequentially.27 A
2:2:1 randomization ratio for the metoprolol, ramipril, and amlodipine groups
was used because comparisons involving amlodipine had increased power because
of a projected early increase in GFR from this medication.25,26 Participants
and investigators were masked to randomized drug but not BP goal.
Three primary treatment comparisons were specified: lower vs usual BP
goal, ramipril vs metoprolol, and amlodipine vs metoprolol. Results of ramipril
vs amlodipine, which was a secondary comparison, have been presented previously.25
Measurement of BP and Kidney Function
At each visit, 3 consecutive seated BP readings were measured using
a Hawksley random zero sphygmomanometer after at least 5 minutes rest,25,27 with the mean of the last 2 readings
recorded. The baseline BP readings were those obtained at the screening visits
prior to randomization. The follow-up BP measurements reported represent the
mean of all BPs measured within a given visit window, including those at interim
visits. The GFR was assessed by renal clearance of iodine I 125 iothalamate
at baseline twice, then at 3, 6, and every 6 months thereafter.28 Serum
and urinary levels of creatinine and protein were measured by a central laboratory
at 6-month intervals.
The primary analysis is based on the rate of change in GFR (GFR slope).
The GFR slope was determined separately during the first 3 months following
randomization (acute slope) and after 3 months (chronic slope). The acute
and chronic phases were distinguished because previous studies indicated that
the AASK interventions have acute effects on GFR that may differ from their
long-term effects on disease progression.25,29-35 The
chronic slope and the mean total slope from baseline (includes both the acute
and chronic phases) were designated as coprimary outcomes. The analysis plan
stipulated that a definitive benefit of a treatment intervention would be
inferred if it is shown to reduce the magnitude of both the chronic and total
mean slopes. The total slope assesses effects of interventions on kidney function
during the study period, while the chronic slope may better reflect long-term
The protocol also designated a main secondary clinical composite outcome,
which included any of the following1: a confirmed
reduction in GFR by 50% or by 25 mL/min per 1.73 m2 from the mean
of the 2 baseline GFRs; ESRD (dialysis or transplantation); or death. The
clinical composite outcome provided the principal assessment of patient benefit.
In contrast with the analysis of GFR slope, which addresses the mean change
in kidney function in all patients, including those with little or no progression,
the analysis of the clinical outcome is based on events of major clinical
relevance, either large declines in kidney function or death.
Urinary protein excretion, expressed as the urinary protein to creatinine
ratio from a 24-hour urine collection, was a secondary outcome. All cardiovascular
events including cardiovascular deaths and hospitalizations for myocardial
infarctions, strokes, heart failure, revascularization procedures, and other
hospitalized cardiovascular events were reviewed and classified by a blinded
end points committee according to a prespecified protocol.
Because of acute changes in GFR at discontinuation of amlodipine, data
from participants assigned to amlodipine were censored at termination of this
arm in September 2000. This required slightly different strategies for each
treatment group comparison. The BP group comparison retained all data through
the end of the study in the 80% of patients in the ramipril or metoprolol
groups but censored data on September 2000 for patients randomized to amlodipine
(giving a median GFR follow-up of 3.8 years). Data was retained to the end
of the study in both groups for the ramipril vs metoprolol comparison (median
GFR follow-up, 4.1 years), and data was censored in September 2000 in both
groups for the amlodipine vs metoprolol comparison (median GFR follow-up,
The primary renal function analysis was based on a mixed-effects model36,37 with random intercepts, acute slopes,
and chronic slopes, and with fixed effects for estimation of the mean acute,
chronic, and total slopes within each of the 6 cells in the 2 × 3 factorial
design, adjusting for clinical center and 5 prespecified baseline covariates:
proteinuria (log urinary protein to creatinine ratio), history of cardiovascular
disease, mean arterial pressure, sex, and age. The mean total slopes were
computed as time-weighted averages of the mean acute and chronic slopes, and
expressed from baseline to 3 years for the amlodipine vs metoprolol comparison
and from baseline to 4 years for the lower vs usual BP and ramipril vs metoprolol
comparisons. Because the effects of the BP and drug interventions were similar
at each level of the other intervention for both the chronic and total GFR
slopes (ie, no interaction between the BP groups and drug interventions),
we report analyses of the main effects for both interventions. Thus, the BP
group comparisons are averaged across the 3 drug groups according to the 2:2:1
randomization ratio, and the drug group comparisons are averaged equally across
the 2 BP groups.
The relationships of the treatment comparisons with baseline proteinuria
were investigated by adding continuous interaction terms between ln(urinary
protein to creatinine ratio) and the treatment groups.21,25 If
a statistically significant interaction was detected, the results were then
illustrated by subgroup analyses in participants with baseline urinary protein
to creatinine ratio of higher than 0.22 (n = 357) and 0.22 or less (n = 733).
The value of 0.22 corresponds to a urine protein excretion of approximately
300 mg/d and divides the two thirds of patients with the lowest proteinuria
from the one third with highest proteinuria in accordance with a heavy positive
skewness of the urinary protein to creatinine ratio. Since proteinuria was
inversely associated with GFR at baseline, the interaction of the treatment
groups with baseline GFR was also tested.
The effects of the interventions on the clinical composite outcome,
specific renal events, mortality, and secondary cardiovascular events were
each analyzed by Cox proportional hazards regression model with adjustment
for the same 5 covariates as the analysis of GFR slope. Baseline GFR was included
as an additional covariate in Cox proportional hazards regression models of
time to ESRD and time to ESRD or death. Participants were administratively
censored at loss-to-follow-up (9 patients) or else at the end of the study
or September 2000 by the same strategy used for the primary renal function
analysis. Because fewer participants were randomized to amlodipine than to
the other 2 groups, numbers of events are expressed as rates per patient-year.
Proportions of participants reporting symptoms during follow-up were compared
between treatment groups by logistic regression controlling for reported symptoms
All analyses are intent-to-treat and were performed using SAS versions
6.12 and 8 (SAS Institute Inc, Cary, NC). Two-sided P values
and 95% confidence intervals (CIs) are reported. This is conservative for
the primary analysis because both the chronic and total slopes comparisons
needed to reach significance for a definitive conclusion. To simplify the
presentation and maintain comparability of risk ratios, comparisons of amlodipine
with metoprolol are expressed as risk reductions for metoprolol relative to
amlodipine, although metoprolol was the reference group in the study design.
Based on 1094 patients and assuming a mean GFR slope of −4 mL/min
per 1.73 m2 per year in the usual BP group, the study was projected
to have 99%, 79%, and 87% power to detect a 30% reduction in GFR slope for
the BP comparison for analyses of the chronic slope, total slope, and clinical
composite outcome, respectively. Assuming a mean GFR slope of −4 mL/min
per 1.73 m2 per year in the metoprolol group, the projected power
was 88%, 99%, and 98% for these same 3 outcomes to detect a 30% reduction
in GFR slope for amlodipine vs metoprolol, and 97%, 69%, and 79% for ramipril
vs metoprolol, respectively. Based on the AASK pilot study and other studies,
the power calculations assumed a 2 mL/min per 1.73 m2 greater acute
GFR decline for the lower vs the usual BP goal, a 2 mL/min per 1.73 m2 greater acute GFR increase for amlodipine vs metoprolol, and a 2 mL/min
per 1.73 m2 per year greater acute decline for ramipril vs metoprolol.29,30,32,34
Baseline and Treatment Characteristics
Baseline patient characteristics were similar in the 2 BP groups and
the 3 drug groups (Table 1). Baseline
GFR was inversely associated with proteinuria (Spearman R, −0.46; P<.001).
After randomization, BP decreased from 152/96 to 128/78 mm Hg in the
lower BP group and from 149/95 to 141/85 mm Hg in the usual blood BP goal
group (Table 2). A mean separation
of approximately 10 mm Hg mean arterial pressure was maintained throughout
most of the follow-up period. Participants were prescribed more antihypertensives
for the lower than the usual BP goal, but there were no significant differences
between drug groups in the total number of antihypertensives or in the percentage
of participants receiving the highest doses of the randomized study drug.
Follow-up systolic BP was 2 mm Hg lower for amlodipine than for the
other drug groups; otherwise BP measurements were similar between the drug
groups. There was a slightly lower use of the fifth-line agent (minoxidil)
in the amlodipine group than in the metoprolol group, but the results of all
drug group comparisons of the primary and secondary outcomes were essentially
unchanged after controlling for follow-up mean arterial pressure and mean
number of add-on (levels 2 to 5) drugs as covariates.
Lower vs Usual BP. During the acute phase, the mean (SE) GFR decline was 1.82 (0.54) mL/min
per 1.73 m2 per 3 months greater in the lower BP than the usual
BP group (P<.001) (Figure 2 and Table 3).
However, mean GFR decline did not differ significantly between the lower and
usual BP groups during either the chronic phase (2.11 [0.16] vs 2.32 [0.17]
mL/min per 1.73 m2 per year; P = .33)
or the total follow-up period from baseline to 4 years (2.21 [0.17] vs 1.95
[0.17] mL/min per 1.73 m2 per year; P =
.24). The 95% CIs for the differences in mean GFR slope between the BP groups
were −0.21 to 0.64 mL/min per 1.73 m2 per year for the chronic
slope and −0.68 to 0.17 mL/min per 1.73 m2 per year for the
Ramipril vs Metoprolol. The mean GFR decline was slower in the ramipril than the metoprolol
group during the acute phase (0.23 [0.44] vs 1.73 [0.40] mL/min per 1.73 m2 per 3 months; P = .01) and the total follow-up
period from baseline to 4 years (1.81 [0.17] vs 2.42 [0.17] mL/min per 1.73
m2 per year; P = .007). However, the mean
chronic GFR slopes did not differ significantly between the 2 groups (1.87
[0.17] vs 2.12 [0.17] mL/min per 1.73 m2 per year; P = .26).
Amlodipine vs Metoprolol. GFR declined faster in the amlodipine than the metoprolol group during
the chronic phase (3.22 [0.33] vs 2.33 [0.20] mL/min per 1.73 m2 per
year; P = .02), but during the acute phase GFR increased
faster in the amlodipine vs the metoprolol group (4.03 [0.64] vs −1.73
[0.40] mL/min per 1.73 m2 per 3 months; P<.001).
Due to this acute effect, the total GFR decline to 3 years was significantly
slower in the amlodipine group than the metoprolol group (1.60 [0.34] vs 2.68
[0.20] mL/min per 1.73 m2 per year; P =
Clinical Composite Outcome Analysis
Lower vs Usual BP. The numbers of events (rate/participant year) for the main clinical
composite outcome (declining GFR events, ESRD, or death) were 173 (rate, 0.081)
and 167 (rate, 0.076) in the lower and usual BP groups. After adjustment for
the prespecified covariates, there were no significant differences between
the BP groups in the risk of the clinical composite outcome (risk reduction
for lower BP goal, 2%; 95% CI, −22% to 21%; P =
.85), the combined kidney end points of a declining GFR event or ESRD, the
combined hard clinical end points of ESRD or death, or ESRD alone (Table 4).
Ramipril vs Metoprolol. A total of 126 (rate, 0.069) and 155 (rate, 0.087) patients in the ramipril
and metoprolol groups reached the main clinical composite outcome during the
full follow-up period. The risk reduction for ramipril vs metoprolol was 22%
(95% CI, 1%-38%; P = .04). Similar risk reductions
of 21% to 22%, which were not statistically significant, were seen for ESRD
alone and for the combined end points of declining GFR events or ESRD, and
ESRD or death.
Metoprolol vs Amlodipine. A total of 117 (rate, 0.079) and 59 (rate, 0.082) patients in the metoprolol
and amlodipine groups reached the main clinical composite outcome by September
2000. There was no significant difference between the amlodipine and metoprolol
groups in the main clinical composite outcome (risk reduction for metoprolol
vs amlodipine, 20%; 95% CI, −10% to 41%; P =
.17) or in declining GFR events or ESRD combined. However, the metoprolol
group had a significantly lower risk than amlodipine for ESRD or death (P = .003) and for ESRD alone (P<.001).
Effects of Baseline Proteinuria and GFR
Baseline proteinuria was a strong predictor of GFR decline. For all
treatment groups combined, the mean (SE) chronic slope was −1.35 (0.15)
mL/min per 1.73 m2 per year if baseline urinary protein to creatinine
ratio was 0.22 or less compared with −4.09 (0.25) mL/min per 1.73 m2 per year if baseline urinary protein to creatinine ratio was higher
than 0.22 (P<.001).
Drug Group Comparisons. The difference in mean GFR decline between the amlodipine and metoprolol
groups was significantly related to baseline proteinuria for the acute and
total GFR slope (Figure 3A and 3C).
These interactions reflect the presence of a large acute increase in GFR with
amlodipine for participants with baseline urinary protein to creatinine ratio
of 0.22 or less but not in participants with urinary protein to creatinine
ratio higher than 0.22. The total decline in GFR to 3 years was 1.98 (SE,
0.43) slower for amlodipine than metoprolol if baseline urinary protein to
creatinine ratio was 0.22 or less, but was 1.29 (SE, 0.75) mL/min per 1.73
m2 per year faster for amlodipine than metoprolol if baseline urinary
protein to creatinine ratio was higher than 0.22. The interaction of baseline
proteinuria with the amlodipine vs metoprolol comparison was not significant
for either the chronic GFR slope or the clinical composite outcomes, although
the effects favoring metoprolol over amlodipine tended to be larger for higher
baseline proteinuria. For participants with baseline urinary protein to creatinine
ratio of higher than 0.22, the metoprolol group had risk reductions compared
with amlodipine of 38% (95% CI, 6%-59%; P = .03)
for the main composite outcome and 46% (95% CI, 15%-66%; P = .008) for ESRD or death.
Consistent with the association of higher GFR with lower proteinuria
at baseline, there were also significant interactions between baseline GFR
and the amlodipine vs metoprolol comparison for the acute (P = .003) and total GFR slopes (P<.001),
such that the amlodipine group had smaller mean GFR declines compared with
metoprolol for patients with higher baseline GFR but larger mean GFR declines
compared with metoprolol for patients with lower baseline GFR.
The level of baseline proteinuria did not influence the comparison of
ramipril to metoprolol (interaction P = .51, .32,
and .61 for the chronic slope, total slope, and main clinical composite outcome,
BP Group Comparison. The BP group comparison also depended significantly on the level of
baseline proteinuria for the acute slope, total slope, and main clinical composite
outcome (P = .007) but not for the chronic slope
(Figure 3B and 3D). For each outcome,
there were slight trends that tended to favor the lower BP goal over the usual
goal in participants with higher proteinuria and opposite trends in participants
with little or no proteinuria. However, with the exception of the acute slope,
the BP comparison for the aforementioned outcomes was not significantly different
within either the lower (baseline urinary protein to creatinine ratio ≤0.22)
or higher (baseline urinary protein to creatinine ratio >0.22) proteinuria
strata. There was a corresponding trend for an interaction of the BP-group
comparison with baseline GFR for the total GFR slope (P = .07) favoring the usual goal over the lower goal for patients with
higher baseline GFR with the opposite pattern for patients with lower baseline
GFR (data not shown).
Proteinuria (geometric mean urinary protein to creatinine ratio) increased
by 58% for the amlodipine group and declined by 14% in the metoprolol group
between baseline and 6 months (P<.001) (Figure 4). Proteinuria increased by 7% in
the usual BP group and decreased by 17% in the lower BP group during the first
6 months. These differences between treatment groups persisted throughout
the study. Follow-up proteinuria was slightly lower in the ramipril than the
metoprolol group but not significantly (P = .06 for
the comparison of total change over 4 years).
There were no significant differences in all-cause mortality, cardiovascular
mortality, or first cardiovascular events (defined as cardiovascular mortality
or first cardiovascular hospitalizations) between the treatment groups (Table 5). Proportions of patients reporting
adverse symptoms (including hypotensive symptoms) were similar in the 2 BP
groups. The proportions of participants reporting angioedema and cough were
highest in the ramipril group, although the proportion reporting edema was
higher in the amlodipine group. Hyperkalemia was reported for 3 participants
randomized to the ramipril group and 1 randomized to metoprolol.
The AASK is the first published large-scale trial to our knowledge that
examines both the effect of 3 different antihypertensive regimens as well
as the effect of 2 BP goals on decline in kidney function in a population
with chronic kidney disease attributed to hypertensive nephrosclerosis.38
Treatment of study participants to a lower than usual mean BP of 128/78
mm Hg did not significantly reduce either the mean rate of GFR decline or
the risk of the clinical composite outcome compared with usual BP goal with
a mean achieved BP of 141/85. The AASK, with its larger sample size and wider
BP separation, extends previous negative findings regarding the level of BP
reduction and change in GFR observed in smaller samples of both African Americans
and non–African Americans with nonproteinuric kidney disease.7,39,40
The average rate of decline in GFR in both treatment groups was approximately
2 mL/min per 1.73 m2 per year. This average rate of GFR decline
is similar to or slower than earlier trials of hypertensive nephrosclerosis40 and slower than other common progressive kidney diseases.7,41,42 The relatively slow
mean GFR decline reduced the power of the primary analysis of GFR slope. Nonetheless,
the upper limits of the 95% CIs for the BP comparison exclude a risk reduction
for the lower BP goal larger than 21% for the clinical composite outcome and
31% for ESRD alone. While a benefit smaller than these limits cannot be excluded,
the upper confidence limits are substantially smaller than the effects that
would be estimated from observational studies given the large separation in
BP that was achieved between the AASK BP groups.43 Because
randomized comparisons more accurately evaluate causal relationships,44,45 this discrepancy suggests that relationships
observed between BP level and rates of ESRD in nonrandomized studies have
overestimated the effect of lowering BP.
Mean BP during follow-up in the usual BP group was 141/85 mm Hg, which
is similar to the level recommended to prevent cardiovascular target organ
damage and is less than that achieved by more than 70% of individuals being
treated for hypertension.46 This study's finding
of a failure to further slow progression of kidney disease by reducing BP
below this level does not diminish the importance of maintaining BP in accordance
with the current guidelines.18 We do not interpret
the apparent lack of an effect of the lower BP goal to slow decline in GFR
(and reduce risk for clinical end points) to illustrate that BP lowering is
not important for preserving kidney function. Our study did not test the hypothesis
that treatment vs no treatment of hypertension preserves kidney function.
Nevertheless, our data suggest that once BP is lowered to a given level, additional
risk factors are important in patients with chronic kidney disease resulting
Although there was no significant effect of the BP intervention on GFR
slope or clinical events in all patients or in subgroup analyses by baseline
proteinuria strata, there were significant interactions with a trend favoring
the lower BP goal in participants with higher baseline proteinuria and an
opposite trend in participants with little or no proteinuria. This is consistent
with the Modification of Diet in Renal Disease results that showed a favorable
trend for the lower BP goal in participants with baseline proteinuria of higher
than 1 gram per day but not at lower levels of proteinuria.8 However,
because proteinuria was inversely correlated with GFR at baseline, it is possible
that the dependence of the BP comparison on baseline proteinuria in the AASK
reflects a larger hemodynamic effect in patients with higher baseline GFR
rather than true differences in clinically relevant outcomes. This study was
not powered to detect differences in the rate of myocardial infarction, stroke,
or death. However, we found no evidence of differences in the rates of these
events between the randomized BP groups.
The primary analysis of GFR slope did not establish a definitive difference
among the 3 drug regimens. However, significant benefits of ramipril vs metoprolol
(reported here) and amlodipine25 on the main
clinical composite outcome and the results of other secondary analyses suggest
that ramipril slows hypertensive kidney disease progression compared with
the other 2 regimens. Secondary analyses also suggest that metoprolol may
improve renal outcome compared with amlodipine, particularly in participants
with higher proteinuria.
Comparisons of amlodipine with the other drug groups were complicated
by a large acute increase in GFR for amlodipine in the 3 months after randomization.
Due to this acute effect, which was likely a hemodynamic response without
clinical significance, beneficial effects of ramipril and metoprolol vs amlodipine
on GFR decline after 3 months did not lead to corresponding beneficial effects
on the total mean slope from baseline to the end of the study (Figure 2 and Figure 3).
However, compared with amlodipine, ramipril significantly reduced the risk
of the main clinical composite,21 and both
ramipril and metoprolol reduced the risk of ESRD and of ESRD and death combined
(Table 4). The latter 2 outcomes
were probably less sensitive to the acute effect, because they are based on
clinical end points independent of GFR measurement. In the subgroup of patients
with baseline urinary protein to creatinine ratio of more than 0.22 (urinary
protein, 300 mg/d), the acute effect was negligible and each of the slope-based
and time-to-event outcomes were in agreement, indicating consistent advantages
for ramipril and metoprolol vs amlodipine.
The AASK was designed to compare 3 active drug regimens and did not
have a placebo control. In a placebo controlled trial of participants with
diabetic nephropathy and proteinuria that included an amlodipine arm, however,
no difference was noted between placebo and amlodipine on ESRD, death, or
doubling of serum creatinine, and trends in proteinuria change were the same
as AASK for amlodipine.47,48
In contrast with the comparisons involving amlodipine, the evidence
for benefit of ramipril vs metoprolol was noted in the full AASK cohort, irrespective
of baseline proteinuria. However, the conclusion of the beneficial effect
of ramipril compared with metoprolol is less definitive because the chronic
slope was not significant. Several clinical trials of participants with proteinuria
and primary glomerular disease show beneficial effects of ramipril.49 Data from AASK23,50 extend
these results to participants with hypertensive glomerulopathy and minimal
proteinuria.51,52 Evidence that
angiotensin-converting enzyme inhibitors and angiotensin receptor blockers
lower BP to a lesser extent in African Americans than others, when used as
monotherapy, taken together with the paucity of prospective clinical end point
data, has resulted in less use of such agents in African Americans.2,51,52 The AASK is the first
outcome trial to demonstrate a renoprotective effect of angiotensin-converting
enzyme inhibitor in an African American population.
We conclude that although BP reduction to levels below current guidelines
for cardiovascular risk reduction are achievable, our results do not support
additional reduction as a strategy to prevent progression of hypertensive
nephrosclerosis. Our results do support recommendations that angiotensin-converting
enzyme inhibitors should be considered as first line therapy over β-blockers
and dihydropyridine calcium channel blockers in these patients. Moreover, β-blockers
may be more effective than dihydropyridine calcium channel blockers in slowing
progression among patients with proteinuria.
US Renal Data System. USRDS 1999 Annual Data Report. Bethesda, Md: National Institutes of Health, National Institute of
Diabetes and Digestive and Kidney Diseases; 1999. Available at: http://www.usrds.org/chapters/begin.pdf. Accessibility verified October 3, 2002.
Rahman M, Douglas JG, Wright JT. Pathophysiology and treatment implications of hypertension in the African-American
population. Endocrinol Metab Clin North Am.1997;26:125-144.Google Scholar
Klag MJ, Whelton PK, Randall BL.
et al. Blood pressure and end-stage renal disease in men. N Engl J Med.1996;334:13-18.Google Scholar
Whelton PK, Perneger TV, He J, Klag MJ. The role of blood pressure as a risk factor for renal disease. J Hum Hypertens.1996;10:683-689.Google Scholar
National High Blood Pressure Education Program Working Group. 1995 Update of the working group reports on chronic renal failure and
renovascular hypertension. Arch Intern Med.1996;156:1938-1947.Google Scholar
Woods JW, Blythe WB, Huffines WD. Management of malignant hypertension complicated by renal insufficiency. N Engl J Med.1974;291:10-14.Google Scholar
Klahr S, Levey AS, Beck GJ.
et al. The effects of dietary protein restriction and blood-pressure control
on the progression of chronic renal disease. N Engl J Med.1994;330:877-884.Google Scholar
Peterson JC, Adler S, Burkart JM.
et al. Blood pressure control, proteinuria, and the progression of renal disease. Ann Intern Med.1995;123:754-762.Google Scholar
Tight blood pressure control and risk of macrovascular and microvascular
complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group. BMJ.1998;317:703-713.Google Scholar
Klag MJ, Whelton PK, Randall BL.
et al. End-stage renal disease in African-American and white men: 16-year
MR-FIT findings. JAMA.1997;277:1293-1298.Google Scholar
Peterson JC, Adler S, Burkart JM.
et al. Blood pressure control, proteinuria, and the progression of renal disease. Ann Intern Med.1995;123:754-762.Google Scholar
Shulman NB, Ford CE, Hall WD.
et al. Prognostic value of serum creatinine and effect of treatment of hypertension
on renal function. Hypertension.1989;13:I80-I93.Google Scholar
Rostand SG, Brown G, Kirk KA, Rutsky EA, Dustan HP. Renal insufficiency in treated essential hypertension. N Engl J Med.1989;320:684-688.Google Scholar
Walker WG, Neaton JD, Cutler JA, Neuwirth R, Cohen JD. Renal function change in hypertensive members of the Multiple Risk
Factor Intervention Trial. JAMA.1992;268:3085-3091.Google Scholar
Striker GE. Kidney disease and hypertension in blacks. Am J Kidney Dis.1992;20:673.Google Scholar
Psaty BM, Furberg CD, Kuller LH.
et al. Association between blood pressure level and the risk of myocardial
infarction, stroke, and total mortality. Arch Intern Med.2001;161:1183-1192.Google Scholar
Stamler J, Stamler R, Neaton JD.
et al. Low risk-factor profile and long-term cardiovascular and noncardiovascular
mortality and life expectancy: findings for 5 large cohorts of young adult
and middle-aged men and women. JAMA.1999;282:2012-2018.Google Scholar
The sixth report of the Joint National Committee on prevention, detection,
evaluation, and treatment of high blood pressure. Arch Intern Med.1997;157:2413-2446.Google Scholar
Bakris GL, Williams M, Dworkin L.
et al. Preserving renal function in adults with hypertension and diabetes. Am J Kidney Dis.2000;36:646-661.Google Scholar
Hansson L, Zanchetti A, Carruthers SG.
et al. Effects of intensive blood pressure lowering and low dose aspirin in
patients with hypertension: principal results of the Hypertension Optimal
Treatment (HOT) randomized trial. Lancet.1998;351:1755-1762.Google Scholar
Hebert LA, Kusek JW, Greene T.
et al. Effects of blood pressure control on progressive renal disease in blacks
and whites. Hypertension.1997;30:428-435.Google Scholar
Bakris GL, Mangrum A, Copley JB, Vicknair N, Sadler R. Effect of calcium channel or beta-blockade on the progression of diabetic
nephropathy in African Americans. Hypertension.1997;29:744-750.Google Scholar
Jafar TH, Schmid CH, Landa M.
et al. Angiotensin-converting enzyme inhibitors and progression of nondiabetic
renal disease: a meta-analysis of patient-level data. Ann Intern Med.2001;135:73-87.Google Scholar
Randomised placebo-controlled trial of effect of ramipril on decline
in glomerular filtration rate and risk of terminal renal failure in proteinuric,
non-diabetic nephropathy. Lancet.1997;349:1857-1863.Google Scholar
Agodoa LY, Appel L, Bakris GL.
et al. Effect of ramipril vs amlodipine on renal outcomes in hypertensive
nephrosclerosis: a randomized controlled trial. JAMA.2001;285:2719-2728.Google Scholar
Wright Jr JT, Kusek JW, Toto RD.
et al. Design and baseline characteristics of participants in the African
American Study of Kidney Disease and Hypertension (AASK) Pilot Study. Control Clin Trials.1996;17(suppl 4):3S-16S.Google Scholar
Wright Jr JT, Agodoa L, Contreras G.
et al. Successful blood pressure control in the African American Study of
Kidney Disease and Hypertension. Arch Intern Med.2002;162:1636-1643.Google Scholar
Lewis J, Agodoa L, Cheek D.
et al. Comparison of cross-sectional renal function measurements in African
Americans with hypertensive nephrosclerosis and of primary formulas to estimate
glomerular filtration rate. Am J Kidney Dis.2001;38:744-753.Google Scholar
Hall WD, Kusek JW, Kirk KA.
et al. Short-term effects of blood pressure control and antihypertensive drug
regimen on glomerular filtration rate: the African American Study of Kidney
Disease and Hypertension Pilot Study. Am J Kidney Dis.1997;29:720-728.Google Scholar
Bakris GL, Copley JB, Vicknair N, Sadler R, Leurgans S. Calcium channel blockers versus other antihypertensive therapies on
progression of NIDDM associated nephropathy. Kidney Int.1996;50:1641-1650.Google Scholar
De Cesaris R, Ranieri G, Filitti V, Andriani A, Bonfantino MV. Effects of atenolol and enalapril on kidney function in hypertensive
diabetic patients. J Cardiovasc Pharmacol.1993;22:208-214.Google Scholar
ter Wee PM, De Micheli AG, Epstein M. Effects of calcium antagonists on renal hemodynamics and progression
of nondiabetic chronic renal disease. Arch Intern Med.1994;154:1185-1202.Google Scholar
Bakris GL, Weir MR. Angiotensin-converting enzyme inhibitor-associated elevations in serum
creatinine: is this a cause for concern? Arch Intern Med.2000;160:685-693.Google Scholar
Modification of Diet in Renal Disease Study Group. Short-term effects of protein intake, blood pressure, and antihypertensive
therapy on glomerular filtration rate in the Modification of Diet in Renal
Disease Study. J Am Soc Nephrol.1996;7:2097-2109.Google Scholar
Ruggenenti P, Perna A, Gherardi G.
et al. Renoprotective properties of ACE-inhibition in non-diabetic nephropathies
with non-nephrotic proteinuria. Lancet.1999;354:359-364.Google Scholar
Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics.1982;38:963-974.Google Scholar
Diggle PJ, Liang K, Zeger SL. Analysis of Longitudinal Data. Oxford, England: Clarendon Press; 1994.
Fogo A, Breyer JA, Smith MC.
et al. Accuracy of the diagnosis of hypertensive nephrosclerosis in African
Americans: a report from the African American Study of Kidney Disease (AASK)
Trial. Kidney Int.1997;51:244-252.Google Scholar
Estacio RO, Jeffers BW, Gifford N, Schrier RW. Effect of blood pressure control on diabetic microvascular complications
in patients with hypertension and type 2 diabetes. Diabetes Care.2000;23(suppl 2):B54-B64.Google Scholar
Toto RD, Mitchell HC, Smith RD.
et al. "Strict" blood pressure control and progression of renal disease in
hypertensive nephrosclerosis. Kidney Int.1995;48:851-859.Google Scholar
Lewis EJ, Hunsicker LG, Bain RP.
et al. The effect of angiotensin-converting-enzyme inhibition on diabetic
nephropathy. N Engl J Med.1993;329:1456-1462.Google Scholar
Maschio G, Alberti D, Janin G.
et al. Effect of the angiotensin-converting-enzyme inhibitor benazepril on
the progression of chronic renal insufficiency. N Engl J Med.1996;334:939-945.Google Scholar
Klag MJ, Whelton PK, Randall BL.
et al. Blood pressure and end-stage renal disease in men. N Engl J Med.1996;334:13-18.Google Scholar
Smith GD, Phillips AN. Confounding in epidemiological studies: why "independent" effects may
not be all they seem. BMJ.1992;305:757-759.Google Scholar
Pocock SJ. The justification for randomized clinical trials. In: Clinical Trials: A Practical Approach. New
York, NY: Wiley; 1983:50-65.
Burt VL, Culter JA, Higgins M.
et al. Trends in the prevalence, awareness, treatment, and control of hypertension
in the adult US population: data from the health examination surveys, 1960
to 1991. Hypertension.1995;26:60-69. [published correction appears in Hypertension . 1996;27:1192].Google Scholar
Lewis EJ, Hunsicker LG, Clarke WR.
et al. Renoprotective effect of the angiotensin-receptor antagonist irbesartan
in patients with nephropathy due to type 2 diabetes. N Engl J Med.2001;345:851-860.Google Scholar
Guasch A, Parham M, Zayas CF.
et al. Contrasting effects of calcium channel blockade versus converting enzyme
inhibition on proteinuria in African Americans with non-insulin-dependent
diabetes mellitus and nephropathy. J Am Soc Nephrol.1997;8:793-798.Google Scholar
Ruggenenti P, Perna A, Benini R, Remuzzi G. Effects of dihydropyridine calcium channel blockers, angiotensin-converting
enzyme inhibition, and blood pressure control on chronic, nondiabetic nephropathies. J Am Soc Nephrol.1998;9:2096-2101.Google Scholar
Locatelli F, Carbarns IR, Maschio G.
et al. Long-term progression of chronic renal insufficiency in the AIPRI Extension
Study. Kidney Int Suppl.1997;63:S63-S66.Google Scholar
Exner DV, Dries DL, Domanski MJ, Cohn JN. Lesser response to angiotensin-converting-enzyme inhibitor therapy
in black as compared with white patients with left ventricular dysfunction. N Engl J Med.2001;344:1351-1357.Google Scholar
Carson P, Ziesche S, Johnson G, Cohn JN. Racial differences in response to therapy for heart failure: analysis
of the vasodilator-heart failure trials. J Card Fail.1999;5:178-187.Google Scholar