Major Outcomes in Moderately Hypercholesterolemic, Hypertensive Patients Randomized to Pravastatin v

1
ALLHAT
Major Outcomes in Moderately Hypercholesterolemi
c, Hypertensive Patients Randomized
to Pravastatin vs Usual Care
U.S. Department of Health and Human Services
National Institutes of Health
The ALLHAT Collaborative Research Group
Sponsored by the National Heart, Lung, and BIood
Institute (NHLBI)
National Heart, Lung, and Blood Institute
JAMA, December 18, 2002 Vol. 288, No. 23, pp
2998 - 3007
2
Primary ObjectiveALLHAT - LLT
To determine whether pravastatin compared with
usual care reduces all-cause mortality
in moderately hypercholesterolemic,
hypertensive, older participants with at least
one additional CHD risk factor.
40 mg/d
3
Randomized Design
Amlodipine Chlorthalidone Doxazosin Lisinopril
Consent / Randomize(N 42,418)
High-risk hypertensive patients
Eligible for lipid-lowering
Not eligible for lipid-lowering
Consent / Randomize (N 10,355)
Pravastatin Usual care
Follow until death or end of study (max 7.8 yr,
average 4.8 yr).
4
Secondary Outcomes

• CHD
• Cause-specific mortality
• Total and site-specific cancers

5
Major Subgroups (a priori)

• Age 65
• Women
• African-Americans
• Type 2 Diabetes

6
Major Subgroups(post hoc)

• CHD at Baseline
• No CHD and
• LDL-C 130 mg/dL
• No CHD and
• LDL-C lt 130 mg/dL

7
Eligibility Criteria
Eligible for and enrolled in antihypertensive
trial Triglyceride level lt350 mg/dL
8
Eligibility Criteria
Moderate hypercholesterolemia LDL 120-189
mg/dL without CHD LDL 100-129 mg/dL with
known CHD
9
Exclusion Criteria

• Continuing use of prescribed lipid-lowering
  agents
• Contraindications to statins
• Secondary cause of elevated serum cholesterol
• ALT gt 2 times ULN

10
Clinical Sites in LLT

• 513 clinical sites
• United States, Canada, Puerto Rico, US Virgin
  Islands
• VA, private group general medicine practices,
  community health centers, HMOs, specialty
  practices
• Variety of research experience

11
Sample Size Assumptions Statistical Methods

• 84 power to detect 20 reduction in risk for
  primary outcome
• 2-sided a 0.05
• Analysis according to intent to treat
• Cumulative event rates Kaplan-Meier
• Differences between event curves
• log-rank tests Cox proportional
• hazards (PH) model
• PH assumption tested by log-log plots, tests with
  treatment by time interaction

12
Vital Statusat Closeout

• 8776 (84.8) known alive
• 1272 (12.3) confirmed deaths
• 55 ( 0.5) confirmation of death
  pending
• 206 ( 2.0) lost
• 46 ( 0.4) refused

Observed/expected person-years 99
Pravastatin and usual care groups similar with
respect to vital status at closeout.
13
Baseline Characteristics
BL baseline
14
Percent on Statin
100.0
88
84
83
80.0
60.0

40.0
26
17
8
20.0
0.0
2 Years
4 Years
6 Years
848
3063
4353
Pravastatin, N
824
2996
4295
Usual Care, N
N gives the total number of participants at the
visit.
15
Lipid Results
BL baseline
16
Total Cholesterol
230
220
4
Usual Care
8
210
Total Cholesterol in mg/dL
11
200
15
17
190
20
180
Pravastatin
Percent decrease from baseline.
170
0
2
4
6
No. of Participants
Year of Blood Draw
912
2998
4102
5134
Pravastatin

854
2781
3763
5139
Usual Care
17
LDL-C
150
7
140
11
130
LDL-C in mg/dL
16
Usual Care
120
23
28
110
30
Percent decrease from baseline.
Pravastatin
100
0
2
4
6
Year of Blood Draw

18
Cause-Specific MortalityNumber of Deaths
(6-Year Rate per 100 Participants)
19
All-Cause Mortality
15
12
Pravastatin
9
Usual Care
Cumulative Mortality Rate,
6
RR 0.99 95 CI (0.89, 1.11) p
0.88 Pravastatin/Usual Care
3
0
2
3
4
6
1
5
0
Time to Death, years
20
All-Cause Mortality
Pravastatin/Usual Care
Relative Risk and 95 Confidence Intervals

0.50
1
2
Favors Pravastatin Favors Usual
Care
21
All-Cause Mortality
Pravastatin/Usual Care
Relative Risk and 95 Confidence Intervals
CHD at baseline
0.95 (0.74, 1.23)
0.96 (0.84, 1.11)
No CHD LDL-C 130 mg/dL

No CHD LDL-C lt 130 mg/dL
1.18 (0.90, 1.56)
0.50
1
2
Favors Pravastatin Favors
Usual Care
22
CHD (Nonfatal MI CHD Death)
15
12
Pravastatin
Usual Care
9
Cumulative CHD Event Rate,
6
RR 0.91 95 CI (0.79, 1.04) p
0.16 Pravastatin/Usual Care
3
0
5
1
2
4
6
3
0
Time to CHD Event, years
No. at Risk
992
1966
3546
4543
4761
4962
5170
Pravastatin
988
1960
3523
4558
4782
4971
5185
Usual Care
23
CHD (Nonfatal MI CHD Death)
Pravastatin/Usual Care
Relative Risk and 95 Confidence Intervals
0.91 (0.79, 1.04)
Total
0.83 (0.65, 1.06)
Age 55-64 y
0.94 (0.80, 1.12)
Age 65 y
0.84 (0.71, 1.00)
Men
1.02 (0.81, 1.28)
Women

0.73 (0.58, 0.92)
Black
1.02 (0.86, 1.21)
Non-Black
0.89 (0.71, 1.10)
Type 2 Diabetes
0.92 (0.76, 1.10)
No Diabetes
0.50
1
2
Favors Pravastatin Favors Usual Care
There is a Black/Non-Black by Treatment
interaction (p 0.025).
24
CHD (Nonfatal MI CHD Death)
Pravastatin/Usual Care
Relative Risk and 95 Confidence Intervals

0.50
1
2
Favors Pravastatin Favors Usual Care
25
Cancer Incidence
The numbers dont add to the total because some
participants have more than one type of cancer.
26
All-Cause Mortality
Pravastatin/Usual Care
Relative Risk and 95 Confidence Intervals
0.99 (0.89, 1.11)
Total
0.91 (0.71, 1.16)
Doxazosin
1.03 (0.86, 1.23)
Chlorthalidone

1.06 (0.84, 1.35)
Amlodipine
0.95 (0.75, 1.21)
Lisinopril
0.50
1
2
Favors Pravastatin Favors Usual Care
27
Meta-Analysis of Large Long-Term Statin
Trialsand Impact of ALLHAT - LLT
4S, WOSCOPS, CARE, LIPID, AFCAPS/TexCAPS,
Post-CABG, HPS, LIPS Percent change in total
cholesterol
28
Meta-Regression AnalysisAll-Cause Mortality
0.6
Post
-
CABG
AFCAPS
0.4
0.2
ALLHAT
-
LLT
CARE
LIPS
-0
OR (TRT/CTL)
HPS
-0.2
ln
-0.4
4S
LIPID
WOSCOPS
-0.6
-0.8
0
5
10
15
20
25
30

?
TC (CTL
-
TRT)
Relation of total cholesterol (TC) differential
in active treatment (TRT) versus control group
(CTL) to log odds ratio for mortality.
29
Meta-Regression AnalysisCHD (Nonfatal MI CHD
Death)
0.4
Post
-
CABG
0.2
ALLHAT
-
LLT
LIPS
0
CARE
LIPID
-0.2
OR (TRT/CTL)
HPS
-0.4
4S
ln
-0.6
WOSCOPS
-0.8
AFCAPS
-1
0
5
10
15
20
25
30

?
TC (CTL
-
TRT)
Relation of total cholesterol (TC) differential
in active treatment (TRT) versus control group
(CTL) to log odds ratio for CHD events.
30
ConclusionsALLHAT - LLT
ALLHAT pravastatin and usual care groups both
attained substantial cholesterol reductions,
resulting in a relatively modest cholesterol
difference between them.
31
ConclusionsALLHAT - LLT
Accordingly, ALLHAT found only a small decrease
in CVD event rates (non-significant) for
pravastatin compared with usual care and no
difference in mortality.
32
ConclusionsALLHAT - LLT
The study results do not alter current
cholesterol treatment guidelines, which are
based on a series of clinical trials with larger
cholesterol reductions than those observed in
ALLHAT. Thus, cholesterol lowering by lifestyle
changes and drug treatment is recommended
to reduce CVD morbidity and mortality.