Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial - ALLHAT Lipid Lowering Trial


This trial evaluated Pravastatin vs usual care for the treatment of moderate hypercholesterolemia


Pravastatin reduces all-cause mortality vs usual care in older, moderately hypercholesterolemic, hypertensive patients with >=1 CHD risk factor.

Study Design

Study Design:

Patients Enrolled: 10,355
NYHA Class: Not reported
Mean Follow Up: mean 4.8 years (up to 8 years)
Mean Patient Age: 55+ (mean age 66)
Female: 49%
Mean Ejection Fraction: Not reported

Patient Populations:

Men and women age >=55 years with SBP >=140mmHg and/or DBP >=90 mmHg or took medication for hypertension and had at least one additional risk factor for CHD (prior MI or stroke, LVH by ECG or echocardiogram, type 2 diabetes, current smoking, low HDL); fasting LDL-C level of 120-189 mg/dL with no known CHD, or 100-129 mg/dL with known CHD; and fasting triglyceride levels <350 mg/dL.


Currently receiving lipid-lowering therapy, taking large doses of niacin, or taking probucol in the last year; known intolerance of statins or significant liver or kidney disease or other contraindications for statin therapy; or known secondary cause of hyperlipidemia.

Primary Endpoints:

All-cause mortality

Secondary Endpoints:

Composite of fatal CHD or nonfatal MI (CHD events), cause-specific mortality, total and site-specific cancers, Q-wave MI

Drug/Procedures Used:

Patients randomized to pravastatin (40 mg/day, n=5170) or to usual care (n=5185). The usual care group was treated for LDL-C lowering according to the discretion of their primary care physicians.

Concomitant Medications:

Patients were advised to follow the NCEP Step I diet.

Principal Findings:

In the usual care arm, crossovers to statin treatment increased from 8% at year 2 to 17% by year 4. Use of any lipid-lowering drugs during the trial in the usual care arm was reported in 32% of patients with CHD and 29% without CHD. Total cholesterol levels decreased by 17.2% in the pravastatin arm and by 7.6% in the usual care arm by year 4. LDL-C was reduced by 28% with pravastatin vs 11% with usual care. The primary endpoint of all-cause mortality did not differ between the two arms (14.9% with pravastatin vs 15.3% with usual care at 6 years, RR 0.99, p=0.88) nor did the secondary endpoints of CHD events (9.3% vs 10.4%, RR 0.91, p=0.16), stroke (5.3% vs 5.8%, RR 0.91; p=0.31), heart failure (6.0% vs 6.2%, RR 0.99; p=0.89) or cancer (9.6% vs 9.3%, RR 1.03, p=0.66). Pravastatin showed a significantly greater benefit in blacks than in nonblacks for the CHD events endpoint (RR 0.73 vs 1.02, p=0.03). No significant heterogeneity was observed for any outcomes with regard to age, sex, or history of diabetes.


The ALLHAT-LLT showed that despite a moderate reduction in total and LDL cholesterol with pravastain treatment, there was no difference in mortality, CHD or stroke compared with usual care treatment for moderate hypercholesterolemia. The lack of clinical benefit with statin therapy is in contrast to the results observed in other large statin trials such as 4S, CARE, LIPID, and PROSPER, all of which demonstrated a benefit in morbidity and/or mortality with statin therapy. While total and LDL cholesterol were statistically significantly lower with pravastain treatment, the magnitude of teh clinical difference compared to the usual care arm was less than expected at the onset of the trial. Part of the reason for the lower than expect cholesterol difference and the lack of benefit in CHD and mortality between the two arms may be due to the relatively high incidence of statin use in the usual care arm (17% by year 4). Furthermore, patients with the highest cholesterals in whom the benefits may have been the greatest, were likely among those treated with statins in the usual care arm. While treatment in the usual care arm was according to the discretion of the primary care physician, “vigorous” cholesterol-lowering therapy in the usual care group was discouraged unless warranted by a change in clinical circumstances. Nonetheless, 17% of patients in the usual care arm received statin therapy by year 4 and 30% started on a lipid-lowering drug during the trial despite only 5% of patients sustained an event that precipitated the lipid-lowering therapy. The lack of difference in cholesterol reduction likely explains much of the lack of difference in clinical outcomes. Patients in the ALLHAT-LLT trial differed in many regards from patients enrolled in other large scale statin trials: all patients were also receiving hypertensive therapy from the main ALLHAT trial and there was a larger proportions of elderly, women, blacks, and Hispanics than most other statin trials. In a meta-analysis of 9 large statin trials including ALLHAT-LLT, total cholesterol was reduced 18% with statin therapy, CHD events were reduced 27% (95% CI, 23%-31%) and mortality was reduced 14% (95% CI, 10%-18%). Given the significant findings of the meta-analysis and the probable explanation of statin use in the usual care arm for the lack of benefit in ALLHAT-LLT, the current guidelines should continued to be followed for lipid-lowering therapy for the prevention and treatment of cardiovascular disease.


JAMA 2002;288:2998-3007.

Keywords: Stroke, Neoplasms, Hyperlipidemias, Cholesterol, LDL, Hydroxymethylglutaryl-CoA Reductase Inhibitors, Diabetes Mellitus, Type 2, Risk Factors, Hypercholesterolemia, Hispanic Americans, Smoking, Heart Failure, Pravastatin, Triglycerides, Hypertension, Fasting

< Back to Listings