ALLHAT-LLT Elderly

Published

Han BH, et al. "Effect of Statin Treatment vs Usual Care on Primary Cardiovascular Prevention Among Older Adults: The ALLHAT-LLT Randomized Clinical Trial". JAMA Internal Medicine. 2017. 177(7):955-965.
PubMed • Full text • PDF

Clinical Question

In the ALLHAT-LLT study, did patients aged ≥65 years with moderate hyperlipidemia and hypertension but without atherosclerotic cardiovascular disease (ASCVD) benefit from the addition of statin therapy in terms of all-cause mortality, when compared to usual care alone?

Bottom Line

In the ALLHAT-LLT study, statins given for primary prevention had no impact on all-cause or CV mortality in elderly patients with moderate hyperlipidemia and hypertension but without ASCVD.

Major Points

Previous studies have quantified that 28% of patients 75-79 and 22% of patients 80 and older take statins for primary prevention, and statin use for primary prevention in patients older than 79 has increased from 8.8% in 1999-2000 to 34.1% in 2011-2012. A 2015 Markov model reported that while statins may be cost effective for primary prevention in patients 75 years and older, any benefit could be easily offset by an increase in adverse events.

The original ALLHAT trial was published in 2002, and included patients over the age of 55, with or without a history of heart disease. ALLHAT contained a lipid-lowering substudy (ALLHAT-LLT) which randomized a subset of patients in the antihypertensive component of the trial to either pravastatin or usual care. ALLHAT-LLT Elderly presents an analysis of the elderly patients in ALLHAT-LLT. Among the 10,355 participants in ALLHAT-LLT, 2,867 patients met the chief eligibility criteria of age ≥65 years and absence of ASCVD at baseline. Baseline and demographic data was well balanced between groups including rates of antihypertensive use and blood pressure. At 6 years, no difference was seen in all-cause mortality (1.3% in the pravastatin group vs. 1.2% in usual care; P=0.09) or CV mortality (0.9% vs. 0.9%; P=0.36). Subset analyses, including among patients above and below 75 years of age, showed no difference between groups.

The authors conclude that statin therapy has no role in the primary prevention of all-cause mortality or CV events in older patients with hyperlipidemia and hypertension but without ASCVD. Similar results were seen in older patients who participated in JUPITER and HOPE-3.

Guidelines

2016 Canadian Cardiovascular Society Guidelines for the Management of Dyslipidemia for the Prevention of Cardiovascular Disease in the Adult, adapted:

• Consider pharmacotherapy with statin if:
  • High FRS (≥ 20%) [Strong Recommendation; High-Quality Evidence], or
  • Intermediate FRS (10%-19%), and
    • LDL-C ≥ 3.5 mmol/L, OR non-HDL-C ≥ 4.3 mmol/L, OR ApoB ≥ 1.2 g/L, OR men ≥ 50 and women ≥ 60 years and 1 additional CVD risk factor [Strong Recommendation; High-Quality Evidence]
• Target for these patients:
  • LDL-C < 2.0 mmol/L or > 50%↓, ApoB < 0.8 g/L, or non-HDL-C < 2.6 mmol/L

Design

• Post hoc analyses of a randomized, open-label, controlled trial
• N=2,867
  • Pravastatin (n=1,467)
  • Usual Care (n=1,400)
• Setting: 513 sites primarily across the United States
• Enrollment: 1994-2002
• Mean follow-up: 4.77 years
• Analysis: Intent-to-treat
• Primary outcome: All-cause mortality at 6 years

Population

Inclusion Criteria

• Ambulatory
• Age ≥55 years
  • Age ≥65 years included in the current analysis
• Stage 1 or 2 hypertension plus ≥ 1 Chronic Heart Disease risk factors:
  • Current cigarette smoking
  • Type 2 diabetes mellitus
  • Left Ventricular Hypertrophy in past 2 yrs
  • HDL cholesterol under 35 mg/dL [0.9 mmol/L] twice in past 5 yrs
  • 2 fasting LDL-C readings of 120-189 mg/dL [3-5 mmol/L]
  • Fasting triglycerides < 350 mg/dL [< 9 mmol/L]

Exclusion Criteria

• Currently lipid-lowering therapy
• Statin intolerance
• Manifested significant liver/kidney disease
• Known secondary cause of hyperlipidemia
• Evidence of ASCVD (previous stroke/MI, CHD, PVD, CVD, history of angina pectoris, any arterial stenosis)

Baseline Characteristics

Pravastatin group displayed

• Demographics: mean age 71.3, 48% female, White non-hispanic 39%, Black non-hispanic 35%, White hispanic 16.7%, Black hispanic 4%, Other 5%, mean years of education 10
• Co-morbidities: Current smokers 23.2%, Type 2 Diabetes 51%
• Physiologic parameters: mean Blood Pressure 147 / 83 mm Hg, mean fasting glucose 130.9 mg/dL [7.3 mol/L], mean serum cholesterol 225 mg/dL [5.8 mmol/L], mean LDL-C 147.7 mg/dL [3.8 mmol/L], mean HDL-C 47.2 mg/dL [1.2 mmol/L], mean fasting triglycerides 150.3 mg/dL [3.9 mmol/L]
• Anthropomorphics: mean BMI 29.5, BMI >30 40.8%
• Medication use: Female taking estrogens 11%, ASA 26%, Antihypertensives 90%
  • Antihypertensive Randomization, no. (%):
    • Chlorthalidone: 35.9% vs. 36.2%
    • Amlodipine: 22.6% vs. 22.9%
    • Lisinopril: 20.9% vs. 20.2%
    • Doxazosin: 20.7% vs. 20.6%

Interventions

• In the ALLHAT-LLT substudy, a subset of patients in the antihypertensive component of the study were randomized to either pravastatin or usual care.
• Pravastatin
  • Initially patients were started on pravastatin 20 mg and titrated upward to a 25% decrease in baseline LDL-C
  • After the first 1000 patients this was changed to a starting dose of 40 mg
• Usual care
  • Treated for LDL-C based on discretion of primary care physician

Outcomes

Comparisons are Pravastatin group vs. Usual care.

Primary Outcomes

All-cause mortality

1.3% vs. 1.2% (ARI = 0.1%, 95% CI 1.18 (0.97-1.42), P = 0.9)

Secondary Outcomes

Cardiovascular disease (CVD) deaths

0.9% vs. 0.9% (ARR = 0%, HR = 1.14 (0.86-1.52), P = 0.36)

Coronary heart disease (CHD) deaths

0.6% vs. 0.7% (ARR = 0.1%, HR = 0.97 (0.65-1.44), P = 0.87)

Stroke deaths

0.3% vs. 0.4% (ARR = 0.1%, HR = 1.36 (0.67-2.78), P = 0.40)

Non-CVD deaths

1.1% vs. 1.0% (ARI = 0.1%, HR = 1.21 (0.93-1.59), P = 0.16)

Unknown cause of death

0.5% vs. 0.3% (ARI = 0.2%, HR = 1.14 (0.54-2.39), P = 0.57)

Fatal CHD and nonfatal MI

0.9% vs. 1.0% (ARR = 0.1%, HR = 0.81 (0.63-1.05), P = 0.12)

Stroke, fatal and nonfatal

0.9% vs. 0.8% (ARI = 0.1%, HR = 1.06 (0.76-1.49), P = 0.72)

Heart failure, hospitalized or fatal

1.0% vs. 1.1% (ARR = 0.1%, HR = 1.00 (0.73-1.36), P = 0.98)

Cancer, fatal and nonfatal

1.1% vs. 1.0% (ARI = 0.1%, HR = 1.14 (0.88-1.46), P = 0.32)

Subgroup Analysis

A comparison of participants aged 65-74 years vs. ≥75 years showed no significant difference.

Adverse Events

About half of patients discontinuing pravastatin cited a reason relating to adverse reactions, however specific information regarding adverse reactions was not recorded.

Criticisms

• Post-hoc secondary analysis of a subgroup of the ALLHAT-LLC trial
• Excluded patients receiving lipid lowering agents at baseline
• Lack of specific adverse event data
• Likely underpowered due to subgroup analysis
• Open label design
• ALLHAT-LLT did not account for non-pharmacologic means of lowering cholesterol (eg, lifestyle/diet/exercise)
• Patients were initially started on pravastatin 20 mg and titrated upward. After about 1,000 patients this was changed to 40 mg as initial therapy, which had the potential to confound results.
• Some of the data may be confounded from patients in the usual care group starting on lipid lowering therapy after the beginning of the trial, and a decrease in adherence from the pravastatin group.
• This study uses ALLHAT-LLT data from 2002. Criteria for diagnosing ASCVD and standards of treatment for comorbidities have since changed. Although treatments groups were represented similarly at baseline, this 15-year time gap could affect the applicability of this data to modern statin protocols and use.

Funding

• Funded by the NHLBI
• Medications provided by pharmaceutical companies:
  • Pfizer (amlodipine and doxazosin)
  • Astrazeneca (atenolol and lisinopril)
  • Bristol-Myers Squibb (pravastatin)