Post hoc analysis turns up no benefit at ages 65-74
In the overall neutral open-label trial, analysis restricted to participants ages 65 and older, showed that randomization to pravastatin (Pravachol) likewise didn't impact the primary endpoint of all-cause mortality during 6 years, Benjamin Han, MD, MPH, of the New York University School of Medicine in New York City, and colleagues reported online in JAMA Internal Medicine.
In the 65- to 74-year-olds, the mortality rate was 15.5% on pravastatin and 14.2% with usual care (HR 1.08, P=0.55). For adults 75 years and older, the trend actually neared significance in the wrong direction (24.5% pravastatin vs 18.5%, HR 1.34, P=0.07).
Coronary heart disease events came out similar between groups, including after multivariable regression, with no treatment by age interaction.
The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial-Lipid-Lowering Trial -- nested within the ALLHAT hypertension trial -- included 2,867 ambulatory adults, ages 65 and older, for the analysis out of the overall population 55 and older with hypertension and at least one additional heart disease risk factor. No one had baseline atherosclerotic cardiovascular disease (ASCVD) or baseline statin use.
An accompanying editor's note by Gregory Curfman, MD, of Harvard Medical School in Boston, acknowledged statin risks that "may be particularly problematic in older people" and concluded the ALLHAT-LLT results "should be considered before prescribing or continuing statins for patients in this age category."
Physicians contacted by MedPage Today were universally skeptical that the analysis should have any clinical impact.
"I think the most important part of this report may be the section on Limitations, which states that it is a post hoc secondary analysis of a trial of a subgroup of patients. I always tell students to avoid this type of analysis," commented Daniel Blumenthal, MD, MPH, president of the American College of Preventive Medicine.
James Stein, MD, director of the Preventive Cardiology Program at the University of Wisconsin in Madison, pointed to the "null bias due to the small difference in achieved LDL-C" between groups.
While agreeing that post hoc subgroup analyses can be misleading, Noel Bairey-Merz, MD, director of the Preventive Cardiac Center at Cedars-Sinai Medical Center in Los Angeles, noted that "given the declining CVD mortality rate in the U.S., particularly in the over 65-year-olds, due to improved public health (less smoking) and healthcare (Medicare), demonstration of mortality benefit of anything now is rare."
Kim Williams, MD, of Rush University in Chicago and a past president of the American College of Cardiology, pointed to the moderate 40-mg pravastatin dose used, too. While the findings might seem to contradict the ACC/American Heart Association lipid guidelines, he noted that the study was consistent with a 15% to 30% relative reduction in coronary events, although not significant.
"This suggests a type II statistical error (accepting the null hypothesis when it is actually false) and that a larger study would have been positive," he told MedPage Today by email.
Chris Cannon, MD, of Brigham and Women's Hospital who has been involved in key lipid trials, noted that about one-third of the usual care group ended up on statins. In contrast to the modest sample size and lack of placebo control of the ALLHAT-LLT analysis, he pointed to pooled data meta-analysis from the HOPE-3 and JUPITER double-blind, placebo-controlled primary prevention trials with five times more seniors.
"It shows clear benefit," he said. "Which would you believe?"
One of the authors of that analysis, Paul Ridker, MD, also of Brigham and Women's Hospital, noted no heterogeneity across age groups in that analysis -- under 65, 65 to 69, and 70 or older.
"Of course, when our paper came out last month, there was no media coverage and we did not issue a press release because we did not think a paper confirming what is already known was particularly newsworthy," he said in an email to MedPage Today.
The study shouldn't undermine the value of statins for older adults, Stein suggested.
"Statins clearly reduce risk of myocardial infarction and stroke among older adults without established ASCVD," he added. "I treat a lot of older patients and the vast majority would be very pleased to avoid a heart attack or stroke, even if it's not clear if they will live longer. Many studies have addressed this issue and are included in a meta-analysis from 2013, which included ALLHAT-LLT. There are subgroup analyses of large statin trials showing similar results.
"An open question is at what age should we stop screening and treating lipids for primary prevention. It's not known, though epidemiological data suggest it may be around 85 years of age. Medical care and decision-making always should be a shared interaction with patients, but especially very old patients who may have competing risks, polypharmacy, and limited lifespan," Stein stated.
An Australian trial, STAREE, comparing statins versus placebo in people over age 70 is underway, with results expected in 2020.
Han and co-authors and Curfman disclosed no relevant relationships with industry.
JAMA Internal MedicineSource Reference: 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 Intern Med 2017; DOI: 10.1001/jamainternmed.2017.1442
JAMA Internal MedicineSource Reference: Curfman G "Risks of statin therapy in older adults" JAMA Intern Med 2017; DOI: 10.1001/jamainternmed.2017.1457.