Primary Prevention Statins in Older Adults: Personalized Care for a Heterogeneous Population

Deirdre O'Neill, MD, MSc; Neil J. Stone, MD; Daniel E. Forman, MD


J Am Geriatr Soc. 2020;68(3):467-473. 

In This Article

Case Study 2

Patient B.H., an 82-year old man, presents with a past medical history of hypertension, hyperlipidemia, and benign prostatic hypertrophy. He takes losartan 50 mg/day, rosuvastatin 40 mg/day, and tamsulosin .4 mg/day. He is very high functioning, lives alone, is independent in all of his ADLs and instrumental ADLs, and just returned from hiking Machu Pichu. He has a remote smoking history, quitting more than 30 years ago, drinks one glass of red wine every day, and uses no recreational drugs.

B.H. has concerns about CV risk reduction. When asked about his own individual goals, the patient says he wants to avoid the fate of his parents, who each suffered devastating strokes in their 60s. He also speaks of the importance of his current quality of life, wishing to remain highly active for as long as possible.

Physical examination shows blood pressure 118/75 mm Hg, heart rate 56 bpm and regular. He looks well and has normal cardiopulmonary findings. Laboratory tests include HbA1c 5.8%, total cholesterol 180 mg/dL, LDL-C 70 mg/dL, HDL-C 60 mg/dL, and triglycerides 95 mg/dL.

Case 2 Summary & Discussion

B.H. is a highly functional so-called super-ager. Current estimated life expectancy at the age of 85 years in the United States is approximately 6 years, with a 75% chance of living 3 years and a 25% chance of living 10 years. The variable depends on functional status and multimorbidity.[35] With B.H.'s high functional state and low burden of comorbidities, his anticipated life expectancy is likely at the higher end or greater than these estimates, and he has expressed the importance of remaining as functional as possible for the remainder of his life. Yet despite his overall health and vigor, B.H.'s age, sex, hypertension, hyperlipidemia, and family history all indicate that his CV risk is significant. Unfortunately, current CV risk estimates, such as the Framingham risk score and ASCVD score, are not validated in individuals of B.H.'s age, undercutting the reliability of his CVD risk estimate.

Just as in case study 1, clinical context and current evidence must be integrated to facilitate shared decision making and personalized care. Age confers high CVD risk, but B.H. still has a reasonable mean life expectancy. His goal is to retain the highest level of function possible, and it is best achieved by avoiding myocardial infarction (MI) and stroke. Meta-analyses, sub-studies, and observational studies on primary prevention statin use in older adults suggest benefit, particularly for stroke and MI event reduction.[5–7] However, this evidence is based on younger cohorts of older adults, with an absence of literature to base such a decision in an 80-year-old super-ager such as B.H. In fact, a multidisciplinary panel of experts recently reviewed the evidence for primary prevention statins in older adults, concluding that existing data were insufficient to fully determine benefits and harms in those 75 years and older.[36]

Savarese et al published a meta-analysis on primary prevention statin use in older adults, defined as adults 65 years or older (N = 24 000 patients; mean age = 73 years, and mean follow-up = 3.5 years).[5] MI occurred in 2.7% of those on a statin, compared with 3.9% of those on placebo. Thus statins reduced the risk of MI by 39.4% compared with placebo (RR = .6; 95% confidence interval [CI] = .43-.84), with a number needed to treat of 84. Stroke occurred in 2.1% of those on a statin, compared with 2.8% of those on placebo over the follow-up period, reducing the risk of stroke by 23.8% compared with placebo (RR = .76; 95% CI = .62-.92), with a number needed to treat of 143. Statins did not significantly reduce the risk of all-cause or CV death compared with placebo (RR = .94; 95% CI = .85–1.03, and RR = .907; 95% CI = .68–1.19, respectively).

Teng et al completed another meta-analysis on primary prevention using statins in older adults, again using a cutpoint of 65 years or older (N = 26 000; mean age = 72.7 years).[6] Statins were associated with a significant reduction in major adverse CV events (RR = .82; 95% CI = .74-.92), nonfatal MI (RR = .75; 95% CI = .59-.94), and total MI (RR = .74; 95% CI = .61-.90), with no significant difference in fatal MI, stroke, or all-cause mortality.

More recently, Ridker et al published a meta-analysis of age-specific primary prevention with statins using data from the JUPITER and HOPE-3 trials.[7] The JUPITER trial included roughly 5600 subjects older than 70 years (32% of the population), and the HOPE-3 trial included 3086 subjects older than 70 years (24% of the population). Subgroups younger than 65 years, 65 to 70 years, and 70 years or older all showed significant pooled estimates favoring statin use, with a 26% RR reduction in those older than 70 years for the end point of nonfatal stroke or MI or CV death. Of significance, however, the authors note that the number of individuals aged 80 years or older in this meta-analysis was modest, potentially limiting the generalizability and statistical significance of these findings to the oldest old.

If our patient B.H. was younger than 80 years, with the same personal goal of preventing an initial CV event, data indicate benefit of statin therapy. This is more complicated for B.H. because data on individuals aged 80 years or older are sparse. Additionally, 80-year-old super-agers like B.H. are exceptional, such that the generalizability of the literature to a patient like B.H. is even less certain. Furthermore, risks attributable to statins also need to be considered, especially for older adults who fear that statin-induced side effects may undercut their activity and independence. Although never substantiated by clinical trials, the association between statin use and myalgias as well as statins and increased susceptibilities to diabetes, Alzheimer's disease, and even cancer have all been reported, and all remain controversial.

Notably, the meta-analysis described previously by Savarese et al did not show differences in cancer onset between those treated with statin or placebo over the 3.5-year follow-up.[5] Likewise, the meta-analysis by Teng et al showed no differences in myalgias, transaminitis, new-onset diabetes, or significant adverse events with statins.[6] The risk of diabetes with statin use appears to be mostly in those already at risk of developing diabetes because the two diseases have many overlapping risk factors, and diabetes onset is only 2 to 4 months earlier in statin-treated patients as opposed to those not on statin therapy, a duration that lacks clinical significance.[37]

Likewise, randomized controlled trials and observational trials do not support a causal effect of late-life statin use and cognitive decline in cognitively normal or impaired individuals.[38–41] The ACC/AHA reviewed statin safety data and addressed it in the 2013 cholesterol guideline and showed no randomized controlled trial evidence of statin-associated cognitive change or risk of dementia.[42] Recent data from the HOPE-3 randomized controlled trial in a population with a mean age 74 years showed that candesartan with hydrochlorothiazide, rosuvastatin, or their combination did not significantly affect cognitive decline in this study of older adults.[43] Additionally, autopsy studies show no significant difference in Alzheimer's disease onset or in burden of plaques and tangles on autopsy study in those on a statin, compared with those not on statin therapy.[44,45]

A CAC score could be performed in the case of B.H., just as it was in case 1. The value of the CAC score in this case would be greatest if it was zero because this may aid in the decision not to prescribe statin therapy in this older, very functional adult, who could be impacted by a statin-adverse effect such as myalgias.

Primary prevention statin benefits are likely to be desirable for B.H. Furthermore, he has been tolerating a high-dose statin for years, and there is no mandate to reduce statin doses for patients who are tolerating them well (unless a new drug is started with a potential for drug-drug interaction and/or elevated statin bioavailability). Overall evidence suggests statins are safe and generally well tolerated in older adults, but corroborating data from trials tend to include adults who are more robust than many seen in the average clinical practice. Additionally, statins have been associated with a higher risk of myopathy and other muscle effects in adults treated with high-dose statin therapy, especially with simvastatin.[46] Just as in case study 1, incorporating the patient's estimated life expectancy and goals of treatment, as well as discussing potential statin-associated side effects, is an important part of the shared decision-making process.