Are Statins “ALL-THAT” for Primary Prevention in Older Adults? A Second Look at ALLHAT-LLT

Statins reduce atherosclerotic cardiovascular disease (ASCVD) risk.^1,2 However, the utility of statin therapy in older adults — particularly in those age 75 years and older — remains controversial. The 2013 ACC/AHA Cholesterol Guideline does not provide specific recommendations for older adults citing a lack of evidence.³ Furthermore, the Pooled Cohort Equations estimates 10-year and lifetime ASCVD risk in adults between 40 and 75 years of age. The need for additional evidence to guide clinical practice is underscored by the fact that one-in-three older Americans report taking a statin.

Given the paucity of prospective data, Han et al⁴ conducted a post-hoc analysis of the ALLHAT-LLT trial.⁵  Participants in the ALLHAT-LLT trial enrolled in the parent Antihypertensive and Lipid Lowering treatment to prevent Heart Attack Trial (ALLHAT), a large landmark study conducted between 1994 and 2002. Eligible patients included those at least 55 years of age with hypertension, at least 1 additional risk factor, and a baseline low-density lipoprotein cholesterol (LDL-C) between 120 and 189 mg/dL in those without ASCVD or 100 to 130 mg/dL in those with ASCVD. Patients already receiving lipid-lowering therapies, with known statin intolerance, significant renal or hepatic disease, or secondary cause of dyslipidemia were excluded. A total of 10,355 patients were enrolled in ALLHAT-LLT and randomized in a 1:1, unblinded fashion to pravastatin 40 mg daily or usual care. Notably, between group differences in total cholesterol (-9.6%) and LDL-C (-16.7%) during the study was quite modest due, in part, to cross-over from the usual care arm. After a mean 4.8 years of follow-up, there was no difference in the primary outcome of all-cause mortality or any secondary endpoints. The lack of benefit from pravastatin was similar in patients with and without ASCVD.

Of 10,355 patients enrolled in ALLHAT-LLT, 2867 (28%) patients were at least 65 years old without prior ASCVD.  It is this subpopulation that was the focus of the Han et al. analysis.⁴ The mean age of this cohort was 71 years and approximately half were female. Mean baseline LDL-C was 148 mg/dL but 20% had an LDL-C less than 130 mg/dL. Despite a difference in mean achieved LDL-C of 19 mg/dL favoring pravastatin (109.1 vs. 128.8 mg/dL), there was no reduction in all-cause mortality and the cardiovascular event rates was similar in the pravastatin and usual care groups.

The authors next turned their attention to patients age 65-74 years and those 75 years and older. In the 75 years or older cohort, there were numerically more deaths in the pravastatin group than usual care group (92 vs. 65; hazard ratio, 1.34; 95% confidence interval, 0.98-1.84; p = 0.07). After multivariate adjustment, the hazard for all-cause death was 1.36 (95% confidence interval, 0.98-1.89) for pravastatin vs. usual care in the 75 years or older cohort and 1.15 (95% confidence interval, 0.82-1.33) in the 65-74 years cohort. The p-value for interaction was not statistically significant (p-interaction = 0.24). Notably, specific safety data regarding adverse effects were not reported, although the incidence of stroke, heart failure, and cancer were similar between groups.

The authors concluded that statin therapy does not provide benefit to persons age 65 years or older and that statin therapy may be harmful in persons age 75 years or older, based on a nonsignificant trend toward increased mortality.

The strengths of this analysis include the use of prospectively collected data and strong statistical adjustments for potential confounders. The authors have identified an important area of clinical practice that lacks strong evidence.  However, there are several important limitations that must be considered. First, cross-over in the parent ALLHAT-LLT trial confounded the original study. The same limitation applies to this post-hoc analysis.  The difference in LDL-C lowering between the pravastatin and usual care groups was rather modest (< 20mg/dL). It is possible that a cardiovascular benefit only manifests with more potent LDL-C lowering.

Second, the investigators chose a liberal interpretation of their findings, emphasizing the trend towards increased mortality in the subgroup of patients age 75 years or older. However, the p-value for interaction was not statistically significant and it is impossible to conclude that pravastatin increases mortality. The biological rationale to explain why statins might increase in all-cause mortality among older adults is weak. It is unlikely that more than 1 or 2 cases of rhabdomyolysis — the statin adverse effect with the highest risk of mortality — occurred in the analyzed subgroup. Based on the reported incidence of rhabdomyolysis in statin-treated patients (2 to 3 per 100,000 patient-years), if we treated 10,000 patients for 5 years with a high-intensity statin, we would expect to see 5 cases of myopathy, which, if left unaddressed, could progress to 1 case of rhabdomyolysis.⁸  Furthermore, it is seems unlikely that a more rapid cognitive decline in the pravastatin group would result in an increase in all-cause mortality. The PROSPER trial did not find any significant effects of pravastatin on cognitive function.⁹ Moreover, adverse effects on cognition are at least two degrees of separation from death (e.g. impaired cognition causes falls or some other undesirable event, that, in turn, causes death) and, therefore, unlikely to play a major role.

Table 1. Effects of Statin Therapy on All-Cause Mortality in Older Adults

* reported for the primary prevention subgroup

The results of the ALLHAT-LLT subgroup analysis add little to the existing literature. In the PROSPER trial which enrolled patients age 70-82 years old with and without established cardiovascular disease, pravastatin significantly reduced the risk of the composite of coronary heart disease death, nonfatal myocardial infarction, and nonfatal stroke (14.1% vs. 16.2%; hazard ratio = 0.85; 95% CI, 0.74-0.97; p = 0.01). The effect of pravastatin on all-cause mortality was not significantly different from that of placebo (10.3% vs. 10.5%; hazard ratio = 0.97; 95% CI, 0.83-1.14; p = 0.74). A recent meta-analysis which included 25,952 older adults also found significant reductions in atherosclerotic cardiovascular disease with statin use.¹⁰ The risks of myalgias, new onset diabetes, and all-cause mortality were not different between statin users and non-users.

Unfortunately, these data neither confirm nor refute any hypotheses regarding the cardiovascular and non-cardiovascular effects of statin therapy in older adults. It is possible that the anti-atherosclerotic benefits of statin therapy, which may not manifest for several years, are less relevant in older adults. Older adults with no prior history of ASCVD are inevitably more likely to die from non-ASCVD causes. Should statins still be recommended in older adults, particularly those at least 75 years of age, for primary prevention? What do you think?