Author + information
- aNorthwestern University Feinberg School of Medicine, Chicago, Illinois
- bUniversity of California, San Diego School of Medicine, La Jolla, California
- ↵∗Reprint requests and correspondence:
Dr. Michael H. Criqui, University of California, San Diego School of Medicine, 9500 Gillman Drive, La Jolla, California 92093-0607.
Peripheral artery disease (PAD) affects approximately 8 million people in the United States and 200 million people worldwide (1,2). Improved medical therapy for cardiovascular disease (CVD) has reduced overall rates of CVD events. However, people with PAD remain at higher risk of CVD events compared with people without PAD. Reducing CVD events in the large and growing number of people living with PAD is an important public health challenge.
However, the high prevalence of asymptomatic or unrecognized PAD is a barrier to successful reduction of CVD events in all people with PAD. For example, in the PARTNERS (PAD Awareness, Risk, and Treatment: New Resources for Survival) study, 6,979 men and women in primary care medical practices who were either age ≥70 years or 50 to 69 years and had a history of diabetes or smoking were screened for PAD with the ankle brachial index (ABI) (3). Overall, 1,865 (29%) of those screened had an ABI <0.90, consistent with PAD. Yet, 823 (44%) of those found to have an ABI <0.90 had no prior diagnosis of PAD. Most patients with undiagnosed PAD were asymptomatic. Successfully reducing CVD events in all people with PAD requires identification of those who are undiagnosed.
Because of the high prevalence of asymptomatic and unrecognized PAD and the high rate of CVD events in PAD (4), the American Heart Association (AHA) and the American College of Cardiology (ACC) recommend screening high-risk groups (such as those age ≥70 years and those age 50 to 69 with history of diabetes or smoking) with the ABI (4). However, widespread measurement of the ABI in high-risk groups is not uniformly recommended. In 2013, the U.S. Preventive Services Task Force concluded that insufficient evidence exists to recommend for or against ABI screening as a cardiovascular risk assessment in adults (5). The conclusion of the U.S. Preventive Services Task Force was based in part on the lack of randomized trial evidence that reduction of CVD risk factors reduces CVD rates in people found to have PAD on the basis of a screening test. This conclusion may have reduced widespread uptake of ABI screening in medical practices in the United States.
To determine whether statin therapy may reduce CVD events even in people with asymptomatic and unrecognized PAD, in this issue of the Journal, Ramos et al. (6) use an observational study design to report the association of statin therapy with all-cause mortality and major CVD events in people with asymptomatic PAD (6). Participants were identified from primary care practices in Catalan, where 1,365 general practitioners performed an ABI in 74,280 patients age 35 to 85 years between 2006 and 2011. Of the 74,280 patients with ABI testing, 12,119 (16%) had an ABI <0.95, which is consistent with PAD, and were determined to be asymptomatic on the basis of absence of a previous diagnosis of cardiovascular disease, medical record mention of intermittent claudication symptoms, and treatment for intermittent claudication. All-cause mortality and major cardiovascular events were determined for 5,480 persons: 2,740 statin users with asymptomatic PAD who were propensity score matched to 2,740 nonstatin users with asymptomatic PAD. Over a median follow-up of 3.6 years, a first major cardiovascular event occurred in 201 of 2,740 (7.3%) of asymptomatic PAD patients started on a statin versus 245 of 2,740 (8.9%) of asymptomatic PAD patients who were not started on a statin. All-cause mortality rates were 263 of 2,740 (9.6%) and 316 of 2,740 (11.5%), respectively. The authors conclude that their findings support the use of the ABI to screen asymptomatic individuals and to start statin therapy in people found to have an ABI <0.95.
Advocates of widespread statin use will find the above considerations rather academic. A meta-analysis of 27 randomized trials with 134,537 participants showed that persons free of vascular disease who are at the lowest CVD risk, <1%/year for a major vascular event, had the same relative risk reduction with statins as the highest-risk group (7): about 20%, the same relative risk reduction as in the current study. This 20% relative risk reduction in the meta-analysis was essentially constant across all risk levels, perhaps raising the question, “who shouldn’t be treated with statins?” However, the more important metric is the absolute risk reduction. The number-needed-to-treat (NNT) is 1 divided by the absolute risk reduction. Unlike the essentially constant relative risk reduction, the absolute risk reduction increases sharply with increasing risk, whereas the corresponding NNT gets smaller, leading to greater cost-effectiveness. In this study, the absolute risk reduction in asymptomatic PAD was similar to that in secondary prevention trials. This reflects both the known elevated risk of asymptomatic PAD independent of CVD risk factors (8), as well as the elevated risk posed by diabetes, which was present in a striking 72% of patients in this study.
It is unlikely that findings from the study by Ramos et al. (6) will change clinical practice. First, the AHA/ACC guidelines on cholesterol treatment already suggest that people with PAD should be treated with cholesterol-lowering therapy (9). This recommendation is not limited to people who have symptoms. Second, the AHA/ACC guidelines also recommend statin therapy for most people with diabetes mellitus and for others who are at high risk of atherosclerotic CVD (9).
Widespread ABI screening could be potentially useful if it identified a large number of individuals with a low ABI who would otherwise not qualify for cholesterol-lowering therapy. However, the results reported by Ramos et al. (6) suggest that most patients in their study qualified for statin therapy even before the ABI measurement. A more useful analysis to address the utility of ABI screening would be assessing the association of statin therapy with cardiovascular events among people who have a low ABI and no other indication for statins. However, available evidence suggests that the number of people with a low ABI and no other indication for cholesterol-lowering therapy is small. Therefore, the results reported by Ramos et al. (6) should not be construed as justifying widespread ABI screening.
The authors are to be commended for measuring the ABI in such a large group of patients in primary care medical practices. The strong association of statin therapy with reduced all-cause mortality and cardiovascular events underscores the likely benefits of statin therapy, because confounding by indication might reduce the observed benefit, even with propensity matching. Other potential benefits of ABI screening include potentially clarifying vague leg symptoms such as numbness, weakness, and fatigue, as well as the identification of additional CVD risk that might motivate better adherence to therapy.
A randomized trial of ABI screening may be indicated. However, first clinicians should optimize therapy for people with clear indications for statin therapy, including those with diabetes mellitus. The number of people with a low ABI who do not have another indication for ABI screening would need to be substantial to justify a clinical trial.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2016 American College of Cardiology Foundation
- Hirsch A.T.,
- Haskal Z.J.,
- Hertzer N.R.,
- et al.
- Moyer V.A.,
- on behalf of the U.S. Preventive Services Task Force
- Ramos R.,
- García-Gil M.,
- Comas-Cufí M.,
- et al.
- Mihaylova B.,
- Emberson J.,
- Blackwell L.,
- et al.,
- for the Cholesterol Treatment Trialists’ (CTT) Collaborators
- Goff D.C.,
- Lloyd-Jones D.M.,
- Bennett G.,
- et al.