October 3rd, 2013

ABI-Based Screening for PAD: Which Guidelines Are Right?


The United States Preventive Services Task Force (USPSTF) recently concluded that the existing evidence base is insufficient to weigh the benefits and harms of using the ankle–brachial index (ABI) to screen patients for peripheral artery disease (PAD) and assess their related risk for cardiovascular disease. Mark Creager and Rita Redberg offer differing analyses of this conclusion.

Creager: The new USPSTF recommendation on ABI screening for PAD contradicts the ACCF/AHA practice guidelines for PAD, which state that “the resting ABI should be used to establish the lower-extremity PAD diagnosis in patients with suspected lower-extremity PAD, defined as individuals with 1 or more of the following: exertional leg symptoms, nonhealing wounds, age 65 and older, or 50 years and older with a history of smoking or diabetes” (class I recommendation, level of evidence B).

PAD is a clinical manifestation of, not a risk factor for, atherosclerosis. It affects more than 7 million U.S. adults, including 15% to 20% of people age 65 or older, and is associated with more than a 2.5-fold risk for death from cardiovascular causes. PAD is also especially prevalent among patients with diabetes and among smokers. About half of patients with PAD do not have leg symptoms yet remain at high risk for adverse cardiovascular events. Patients with PAD often have a history of coronary or cerebrovascular disease, but even those who do not have such a history remain at increased risk for myocardial infarction and stroke.

Therefore, an important reason to detect PAD in an otherwise asymptomatic patient is to elicit findings of occult atherosclerosis and initiate risk factor–modifying therapies. Data from the National Health and Nutrition Examination Survey indicate that millions of patients with PAD, many of whom are asymptomatic and have no known coronary or cerebrovascular disease, are not taking a statin, an ACE inhibitor, or an antiplatelet drug.

The ABI is very easy to assess as part of a physical examination, using the same knowledge and skills required to assess blood pressure, except that a handheld Doppler device is used to measure ankle pressures. It is neither costly nor harmful. The ABI, calculated for each leg as the ratio of the systolic pressure at the ankle to that of the brachial artery, is considered abnormal if ≤0.90. Its diagnostic accuracy to detect PAD based on the area under the receiver-operating curve is 0.87­–0.95. The ABI enables further characterization of risk beyond the Framingham risk score (FRS); 1 in 5 men and 1 in 3 women would have their FRS risk category reclassified by inclusion of the ABI.

By focusing rigidly on screening asymptomatic patients, the USPSTF limited its analysis and failed to consider important studies that reported outcomes of both symptomatic and asymptomatic PAD patients. The excluded studies have found that the risk for adverse cardiovascular events is high in asymptomatic patients with PAD, though likely less than in symptomatic patients. Similarly, the USPSTF excluded articles on the efficacy of treatment, such as statins or antiplatelet drugs, in reducing the risk for cardiovascular events, as the studies primarily included symptomatic PAD patients. I acknowledge that we lack large randomized trials of outcomes resulting from treatment of asymptomatic patients with PAD detected by ABI screening. Pending such trials, it is reasonable to infer — from studies showing a twofold increased risk for cardiovascular death in patients with asymptomatic PAD than in patients without PAD — that risk factor–modifying therapies known to be effective in symptomatic patients are likely to benefit asymptomatic PAD patients.

Given the evidence I have outlined, I concur with the ACCF/AHA PAD guidelines, which state that ABI screening should be performed in patients age 65 or older, and patients age 50 or older who have a history of smoking or diabetes.


Redberg: By definition, screening means looking for signs of subclinical disease in people who lack symptoms. The bar is high, and should remain so, before a screening test is used to justify an intervention in an asymptomatic patient. Generally, the rationale for an intervention is to help people either feel better or live longer. Given that asymptomatic people already feel perfectly well, we must offer something that helps them live longer if we are going to intervene. Any screening test should meet these criteria:

  1. It must be widely available and inexpensive.
  2. It must be safe.
  3. It should lead to therapy that can be guided only by use of that test.
  4. The therapy prompted by the test should yield better clinical outcomes.

The ABI meets the first two criteria, but it fails on the last two. We must counsel all patients in risk reduction: All smokers should be counseled to quit, and all patients should be encouraged to eat a heart-healthy diet and get regular physical activity. A patient’s ABI score would not change this advice. The USPSTF recommendations focus on screening and therefore require evidence of benefit, as they should. As the USPSTF notes, no data show improved outcomes after ABI screening.

It is hard to justify ABI screening without any evidence of clinical benefit. We clinicians (and our patients) are busy, so we should spend our time on what matters to patients, in terms of improving outcomes. Using that time to discuss, for example, the benefits of a heart-healthy diet and regular physical activity is much more productive than ABI screening would be in reducing patients’ risks for death, PAD, and CV disease. Clearly, we need to study whether ABI screening improves clinical outcomes before we proceed with using this test in clinical practice.


With which set of PAD guidelines are you more closely aligned: those from the USPSTF or those from the ACCF/AHA?

2 Responses to “ABI-Based Screening for PAD: Which Guidelines Are Right?”

  1. Karen Politis, MD says:

    Since ABI screening is safe, easy and relatively quick, I think it could be a good tool to jolt a patient into lifestyle changes. Some patients will ignore good advice, but seeing indisputable numbers that prove that their legs are in danger, they might decide to change course. All’s fair in love and war?

  2. Andre Paixao, MD says:

    I tend to agree with Dr. Redberg. The idea of screening asymptomatic patients for a condition where treatment is mostly aimed at reducing symptoms seems counterproductive. If the purpose of obtaining ABIs is to improve cardiovascular risk classification, it should then be held to the same standards as other biomarkers such as CRP and coronary calcium scores. As the USPSTF paper points out, the impact of ABIs on c-statistics, IDIs and NRIs have been modest at best.