Peripheral arterial disease and subsequent cardiovascular mortality. A strong and consistent association.

In this issue of Circulation, Davey Smith and colleagues1 report the association of intermittent claudication with subsequent mortality among more than 18,000 men in the Whitehall Study. Intermittent claudication was assessed by self-administration of the brief and simple London School of Hygiene and Tropical Medicine questionnaire, better known as the Rose Questionnaire after its author.2 This questionnaire is the epidemiological standard for assessment of angina, infarction pain, and intermittent claudication. The Whitehall Study confirms earlier reports of a doubling of mortality and a significant reduction of life expectancy in subjects with symptoms of peripheral arterial disease (PAD).3-7 Most of this increase in mortality is due to cardiovascular disease. A population-based study from our institution confirmed that Rose claudication, called probable clau-dication in the Whitehall Study, doubled the risk of See p 1925 mortality.8 However, we also assessed large-vessel PAD at baseline with highly valid and reliable non-invasive tests (i.e., flow velocity by Doppler ultra-sound and segmental blood pressure ratios).8 We had earlier published an analysis showing a high speci-ficity value and high positive and negative predictive values of Rose claudication for large-vessel PAD but a low sensitivity value.9 Because of misclassification due mostly to this low sensitivity, we anticipated that large-vessel PAD measured noninvasively would show a better association with mortality; indeed, the relative risk of 1.9 in our study for Rose claudication was nearly quadrupled to 7.5 when large-vessel PAD measured noninvasively was used to predict subsequent mortality.8 After adjustment for other risk factors and exclusion of prevalent disease at baseline, the relative risk of large-vessel PAD for mortality was 4.5. Interestingly, after exclusion of subjects with base-line ischemia in the Whitehall Study, the association of "possible" claudication (or exercise calf pain that could disappear while walking) with mortality was stronger than the association for probable claudica-tion. We had earlier reported that defining claudica-tion as any exercise calf pain not present at rest (i.e., possible or probable claudication) doubled the sensitivity for large-vessel PAD,9 albeit at the expense of some loss of specificity and positive predictive value, and, interestingly, doubled the relative risk of clau-dication for subsequent mortality.8 It appears that there is a new lesson to be learned and an old lesson to be emphasized from these new analyses. The new lesson is that, as the Whitehall investigators note, the atypical or possible claudication group "contains many genuine cases."1 This may not surprise clinicians who have long recognized …