Practical equations to predict claudication pain distances from a graded treadmill test

Abstract

Treadmill testing is used to estimate the severity of claudication, but routine use is not practical or cost-effective in all settings. Thus, the purposes of this study were: (1) to develop and cross-validate prediction equations for treadmill claudication pain distances in a heterogeneous cohort of peripheral arterial occlusive disease patients, and (2) to determine if the regression equations were more accurate in assessing claudication distances than self-reported distances of patients. Medical history, vital signs, resting ankle/brachial systolic pressure index (ABI), and claudication distances during a graded treadmill test were obtained on a validation group of 178 claudicants and on a cross-validation group of 94 claudicants. The independent predictors of claudication pain distances of the validation group were ABI, body mass index, gender, and current smoking status, with multiple correlation coefficients of R = 0.73 and R = 0.82 for the distances to onset and to maximal pain, respectively. These equations were successfully cross-validated on an independent group of claudicants, as the predicted distances to onset of claudication pain (167.2 +/- 102.6 m) and to maximal pain (354.6 +/- 154.3 m) were similar (p = 0.99) to measured distances (169.1 +/- 127.8 m and 356.6 +/- 181.0 m, respectively). However, the self-reported distances to onset (89.5 +/- 126.3) and to maximal claudication pain (189.2 +/- 284.3) were 1-2 blocks less than either the measured or predicted distances (p < 0.01). It is concluded that claudication pain distances during an incremental treadmill test can be more accurately estimated from a composite of variables obtained during medical screening than by relying on the self-report of patients. Consequently, in clinical settings where treadmill testing is impractical, the functional severity of claudication can be assessed without exercise testing.