The postexercise ankle-brachial index (ABI) is recommended in patients with normal resting ABI when peripheral artery disease (PAD) is suspected. The aims of this study were to determine the comparative diagnostic accuracy of the resting and postexercise ABI for detecting PAD, and, the effect of the presence of diabetes on these. Three methods of interpretation currently in use were also investigated: a reduction in postexercise ABI by >20% compared to resting ABI, an ABI value of ≤0.90 postexercise, or a reduction in systolic ankle pressure of >30 mmHg postexercise. This retrospective study used colour duplex ultrasound (CDU) as the reference standard. In 278 limbs (whole group), the resting ABI had an overall area under the curve (AUC) of 0.71, with the postexercise ABI yielding a similar diagnostic accuracy of AUC 0.72. In the non-diabetes group ( n=171), the resting ABI had an overall AUC of 0.74 and the postexercise ABI had a similar AUC of 0.76. In the diabetes group ( n=107), overall accuracy was reduced compared to the non-diabetes group, with the resting ABI having an overall AUC of 0.65 and the postexercise ABI yielding a similar accuracy with an AUC of 0.64. The overall diagnostic accuracy of the postexercise ABI for diagnosing PAD was not greatly improved compared to resting ABI. Given the lower overall diagnostic accuracy in the diabetes group, both the resting and the postexercise ABI results in diabetes populations should be interpreted with caution. There is a risk of undiagnosed disease if relying on these results alone to determine lower limb vascular status.
Keywords: ankle–brachial index (ABI); peripheral artery disease (PAD); postexercise ABI; sensitivity; specificity; stress testing.