LAUSR

Toe–brachial indices (TBI) are being integrated into international guidelines1,2 due to their utility as an adjunct test for peripheral artery disease (PAD), particularly in diabetes populations.3,4 The addition of an exercise protocol has been demonstrated to improve diagnostic accuracy of the traditional ankle–brachial index (ABI).5 However, there has been limited investigation of the effect of exercise on the accuracy of TBI. Therefore, this pilot study aimed to determine the diagnostic test accuracy of the TBI for PAD following an exercise protocol in participants with suspected PAD. This prospective cross-sectional study ran between 2019 and 2020 in a private vascular laboratory in Newcastle, Australia. Adults exhibiting signs or symptoms consistent with PAD were recruited. Exclusion criteria included inability to adhere to testing protocol (e.g., treadmill walking), conditions preventing toe or brachial pressure measurement (e.g., vasoneural disorders, connective tissue disorders, hallux wound or bilateral mastectomy). Informed written consent was obtained, and ethics approval was in place (H-2010-1230). One limb per participant was included in the study – the symptomatic limb, or where both limbs were symptomatic, the right limb. One vascular sonographer conducted all assessments, with inter-rater reliability previously determined to be acceptable.6 All measurements were taken in a temperature-controlled clinical room (23–25°C). Pretesting participant protocols and the measurement procedure for pressure measurements for the vascular laboratory are consistent with current guidelines and as previously described.1,4 A Hadeco Smartdop 45 (Hadeco, Kawasaki, Japan) with a photoplethysmography probe was used to measure systolic toe pressure (TP). Participants walked on a treadmill at a maximum of 3 km/ hour and a 10° incline for up to 5 minutes. If the participant experienced excessive lower extremity discomfort, angina, or dyspnoea, exercise was ceased. Toe, ankle, and brachial pressures were then immediately recorded in a supine position. Subsequently, visualisation of arteries and peak systolic velocities were obtained from the distal aorta to the foot via colour duplex ultrasound (reference standard) and were completed in the same session. Velocity ratios for grading of stenoses have previously been reported.6 Descriptive statistics and measures of sensitivity, specificity, and positive and negative likelihood ratios with 95% CIs were calculated for TBI for detecting PAD – defined as one or more lower limb arteries with stenosis of ⩾ 50%.7 The diagnostic threshold for PAD using TBI was considered to be < 0.70. Interpretation of postexercise TBI was evaluated firstly by using the diagnostic threshold of < 0.70 for PAD, and secondly, if the postexercise value was > 20% lower than the resting TBI. A receiver operating characteristic (ROC) curve was performed for TBI and postexercise TBI (online Supplementary file) and the area under the curve (AUC), calculated using IBM SPSS statistical software, Version 24 (IBM Corp., Armonk, NY, USA). Participant characteristics, vascular testing results, and diagnostic accuracy results are presented in Table 1 (n = 50). ABI mean values fell within a normal range for resting and postexercise (1–1.3). TBI mean values were lower postexercise compared to resting values (mean TBI: 0.66 resting, 0.55 postexercise), which is similar to a previous study.8 Using a diagnostic threshold of < 0.70 for PAD, AUC values for postexercise TP (0.88, 95% CI 0.78 to 0.97) and TBI (0.95, 95% CI 0.88 to 1.02) were higher than for resting measures (TP: 0.76, 95% CI 0.62 to 0.90; TBI: 0.83, 95% CI 0.69 to 0.97). The negative likelihood ratio for postexercise TBI (0.05, 95% CI 0.01 to 0.37) was important,9 indicating a 20-fold decrease in the odds of having PAD in a patient with a negative result. Sensitivity of postexercise TBI (96.3, 95% CI 81.03 to 99.91) was higher than resting TBI (80.77, 95% CI 60.65 to 93.45). Diagnostic accuracy of postexercise toe–brachial index for identifying peripheral artery disease (PAD): A pilot study