Lower limb vascular assessment for people with diabetes: a multifaceted assessment of objective screening techniques (#121)
Background: Diabetes-related lower limb macro- and microvascular disease is associated foot complications including ulceration and amputation. The nature of this multi-system disease process makes accurate vascular examination difficult. This study evaluated the reliability and diagnostic accuracy of the ankle-brachial index (ABI), toe-brachial index (TBI) and continuous wave Doppler (CDW) assessment in people with diabetes and determined the strength of the relationship of these measures with history of foot complications.
Methods: Ankle and brachial pressure measurements and CWD assessment were performed using handheld Doppler and toe pressures using photoplethesmography. Retesting was performed 7-10 days later. Sensitivity and specificity of the ABI and TBI were determined using colour duplex ultrasound as reference standard and receiver operating characteristic (ROC) analysis was performed to assess the clinical utility of the ABI and TBI. Logistic regression was used to determine the relationship between vascular measurements and history of foot complications.
Results: A total of 389 participants with diabetes were recruited to the various arms of this study. Reliability of CWD was generally poor-to-moderate ranging from k=0.17 (95% confidence interval [CI] -0.15 to 0.49) to k= 0.44 (95%CI: 0.03 to 0.88). Reliability of the TBI and ABI were excellent ranging from ICC 0.75 (95%CI -0.19 to 0.28) to 0.81 (95%LOA -0.23 to 0.25). CWD had the highest sensitivity and specificity for detecting PAD (74% and 93% respectively). ROC analysis demonstrated the TBI had greater clinical efficacy for the diagnosis of PAD (ROC area: 0.75 p<0.01) than the ABI (ROC area: 0.58, p= 0.27). Participants with a TBI of less 0.6 were almost eleven times more likely to have a history of foot complication(OR: 10.73,p<0.05). The ABI and CWD were not independently associated with history of foot complications.
Conclusion: CWD has the highest sensitivity and specificity for PAD however had poor clinical reliability. The TBI and ABI both demonstrated acceptable reliability however the TBI was more effective at detecting PAD and was associated with elevated likelihood of foot complications suggesting it has good clinical utility as an assessment of vascular function and to help determine risk of foot complications in people with diabetes.