An evaluation of non-invasive vascular assessment methods for detecting peripheral arterial disease in the lower limb

This thesis provides an examination of the current evidence base regarding the diagnostic accuracy of non-invasive vascular assessment examination of the lower limb. This project comprised of a systematic review and a further four studies investigating the comparative diagnostic accuracy of non-invasive vascular assessment methods in cohorts at risk of peripheral arterial disease (PAD), the current vascular assessment techniques of Podiatrists in Australia and New Zealand and the reliability of continuous wave Doppler (CWD) assessment performed by Podiatrists. The results of these studies were then used to develop a modified method of lower limb vascular assessment designed to reduce the time burden of performing assessment in clinical practice. The diagnostic accuracy of this method for PAD was then compared to existing international guidelines. Systematic review of studies investigating the diagnostic accuracy of the toe-brachial index (TBI) for detecting PAD, using diagnostic imaging as a reference standard, identified a lack of existing data. Furthermore, of the studies that have been done, we found that there are significant variations in TBI value used to indicate pathology, making results difficult to interpret. Additionally no studies had undertaken investigations of comparative diagnostic accuracy of the TBI and the more widely used ankle-brachial index (ABI) using a valid reference standard. Therefore undertaking a study evaluating the comparative diagnostic accuracy of the TBI and the ABI for detecting PAD was necessary.The diagnostic accuracy of the TBI and ABI were determined in a population at risk of PAD and demonstrated the TBI was a better clinical tests for PAD while the ABI was highly likely to fail to detect the presence of disease. As vascular assessment is also known to be particularly challenging in diabetes cohorts due to the specific clinical presentation of diabetes related PAD. Therefore a case-control diagnostic accuracy study of the ABI, TBI and CWD for diagnosing PAD was performed. Compared to a control group, all tests had lower sensitivity in the group with diabetes with CWD superior diagnostic accuracy in both cohorts.To further explore the nature of lower limb vascular assessment in clinical Podiatry practice a survey of self-reported lower limb vascular screening techniques used by Podiatrists in Australia and New Zealand was undertaken. From this survey, poor alignment of clinical assessment techniques with existing international guidelines was identified. The most commonly employed vascular assessment techniques used by Podiatrists was reported to be CWD using hand-held Doppler while lack of time was reported to be a significant barrier to undertaking objective vascular assessment tests in clinical practice. As a result of these findings, an inter and intra-tester reliability study of hand-held Doppler examination by performed by Podiatrists was undertaken. This showed that the inter and intra-tester reliability of clinical Doppler examination by podiatrists is low and therefore likely to be of limited value for ongoing monitoring of lower limb vascular function. Finally, using the research completed in this thesis combined with the current evidence base, a modified lower limb vascular screening method was devised. The diagnostic accuracy of this modified method for detecting PAD was then compared to the diagnostic accuracy of current international guidelines (American Heart Association Guidelines). This showed that the method had similar diagnostic accuracy to the current guideline, however may be more time effective. The studies presented in this thesis re-enforce the difficulties with non-invasive vascular assessment of the lower limb, particularly in diabetes. The studies also demonstrate that the TBI has good clinical applicability and has good diagnostic accuracy and therefore may be a screening test of choice in populations at risk of PAD.