Effectiveness of bedside investigations to diagnose peripheral artery disease among people with diabetes mellitus: A systematic review

Abstract

AbstractAs a progressive disease process, early diagnosis and ongoing monitoring and treatment of lower limb peripheral artery disease (PAD) is critical to reduce the risk of diabetes‐related foot ulcer (DFU) development, non‐healing of wounds, infection and amputation, in addition to cardiovascular complications. There are a variety of non‐invasive tests available to diagnose PAD at the bedside, but there is no consensus as to the most diagnostically accurate of these bedside investigations or their reliability for use as a method of ongoing monitoring. Therefore, the aim of this systematic review was to first determine the diagnostic accuracy of non‐invasive bedside tests for identifying PAD compared to an imaging reference test and second to determine the intra‐ and inter‐rater reliability of non‐invasive bedside tests in adults with diabetes. A database search of Medline and Embase was conducted from 1980 to 30 November 2022. Prospective and retrospective investigations of the diagnostic accuracy of bedside testing in people with diabetes using an imaging reference standard and reliability studies of bedside testing techniques conducted in people with diabetes were eligible. Included studies of diagnostic accuracy were required to report adequate data to calculate the positive likelihood ratio (PLR) and negative likelihood ratio (NLR) which were the primary endpoints. The quality appraisal was conducted using the Quality Assessment of Diagnostic Accuracy Studies and Quality Appraisal of Reliability quality appraisal tools. From a total of 8517 abstracts retrieved, 40 studies met the inclusion criteria for the diagnostic accuracy component of the review and seven studies met the inclusion criteria for the reliability component of the review. Most studies investigated the diagnostic accuracy of ankle ‐brachial index (ABI) (N = 38). In people with and without DFU, PLRs ranged from 1.69 to 19.9 and NLRs from 0.29 to 0.84 indicating an ABI <0.9 increases the likelihood of disease (but the extent of the increase ranges from a small to large amount) and an ABI within the normal range (≥0.90 and <1.3) does not exclude PAD. For toe‐brachial index (TBI), a threshold of <0.70 has a moderate ability to rule PAD in and out; however, this is based on limited evidence. Similarly, a small number of studies indicate that one or more monophasic Doppler waveforms in the pedal arteries is associated with the presence of PAD, whereas tri‐ or biphasic waveform suggests that PAD is less likely. Several forms of bedside testing may also be useful as adjunct tests and 7 studies were identified that investigated the reliability of bedside tests including ABI, toe pressure, TBI, transcutaneous oxygen pressure (TcPO2) and pulse palpation. Inter‐rater reliability was poor for pulse palpation and moderate for TcPO2. The ABI, toe pressure and TBI may have good inter‐ and intra‐rater reliability, but margins of error are wide, requiring a large change in the measurement for it to be considered a true change rather than error. There is currently no single bedside test or a combination of bedside tests that has been shown to have superior diagnostic accuracy for PAD in people with diabetes with or without DFU. However, an ABI <0.9 or >1.3, TBI of <0.70, and absent or monophasic pedal Doppler waveforms are useful to identify the presence of disease. The ability of the tests to exclude disease is variable and although reliability may be acceptable, evidence of error in the measurements means test results that are within normal limits should be considered with caution and in the context of other vascular assessment findings (e.g., pedal pulse palpation and clinical signs) and progress of DFU healing.