Performance of non-invasive bedside vascular testing in the prediction of wound healing or amputation among people with foot ulcers in diabetes: A systematic review

Introduction: The presence of peripheral artery disease (PAD) confers a significantly increased risk of failure to heal and major lower limb amputation for people with diabetes-related foot ulcer (DFU). Determining performance of non-invasive bedside tests for predicting likely DFU outcomes is therefore key to effective risk stratification of patients with DFU and PAD to guide management decisions. The aim of this systematic review was to determine the performance of non-invasive bedside tests for PAD to predict DFU healing, healing post-minor amputation, or need for minor or major amputation in people with diabetes and DFU or gangrene.Methods: A database search of Medline and Embase was conducted from 1980 to 30 November 2022. Prospective studies that evaluated non-invasive bedside tests in patients with diabetes, with and without PAD and foot ulceration or gangrene to predict the outcomes of DFU healing, minor amputation, and major amputation with or without revascularisation, were eligible. Included studies were required to have a minimum 6-month follow-up period and report adequate data to calculate the positive likelihood ratio (PLR) and negative likelihood ratio for the outcomes of DFU healing, and minor and major amputation. Methodological quality was assessed using the Quality in Prognosis Studies tool.Results: From 14,820 abstracts screened 28 prognostic studies met the inclusion criteria. The prognostic tests evaluated by the studies included: ankle-brachial index (ABI) in 9 studies; ankle pressures in 10 studies, toe-brachial index in 4 studies, toe pressure in 9 studies, transcutaneous oxygen pressure (TcPO2) in 7 studies, skin perfusion pressure in 5 studies, continuous wave Doppler (pedal waveforms) in 2 studies, pedal pulses in 3 studies, and ankle peak systolic velocity in 1 study. Study quality was variable. Common reasons for studies having a moderate or high risk of bias were poorly described study participation, attrition rates, and inadequate adjustment for confounders. In people with DFU, toe pressure =30 mmHg, TcPO2 =25 mmHg, and skin perfusion pressure of =40 mmHg were associated with a moderate to large increase in pretest probability of healing in people with DFU. Toe pressure =30 mmHg was associated with a moderate increase in healing post-minor amputation. An ABI using a threshold of =0.9 did not increase the pretest probability of DFU healing, whereas an ABI <0.5 was associated with a moderate increase in pretest probability of non-healing. Few studies investigated amputation outcomes. An ABI <0.4 demonstrated the largest increase in pretest probability of a major amputation (PLR =10).Conclusions: Prognostic capacity of bedside testing for DFU healing and amputation is variable. A toe pressure =30 mmHg, TcPO2 =25 mmHg, and skin perfusion pressure of =40 mmHg are associated with a moderate to large increase in pretest probability of healing in people with DFU. There are little data available evaluating the prognostic capacity of bedside testing for healing after minor amputation or for major amputation in people with DFU. Current evidence suggests that an ABI <0.4 may be associated with a large increase in risk of major amputation. The findings of this systematic review need to be interpreted in the context of limitations of available evidence, including varying rates of revascularisation, lack of postrevascularisation bedside testing, and heterogenous subpopulations.