RESUMO
INTRODUCTION: Identifying peripheral arterial disease (PAD) during diabetic foot screening (DFS) is crucial in reducing the risk of diabetic foot ulcerations and lower limb amputations. Screening assessments commonly used include absolute systolic toe pressure (ASTP) and toe-brachial index (TBI). There is a lack of research defining the threshold values of both assessment methods. We aimed to compare the accuracy of ASTP and TBI and establish optimal threshold values of ASTP and TBI with reference to the internationally accepted ankle-brachial pressure index (ABPI) screening test, for a multiethnic diabetic population in Singapore. METHODS: A retrospective, observational study of DFS results from January 2017 to December 2017 was conducted. Receiver operating characteristic analysis was conducted for ASTP and TBI using the internationally accepted ABPI cut-off value of ≤0.9 to indicate PAD. RESULTS: A total of 1,454 patients with mean (standard deviation) age of 63.1 (12.4) years old were included. There were 50.8% men and 49.2% women, comprising 69.7% Chinese, 13.5% Indian, 10.1% Malay and 6.7% other ethnicities. Areas under the curve for ASTP and TBI were 0.89 (95% confidence interval [Cl] 0.85-0.94) and 0.94 (95% Cl 0.90-0.98), respectively, and the difference was statistically significant (P<0.001). Derived optimal threshold values to indicate ABPI≤0.9 for ASTP and TBI were <95.5mmHg (specificity 0.86, sensitivity 0.84) and <0.7 (specificity 0.89, sensitivity 0.95), respectively. CONCLUSION: ASTP or TBI may be used to detect ABPI-determined PAD in DFS. The optimal threshold values derived from a multiethnic Asian diabetic population were <95.5mmHg for ASTP and <0.7 for TBI.
