The aim of this study was to investigate the sensitivity and specificity of changes of the ankle/brachial pressure index (ABI) and changes in absolute ankle pressure values to detect restenosis in patients who underwent femoropopliteal percutaneous transluminal angioplasty (PTA). In total, 171 patients were followed up prospectively for 12 months; sensitivity and specificity of Doppler-based diagnosis were calculated with duplex scanning as the gold standard. The criteria for restenosis were: (1) a loss of 50% of the ABI increase or (2) loss of 50% of the absolute ankle systolic pressure, gained by PTA. For both criteria, different cut-off points (minimum increase of ABI or ankle pressure gained by PTA) were evaluated. The overall sensitivity and specificity of the ABI criterion was 67% and 80%, respectively. The introduction of cut-off points (the minimum ABI increase gained by PTA), ranging between > or = 0.13 and > or = 0.35, did not markedly improve the results. The overall sensitivity and specificity of the absolute ankle pressure criterion again was poor (59% and 81%). With the introduction of cut-off points (the minimum increase of absolute ankle pressure gained by PTA) ranging between > or = 15 mm Hg and > or = 20 mm Hg, the sensitivity and specificity of the criterion improved to acceptable 92% and 96%, respectively. It is concluded, that in the long-term follow-up of PTA patients, the "loss of 50% ankle pressure" criterion will detect restenosis with reasonable accuracy in those patients, in whom an increase in systolic ankle pressure > or = 20 mm Hg is warranted.