Early reocclusion after successful thrombolysis is related to lesion length and roughness

BACKGROUND: Reocclusion is a significant problem after thrombolysis. Results of previous studies conflict regarding the association of various features of postlytic lesions that might predict reocclusion. METHODS: A computer-assisted algorithm was therefore used to quantitatively measure edge roughness in the 90-minute postlysis angiogram of 84 patients receiving recombinant tissue plasminogen activator within 6 hours of chest pain. RESULTS: Twenty-five patients had reocclusion, and 59 did not. The baseline angiogram showed no differences between these two groups with respect to minimal dimensions or relative percentage of stenosis. Length was greater in the reocclusion group (12.2 +/- 5.0 vs 10.0 +/- 4.2 mm, P < 0.05). Three of four roughness indices based on curvature analysis indicated greater roughness in those patients with reocclusion. These differences were largely due to the increased length of these lesions. The scaled edge-length ratio, an index of roughness that is independent of length, was, however, significantly greater in the reocclusion group (1.15 +/- 0.10 vs 1.09 +/- 0.08, P < 0.006). Multiple regression analysis showed that lesion length, the scaled edge-length ratio, and the number of features (invaginations and evaginations) per cm correlated independently with the risk for reocclusion. CONCLUSIONS: The length and roughness of postlytic residual lesions are determinants of reocclusion.