Presence of ultrastructural arterial lesions in muscle and skin vessels of patients with CADASIL.

Abstract

ResponseWe are grateful to Drs Norris and Bladin for their interest in our review of the relative merits of ticlopidine and aspirin, 1 but we are not sure whether there is a baby in the bathwater at all.We feel less confident than they do that the point estimate of the overall advantage of ticlopidine over aspirin (8% with 3500 patients) would remain the same with greater numbers and would then reach statistical significance.This is what 95% confidence intervals (-4% to 19%) are all about: we simply don't know.By the way, we could not find a P of <.OO7 in the paper about the Ticlopidine Aspirin Stroke Study (nor indeed any significant difference for any outcome event that included more than stroke alone) 2 ; is this perhaps a subgroup analysis that did not survive the final stages of publication?What the report does contain is a table of side effects, totalling 62% in the ticlopidine group and 53% in the aspirin group (with 1300 mg!).These proportions have not been weighted for severity, but this uncertainty can be interpreted both ways, as it can for efficacy.