Abstract
To the Editor: The Multicenter Diltiazem Postinfarction Trial Research Group (Aug. 18 issue)1 reported no overall effect of diltiazem on cardiac events and mortality in a randomized trial in 2466 patients with previous infarction. The authors suggested that this lack of superiority to placebo may have been due to an apparently striking (P = 0.0042) bidirectional effect, depending on the presence of left ventricular dysfunction. Although they are cautious in the analysis and interpretation of the trial, they nevertheless state that these data constitute "strong evidence" of a treatment-by-subset interaction and finally conclude ambiguously that "diltiazem is not appropriate for . . .