Long-term Effects of Angiopeptin Treatment in Coronary Angioplasty: Reduction of Clinical Events but not Angiographic Restenosis
Serruys, Patrick W. MD PhD
Author Information
Professor of Interventional Cardiology, Erasmus University, University Hospital Dijkzigt, Rotterdam, Netherlands.
 
To the Editor:
 
We read with great interest the results of the European Angiopeptin trial, which demonstrated that angiopeptin significantly decreases the incidence of clinical events after percutaneous transluminal coronary angioplasty (PTCA), principally the rate of revascularization procedures, while having no significant effect on angiographic variables. We believe that the apparent discordance between clinical and angiographic outcome demonstrated in this study is a spurious finding and a direct result of some of the inherent logistic limitations associated with restenosis trials. In particular, we believe it can be simply explained by the timing of randomization and follow-up duration.

 
Because previous animal studies had shown loss of therapeutic efficacy if angiopeptin was not begun before the procedure, randomization in this trial was performed before PTCA. However, this had the effect that the subsequent intention-to-treat analysis contained 54 (20%) patients in the placebo group and 44 (16%) patients in the angiopeptin-treated group who either had no PTCA or an unsuccessful PTCA. Although one can argue that all failure between first balloon inflation and the end of the procedure could have been influenced by the trial medication and should therefore be counted as clinical end points, the aim of this trial was to study the effect of angiopeptin on the inhibition of neointimal hyperplasia after balloon wall injury, it would have seemed reasonable to exclude from the analysis of the main clinical end points, those patients in whom no balloon inflation had occurred. It is not clear what effect this may have had on the subsequent analysis with regard to clinical outcome particularly as 10 patients in the placebo group versus 3 in the angiopeptin-treated group had subsequent elective coronary artery bypass grafting.

 
The timing of angiographic and clinical follow-up also could have played a substantial role in the apparent discordance. Quantitative coronary angiography was performed at 6 months' follow-up or earlier if symptoms recurred, but the clinical follow-up and the incidence of clinical events was at 12 months. Although this may have led to the inclusion of any dilatations triggered by the restudy it appears to have introduced another problem in that Figure 1, showing the event-free survival curves, shows the greatest divergence between the two groups to occur after the 6-month angiographic follow-up. We estimate from the graph that the event-free survival rates for angiopeptin and placebo, at the time of the QCA data, to be 73% and 79%, respectively, which is not statistically significant. It is not clear why the curves subsequently diverge so markedly after the 6-month angiographic follow-up, especially as previous studies have shown a plateau in the incidence of restenosis after 6 months.

 
We thus believe that the discordance between quantitative angiography and clinical outcome in this study is a "pseudodiscordance," precipitated by variations in the timing of randomization, the intention-to-treat analysis, and differences between the timing of quantitative angiography and clinical follow-up. One way of verifying this would be to reanalyze the data on the basis of successful PTCAs with clinical and angiographic follow-up to 6 months.

 
Patrick W. Serruys, MD, PhD, Professor of Interventional Cardiology, Erasmus University, University Hospital Dijkzigt, Rotterdam, Netherlands.