Optimal activated clotting time during percutaneous coronary intervention.

Percutaneous Coronary Intervention To the Editor: We read with interest the article by Chew et al1 in the February 20, 2001, issue of Circulation, but we were disappointed that such eminent investigators could confuse associations arising from heterogeneous pooled data with hard evidence on which scientific conclusions can be drawn. In their Conclusions, they state that their meta-analysis “. . .identifies an optimal level of ACT [activated clotting time] for patients undergoing PCI [percutaneous coronary intervention] in the range of 350 to 375 s.” Their data, however, merely show that when aiming for an ACT of 300 to 350 s, an ACT of ,300 s or .475 s is a marker for increased events. It is quite possible that the level of anticoagulation has nothing to do with the increase in events. Our conclusions from these data were that a high ACT after standard-dose heparin most likely reflects low body weight, a well-recognized cause of increased events.2 Logistic regression may have helped clarify the relative importance of low body weight versus ACT in determining events in this group. Similarly, a low ACT with a standard heparin dose (heparin resistance) is a marker of systemic acute-phase response and would be associated with more actively inflamed plaques and thus a higher acute event rate, independent of anticoagulation level. Only 85% of patients had an ACT measured, and it was ,300 s in .50% of subjects. Measurement and maintenance of ACT were obviously not an important part of the study protocols. Also, non-target ACTs may have been associated with poor operators and related complications. There was also marked heterogeneity of the patient populations and heparin protocols in the different trials. The composite end points varied from 10.8% in the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) study3 to 17.8% in the ReoPro and Primary PTCA Organization and Randomized Trial (RAPPORT).4 Major bleeding rates also varied significantly, from 2.2% in EPISTENT3 to 9.5% in RAPPORT.4 Because the heparin regimens varied from fixed bolus2 to weight-adjusted bolus with titration3 with or without postprocedural infusion,3,5 it is possible that non-target ACTs may have been more prevalent in trials with higher event rates. Any correlation between ACT and end points should have been adjusted in a multivariate manner for the trial. In their Discussion, the authors mentioned that data from well-designed, prospective, randomized trials have contradicted the conclusions of this meta-analysis. Because of the difficulty with confounding variables, nonprospective data pooling must be regarded as inferior evidence to these trials, irrespective of the numbers involved, and cannot support changes in clinical practice.