Author + information
- C. Michael Gibson, MS, MD, FACC⁎ ( and )
- Eugene Braunwald, MD, MACC
- ↵⁎350 Longwood Avenue, First Floor, Boston, Massachusetts 02115
We thank Dr. Valgimigli et al. for their careful review of the PROTECT–TIMI-30 (Randomized Trial to Evaluate the Relative PROTECTion against Post-PCI Microvascular Dysfunction and Post-PCI Ischemia among Anti-Platelet and Anti-Thrombotic Agents–Thrombolysis In Myocardial Infarction-30) study (1). The corrected Thrombolysis In Myocardial Infarction (TIMI) frame count (cTFC) measures the time that it takes for blood to traverse the entire length of the artery and reflects the average velocity throughout the entire length of the artery. In contrast, a Doppler velocity wire measures the velocity at a single point in the artery. As shown in the angiogram that accompanies our report (1), blood velocity can vary significantly throughout the length of the artery, and it is therefore no surprise that the cTFC (average velocity) and Doppler velocity (velocity at a single point) measures may be discordant in their assessment of basal blood velocity given the heterogeneity in velocity throughout an artery.
The ability to speed up this basal blood velocity is the coronary flow reserve (CFR). The cTFC measures how much faster blood passes through the same length of the entire artery before and after adenosine. In contrast, a Doppler velocity wire measures whether the velocity accelerated at a single point in the artery. As opposed to potential discrepancies in basal velocity, the CFR based on the cTFC has been related to Doppler velocity measures (r = 0.88) (2).
Although the basal cTFC has been related to mortality and post-percutaneous coronary intervention (PCI) events, Valgimigli et al. correctly point out that CFR by the cTFC method has not been related to clinical outcomes (3), and it should be added that Doppler velocity measures have not been related to outcomes as well. In contrast, the TIMI myocardial perfusion grade (TMPG) was related to ischemic and clinical outcomes following PCI (3).
Dr. Valgimigli et al. point out that CFR could not be assessed in patients with a failed PCI procedure in whom there was abrupt vessel closure, emergency coronary artery bypass graft surgery, or thrombotic closure with bailout to eptifibatide. To account for these failed procedures, an additional analysis was stated a priori in detail in the statistical analysis plan prior to unblinding in which the lowest (i.e., worst) CFR from the prior ESPRIT (Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrilin Therapy) substudy (0.55) was imputed to those patients with a failed procedure. When patients with a failed procedure were included in the analysis, no difference was seen between bivalirudin and eptifibatide in the primary end point (median of 1.43 for bivalirudin vs. 1.33 for pooled eptifibatide arms, p = 0.13). To comply with the requests of the reviewers, however, the analysis in the published report included only patients with a successful PCI and does not account for patients with a failed procedure.
Given the potential for clopidogrel pretreatment to confound the efficacy outcomes in the trial, randomization in the trial was stratified by clopidogrel pretreatment, and it was prespecified that the key analyses would be adjusted for clopidogrel pretreatment. In a prespecified analysis, there was a significant interaction term between clopidogrel pretreatment and the randomization arm with respect to the end point of death or myocardial infarction (MI) (p = 0.031 for interaction between clopidogrel pretreatment and randomization arm). Among patients treated with clopidogrel for ≤6 h before angiography, the risk of death/MI was significantly lower among patients treated with eptifibatide compared with bivalirudin (5.0% vs. 10.3%, n = 548, p = 0.022). In the smaller cohort of 309 patients treated with clopidogrel for >6 h, no significant difference existed in the incidence of death/MI between eptifibatide and bivalirudin (9.5% vs. 6.1%, p = 0.31). Among troponin-positive patients treated for ≤6 h with clopidogrel, the risk of death/MI was significantly lower among patients treated with eptifibatide (4.1%, n = 197 vs. 13.2%, n = 106, p = 0.003).
Please note: The PROTECT trial was supported in part by a grant from Millennium Pharmaceuticals, Cambridge, Massachusetts, and Schering-Plough Research Institute, Kenilworth, New Jersey. Dr. Gibson has received grant support and honorarium and has served as a consultant for the Medicines Company, Millennium Pharmaceuticals, and Schering-Plough. Drs. Gibson and Braunwald have also received research and grant support from Sanofi Aventis. Reagents and support for biomarker testing were provided by Dade-Behring, Inc., Deerfield, Illinois.
- American College of Cardiology Foundation
- Gibson C.M.,
- Morrow D.A.,
- Murphy S.A.,
- et al.