CORRESPONDENCE: LETTER TO THE EDITOR
Restenosis, Statistics, and Reasonable Inferences
Stephen G. Ellis, MD*,
Jeffrey J. Popma, MD,
Gregg W. Stone, MD and
Mary E. Russell, MD, FACC
* Sones Cardiac, Department of Cardiovascular Medicine, The Cleveland Clinic, Lerner College of Medicine, Case Western Reserve University, Desk F-25, 9500 Euclid Avenue, Cleveland, OH 44195-0001 (Email: elliss{at}ccf.org).
We appreciate the accompanying editorial (1) to our recent report entitled "Relationship Between Angiographic Late Loss and Target Lesion Revascularization After Coronary Stent Implantation: Analysis From the TAXUS-IV Trial" (2) but we wish to clarify several apparent misconceptions. Our principal messages are that in the era of drug-eluting stents (DES), it is not only the mean value but also the shape of the distribution curve (variance and skewedness) that will determine the population target lesion revascularization (TLR), and that with a homogeneous response in-stent late losses up to about 0.75 mm may provide acceptable clinical results (2). That DES may have rightward skewed late loss histograms has been previously reported (3). The fact that the individual patient late loss/TLR relationship is curvilinear with an apparent inflection point, rather than linear, had not been reported and is a novel and unique observation that has since been replicated in several other DES trials (DELIVER, ENDEAVOR-II). Most importantly, given the rightward skew in patient population data seen in all these trials (including the pivotal SIRIUS trial of the sirolimus-eluting stent), a certain lower level or "floor" of TLR may be unavoidable, somewhat independent of the mean late loss.
Moreover, recent data reported at the most recent American College of Cardiology meetings substantiate our findings. The ENDEAVOR-II trial, with a considerably higher mean late loss (0.62 mm) but with rather a homogeneous effect (standard deviation of late loss 0.46), reported a low TLR of 4.6% with the ABT-578-eluting stent. The large-scale REALITY trial, comparing the sirolimus-eluting and paclitaxel-eluting stents, found significantly greater in-stent late loss with the latter (0.09 vs. 0.31 mm, respectively, p < 0.001), but nearly identical eight-month TLR rates (5.0% vs. 5.4%, p = 0.81). Thus, given the patient and lesion complexity studied in the pivotal DES trials to date, an "acceptable" TLR can be achieved with a relatively high in-stent late loss, providing that a homogeneous response is seen.
In addition, other variables beyond angiographic late loss that may affect TLR rates must be considered, including the inaccuracies and variability of quantitative measures of late loss. Variable thresholds of patient angina perception, follow-up ischemia detection, and practice to refer patients for subsequent repeat catheterization and revascularization procedures will also minimize the practical relevance of any differences in late loss. In the end, probably all would agree that, up to a point, less late loss is better, but the range of late loss that is associated with infrequent clinically driven TLR may be wider than previously believed. Whether this relationship holds in more complex and challenging lesion subsets will be analyzed in the TAXUS-V and -VI studies.
 |
References
|
|---|
- Kereiakes DJ, Kuntz RE, Mauri L, Krucoff MW. Surrogates, substudies, and real clinical end points in trials of drug-eluting stents J Am Coll Cardiol 2005;45:1206-1212.[Free Full Text]
- Ellis SG, Popma JJ, Lasala JM, et al. Relationship between angiographic late loss and target lesion revascularization after coronary stent implantationanalysis from the TAXUS-IV trial. J Am Coll Cardiol 2005;45:1193-1200.[Abstract/Free Full Text]
- Lemos PA, Mercado N, van Domburg RT, Kuntz RE, O’Neill WW, Serruys PW. Comparison of late luminal loss response pattern after sirolimus-eluting stent implantation or conventional stenting Circulation 2004;110:3199-3205.[Abstract/Free Full Text]
Top
References