Late Loss and TLR in Challenging Vessels

Dr. Lasala received the Outstanding Cardiology Alumnus Award from Yale University School of Medicine in 2000, where he did his cardiology and interventional cardiology fellowships. Currently, Dr. Lasala is Associate Professor of Medicine, Washington University School of Medicine; Director of Interventional Cardiology and Medical Director of the Cardiac Catheterization Lab, Barnes-Jewish Hospital, St. Louis, Missouri. Dr. Lasala is the Primary Investigator for over 33 research trials and the Co-Investigator for 13 studies, including ARRIVE. He has had over 40 medical peer-reviewed manuscripts published and has various editorial responsibilities, including the American Heart Journal, American Journal of Cardiology, Journal of American College of Cardiology, Circulation and Catheterization & Cardiovascular Intervention. Special research interests include congenital heart disease and stereotaxis. What is late lumen loss and how is it measured? Late lumen loss is the difference in millimeters between the diameter of a stented segment post-procedure compared with the follow-up angiogram at six or nine months. We expect some amount of tissue to grow in after implantation of the stent, and we want some neointimal growth to cover the stent, but with enough suppression to reduce restenosis. It is important to note that late lumen loss does take into account the worst-case scenario within the analysis region and therefore may not be uniform. What are late loss models and what do they tell us? There are many accepted models that analyze late lumen loss in DES trials. One of these statistical models, Ellis et al. (JACC Vol. 45, No. 8, 2005), studied in the TAXUS trials, demonstrates a curvilinear relationship between late lumen loss and risk of having a target lesion revascularization (TLR). Another model, the Mauri, Kuntz, et al. model, in the SIRIUS trials, correlates late lumen loss primarily to binary angiographic restenosis(BAR), another secondary endpoint that represents vessel narrowing of greater than or equal to 50%. Drs. Mauri and Kuntz provide an elegant description of the monotonic relationship between late loss and BAR that is relatively constant from trial to trial. This model suggests that as late loss increases, BAR increases in a predictable fashion. The Ellis et al. model supports a curvilinear relationship between late lumen loss and TLR. In other words, increase in risk of TLR is not linear over the entire range of the late lumen loss. The probability of TLR is low and only increases significantly when late loss reaches a moderate level, which, as suggested in the TAXUS trial data, is around 0.5 mm to 0.6 mm. This model demonstrates that you can accommodate a variable range of tissue growth before seeing any meaningful differences in TLR, which was demonstrated in the TAXUS trials, SIRIUS studies and in the REALITY trial as follows: TAXUS IV In-stent late loss at 9 months is 0.39 mm with a TLR rate of 3.0%. SIRIUS In-stent late loss at 9 months is 0.17 mm with a TLR rate of 4.1%. REALITY In-stent late loss at 8 months for the TAXUS® Stent was 0.31 mm with a TLR rate of 5.4%. For the Cypher® Stent, in-stent late loss at 8 months was 0.09 mm with a TLR rate of 5.0%. This relatively flat response in the initial range of late loss suggested a weak correlation of late loss as a secondary to the clinical endpoint of TLR. Most of the late loss amounts observed in various DES trials are, however, in the range of 0.6 mm or less. This is the reason why it is difficult to show differences between sirolimus and paclitaxel in TLR rates despite the fact that both have different late loss measurements. As with all statistical models, both the Ellis et al. and the Mauri, Kuntz et al. models have limitations based on how they were designed and data points analyzed. What is the value of late lumen loss as a secondary endpoint? Why is TLR the most relevant endpoint to physicians and patients? Late lumen loss as a surrogate endpoint for restenosis is used in DES trials because it significantly reduces the number of patients that need to be recruited. For example, using late loss as a secondary endpoint in non-inferiority trials allows investigators to use fewer patients than the number studied in the TAXUS and SIRIUS clinical trials. Also, late loss is used in DES trials to evaluate durability of DES over time. As both SIRIUS and TAXUS trials require long term follow-up greater than one year, late loss may be most appropriately used to see if any changes occur over time that would signal late catch-up in lumen narrowing. Lastly, patients are more informed and aware of DES technology, which can make it challenging for an interventionalist to randomize a patient to a bare metal stent control arm. TLR is the most relevant endpoint to physicians and most importantly, to patients. As interventionalists, we are primarily concerned with whether our patients are going to come back with recurrent symptoms. By definition, TLR is what ultimately brings the patient back. The term headroom is often used relative to late lumen loss as vessel size decreases. What does it mean and what data exist to show what happens to late lumen loss as vessel size decreases? Headroom is another term for buffer. Smaller vessels have less headroom, which means that they cannot accommodate as much in-growth of tissue compared with larger diameters. In the TAXUS IV trial, we saw a decrease in late lumen loss as the size of the vessel decreased (see Figure 1). The variable late loss in TAXUS IV may, in theory, be explained by dose density. As vessel diameter decreases, the struts of similar stents are closer together. This increases the concentration of the drug at the site of the lesion and therefore may result in less neointimal proliferation. The decrease in late loss in TAXUS IV may suggest that as dose increases, so may its efficacy a concept worthy of continued study. Vessel diameter doesn’t appear to be the sole contributing factor of TLR or restenosis. How do vessel length and lesion type play a role? The three most significant predictors of restenosis are: 1) lesion length and/or stented length, 2) inverse relationship to vessel diameter and 3) presence or absence of diabetes as seen in various clinical trials. For example, the TAXUS V small vessel cohort at 9 months showed a TLR rate of 10.4% with increased lesion lengths greater than or equal to 16.44 mm. In the SIRIUS 2.25 mm registry, the third tertile (lesion lengths greater than or equal to 13.44 mm) showed a TLR rate of 12.6%. Both studies suggest that vessel length and lesion type play a significant role in restenosis and thus subsequently increase TLR. Neither late lumen loss nor TLR matter if you cannot get the stent to the lesion. Why does deliverability become more important as vessel complexity increases? If I perform five or six angioplasties in one day, statistically there are likely going to be one or two during which I’ll struggle with delivery. It is clear that delivery is now the rate-limiting step in interventional procedures. Regardless of the type of lesion you are treating, flexible and deliverable stents are especially important in dealing with small vessels. The safety and efficacy of the TAXUS Express2 Stent have not been established in vessels smaller than 2.5 mm. The safety and efficacy of the Cypher Stent have not been established in vessels smaller than 2.5 mm. TAXUS is a trademark of Boston Scientific. Cypher is a trademark of Cordis Corp. Sponsored by Boston Scientific Corporation.

NULL