Comparison of Late Luminal Loss Response Pattern After Sirolimus-Eluting Stent Implantation or Conventional Stenting
Background— We investigated the pattern of late luminal loss after sirolimus-eluting or bare stent implantation.
Methods and Results— The study population comprised 238 patients treated with sirolimus-eluting stents and 526 patients treated with conventional stents. The distribution of late loss of sirolimus stents was largely skewed to the right and differed from the distribution for bare stents. When divided according to the presence of binary restenosis (diameter stenosis >50%), restenotic lesions in the bare stent group (26.0%) had a late loss of 1.40±0.64 mm and in the sirolimus group (7.9%) of 1.16±0.76 mm. Nonrestenotic lesions in the bare stent group had a late loss of 0.58±0.44 mm, whereas the late loss of nonrestenotic lesions in the sirolimus group remained close to zero (−0.05±0.33 mm). Differences between poststenting and follow-up measurements in the sirolimus group (late loss) resembled variations observed in repeated angiographic measurements, as assessed from a random sample of 30 segments measured repeatedly. After multivariate adjustment, stent type did not influence the degree of late loss in restenotic lesions. However, nonrestenotic bare stents had a significantly larger estimated luminal loss (0.58 mm; 95% CI, 0.52 to 0.65) than sirolimus-eluting stents, for which the predicted late loss was almost 0 (−0.04 mm; 95% CI, −0.10 to 0.02).
Conclusions— The pattern of late loss after sirolimus-eluting stent implantation follows a peculiar behavior, different from lesions treated with conventional stents. Whether this is explained by an unusual statistical distribution or a biological all-or-none response of restenosis after sirolimus-eluting stenting remains to be investigated.
Received December 1, 2003; de novo received March 4, 2004; revision received June 30, 2004; accepted July 14, 2004.
The pathophysiological processes involved in lumen renarrowing after percutaneous coronary intervention have largely been studied over the past decades. However, it remains a matter of debate whether restenosis simply represents an extreme form of the “normal” vessel healing response after mechanical dilatation or if it is related to specific mechanisms that ultimately lead to vessel renarrowing.1–7 Although several cutoff criteria have been proposed to dichotomize patients with restenosis from those without restenosis, it has been widely recognized that, to some extent, late luminal reduction is a ubiquitous phenomenon, occurring even in those categorized as not having binary restenosis.1,7
Recently, drug-eluting stents with the antiproliferative agent sirolimus have proved to reduce neointimal growth markedly in clinical trials. In the First-in-Man (FIM) study8 and the Randomized Study With the Sirolimus-Eluting Bx Velocity Balloon-Expandable Stent in the Treatment of Patients With De Novo Native Coronary Artery Lesions (RAVEL),9,10 no cases of binary angiographic restenosis were seen after sirolimus-eluting stent (SES) implantation. Restenosis (diameter stenosis >50% at follow-up) after SES implantation occurred in 9% of cases (compared with 36% with bare stents; P<0.001) in the recent randomized Sirolimus-Eluting Bx Velocity Balloon Expandable Stent Trial (SIRIUS).11 However, SIRIUS included more complex patients than RAVEL and FIM, which could at least partially account for the occurrence of restenosis in some cases. Moreover, we have recently shown that post-SES restenosis frequently occurs in association with particular local conditions in complex cases.12 Nonetheless, the late luminal renarrowing response after SES implantation currently is poorly understood. In the present study, we examine the pattern of late luminal loss after SESs compared with conventional bare metal stent implantation.
Study Design and Patient Population
Since April 2002, SESs (Cypher; Johnson & Johnson-Cordis Unit, Cordis Europa NV) have been the device of choice for every percutaneous procedure in our institution, as detailed elsewhere.13 The final interventional strategy was left to the discretion of the operator, as was the use of periprocedural glycoprotein IIb/IIIa inhibitors and antithrombotic medications.
Patients treated with SESs were considered candidates for angiographic reevaluation if presenting with at least one of the following: (1) treatment of acute myocardial infarction, (2) treatment of in-stent restenosis, (3) use of a very small SES (2.25-mm nominal diameter), (4) treatment of the left main coronary, (5) treatment of chronic total occlusion (>3 months), (6) adjacent stented segment >36 mm, and (7) bifurcation stenting (SES implanted in the both the main vessel and the side branch). Patients not undergoing a repeated intervention in the first month and without a formal medical contraindication were considered eligible for angiographic follow-up between 6 and 8 months. Importantly, although all such patients were approached for angiographic follow-up, patient refusal was not considered an exclusion criterion to be treated with an SES. Angiographic restudy was not requested for nonresidents of the Netherlands.
During the first 6 months of enrollment, a total of 362 consecutive patients had at least one of the above criteria (57% of all patients treated with SESs in the period). Of these, 2 patients moved to another country, 10 patients had died at the 6-month follow-up, 6 patients had repeated intervention before 30 days, and 3 patients were considered to have a medical contraindication to the angiographic follow-up. Of the remaining 341 patients, angiographic follow-up (204±34 days) was obtained from 238 patients (70% of eligible patients), who make up the present study population. A total of 441 lesions were treated and included in the present report.14 Apart from being older (63±12 versus 60±12 years; P=0.02), excluded patients had baseline characteristics similar to those of patients with angiographic follow-up. The hospital ethics committee approved this protocol, and written informed consent was obtained from every patient.
To better evaluate the angiographic outcomes of patients treated with an SES, a control group for comparison was composed of patients treated with bare metal stents included in the Evaluation of Oral Xemilofiban in Controlling Thrombotic Events (EXCITE) trial, the study design and main results of which have been reported elsewhere.15 Briefly, EXCITE randomized a total of 7232 patients to the oral glycoprotein IIb/IIIa inhibitor xemilofiban or placebo, which was administered before percutaneous coronary revascularization and maintained for up to 6 months. The trial included a substudy of 526 patients (734 lesions) treated with coronary stenting, for whom baseline, postprocedure, and 6-month angiographic reevaluation was obtained and who made up the control population of the present study. Because of the negative results of the trial in preventing restenosis, all patients were included in the present report regardless of the allocated treatment.
Quantitative Coronary Angiography
Coronary angiograms were obtained before intervention, after stenting, and at follow-up. The projection showing the maximal degree of stenosis (worst view) was selected at baseline and used for the subsequent analysis. Quantitative coronary angiographic analysis was performed as previously described with a validated computer-based edge-detection system (CASS II, Pie Medical).16 Interpolated reference diameter, minimal luminal diameter, and diameter stenosis were measured at all time points. Late loss was calculated as the difference between the minimal luminal diameter after stenting and at follow-up. The target lesion was defined as the entire segment involving the implanted stent and the 5-mm distal and proximal borders adjacent to the stent.
Categorical variables were expressed as percentages and compared by Fisher exact test. Continuous variables were presented by their mean and SD and compared by Student t test. The frequency distribution of late loss was graphically represented with histograms and tested for normality with the Kolmogorov-Smirnov goodness-of-fit test. The method proposed by Bland and Altman13 was used to evaluate the variation in the measurements of luminal diameters after the procedure and at follow-up, as well as the variations between repeated measurements in a random sample of segments in the index angiogram. In the Bland-Altman method, the average of the 2 measurements for an individual lesion is plotted against the difference between them, with 95% limits of agreement being calculated to evaluate the measurement concordance. Also, the regression line of difference on average was depicted in the graphs, as well as the correspondent R2 value. Multivariate estimates of late loss were calculated by general linear models adjusted for stent type (bare stents or SESs) and for the following baseline and procedural variables (which differed between the study groups): gender, diabetes, clinical syndrome at presentation, treated vessel, lesion length, reference diameter, and postprocedure minimal luminal diameter.
Patients treated with an SES had a higher risk profile for restenosis than the control group with bare stents, according to previously proposed risk factors (Table 1). Specifically, in the SES group, the lesion was longer (16.1±11.8 versus 10.0±7.5 mm; P<0.01), reference vessel diameter was smaller (2.50±0.61 versus 2.80±0.59 mm; P<0.01), and postprocedural minimal luminal diameter was smaller (2.13±0.58 versus 2.43±0.54 mm; P<0.01). Diabetes tended to be more prevalent among patients treated with an SES, although not reaching statistical significance (22% versus 17%; P=0.1). Nevertheless, the total binary restenosis rate was significantly lower (7.9% versus 26.0%; P<0.01) and the overall late lumen loss was significantly smaller (0.04±0.49 versus 0.80±0.61; P<0.01) in the SES group than in the bare stent group.
Frequency Distribution of Late Loss
Overall, patients treated with bare stents had an average late loss of 0.80 mm (range, −0.75 to 3.48 mm), and patients treated with SES had an average late loss of 0.04 mm (range, −1.12 to 2.97 mm; P<0.01 versus bare stents). Although presenting an apparent bell-shaped pattern (Figure 1), the frequency distribution of late loss values strongly deviated from normality for both stent groups (P<0.001 by Kolmogorov-Smirnov test for normality for both). The frequency distribution of late loss was analyzed after the lesions were separated into 2 categories according to several proposed definitions for binary restenosis.4,17,18 Also, lesions were separated according to the previously described intersection point between lesions with and without excessive late loss after conventional stenting (ie, minimal luminal diameter at follow-up of 1.09 mm).7 All analyses gave similar findings, and further results of this study are presented for the most commonly used criterion of binary restenosis, which defines restenosis as a luminal diameter stenosis >50% at follow-up.
The frequency of late loss values of restenotic lesions among control subjects (26.0% of lesions; mean late loss, 1.40±0.64 mm) followed a bell-shaped format with a tendency to deviate from normality, although not reaching statistical significance (P=0.1 by Kolmogorov-Smirnov test for normality) (Figure 2), whereas the late loss of lesions with binary angiographic restenosis after SES implantation (7.9% of lesions) had a mean late loss of 1.16±0.76 mm (P=0.02 versus bare stents) and presented an uncharacteristic distribution pattern (Figure 2). Because lesions presenting with total occlusions at follow-up (TIMI flow 0 or I) may be associated with distinct physiopathological processes of lumen renarrowing (eg, thrombosis instead of progressive neointimal proliferation)7 and because of the low incidence of late total occlusions in both groups (0.9% for SES versus 4.1% for bare stent; P<0.01), restenotic lesions were also analyzed after exclusion of occlusions. Among control subjects, the frequency of late loss for nonoccluded restenotic lesions presented a bell-shaped format resembling a normal distribution (P=0.5 by Kolmogorov-Smirnov test for normality), with an average lumen loss of 1.24±0.52 mm (Figure 2). In the SES group, nonoccluded restenosis presented an uncharacteristic distribution pattern (mean late loss, 0.96±0.64 mm; P<0.01 versus bare stents) (Figure 2).
Nonrestenotic lesions in the bare stent group presented a frequency distribution of late loss close to normality (P=0.3 by Kolmogorov-Smirnov test for normality), with an average late loss of 0.58±0.44 mm (Figure 3). However, nonrestenotic lesions in SES presented a mean late loss close to 0 (−0.05 mm; SD, 0.33 mm) and a frequency distribution also close to normality (P=0.5 by Kolmogorov-Smirnov test for normality) (Figure 3).
Repeatability of Measurements
Previous studies from our institution have evaluated the medium-term repeatability of quantitative angiographic measurements. For this analysis, repeated images were obtained within an interval of some minutes (acquisition performed at the beginning and end of the catheterization), with the x-ray system repositioned at the same projection after being moved to acquire other views.16,19 Medium-term repeatability accuracy was defined as the average difference between the measurements of both images; medium-term repeatability precision was defined as the SD of the differences between repeated measurements. The difference between both measurements has previously been reported to average 0.03 mm (medium-term repeatability accuracy), with an SD of 0.18 mm (medium-term repeatability precision).16,19 Coincidentally, the calculations performed to assess accuracy and precision of repeated measurements are the same as those used for late loss (ie, the difference of luminal diameters between 2 sequential angiograms) and its SD, respectively. Interestingly, the late loss and SD of lesions without binary restenosis in the SES group (and not in the bare stent group) were similar to the accuracy and precision of repeated measurements of the same vessel segment. These findings suggest that the values of late loss for lesions with no restenosis after SES implantation may potentially reflect solely the variability of repeated measurements, with a minimal (or absent) component resulting from actual neointimal accumulation (or vessel enlargement). Conversely, even when not classified as restenotic according to binary criteria, lesions treated with bare stents did present some extent of luminal loss, implying that mild neointimal proliferation may also be present in nonrestenotic lesions after conventional stenting.
To further evaluate this concept, we analyzed in our patients the intraprocedural measurement repeatability of a random sample of 30 vessel segments not related to the target lesion. In this analysis, the selected segments were measured in the same projection at the beginning and end of the index procedure; all paired measurements were done blindly without knowledge of the matched values. The vessel diameters differed between baseline and the end of the procedure by −0.02 mm (95% CI, −0.59 to 0.56 mm) (repeatability accuracy) with an SD of 0.29 mm (repeatability precision). The corresponding Bland-Altman plot is depicted in Figure 4. For comparison, the Bland-Altman plot is also shown for late loss (difference in diameters after stenting and at follow-up) of lesions without restenosis in the SES group (late loss, −0.05 mm; 95% CI, −0.69 to 0.59 mm) and in the bare stent group (late loss, 0.58 mm; 95% CI, −0.28 to 1.45 mm) (Figure 4). The resemblance between the analyses for repeated measurements and late loss in the SES stents was evident and clearly differed from the graph for bare stents. In addition, the regression lines of difference on average (and respective R2 values) are shown and demonstrate a strikingly similar behavior of the values through the range for the repeated measurements and the SES group.
Multivariate Estimates of Late Loss
To correct for baseline and procedural differences between the SES and bare stent groups, we performed a multivariate analysis to estimate the value of late loss after adjustment for potential confounders. Estimates of late lumen loss for patients treated with bare stents or SES are shown in Table 2. Restenotic lesions had a comparable estimated late lumen loss whether occurring in bare stents (1.26 mm; 95% CI, 1.15 to 1.36 mm) or in SESs (1.32 mm; 95% CI, 1.14 to 1.51 mm) (for the effect of stent type, P=0.7), even after exclusion of late total occlusions. Adjusted estimates for nonrestenotic lesions, however, have shown a different behavior for bare stents and SESs. Although nonrestenotic bare stents had a predicted late loss of 0.58 mm (95% CI, 0.52 to 0.65), the predicted late luminal loss of SESs without restenosis was ≈0 (−0.04 mm; 95% CI, −0.10 to 0.02) (for the effect of stent type, P<0.01).
The present study shows that the pattern of angiographic late lumen loss after SES implantation follows a peculiar behavior that differed from lesions treated with conventional stents. The distribution of late loss of SES appeared largely skewed to the right (ie, most late loss tended toward smaller values), which could merely represent a distinct feature explained by an unusual statistical distribution. Nevertheless, the possibility of a biological all-or-none response of restenosis after SES implantation remains to be investigated. Substantial luminal renarrowing occurred in a minority of lesions diagnosed as restenotic, according to binary definitions. In most patients, however, luminal dimensions were maintained, with no late loss at follow-up. In the latter group, eventual differences in luminal measurements between poststenting and follow-up resembled variations expected to occur in repeated angiographic measurements. This pattern of late angiographic outcome differed from that observed after bare stent implantation. After conventional stenting, the late loss of nonrestenotic lesions in bare stents (adjusted estimate, 0.58 mm) was significantly higher than in nonrestenotic SESs, which predicted that late loss was maintained close to 0.
It has been reported that some degree of late loss occurs even for nonrestenotic lesions after percutaneous interventions with bare stents.1,7 In a previous study with conventional stenting, Schomig et al7 have shown that lesions in the lower range of lumen renarrowing still presented a late lumen loss of ≈0.5 mm, a number similar to that observed in our control group. SES implantation, however, has been shown in our series to virtually abolish neointimal formation in nonrestenotic lesions. The elimination of neointima creates a peculiar scenario in which the angiograms of nonrestenotic lesions obtained immediately after the procedure and at follow-up are usually indistinguishable. In this context, the measurements performed to calculate late loss (ie, the difference in luminal diameters between both angiograms) actually mimic repeated measurements of the “same” angiogram, a fact that is readily appreciated by the average “late loss” of ≈0. Moreover, the SD of late loss measurements (0.3 mm) represents the normal fluctuations of repeated measurements performed in the same segment. Interestingly, similar findings were observed in the FIM and RAVEL studies, in which all cases were free of restenosis. In the FIM study,8 late loss was 0.16±0.3 mm; in the RAVEL trial, late loss was reported to be −0.01±0.33 mm9,10 (slow-release formulation for both studies).
It is noteworthy that, because of the relatively limited number of patients treated with SESs in the present study, it is not possible to rule out that systemic or patient-based factors may induce, in rare cases, some extent of mild neointimal proliferation even in the absence of restenosis. However, the possibility that restenosis after SESs may follow an all-or-none response does not necessarily imply the presence of biological drug resistance. Indeed, preliminary reports have shown that post-SES restenosis is associated with a variety of lesion- and stent-related conditions,12,20 which underscores the importance of local mechanisms in the pathophysiology of the condition. Furthermore, the fact that neointimal proliferation in SES restenosis is very localized (instead of diffuse) and frequently bordered by segments with no evidence of neointima12 indicates that an absolute lack of drug action is unlikely in most cases.
In addition, because of the low incidence of lesions presenting actual luminal renarrowing after SES implantation, it was not possible to identify the distribution pattern of late loss in this group of lesions. Thus, lesions were separated into 2 groups according to the usual cutoff definition of binary restenosis. Further analyses with increased numbers of lesions are warranted to validate the present results. A higher rate of angiographic follow-up is desirable to fully evaluate the angiographic outcomes.4 However, the present study enrolled a unique cohort of unselected patients with complex procedures, which differs from patients usually included in randomized trials and may limit the compliance for angiographic restudy. Moreover, the nonrandomized nature of our study with SES precluded the inclusion of an unbiased control group. However, this limitation was partially overcome by the comparison of patients treated with SESs with patients treated with conventional stents enrolled in a recent multicenter trial.
Healthcare funds were received from the Erasmus University Medical Center, Rotterdam, the Netherlands. An institutional grant was given by Cordis, a Johnson & Johnson Co, Miami Lakes, Fla.
Beatt KJ, Luijten HE, de Feyter PJ, van den Brand M, Reiber JH, Serruys PW. Change in diameter of coronary artery segments adjacent to stenosis after percutaneous transluminal coronary angioplasty: failure of percent diameter stenosis measurement to reflect morphologic changes induced by balloon dilation. J Am Coll Cardiol. 1988; 12: 315–323.
Serruys PW, Foley DP, de Feyter PJ. Restenosis after coronary angioplasty: a proposal of new comparative approaches based on quantitative angiography. Br Heart J. 1992; 68: 417–424.
Kuntz RE, Baim DS. Defining coronary restenosis: newer clinical and angiographic paradigms. Circulation. 1993; 88: 1310–1323.
Rensing BJ, Hermans WR, Vos J, Tijssen JG, Rutch W, Danchin N, Heyndrickx GR, Mast EG, Wijns W, Serruys PW. Luminal narrowing after percutaneous transluminal coronary angioplasty: a study of clinical, procedural, and lesional factors related to long-term angiographic outcome: Coronary Artery Restenosis Prevention on Repeated Thromboxane Antagonism (CARPORT) Study Group. Circulation. 1993; 88: 975–985.
Lehmann KG, Melkert R, Serruys PW. Contributions of frequency distribution analysis to the understanding of coronary restenosis: a reappraisal of the gaussian curve. Circulation. 1996; 93: 1123–1132.
Schomig A, Kastrati A, Elezi S, Schuhlen H, Dirschinger J, Dannegger F, Wilhelm M, Ulm K. Bimodal distribution of angiographic measures of restenosis six months after coronary stent placement. Circulation. 1997; 96: 3880–3887.
Sousa JE, Costa MA, Abizaid A, Abizaid AS, Feres F, Pinto IM, Seixas AC, Staico R, Mattos LA, Sousa AG, Falotico R, Jaeger J, Popma JJ, Serruys PW. Lack of neointimal proliferation after implantation of sirolimus-coated stents in human coronary arteries: a quantitative coronary angiography and three-dimensional intravascular ultrasound study. Circulation. 2001; 103: 192–195.
Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M, Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F, Falotico R. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med. 2002; 346: 1773–1780.
Regar E, Serruys PW, Bode C, Holubarsch C, Guermonprez JL, Wijns W, Bartorelli A, Constantini C, Degertekin M, Tanabe K, Disco C, Wuelfert E, Morice MC. Angiographic findings of the multicenter Randomized Study With the Sirolimus-Eluting Bx Velocity Balloon-Expandable Stent (RAVEL): sirolimus-eluting stents inhibit restenosis irrespective of the vessel size. Circulation. 2002; 106: 1949–1956.
Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O’Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med. 2003; 349: 1315–1323.
Lemos PA, Saia F, Ligthart JM, Arampatzis CA, Sianos G, Tanabe K, Hoye A, Degertekin M, Daemen J, McFadden E, Hofma S, Smits PC, de Feyter P, van der Giessen WJ, van Domburg RT, Serruys PW. Coronary restenosis after sirolimus-eluting stent implantation: morphological description and mechanistic analysis from a consecutive series of cases. Circulation. 2003; 108: 257–260.
Lemos PA, Lee C, Degertekin M, Saia F, Tanabe K, Arampatzis CA, Hoye A, van Duuren M, Sianos G, Smits PC, de Feyter P, van der Giessen WJ, van Domburg RT, Serruys PW. Early outcome after sirolimus-eluting stent implantation in patients with acute coronary syndromes: insights from the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) Registry. J Am Coll Cardiol. 2003; 41: 2093–2099.
Lemos PA, Hoye A, Goedhart D, Arampatzis CA, Saia F, van der Giessen WJ, McFadden EP, Sianos G, Smits PC, Hofma SH, de Feyter P, van Domburg RT, Serruys PW. Clinical, angiographic, and procedural predictors of angiographic restenosis after sirolimus-eluting stent implantation in complex patients: an evaluation from the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) Study. Circulation. 2004; 109: 1366–1370.
O’Neill WW, Serruys P, Knudtson M, van Es GA, Timmis GC, van der Zwaan C, Kleiman J, Gong J, Roecker EB, Dreiling R, Alexander J, Anders R. Long-term treatment with a platelet glycoprotein-receptor antagonist after percutaneous coronary revascularization: EXCITE Trial Investigators: Evaluation of Oral Xemilofiban in Controlling Thrombotic Events. N Engl J Med. 2000; 342: 1316–1324.
Reiber JH, Serruys PW, Kooijman CJ, Wijns W, Slager CJ, Gerbrands JJ, Schuurbiers JC, den Boer A, Hugenholtz PG. Assessment of short-, medium-, and long-term variations in arterial dimensions from computer-assisted quantitation of coronary cineangiograms. Circulation. 1985; 71: 280–288.
Serruys PW, Luijten HE, Beatt KJ, Geuskens R, de Feyter PJ, van den Brand M, Reiber JH, ten Katen HJ, van Es GA, Hugenholtz PG. Incidence of restenosis after successful coronary angioplasty: a time-related phenomenon: a quantitative angiographic study in 342 consecutive patients at 1, 2, 3, and 4 months. Circulation. 1988; 77: 361–371.
Reiber JH, van der Zwet PMJ, Koning G, von Land CD, van Meurs B, Gerbrands JJ, Bruis B, van Voorthuisen AE. Accuracy and precision of quantitative digital coronary arteriography; observer-, as well as short- and medium-term variabilities. In: Serruys PW, Foley DP, de Feyter PJ, eds. Quantitative Coronary Angiography in Clinical Practice. Dordrecht, the Netherlands: Kluwer Academic Publishers; 1994.
Colombo A, Orlic D, Stankovic G, Corvaja N, Spanos V, Montorfano M, Liistro F, Carlino M, Airoldi F, Chieffo A, Di Mario C. Preliminary observations regarding angiographic pattern of restenosis after rapamycin-eluting stent implantation. Circulation. 2003; 107: 2178–2180.