Is Stent Underexpansion the Main Cause of In-Stent Restenosis after Sirolimus-Eluting Stent Implantation?: An Intravascular Ultrasound Study

Several studies have demonstrated that Sirolimus-eluting stents(SES, Cypher, Cordis/Johnson and Johnson, Miami, FL) strongly suppress neointimal hyperplasia and effectively reduce in-stent restenosis(ISR).1-4) However, ISR after SES implantation does occur in some cases and restenosis is still a major problem in the era of drug-eluting stent(DES).5), 6) Although attempts have been made to find the causes of ISR after SES implantation and some risk factors have been suggested as the predictors of ISR,4), 6-8) the predictors of ISR in de novo lesion after SES implantation, as assessed by intravascular ultrasound(IVUS), have not been widely established. The aims of present study were to summarize the clinical, angiographic and IVUS findings of patients with ISR after SES implantation and to elucidate the possible mechanism of ISR after SES implantation by using IVUS.

The mean duration of angiographic follow-up was 221±58 days. The overall clinical characteristics of the 27 enrolled patients with restenosis are shown in Table 1. The mean lesion length was 36.7±19.0 mm. In 12 lesions(43%), implantation of multiple overlapping stents was required for achieving full coverage of the diffuse long lesion on the index procedure. The other angiographic and procedural findings are shown in Table 2. The follow-up angiographic findings are shown in Table 3. The ISR pattern was focal in 26 lesions and diffuse ISR was noted in 2 lesions.

Table 1
Baseline clinical characteristics of 27 patients with in-stent restenosis after sirolimus-eluting stents implantation

• Click for larger image

Table 2
Baseline angiographic and procedural characteristics of 28 lesions with in-stent restenosis after sirolimus-eluting stents implantation at the index procedure

• Click for larger image

Table 3
Angiographic analysis of 28 lesions with in-stent restenosis after sirolimus-eluting stents implantation on the follow-up angiogram

• Click for larger image

The follow-up IVUS data was obtained for 21 of 28 ISR lesions. Follow-up IVUS study was not performed for the other 7 lesions at the follow-up angiogram because of vasospasm that was due to radial artery access (n=3), inability to pass the guide-wire in the totally occluded restenotic lesion(n=1), and refusal of the test (n=3). Table 4 shows the IVUS measurements of the stented segments with ISR. For the 21 IVUS-applicable lesions, the stent CSA at the ISR site was <5.0 mm² in 11(52%) lesions. Of the subset of small coronary arteries(RVD <2.8 mm, a total of 8 lesions), the stent CSA at the ISR site was <4.5 mm² in 7(88%) lesions. Overall stent underexpansion was observed in 10(48%) of the 21 IVUS applicable ISR lesions. Edge stenosis that was located in the proximal edge was noted in 3 lesions(14%) on the follow-up IVUS. Reviewing the index procedural angiogram, one lesion was related to residual dissection after the procedure, and the others were related to balloon trauma outside the stent. Stent fracture was observed in 3 ISR lesions (14%). De nole lesions of 2 ISR lesions were severely angulated lesions, and one ISR lesion was located in the gap between two overlapped stents and this lesion was not fully covered by the stents. The examples of stent underexpansion and stent fracture in the patients with ISR are shown in Fig. 1, 2., respectively.

Fig. 1
In-stent restenosis caused by stent underexpansion after sirolimus-eluting stent implantation (SES). The pre-stenting (A) and post-stenting angiogram (B) are shown [overlapped with a SES 3.5×33 mm (dashed line) and a SES 2.75×33 mm (dotted line)]. Follow-up angiography (C) demonstrates 90% stenosis in mid-portion of the distal stent of the mid-left anterior descending artery (LAD). IVUS images are from 10 mm proximally (D), the minimal lumen site (E), and 10 mm distally (F) at follow-up. IVUS demonstrated stent underexpansion (minimal stent CSA=4.0 mm²) at the minimum lumen site.

• Click for larger image

Fig. 2
In-stent restenosis caused by stent fracture after implantation of sirolimus-eluting stent. Pre-stenting (A: left anterior oblique view. B: right anterior oblique view) and the poststenting angiogram (C) after implanting a 2.75×33 mm sirolimus-eluting stent are shown. Follow-up angiography (D) demonstrates 85% stenosis in the proximal portion of the right coronary artery: an arrow indicates the image slice of intravascular ultrasound (IVUS). Stent separation (arrowhead) of mid-portion of the sirolimus-eluting stent was noted on fluoroscopy (lower box in D). The follow-up IVUS images are 5 mm (E) and 2mm (F) proximal to the minimal lumen site (G), and 2 mm (H) and 5 mm (I) distal to the minimal lumen site. An IVUS investigation at the spot of the in-stent restenosis represents non-visualization of struts (G).

• Click for larger image

Table 4
Quantitative IVUS measurements of the IVUS-applicable 21 in-stent restenotic lesions

• Click for larger image

Categorical causes of restenosis after SES implantation by IVUS analysis

To examine the possible mechanical causes, we summarized the 21 IVUS-applicable patients with ISR after SES implantation in Table 5. Stent underexpansion was the most common IVUS findings, and this was observed in about half of the ISR cases [10(48%) out of 21 ISR lesions]. Except for edge stenosis, stent underexpansion was observed in 55% of the intra-stent restenotic lesions. Stent underexpansion was the more prominent finding in the small coronary arteries, and this was observed in 7 out of 8 small coronary arteries. There were 7 lesions(33%) in which profound intimal hyperplasia within the stent occurred that was not related to stent underexpansion, stent fracture or edge stenosis. Five of these 7 lesions were diffuse long lesions that required implanting 2 stents.

Table 5
Summary of the 21 IVUS-applicable patients with in-stent restenosis after sirolimus-eluting stent implantation

• Click for larger image

The main findings of our observational study were: (1) the common IVUS findings of the patients with ISR after SES implantation were stent underexpansion, edge stenosis and stent fracture; (2) stent underexpansion was the most prominent finding, especially in the intra-stent ISR and small coronary arteries; and (3) profound intimal hyperplasia within the stent was observed in one third of the lesions.

In the SIRIUS2) and TAXUS-II trials,12) the restenotic lesions were frequently located near the stent margins or at the site of a gap between stents. Thereafter, the new concept, "the longer, the better" became a widespread strategy for coronary intervention in the DES era. However, recent studies have suggested that the mechanical factors, especially minimal stent CSA, still remain the important predictors of ISR.11), 13), 14) Sonoda et al.11) reported that the optimal minimum stent CSA threshold for SES to predict ISR was 5.0 mm. In addition, Takebayashi et al.13) have reported that stent underexpansion was observed in 67% of ISR. Hong et al.14) have recently reported that the post-procedural minimum stent CSA(<5.5 mm²), as determined by IVUS, was the independent predictor of ISR after SES implantation.14) In our study, the most common IVUS finding in the restenotic lesions was stent underexpansion. Since stent underexpansion usually occurred in the intra-stent and small coronary arteries, we can speculate that stent underexpansion may be the main cause of ISR, especially in the small coronary arteries and intra-stent lesion.

SES fractures have also been recently reported in cases of focal ISR.15-19) The clinical significance of these findings are currently unknown. In our ISR cases, stent fracture associated with ISR was noted in a minority of ISR lesions(3 out of 21 lesions). The mechanism and clinical significance of SES fracture should be investigated in the future.

The mechanism of edge stenosis is different from those of stent underexpansion and fracture.20) The results of the SIRIUS study and observations by other authors have suggested that ISR, especially edge stenosis, were associated with insufficient lesion coverage and edge damage.5), 6) In our study, there were three cases (14%) of edge stenosis on the follow-up angiograms, and they were associated with what was previously mentioned. With employing the stent placement strategy to cover the treated segment "from healthy tissue to healthy tissue", the incidence of this phenomenon has decreased, from 5.7% in the SIRIUS study2) to 3.4% in the E-SIRIUS study,3) and 1.6% in the RESEARCH study.4)

Stent underexpansion, stent fracture or edge restenosis were not related to 7 ISR lesions(33%) in this study. The clinical characteristics of these patients were not very much different from those of the other patients with restenosis; there were 3 cases of diabetes, 3 cases of hyperlipidemia and 3 males among these patients. In addition, there were no small coronary arteries in this subset. Notably, implantation of multiple overlapping stents was required in 5 of 7 patients. Because this subset had no definite risk factors in this analysis, except for long lesion, further evaluation of this subset will be required to understand the ISR of SES.

This study has several limitations. This was a small observational study and pre-intervention IVUS was not performed in all patients. Follow-up IVUS was not performed in all the patients with ISR after SES implantation. Because there was no control group to compare with the ISR group, it was difficult to definitely explain the possible causes of ISR after SES implantation.

IVUS is a useful tool for evaluating the lesions of intra-stent(e. g., a precise description of stent morphology), stent expansion and the relationship of the stent and the vessel wall. The mechanical factors on IVUS, such as stent expansion, stent fracture or non uniform strut distribution, have been in the spotlight more often during the DES era than the systemic factors and they will be investigated in future studies. However, all the ISR lesions were not explained by the mechanical factors. To reduce the incidence of ISR after SES implantation and to understand it, further studies about mechanical factors together with the other systemic factors will be required.