Journal of Invasive Cardiology

ABSTRACT: Background. Stents are used to treat the first and even successive episodes of in-stent restenosis (ISR). Methods. In 18 patients (19 lesions), intravascular ultrasound (IVUS) was performed after placement of a stent for a nonrestenotic lesion after the second stent was used to treat the first episode of ISR and after the third stent was used to treat the second episode of ISR. The duration between the first and second stent was 355 ± 374 days, and between the second and third stent was 330 ± 279 days. The duration of follow up after the third stent was 307 ± 145 days. High-pressure inflation (> 14 atm) was performed for 69% (11/16) of patients when treating the first episode of ISR, and all patients when treating the re-ISR (p = 0.018). Results. Nevertheless, vessel area and final minimal stent area (MSA) did not increase with successive restenting, and the ratio of minimum stent diameter to nominal stent size suggested that chronic stent underexpansion persisted. MSA > 5 mm² was noted in 54% after the first stent, 35% after the second stent, and 42% after the third stent (p = 0.6). After the third stent, the rate of target lesion revascularization was 26% (5/19) and target vessel revascularization was 37% (7/19); there was 1 nonfatal myocardial infarction. Thus, the overall rate of major adverse cardiac events was 42%. Conclusion. While serial restenting a recurrent ISR lesion was not associated with progressive decrease in MSA, it was still associated with chronic stent underexpansion and a high rate of adverse events. Attention should be directed to achieving better stent expansion to minimize subsequent recurrences.

J INVASIVE CARDIOL 2007;19:506–509
Key Words: in-stent restenosis

In-stent restenosis (ISR) has been treated with balloon angioplasty, directional coronary atherectomy, laser angioplasty, rotational atherectomy, repeat bare-metal stenting (BMS), brachytherapy, and most recently, drug-eluting stent (DES) implantation. Reports have shown DES to be comparable with or superior to brachytherapy,^1–3 and DES implantation has become the preferred treatment in many institutions. Stent underexpansion is an important cause of both firsttime and recurrent ISR.^4–7 Because the treatment of recurrent ISR frequently involves repeat stent implantation, the purpose of this study was to: (1) evaluate frequency of stent underexpansion in recurrent ISR; (2) assess the acute results after implanting a third DES to treat a second episode of ISR; and (3) evaluate the outcome of DES implantation to treat the second ISR episode.

Methods
Study population. We identified 18 patients (19 lesions) who had high-quality intravascular ultrasound (IVUS) imaging: (1) at the time of placement of a stent for a nonrestenotic lesion; (2) when the first episode of ISR was treated with restenting; and (3) when the second episode of ISR was treated with implantation of a third stent. De novo stent types were 17 BMS, 1 Cypher™ stent (Cordis Corp., Miami Lakes, Florida), and 1 unknown stent type. The second stent types (those used to treat the first episode of ISR) were: 10 Cypher stents, 3 Taxus^® stents (Boston Scientific Corp., Natick, Massachusetts), 4 BMS, and 2 unknown stent types. The third stent types (those used to treat the second episode of ISR) were: 13 Cypher stents, 5 Taxus stents, and 1 BMS. Two patients had previous bypass surgery, and brachytherapy was used in 4 cases to treat the first ISR and in 2 cases to treat the second ISR episode. All patients who presented with a second episode of ISR and who underwent stent implantation had chest pain, and 2 patients had a positive stress test. DES selection was at the discretion of the individual operator who also selected DES diameter (based on preinterventional IVUS measurements of stent diameter to reference lumen diameter = 1:1) and post-implantation inflation strategies. Follow up was conducted by hospital chart review or telephone interview in all patients. The clinical endpoints were all-cause mortality, myocardial infarction and revascularization of the target lesion after the third stent.

Angiographic analysis. Coronary angiography was performed after intracoronary administration of 0.2 mg nitroglycerin. All angiograms were analyzed with an automated edge-detection algorithm (CAAS II, Pie Medical, Neptune, New Jersey), independently using standard protocols. Minimum luminal diameter, percent (%) diameter stenosis and reference lumen diameter were measured before and after intervention.
ISR was defined as > 50% angiographic diameter stenosis at follow up and classified as previously reported.8 ISR lesion length was measured from shoulder-to-shoulder; in the setting of multifocal ISR, the length of the individual ISR foci were added. IVUS analysis. All IVUS examinations were performed using a commercially available IVUS system (Boston Scientific) and either 30 Mhz or 40 Mhz transducers; studies were performed after intracoronary administration of 0.2 mg nitroglycerin. The IVUS catheter was advanced 10 mm distal to the target lesion, and imaging was performed retrogradely to the aorto-ostial junction at an automatic pullback speed of 0.5 mm/second. Studies were recorded on 0.5 inch high-resolution s-VHS tapes for offline analysis. Standard IVUS measurements were performed every 1 mm to include both the stent and the proximal and distal stent edges according to criteria of the American College of Cardiology clinical expert consensus document on IVUS9 using planimetry software (TapeMeasure, INDEC Systems, Santa Clara, California). We measured the length of the newly placed stent that overlapped any old stents to assess the multilayered stent strut. Intimal hyperplasia (IH) Cross-sectional area (CSA) was calculated as stent minus lumen CSA. Significant IH was defined as IH CSA ≥ 50% of stent CSA. Focal IH was defined as 10 mm in length or < 25% of total stent length.

Statistical analysis. Descriptive statistics of the demographic data were calculated for all of the 18 patients, and descriptive statistics of the angiographic lesion characteristics, serial procedural details and serial IVUS measurements were calculated for the stent groups. Comparison of continuous variables between the first ISR and second ISR was conducted by using the paired t-test or nonparametric signed rank test, while normality assumption was violated; the categorical variable was compared by using the symmetry test.
Repeated measures analysis was performed to examine the change of serial procedural characteristics and the change of serial IVUS measurements over 3 stent time points. CATMOD
(categorical data modeling) and GLM (general linear modeling) procedures were used as appropriate.

Results
Clinical and angiographic findings. Demographics and clinical characteristics are summarized in Table 1. The duration between the first and second stent (between placement
of a stent for a nonrestenotic lesion and first episode of ISR) was 355 ± 374 days, and was 330 ± 279 days between the second and third stent (between the first and second episodes of ISR).
Angiographic lesion characteristics are shown in Table 2. The reference vessel diameter was 2.54 ± 0.59 mm. The angiographic classification of ISR was mostly focal: 61% (11/18) of first ISR episodes versus 89% (17/19) of second ISR episodes (p = 0.4). Similarly, ISR lesion length was shorter after the second ISR episode than after the first ISR episode (11.61 ± 11.69 mm in first ISR versus 6.42 ± 4.32 mm in second ISR; p = 0 . 0 3 9 ).
Procedural characteristics are shown in Table 3. The number of stents (p = 0.5) and the implanted stent length (p = 0.3) were similar from the first to the second to the third procedure. Maximal balloon diameter was similar (p = 0.9), and inflation pressures > 14 atm were used for 69% (11/16) of patients when treating the first episode of ISR and all patients when treating the second episode of ISR (p = 0.018). The use of the Cutting Balloon^® (Interventional Technologies, Letterkerry, Ireland) was higher to treat first ISR (p = 0.014) and more frequent use of additional noncompliant balloons to treat second ISR (p = 0.043).
The angiographic diameter stenosis measured 64 ± 16% at the time of the first episode of ISR, 12 ± 5% after implanting the second stent, 66 ± 15% at the time of the second episode
of ISR and 12 ± 6% after implanting the third stent. IVUS intrastent comparisons. IVUS findings are shown in Table 4. External elastic membrane area, persistent plaque area, and mean stent area were similar after each successive restenting . IH was completely excluded during the treatment of both the first and second episode of ISR. The minimum stent CSA was smaller after the second stent implantation (to treat the first episode of ISR) than after the original stent was implanted (p = 0.004) when only 69% of lesions were postdilated with high-pressure inflations; however, it increased again after the third stent was implanted (p = 0.067) when all lesions were postdilated with high-pressure inflations (Figure 1). Nevertheless, a minimal stent area (MSA) < 5 mm2 was seen in 46% after the first stent, 65% after the second stent, and 58% after the third stent (p = 0.6). Although the maximum i n f l ation pressure increased during successive interventions (p = 0.018), the ratio of IVUSmeasur ed minimum stent diamet er to nominal stent size did not increase, even after the third stent implantation (p = 0.9).

Clinical outcomes. The follow-up duration of the study population was 307 ± 145 days (median 328 days, range 119–465 days) after the third stent was implanted to treat the second ISR episode. No deaths were recorded, 1 patient had a nonfatal myocardial infarction, and no patient had acute or subacute stent thrombosis. However, the rate of target lesion revascularization was 5/19 (26%), and the rate of target vessel revascularization was 7/19 (37%). Repeated treatment for a third episode of ISR was required in 2 patients at 6 months, 1 patient at 8 months, 1 patient at 13 months, and 1 patient at 16 months. All 5 of these patients had Cypher stents implanted to treat the second episode of ISR. MSA < 5 mm² was found in 40% (2/5) of lesions in patients who required repeat intervention versus 64% (9/14) in patients who did not require repeat intervention (p = 0.6). Overall, the cumulative rate of major adverse clinical events was 42%.