Stenting has become increasingly common in the endovascular treatment of IA disease, as stents improve the early outcomes after suboptimal or failed PTA [21
]. Prior studies of selective IA stenting for de novo
lesions have reported a wide range of long-term primary patency rates ranging from 59–86% at 3 years to 49–75% at 5 years, indicating that IA- ISR is frequently encountered during long-term follow up [7
]. However, the optimal treatment for IA-ISR remains uncertain. In this study, we describe excellent mid-term results after endovascular treatment of IA-ISR.
Several patient and procedural variables have previously been associated with lower rates of restenosis after IA stenting. These include absence of diabetes, smoking, or renal insufficiency, short and less complex lesions, larger vessel diameter, lesion location (common IA vs. external IA), and intervention driven by claudication versus CLI as well as possibly younger age [29
]. However, Davies et al. [33
] recently reported that recurrence of iliac disease after initial intervention was associated with younger age and female gender, likely due to smaller vessel diameter. Our current cohort had many favorable characteristics including a relatively low prevalence of diabetes and renal insufficiency, young age, short lesion length, and lesions that were predominantly TASC A/B and ISR type I/II. Contrary to prior studies, IA-ISR in this series occurred in subjects who otherwise would be considered at “low risk” for IA-ISR. However, 63% of the subjects were women, supporting the possible role of gender in IA restenosis.
Earlier studies such as Kropman et al. used primarily angioplasty alone to treat in-stent restenosis (only 14% of lesions had stent replacement) while Davies et al. and this study utilized adjunctive stenting in 70% and 66% of the lesions, respectively. Of note, all three of these studies reported primary patency rates >80% at 1 year and primary-assisted patency rates >90% at 1 year, suggesting overall excellent results of endovascular intervention for IA-ISR [33
In comparison with infrainguinal endovascular interventions, the treatment for IA-ISR is conventionally limited to balloon/cutting balloon angioplasty or stenting, as other modalities such as laser, rotational, and directional atherectomy techniques are not routinely employed. The current data are too limited to identify the superiority of a particular modality for treatment of IA-ISR. In a retrospective analysis of 14 IA-ISR lesions in 12 patients, Tsetis et al. [35
] demonstrated 100% primary patency at a mean follow-up of approximately 2 years with cutting balloon angioplasty. In our series, the utilization of cutting balloon angioplasty was lower without significant differences among the various ISR classifications.
Schurmann et al. [9
] previously reported long-term outcomes of 110 patients who underwent stent placement for occlusive IA disease. During a mean follow-up of 5.8 years, 23 patients underwent repeat endovascular intervention for restenosis, and 9 of those 23 patients required repeat stent placement. Subsequently, 12 of those 23 patients developed recurrent restenosis prompting another endovascular intervention or surgery. Procedural and lesion characteristics were not reported, nor were the outcomes comparing PTA and restenting. All of the IA-ISR lesions in our series were initially treated with standard PTA, and selective restenting after unsatisfactory PTA was performed in 66% of the lesions. Although type 2 ISR lesions were more likely to require stenting (13 of 16 lesions), the overall benefit of stenting could not be discerned among various patterns of ISR due to small numbers in each group. However, given the high rate of stenting (66%) required after initial suboptimal angioplasty, our experience suggests that repeat stenting is often required for IA-ISR rather than PTA alone.
There is increasing evidence regarding utilization of covered stents in peripheral interventions [36
]. Covered stents may lower the incidence of ISR by excluding plaque and endothelium as well as preventing the migration and proliferation of vascular smooth muscle and inflammatory cells through the stent struts [37
]. In our current series, 27 (65.7%) of the IA-ISR lesions required restenting, and covered stents were placed in 19 of those lesions. The COBEST trial found that for complex, de novo
IA lesions (TASC C and D), implantation of a covered stent was associated with significantly less binary restenosis at 18 months compared to uncovered stents, and there was a trend towards less stent occlusion [14
]. However, the superiority of covered stenting compared with uncovered stent placement has not yet been proven for IA-ISR treatment. In a recent analysis of femoral–popliteal ISR treatment with excimer laser atherectomy and PTA followed by implantation of VIABAHN Endoprosthesis, a modest primary patency rate of 48% was reported; however, the TLR was low at 17.4% [38
]. Our analysis did not demonstrate a difference in patency between placement of a covered stent versus an uncovered stent or any stent versus PTA, but this subgroup analysis was limited by small overall size of the cohort.
In our current series, the endovascular treatment of IA-ISR was associated with a high procedural success of 100% and a low rate of procedural complications (4.2%), which compares favorably with results in the primary endovascular repair of IA disease [2
]. Despite the fact that more than 29% of the patients presented with acute or critical limb ischemia, there was no limb loss or need for surgical revision at 1 year. A minority of patients underwent revascularization primarily due to the findings on DUS, with a goal of maintaining patency of the IA. Progression to total occlusion may make subsequent intervention more complex with a higher likelihood of failure or complications. The high primary and primary-assisted patency rates seen in this series underscore endovascular repair as a viable option for IA-ISR.
The results of this study should be interpreted in the context of its design. The series is a single-center, retrospective study and thus has inherent limitations. There was a potential selection bias in patient selection for endovascular intervention, but in general, our clinical practice is to reintervene only in the presence of symptoms and documented restenosis by noninvasive techniques. As the number of procedures was small, this study lacks sufficient statistical power to make conclusions on the comparative effectiveness of a particular endovascular modality and therefore the results may not be fully generalizable. Long-term follow-up with a larger sample size will be helpful to determine the relative contribution of different ISR patterns to long-term patency.