Author + information
- Dong Yin, MD,
- Akiko Maehara, MD∗ (, )
- Gary S. Mintz, MD,
- Lei Song, MD,
- Matthew T. Finn, MD,
- Raja Hatem, MD,
- Kisaki Amemiya, MD,
- Jeffrey W. Moses, MD,
- Manish A. Parikh, MD,
- Ajay J. Kirtane, MD, SM,
- Michael B. Collins, MD,
- Tamim M. Nazif, MD,
- Khady N. Fall, MD, MPH,
- Ming Liao, MA,
- Philip Green, MD,
- Ziad A. Ali, MD, DPhil,
- Candido Batres, MD,
- Gregg W. Stone, MD,
- Martin B. Leon, MD,
- Masahiko Ochiai, MD, PhD and
- Dimitri Karmpaliotis, MD, PhD
- ↵∗Cardiovascular Research Foundation, 1700 Broadway, 9th Floor, New York, New York 10019
This was a retrospective intravascular ultrasound (IVUS) study of mechanisms of in-stent restenosis (ISR) chronic total occlusions (CTOs). ISR CTOs were defined as TIMI (Thrombolysis In Myocardial Infarction) flow grade 0 within a stented lesion (including 5-mm margins) with estimated occlusion duration >3 months by clinical symptoms (1). From February 2003 to May 2016, there were 77 ISR CTOs in 77 patients treated at NewYork-Presbyterian Hospital (n = 37) and Showa University Northern Yokohama Hospital (n = 40). The ethics committee at each site approved the protocol on the basis of prior written informed consent.
IVUS was performed after guidewire crossing and pre-dilation (if necessary) with a 1.5- to 2.0-mm balloon. In-stent calcium (superficial hyperintensity with acoustic shadow) or attenuated neointimal tissue (attenuation without calcium) was considered to represent neoatherosclerosis.
Primary morphological patterns were: 1) in-stent neointimal hyperplasia (NIH) (excessive NIH without stent underexpansion or neoatherosclerosis); 2) stent underexpansion (minimum stent area [MSA] <5 mm2); 3) proximal new lesion progression (occlusion within 5 mm proximal to the stent edge without excessive in-stent NIH); or 4) in-stent neoatherosclerosis.
The mean patient age was 63.5 years, 91% were men, and 41% presented with unstable angina (the rest with stable angina [50%] or silent ischemia [9%]). In 64%, occluded length was ≥20 mm, with 63% having Rentrop’s collateral grade ≥2. Lesions with grade ≥2 collateral vessels presented later (7.3 ± 4.3 years vs 5.4 ± 5.0 years, respectively, p = 0.08). The ISR CTO was crossed by wire escalation in 74% (n = 57) and dissection or re-entry in 26% (n = 20).
Morphological patterns of ISR CTO were: 1) NIH in 30% (n = 23); 2) stent underexpansion in 22% (n = 17); 3) proximal new lesion in 25% (n = 19); and 4) neoatherosclerosis in 23% (n = 18). In 23 cases in which the ISR CTO was due to NIH, MSA was large (7.0 ± 1.9 mm2) compared with 17 cases with stent underexpansion (MSA 3.9 ± 0.8 mm2) in a normally sized vessel at the site of the original lesion (peristent plaque burden 64.5 ± 11.0%). In 19 cases in which the ISR CTO was due to a proximal new lesion, 2 also had neoatherosclerosis within the nonoccluded stent segment. Among 18 cases in which the ISR CTO appeared to be due to neoatherosclerosis, there was 1 case with stent underexpansion, but because of the very late timing of the occlusion (13.6 years post-stenting), neoatherosclerosis was considered the primary cause. In addition, there was 1 case in which a calcified nodule (subcategory of neoatherosclerosis) caused ISR CTO. Four cases with stent fracture (3 overlap type and 1 complete fracture) were not considered the primary cause of ISR CTO because NIH (n = 2) or neoatherosclerosis (n = 2) appeared diffusely within the old stent and not just at the fracture site.
The median duration from stent implantation to ISR CTO presentation was 6.6 years and was longer in ISR CTOs with biologic versus mechanical causes: 1) NIH, 5.4 years; 2) stent underexpansion, 2.3 years; 3) proximal new lesion, 8.5 years; and 4) neoatherosclerosis, 9.8 years. Patients were divided into tertiles on the basis of the time from implantation (earliest tertile, <3.0 years; second tertile, 3.0 to 8.5 years; latest tertile, >8.5 years) (Figure 1). The prevalence of neoatherosclerosis was significantly higher in tertile 3 than tertile 1 (46.2% vs. 4.0%, p < 0.001), whereas the prevalence of stent underexpansion was significant greater in tertile 1 than tertile 3 (48.0% vs. 3.8%, p < 0.001).
ISR CTOs were drug eluting in 41 (53.2%) versus bare metal in 36 (5.2 ± 3.2 years vs. 8.4 ± 5.4 years, p = 0.003). Bare-metal stent ISR CTOs more often had proximal new lesion progression (p = 0.03).
Seventy-five ISR CTOs (97.4%) were treated with restenting, and the overall final (post–new stent) MSA (measured in 65 cases) was similar among the ISR CTO patterns or techniques used to treat the ISR CTO. In 44%, the restenting MSA was located within the CTO segment, and in the rest, the restenting MSA was located at the distal stent edge. Therefore, in part because of vessel tapering, final restenting MSA was smaller compared with the MSA of the ISR CTO lesion (4.7 mm2 [4.0 to 5.5 mm2] vs. 6.1 mm2 [4.5 to 7.9 mm2]; p < 0.001).
There were 10 cases (13.0%) in which a guidewire passed either partially or completely behind the old stent, and the newly implanted stent “crushed” the old stent. (Antegrade wire escalation was less common as a final successful wiring technique if a guidewire passed either partially or completely behind the old stent: 30% vs. 68.7%; p = 0.03.) Final (post-restenting) MSA between “stent-in-stent” and “stent-crush” interventions was similar: 4.7 mm2 (4.0 to 5.5 mm2) versus 4.5 mm2 (4.2 to 6.0 mm2) (p = 0.91).
The present study demonstrated that ISR CTOs had 4 distinct morphological patterns: NIH, stent underexpansion, proximal new lesion progression, and neoatherosclerosis. Some selection bias was expected because this was a retrospective observational study, and the ISR CTOs represented a complex population. Findings from IVUS-based tissue characterization, especially thrombus and neoatherosclerosis, should be interpreted with caution; neoatherosclerosis is better assessed with optical coherence tomography than IVUS.
Please note: Dr. Yin has received a research grant from Boston Scientific. Dr. Maehara has received institutional grant support from Boston Scientific and St. Jude Medical; is a consultant for Boston Scientific and OCT Medical Imaging; and has received speaking honoraria from St. Jude Medical. Dr. Mintz is a consultant for Boston Scientific and ACIST; has received fellowship and grant support from Volcano, Boston Scientific, and InfraReDx; and has received honoraria from Boston Scientific and ACIST. Dr. Parikh is a member of the Speakers Bureaus of Medtronic, Boston Scientific, Abbott Vascular, Cardiovascular Systems, Inc., St. Jude Medical, and TriReme Medical; and is a member of the advisory boards of Philips, Abbott Vascular, and Medtronic. Dr. Kirtane has received institutional grants to Columbia University and/or the Cardiovascular Research Foundation from Medtronic, Boston Scientific, Abbott Vascular, Abiomed, Cardiovascular Systems, Inc., CathWorks, Siemens, Philips, ReCor Medical, and Spectranetics. Dr. Green is supported by a career development grant award (K23 HL12114) from the National Heart, Lung, and Blood Institute. Dr. Ali has received institutional research grants to Columbia University from St. Jude Medical and Cardiovascular Systems; and is a consultant to St. Jude Medical and ACIST. Dr. Ochiai is a member of the Speakers Bureau of Boston Scientific. Dr. Karmpaliotis has received speaking honoraria from Boston Scientific, Abbott Vascular, and Medtronic; and is a consultant for Vascular Solutions. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation