Longitudinal Deformation of Drug-Eluting Stents: Evaluation by Multislice Computed Tomography

Abstract: Background. Some modifications introduced in the design of the new generation of drug-eluting stent (DES) to improve their flexibility may entail a reduction in their longitudinal strength. This study sought to evaluate the longitudinal deformation of DESs by multislice computed tomography (MSCT). Methods. This study included DESs that could have been potentially deformed by mechanical actions such as: (1) catheter impingement; (2) postdilation; (3) kissing balloon; and (4) intravascular imaging after implantation. Patients on atrial fibrillation or with overlapping stents were excluded. All patients underwent stent length evaluation by MSCT 9-12 months after implantation. Results. Forty-five stents were included: 15 platinum chromium (PtCr-DES), 15 cobalt chromium (CoCr-DES), and 15 stainless-steel (SS-DES). The relative longitudinal deformation by stent type was 6.93 ± 5.82% for PtCr-DES, 6.19 ± 5.79% for CoCr-DES, and 4.03 ± 4.07% for SS-DES (P=.31). Among the mechanical actions studied, only catheter impingement was related to longitudinal stent deformation (P<.01). After adjustment, only catheter impingement (P<.01) and nominal stent length (P=.049) were independently related to longitudinal deformation. There were no stent fractures. Conclusions. Longitudinal deformation of DESs is common in all the studied platforms when subject to longitudinal forces. Guiding catheter impingement is the only mechanical action significantly associated with DES shortening. 

J INVASIVE CARDIOL 2014;26(4):161-166

Key words: cardiac computed tomography, drug-eluting stents

_______________________

Recent-generation coronary stents have been redesigned in order to improve flexibility, deliverability, and conformability to the vessel wall. These modifications have included the introduction of new metal alloys such as cobalt chromium or platinum chromium, reductions of strut thickness, modifications of hoop design, and changes in the number, shape, and orientation of connectors between hoops. However, some of these modifications, especially the changes in connectors, could lead to a reduction of longitudinal strength.^1-3 Particularly, recent publications have suggested that the Element platform (Promus Element and Taxus Element; Boston Scientific) could be more prone to longitudinal deformation than other platforms.^4-7 

 Multislice computed tomography (MSCT) has been proven to be reliable in assessing length and integrity of coronary stents.^8,9 We designed this study to evaluate the longitudinal deformation of DESs in vivo using MSCT as a primary image modality.

Methods

Patient selection. We identified from our institutional angioplasty registry 45 DESs placed in 42 procedures from June to December 2011 that met inclusion and exclusion criteria: 15 Promus/Taxus Element drug-eluting stents (PtCr-DES), 15 consecutive Xience Prime everolimus-eluting stents (CoCr-DES) (Abbott Vascular), and a control group of 15 stainless-steel first-generation drug-eluting stents (SS-DES) (Taxus Liberté, Boston Scientific; and Cypher Select, Cordis Corporation). All underwent stent length evaluation by MSCT 9-12 months after implantation. The inclusion criterion for the study was to be subject to potential longitudinal deformation within the procedure by mechanical actions such as: (1) guiding catheter impingement; (2) bifurcation technique such as final kissing balloon or crush stenting; (3) intravascular ultrasound or optical coherence tomography evaluation immediately after stent implantation; and (4) postdilation. Exclusion criteria were: (1) impaired renal function; (2) contrast media allergy; (3) causes of potential bias of stent length evaluation such as atrial fibrillation or overlapped stents; and (4) patients that did not consent to undergo MSCT. Figure 1 summarizes the key characteristics of each stent design. The study protocol conformed to the ethical guidelines of the Declaration of Helsinki. All patients gave written informed consent, and the study protocol was approved by the hospital ethics committee.

MSCT protocol and image acquisition. The MSCT angiography was performed with a 64-slice computed tomography scanner (Lightspeed VCT 64-slice; GE Medical Systems). Patients with heart rate over 65 bpm received propranolol (1 mg/mL) intravenously for heart rate control. A bolus dose of 80 mL of contrast (iodixanol, 320 mg iodine/1 mL; Visipaque, GE Healthcare) was infused at 5 mL/s into an antecubital vein, followed by flushing with 20-50 mL saline. MSCT data were acquired by the fluoroscopic bolus-tracking technique, started as soon as the signal density level in the ascending aorta reached a threshold of 100 Hounsfield Units (HU). 

Scan parameters were as follow: detector collimation width, 64 x 0.625 mm; gantry rotation time, 350 ms; pitch, 0.2; tube voltage, 120 kV; and use of dose modulation (peak tube current of 600 mA during 65%-85% of the R-R interval and minimal tube current of 300 mA). For estimation of the effective dose, the product of the dose-length product and the chest organ weighting factor (k = 0.014 mSv × (mGy × cm) – 1) was calculated.¹⁰

Reconstruction and analysis of MSCT images. Image reconstruction was retrospectively gated to the electrocardiogram (ECG). Scans were reformatted to show the long axes of the stents for measurement. One experienced observer calculated the stent length on straightened reconstructions. In order to define the extent of individual variation in this measurement, a subgroup of stents was assessed independently by a second observer and the interobserver variation was evaluated. Both observers were blinded to the stent characteristics and longitudinal forces applied. The agreement between both observers was excellent (consistency interclass correlation coefficients [ICC] 0.989 with 95% confidence interval [CI], 0.956-0.997; absolute ICC 0.990 with 95% CI, 0.9604-0.997). 

Evaluation of longitudinal deformation by plain fluoroscopy. Stent length measurements were performed offline by two experienced analysts blinded to stent characteristics and MSCT data, with dedicated software (CAAS II system; Pie Medical). In order to evaluate the reliability of longitudinal stent deformation observations by plain fluoroscopy, the stent length was measured without contrast after longitudinal forces were applied, and was compared with the stent length before implantation (undeployed stent, during stent positioning) in the same angiographic projection. The agreement between observers and between plain fluoroscopy during percutaneous coronary intervention (PCI) and MSCT was then calculated for each group.

Definitions and study endpoints. Catheter impingement was considered only for stents placed at the left main coronary artery or the proximal third of the right coronary artery, if the guiding catheter advanced up to the proximal edge of the stent following its implantation. The main study endpoint was the ratio stent length by MSCT divided by nominal stent length (Figure 2). Relative deformation was defined as the absolute value of the difference between the nominal stent length and the MSCT length divided by the nominal length. Stent fracture was defined as a partial or circumferential gap on visual estimation with a confirmation of HU <300 at the site of separation,¹¹ except for stent segments subject to bifurcation techniques, in which only circumferential gaps were considered to be a fracture.

Statistical analysis. Statistical analysis was performed using PAWS Statistics version 18.0 (SPSS, Inc). Categorical variables are expressed as frequencies and group percentages. Continuous variables are shown as mean ± standard deviation unless otherwise noted. Baseline characteristics were compared using χ² tests for categorical variables and the analysis of variance or Kruskal-Wallis test for continuous variables. Normality of distributions and homogeneity of variances of continuous variables were examined with Shapiro-Wilk test and Levene’s test, respectively. An analysis of covariance (ANCOVA) method with a general linear model was used to evaluate the main endpoint and all other between-group comparisons, using catheter impingement and nominal stent length as covariates. To test the agreement between observers as well as the agreement between MSCT and conventional-x ray, consistency and absolute ICCs were obtained. A P-value <.05 was considered to be significant for all statistical tests used.

Results

Baseline characteristics. Patient demographics and procedural characteristics are detailed in Table 1 and Table 2, respectively. Most stents were implanted in complex lesions. The SS-DES was placed more frequently in the left main coronary artery than the PtCr-DES and CoCr-DES. Nominal length and diameter were similar between all groups (all P>.20). There were no cardiac events (death, myocardial infarction, or target lesion revascularization) between the index procedure and the MSCT.

The median interval between stent placement and the MSCT study was 259 days (range, 206-386 days). Estimated radiation dose was 14.10 ± 4.19 mSv. Stent length could be assessed by MSCT in all patients but 1 (CoCr-DES group) due to movement of the patient during the image acquisition. 

Longitudinal deformation by MSCT. All stents were subject to at least 1 longitudinal force after implantation. However, SS-DESs were subject to catheter impingement more frequently than the other stents (P=.01). Postdilation tended to be more frequent in the PtCr-DES group (P=.05). The overall degree of deformation was higher with stents subject to catheter impingement (mean difference, -7.4%; 95% CI, -11.6 to -3.2; Figure 3). There was no relationship between other mechanical actions and the ratio of deformation (all P>.20). 

 The relative deformation by stent type was 6.93 ± 5.82% for the PtCr-DES group, 6.19 ± 5.79% for the CoCr-DES group, and 4.03 ± 4.07% for the SS-DES group (P=.31). The ratio of deformation (Figure 4) did not differ between groups (P=.26). After adjustment, only catheter impingement (P<.01) and nominal stent length (P=.049) were independently related to the ratio of deformation. There were no stent fractures in any group.

Longitudinal deformation by plain fluoroscopy. The ratio of deformation of 3 patients in the SS-DES group could not be assessed due to technical problems (absence of the same angiographic projection with the stent undeployed and deployed). Moreover, the final stent length could not be defined by any of the operators in 2 cases of the Taxus Liberte stent (SS-DES group) due to poor image quality. Similarly to MSCT, no differences between groups were seen with regard to the ratio of deformation by plain fluoroscopy (P=.53).

The agreement between both fluoroscopy observers was moderate (consistency ICC 0.580 with 95% CI, 0.068-0.850; absolute ICC 0.599 with 95% CI, 0.074-0.860). The agreement between MSCT and plain fluoroscopy was poor for the CoCr-DES group (consistency ICC 0.350 with 95% CI, -0.200-0.732; absolute ICC 0.361 with 95% CI, -0.205-0.740) and moderate for the SS-DES group (consistency ICC 0.454 with 95% CI, -0.203-0.829; absolute ICC 0.478 with 95% CI, -0.224-0.842) and the PtCr-DES group (consistency ICC 0.726 with 95% CI, 0.358-0.899; absolute ICC 0.739 with 95% CI, 0.374-0.905). 

Discussion

This study is the first to evaluate the longitudinal deformation of DES by MSCT. The main findings of this study are: (1) among the mechanical actions studied, only guiding catheter impingement resulted in significant longitudinal stent shortening; (2) all the stent platforms studied deformed longitudinally, independently of the stent type; and (3) the reliability of longitudinal deformation observations by plain fluoroscopy was poor.

Previous studies have sought to evaluate the longitudinal deformation of coronary stents in vitro.^1,12-14 In these bench tests, coronary stents were placed in mandrels and compressed longitudinally. However, the stent struts were not apposed to a vessel wall when compressed, which may definitely enhance their longitudinal stability. Importantly, all these studies were conducted at room temperature rather than body temperature. Moreover, it is really complicated to define the longitudinal forces applied for different operators in different clinical scenarios. Therefore, it is likely that these bench tests may not precisely correlate with the real forces involved in a percutaneous coronary intervention. 

In addition to these bench tests, there have been other studies that evaluated the longitudinal stent deformation in vivo by conventional x-ray.^13-17 Our study, however, showed that the evaluation of longitudinal stent deformation by plain fluoroscopy is imprecise. In fact, agreement between fluoroscopy observers as well as agreement between plain fluoroscopy and MSCT was moderate at best. Interestingly, the agreement between plain fluoroscopy and MSCT data was slightly better for PtCr-DES, which could be related to a higher radiopacity of the platinum-chromium alloy.¹⁸

Among the postimplantation maneuvers studied here, only guiding catheter impingement was related to significant longitudinal deformation (Figure 3). In agreement with our findings, Leibundgut et al¹³ found the left main coronary artery to be the strongest predictor of longitudinal deformation. However, they also found a significant relationship between bifurcations and longitudinal deformation, which in our study was not present. Of note, they defined longitudinal deformation as a subjective visualization of nesting of struts, which evaluates a different concept than length shortening, and may probably be seen in many bifurcations.

Our data show that all stent scaffolds subject to longitudinal forces in vivo were significantly deformed. Thus, although it seems clear that stents with offset peak-to-peak connectors (PtCr-DES) are more prone to be deformed in vitro than other stent families, its implications in a real-world PCI are not yet well understood. Therefore, special attention should be paid in cases with high risk of stent compression (such as proximal left main lesions) not only with PtCr-DES but also with the other DESs. Importantly, significant deformation of the stent scaffold may lead to unequal drug delivery and also to vessel injury, which has been previously associated with higher risk of DES failure.^17,19-21

Study limitations. Although sufficient stent numbers were available for a meaningful analysis, the sample size is the main limitation of this study. Therefore, it may have been underpowered to detect significant associations for other causes of longitudinal deformation. It does, however, have some strengths, as it is the first to address this topic using MSCT as a primary image modality, which we suggested to be more reliable than conventional x-ray. Furthermore, all the studied stents underwent potential deformation in a real scenario rather than bench test.

Conclusion

Our study suggests that longitudinal deformation of DESs is common in all the studied platforms subject to longitudinal forces. Guiding catheter impingement is the only mechanical action significantly associated to DES shortening. Therefore, it is of importance to avoid catheter impingement after stent implantation in coronary ostia independent of the stent type.

References

1. Ormiston JA, Webber B,Webster MW. Stent longitudinal integrity bench insights into a clinical problem. JACC Cardiovasc Interv. 2011;4(12):1310-1317.
2. Kwok OH. Stent “concertina:” stent design does matter. J Invasive Cardiol. 2013;25(6):E114-E119.
3. De Caterina AR, Cuculi F, Banning AP. Axial deformation during coronary stenting: an extreme case. J Invasive Cardiol. 2012;24(6):E122-E123.
4. Shannon J, Latib A, Takagi K, et al. Procedural trauma risks longitudinal shortening of the promus element stent platform. Catheter Cardiovasc Interv. 2013;81(5):810-817. Epub 2012 Nov 8.
5. Leibundgut G, Loffelhardt N, Toma A, et al. Optical coherence tomography of longitudinal stent compression. EuroIntervention. 2012;8(8):989.
6. Foerst J, Foin N, Hettleman B. Longitudinal stent compression demonstrated by angiographic “wedding band” and 3-dimensional optical coherence tomography. JACC Cardiovasc Interv. 2012;5(12):e39-e40.
7. Rigattieri S, Sciahbasi A, Loschiavo P. The clinical spectrum of longitudinal deformation of coronary stents: from a mere angiographic finding to a severe complication. J Invasive Cardiol. 2013;25(5):E101-E105.
8. Halon DA, Gaspar T, Adawi S, et al. Coronary stent assessment on multidetector computed tomography: source and predictors of image distortion. Int J Cardiol. 2008;128(1):62-68.
9. Pregowski J, Kepka C, Kalinczuk L, et al. Comparison of intravascular ultrasound, quantitative coronary angiography, and dual-source 64-slice computed tomography in the preprocedural assessment of significant saphenous vein graft lesions. Am J Cardiol. 2011;107(10):1453-1459.
10. Mayo JR, Leipsic JA. Radiation dose in cardiac CT. AJR Am J Roentgenol. 2009;192(3):646-653.
11. Hecht HS, Polena S, Jelnin V, et al. Stent gap by 64-detector computed tomographic angiography relationship to in-stent restenosis, fracture, and overlap failure. J Am Coll Cardiol. 2009;54(21):1949-1959.
12. Prabhu S, Schikorr T, Mahmoud T, et al. Engineering assessment of the longitudinal compression behaviour of contemporary coronary stents. EuroIntervention. 2012;8(2):275-281.
13. Leibundgut G, Gick M, Toma A, et al. Longitudinal compression of the platinum-chromium everolimus-eluting stent during coronary implantation: predisposing mechanical properties, incidence, and predictors in a large patient cohort. Catheter Cardiovasc Interv. 2013;81(5):E206-E214.
14. Abdel-Wahab M, Sulimov DS, Kassner G, et al. Longitudinal deformation of contemporary coronary stents: an integrated analysis of clinical experience and observations from the bench. J Interv Cardiol. 2012;25(6):576-585.
15. Mamas MA,Williams PD. Longitudinal stent deformation: insights on mechanisms, treatments and outcomes from the Food and Drug Administration Manufacturer and User Facility Device Experience database. EuroIntervention. 2012;8(2):196-204.
16. Kereiakes DJ, Popma JJ, Cannon LA, et al. Longitudinal stent deformation: quantitative coronary angiographic analysis from the PERSEUS and PLATINUM randomised controlled clinical trials. EuroIntervention. 2012;8(2):187-195.
17. Williams PD, Mamas MA, Morgan KP, et al. Longitudinal stent deformation: a retrospective analysis of frequency and mechanisms. EuroIntervention. 2012;8(2):267-274.
18. O’Brien BJ, Stinson JS, Larsen SR, et al. A platinum-chromium steel for cardiovascular stents. Biomaterials. 2010;31(14):3755-3761.
19. Freixa X, Almasood AS, Khan SQ, et al. Decreased risk of stent fracture-related restenosis between paclitaxel-eluting stents and sirolimus eluting stents: results of long-term follow-up. Catheter Cardiovasc Interv. 2012;79(4):559-565.
20. Nakano M, Otsuka F, Yahagi K, et al. Human autopsy study of drug-eluting stents restenosis: histomorphological predictors and neointimal characteristics. Eur Heart J.  2013;34(42):3304-3313.  Epub 2013 Jul 3.
21. Kuramitsu S, Iwabuchi M, Haraguchi T, et al. Incidence and clinical impact of stent fracture after everolimus-eluting stent implantation. Circ Cardiovasc Interv. 2012;5(5):663-671.