Abstract
Objective
Methods for intravascular assessment of the side branch (SB) orifice after stenting are not readily available. The aim of this study is to assess the utility of an en face projection processing for optical coherence tomography (OCT) images for the SB evaluation.
Methods
Measurements of the SB orifice using en face OCT images were validated using a phantom model. Linear regression modeling was applied to estimated area measurements made on the en face images. The SB orifice was then analyzed in 88 patients with bifurcation lesions treated with either a Xience V (EES) or a Resolute Integrity (ZES). The SB orifice area (A) and the area obstructed by struts (B) were calculated, and the %open area was evaluated as (A-B)/A*100.
Results
Linear regression modeling demonstrated observed departures of the intercept and slope was not significantly different from 0 (−0.12 ± 0.22, p=0.59) and from 1 (1.01 ± 0.06, R2=0.88, p=0.87), respectively. In cases without SB dilatation, the %open area was significantly larger in EES group (n=25) than in ZES group (n=32) (89.2% [83.7–91.3] vs. 84.3% [78.9–87.8], P=0.04). A significant difference in the %open area between with and without SB dilatation was demonstrated in the ZES group (91.4% [86.1–94.0] vs. 84.3% [78.9–87.8], P=0.04).
Conclusions
The accuracy of SB orifice measurement on the en face OCT image was validated using a phantom model. This novel approach can quantitatively evaluate the differences in SB orifice area free from struts among different stent types and different treatment strategies in vivo.
Keywords: optical coherence tomography, drug-eluting stent, bifurcation lesion
Introduction
Stenting for coronary bifurcation lesions is associated with a higher incidence of thrombotic events1. Potential mechanisms underlying these thrombotic complications include incomplete stent apposition (ISA)2, 3 and nonapposed struts crossing the side branch (SB) orifice4. A recent study using optical coherence tomography (OCT) demonstrated that, at mid-term follow-up, the neointimal coverage of nonapposed side branch (NASB) struts crossing the orifice was delayed compared with well-apposed struts5. Thus, the evaluation and management of NASB struts may help to prevent thrombotic complications.
Using phantom models and micro-computed tomography (CT)6, 7, several ex vivo studies have evaluated stent struts around SB orifices. These analyses enabled the quantitative evaluation of SB orifices following stent implantation, providing important preliminary information for selecting the optimal stent type and intervention strategy for bifurcation lesions. On the other hand, no method has previously enabled the quantitative evaluation of stent struts around SB orifices in clinical cases. Recently, several studies introduced the utility of three-dimensional (3D) reconstruction of OCT images for the assessment of SB orifice8–10. This 3D methodology might contribute to the optimization of percutaneous coronary intervention (PCI) in bifurcation lesions11 by, for example, showing the best strut cell for rewiring12. However, its utility within the realm of qualitative assessment has been limited so far.
In the present study, we introduce an en face projection OCT processing for the quantitative assessment of SB orifices, and examined the accuracy using a phantom model. We then apply this algorithm to clinical cases of bifurcation stenting to assess the impact of stent design and treatment strategies on the SB orifice area free from NASB struts.
Methods
Quantification of side branch orifice in an en face projection view
To quantitatively evaluate SB orifices, we developed an algorithm using standard cross-sectional OCT data to create en face projection images (Fig. 1). En face images provide the user with a view looking directly into the SB orifice, and therefore are useful for measuring orifice dimensions and features. OCT images were first normalized to the range 0–1 by the maximum and minimum values of the OCT pullback for convenience of processing. Since OCT is performed by helical-scanning a catheter inside the artery, a depth-resolved tissue intensity profile is generated at each angular position, which is also known as an A-line. We projected each 2-D cross-sectional image in polar coordinates into one dimension by averaging intensity values along each A-line between the lumen and a depth of 0.5mm. The lumen boundary was identified by a robust algorithm reported previously13. By combining all the projected A-line profiles of the stented region into one image, we created an en face projection image displayed as a function of pullback distance and the angular position. Because the SB regions have accumulated low intensity, they appear as dark holes in the en face projection view. The exact boundaries of SB can be segmented automatically by extracting the region with the pixel intensity value lower than a predefined threshold (T=0.3) and with the area larger than another threshold (20 pixels). The segmentation was reviewed and corrected manually by examining the SB orifice in the corresponding cross-sectional view and mapping it back to the en face view.
Fig. 1. En face projection view for quantification of side branches.
(A) – (B) OCT images in polar coordinates were projected to one dimension curves by averaging intensity values along each A-line between the lumen and a depth of 0.5mm. The side branch regions have low intensities in the projected curve. (C) By combining all the projected A-line profiles of the stented region into one image, an en face projection image can be created as a function of pullback distance and the angular position. *Guide wire artifact.
Ex vivo phantom model
The accuracy of area measurements obtained from en face views was validated using a bifurcation phantom model with three side holes (Supplement Fig. 1). OCT images were acquired 15 times across each side hole using a commercially available frequency domain (C7-XR OCT Intravascular Imaging System; St. Jude Medical Inc., St. Paul, Minnesota). The side hole orifice area was measured on the en face projection image by an independent investigator who was blinded to the actual size of the side holes. The accuracy of area measurements was evaluated by a linear regression model regressing the actual value on the estimated area measurements, and testing departures of the intercept and slope from 0 and 1 respectively. We also examined if there were any measured area values outside of the 95% prediction interval over whole range of experiment.
In vivo study population
All in vivo study subjects were selected from the Massachusetts General Hospital (MGH) OCT Registry. We identified 137 cases of bifurcation lesions with SB diameter ≥2mm on angiogram and treated with a Xience V (EES: everolimus-eluting stent) (Abbott Vascular, Santa Clara, CA, USA) or Resolute Integrity (ZES: zotarolimus-eluting stent) (Medtronic, Minneapolis, MN, USA) from August 2010 to January 2014. We excluded cases of restenosis (n=20), cases treated with two stents (n=12), and those with poor image quality (n=17). Finally, we included 42 EES and 46 ZES cases in the analysis. The registry protocol was approved by each institution’s ethics committee, and all patients provided written informed consent.
In vivo procedure and OCT image acquisition
All treatment strategies were performed at the discretion of the operators at each institution, including decisions regarding side branch dilatation and/or additional stenting. Cases in which a stent was implanted in the SB were, however, excluded from our study. Off-line quantitative coronary angiography (QCA) was performed to assess the reference and minimal lumen diameters of the target lesions before and after PCI using dedicated software (CASS version 5.10.1, Pie Medical Imaging BV, Maastricht, the Netherlands). OCT images were acquired at the end of PCI using commercially available frequency domain or time domain (M2/M3 Cardiology Imaging System; LightLab Imaging Inc., Westford, Massachusetts) OCT systems. All images were de-identified, digitally stored, and submitted to the MGH-OCT Registry Laboratory for analysis. Quantitative analysis was performed for every cross-sectional frame using proprietary software (St. Jude Medical) by independent, experienced investigators14.
In vivo quantification of side branch open area in an en face projection view
Using the en face projection algorithm, SB orifice area free from stent struts was evaluated in vivo. The stent struts were automatically detected based on a previous method15 and corrected manually, if necessary. The open area can be computed from the segmented SB and stent struts by using stent specific information (i.e. strut width) from the manufacturers. For the entire task, custom designed software was developed in MATLAB (The MathWorks, Inc. Natick, MA) to allow users to visualize, annotate, and review the OCT images of SB and stent struts. The SB orifice area (A) and the area obstructed by struts (B) were calculated on the en face view (Fig. 2). The percent open area was defined as (A−B) / A*100.
Fig. 2. Definitions.
(Left) A representative image of en face OCT reconstruction. *Guide wire artifact. (Right) A: Orifice area; B: Strut area. %Open area was calculated as (A−B) / A*100.
Definitions
The SB orifice width was defined as the length between both corners which divides the SB and main vessel on the cross sectional view. All parameters were evaluated from the proximal edge to the distal edge of the SB orifice, which were identified on the longitudinal OCT images.
Statistical Methods
If normally distributed according to Kolmogorov Smirnov test, categorical data were summarized as counts and proportions (%), and continuous data were summarized as mean ± standard deviation. Otherwise, median values with lower and upper quartiles were reported. Mean values were compared by t-test or one-way ANOVA when the data were normally distributed and by Mann-Whitney’s U test or Kruskal-Wallis test when the data were not normally distributed. Measurement accuracy of en face projection views was evaluated by means of a linear regression model. Independent risk factors for the lower %open area (below median; <85.5%) were determined by means of a multivariate logtistic model. The statistical significance was defined as P<0.05. All statistical analysis was performed with SPSS version 17.0 (SPSS, Chicago, Illinois, USA).
Results
Measurement accuracy of an en face projection view
The measurement values of three side branch holes (actual area: 3.14mm2, 3.97mm2, 4.91mm2) in ex vivo phantom were 3.09 ± 0.22 mm2, 3.79 ± 0.27 mm2, 4.87 ± 0.30 mm2, respectively. The estimated area predicted the value accurately in that the observed intercept was not different from 0 (−0.12434 ± 0.22782, p=0.588) and the slope was not different from 1 (1.00896 ± 0.05596, R-square = 0.88, p=0.874) (Fig. 3).
Fig. 3. Accuracy of area measurement on an en face image.
Horizontal-axis: estimated area of mimic side branch orifice in a phantom model; Vertical-axis: measured area of mimic side branch orifice on an en face projection image.
Baseline characteristics of in vivo assessment
Of the total 88 cases, 31 (17 EES and 14 ZES) cases were treated with SB dilatation and 57 (25 EES and 32 ZES) cases were treated with crossover stenting without SB dilatation (Fig. 4). Baseline characteristics are shown in Table 1. All variables other than the clinical presentation in cases without SB dilatation were comparable between patients treated with EES and those treated with ZES. There was no significant difference in the target vessel or medina classification between lesions treated with EES and those treated with ZES (Table 2). In both pre and post QCA analysis, the minimal lumen diameter was smaller in EES group than in ZES group in cases without SB dilatation. The frequency of use of the kissing balloon technique (KBT) was significantly higher in EES group than in ZES group (82.4% vs. 35.7%, P<0.01) (Table 2).
Fig. 4. Study flow chart.
EES: everolimus-eluting Stent (Xience V); ZES: zotarolimus-eluting stent (Resolute Integrity).
Table 1.
Baseline clinical characteristics
| Side branch dilatation | No side branch dilatation | |||||
|---|---|---|---|---|---|---|
| EES (n=17) | ZES (n=14) | P value | EES (n=25) | ZES (n=32) | P value | |
| Male, n (%) | 14 (82.4) | 10 (71.4) | 0.47 | 24 (96.0) | 29 (90.6) | 0.43 |
| Age (years), mean ± SD | 63.7±18.5 | 62.3±13.7 | 0.81 | 62.8±11.6 | 64.0±9.69 | 0.67 |
| BMI, mean ± SD | 26.0±4.21 | 26.2±3.30 | 0.92 | 26.3±4.52 | 25.0±3.74 | 0.27 |
| Hypertension, n (%) | 10 (58.8) | 8 (57.1) | 0.25 | 20 (80.0) | 22 (68.8) | 0.50 |
| Hyperlipidemia, n (%) | 10 (58.8) | 7 (50.0) | 0.27 | 19 (76.0) | 22 (68.8) | 0.29 |
| Diabetes, n (%) | 7 (41.2) | 5 (35.7) | 0.53 | 7 (28.0) | 9 (28.1) | 0.98 |
| Renal insufficiency, n (%) | 2 (11.8) | 1 (7.1) | 0.50 | 0 (0) | 2 (6.3) | 0.20 |
| Current smoker, n (%) | 2 (11.8) | 2 (15.4) | 0.77 | 7 (28.0) | 7 (23.3) | 0.69 |
| Family history of CAD, n (%) | 5 (29.4) | 2 (14.3) | 0.36 | 3 (12.0) | 1 (3.1) | 0.39 |
| Previous MI, n (%) | 3 (17.6) | 5 (35.7) | 0.25 | 5 (20.0) | 5 (15.6) | 0.63 |
| Previous PCI, n (%) | 3 (17.6) | 1 (7.1) | 0.39 | 7 (28.0) | 7 (21.9) | 0.59 |
| Clinical presentation | 0.76 | <0.01 | ||||
| Stable angina, n (%) | 11 (64.7) | 11 (78.6) | 11 (44.0) | 28 (87.5) | ||
| Acute coronary syndrome, n (%) | 6 (35.3) | 3 (21.4) | 14 (56.0) | 4 (12.5) | ||
BMI: body mass index; CAD: coronary artery disease; EES: everolimus-eluting stent = Xience V; MI: myocardial infarction; PCI: percutaneous coronary intervention; ZES: zotarolimus-eluting stent = Resolute Integrity
Table 2.
Angiographic and procedural characteristics
| Side branch dilatation | No side branch dilatation | |||||
|---|---|---|---|---|---|---|
| EES (n=17) | ZES (n=14) | P value | EES (n=25) | ZES (n=32) | P value | |
| Angiographic characteristics | ||||||
| Target lesion, n (%) | 0.09 | 0.27 | ||||
| LAD | 12 (70.6) | 6 (42.9) | 16 (64.0) | 22 (68.8) | ||
| LCX | 2 (11.8) | 7 (50.0) | 7 (28.0) | 10 (31.3) | ||
| RCA | 2 (11.8) | 0 (0) | 2 (8.0) | 0 (0) | ||
| LMT | 1 (5.9) | 1 (7.1) | 0 (0) | 0 (0) | ||
| Medina classification, n (%) | 0.77 | 0.28 | ||||
| (1,1,1) | 3 (17.6) | 3 (21.4) | 4 (16.0) | 3 (9.4) | ||
| (1,1,0) | 3 (17.6) | 1 (7.1) | 3 (12.0) | 4 (12.5) | ||
| (0,1,1) | 2 (11.8) | 3 (21.4) | 2 (8.0) | 3 (9.4) | ||
| (1,0,1) | 2 (11.8) | 0 (0) | 4 (16.0) | 0 (0) | ||
| (1,0,0) | 1 (5.9) | 2 (14.3) | 4 (16.0) | 9 (28.1) | ||
| (0,1,0) | 5 (29.4) | 4 (28.6) | 8 (32.0) | 12 (37.5) | ||
| (0,0,1) | 1 (5.9) | 1 (7.1) | 0 (0) | 1 (3.1) | ||
| True bifurcation, n (%) | 7 (41.2) | 6 (42.9) | 0.93 | 10 (40.0) | 6 (18.8) | 0.08 |
| Calcified lesion, n (%) | 7 (41.2) | 3 (21.4) | 0.24 | 14 (56.0) | 11 (34.4) | 0.10 |
| Pre procedure QCA | ||||||
| Main vessel | ||||||
| Proximal REF (mm) | 2.94±0.76 | 2.79±0.80 | 0.61 | 2.90±0.82 | 2.99±0.48 | 0.62 |
| Distal REF (mm) | 2.45±0.86 | 2.56±0.59 | 0.72 | 2.35±0.89 | 2.55±0.49 | 0.28 |
| MLD (mm) | 0.99±0.46 | 1.17±0.48 | 0.32 | 0.83±0.43 | 1.12±0.51 | 0.03 |
| %DS | 64.2±16.2 | 57.7±12.5 | 0.26 | 69.6±15.2 | 59.1±17.3 | 0.02 |
| Side branch | ||||||
| REF (mm) | 2.25±0.30 | 2.39±0.43 | 0.29 | 2.23±0.18 | 2.38±0.37 | 0.05 |
| MLD (mm) | 1.43±0.42 | 1.42±0.37 | 0.93 | 1.37±0.48 | 1.47±0.48 | 0.43 |
| %DS | 34.3±15.9 | 38.1±10.6 | 0.44 | 34.5±17.6 | 38.4±16.4 | 0.39 |
| Post procedure QCA | ||||||
| Main vessel | ||||||
| MLD (mm) | 2.35±0.23 | 2.59±0.70 | 0.19 | 2.48±0.51 | 2.87±0.61 | 0.02 |
| %DS | 16.7±9.61 | 18.5±8.02 | 0.58 | 17.8±10.5 | 12.9±8.09 | 0.07 |
| Acute gain (mm) | 1.36±0.48 | 1.46±0.76 | 0.65 | 1.57±0.73 | 1.80±0.66 | 0.26 |
| Side branch | ||||||
| MLD (mm) | 1.20±0.44 | 1.21±0.46 | 0.95 | 1.46±0.36 | 1.57±0.47 | 0.37 |
| %DS | 16.7±9.09 | 20.3±8.17 | 0.29 | 34.9±13.7 | 35.7±18.2 | 0.87 |
| Acute gain (mm) | 0.23±0.06 | 0.21±0.08 | 0.88 | 0.04±0.59 | 0.09±0.49 | 0.76 |
| Procedural characteristics | ||||||
| Main branch | ||||||
| Pre dilatation | 8 (47.1) | 11 (78.6) | 0.07 | 19 (76.0) | 26 (81.3) | 0.63 |
| Post dilatation | 15 (88.2) | 10 (71.4) | 0.24 | 20 (80.0) | 19 (59.4) | 0.10 |
| Stent diameter (mm) | 2.96±0.46 | 3.07±0.59 | 0.55 | 2.86±0.36 | 2.88±0.46 | 0.84 |
| Stent length (mm) | 22.9±7.67 | 21.3±6.99 | 0.53 | 23.2±9.35 | 23.0±8.25 | 0.92 |
| Stent / Vessel ratio | 1.13±0.17 | 1.15±0.32 | 0.77 | 1.06±0.25 | 1.06±0.21 | 0.96 |
| Side branch | ||||||
| Pre dilatation | 5 (29.4) | 1 (7.1) | 0.12 | - | - | - |
| KBT | 14 (82.4) | 5 (35.7) | <0.01 | - | - | - |
| Balloon diameter | 2.13±0.40 | 2.38±0.21 | 0.09 | - | - | - |
| Balloon / Vessel ratio | 0.99±0.24 | 0.91±0.21 | 0.50 | - | - | - |
| Max balloon pressure (atm) | 9.73±5.01 | 12.0±4.00 | 0.34 | - | - | - |
KBT: kissing balloon technique; LAD: left anterior descending artery; LCX: left circumflex artery; LMT: left main trunk; MLD: minimal vessel diameter; QCA: quantitative coronary angiography; RCA: right coronary artery; REF: reference vessel diameter; %DS: percent diameter stenosis
Comparison between cases treated with EES and ZES
The results for quantitative OCT analysis are shown in Table 3 and Fig. 5. On the analysis using en face views, the %open area was significantly larger in cases treated with EES than with ZES (89.2% [83.7 – 91.3] vs. 84.3% [78.9 – 87.8], P=0.04) in cases without SB dilatation. However, the %open area was not significantly different between the EES group and the ZES group (89.0% [87.8 – 90.9] vs. 91.4% [86.1 – 94.0], P=0.33) in the cases with SB dilatation. The difference of %open area was statistically significant between SB dilatation and no SB dilatation in ZES group (91.4% [86.1 – 94.0] vs. 84.3% [78.9 – 87.8], P=0.04), whereas the difference was not significant in EES group (P=0.58). The representative images are shown in Fig. 6. Multivariate analysis showed that the use of ZES was an independent predictor for the smaller %open area (below median; <85.5%, odds ratio 4.85, 95% confidential interval [1.23 – 19.1], P=0.02) in addition to small orifice width in no SB dilatation (Table 4).
Table 3.
OCT comparison of open area between EES and ZES
| Side branch dilatation | No side branch dilatation | |||||
|---|---|---|---|---|---|---|
| EES (n=17) | ZES (n=14) | P value | EES (n=25) | ZES (n=32) | P value | |
| En face view analysis | ||||||
| Orifice area (mm2) | 3.34 [2.33–5.44] | 3.54 [2.45–4.92] | 0.70 | 1.37 [0.97–2.34] | 1.57 [0.91–2.24] | 0.72 |
| Strut area (mm2) | 0.40 [0.25–0.57] | 0.31 [0.18–0.45] | 0.18 | 0.19 [0.11–0.30] | 0.23 [0.19–0.32] | 0.29 |
| Open area (mm2) | 2.80 [2.00–4.95] | 3.47 [2.20–4.55] | 0.79 | 1.17 [0.79–2.18] | 1.29 [0.77–1.91] | 0.60 |
| % Open area | 89.0 [87.8–90.9] | 91.4 [86.1–94.0] | 0.33 | 89.2 [83.7–91.3] | 84.3 [78.9–87.8] | 0.04 |
| Cross sectional analysis | ||||||
| Orifice length (mm) | 2.10 [1.35–2.20] | 1.85 [1.48–2.23] | 0.69 | 1.20 [1.05–1.55] | 1.10 [0.90–1.30] | 0.19 |
| Max orifice width (mm) | 2.11 [1.85–2.44] | 2.28 [1.74–2.95] | 0.28 | 1.76 [1.59–2.28] | 1.85 [1.71–2.28] | 0.38 |
| Malapposed strut* (%) | 10.6 [2.21–25.9] | 9.52 [2.50–12.5] | 0.58 | 5.71 [3.19–10.5] | 4.01 [0.00–9.63] | 0.10 |
Stent strut malapposition was defined as any strut with the distance from the middle of the strut to the adjacent visible surface of the vessel being greater than the sum of the strut thickness plus polymer thickness which were provided by the manufacture.
Fig. 5. Comparison of %open area between EES and ZES.
Comparison of %open area measured on the en face view. Gray column: Xience V (EES); white column: Resolute Integrity (ZES); Y-axis: percent of open area; *P=0.04 vs. ZES with side branch (SB) dilatation.
Fig. 6. Representative cases.
(A) – (C) A case of no side branch dilatation with Xience V (EES). No strut was seen at the center of SB orifice (*) in the longitudinal reconstructed view (A), cross sectional view (B), and en face view (C). (D) – (F) A case of no side branch dilatation with Resolute Integrity (ZES). Two struts (white arrow) in the SB orifice were visible in the cross sectional view (E). The shape of crossing struts was well depicted in the en face view (F). (G) – (I) A case of SB dilatation after main vessel stenting with Resolute Integrity (ZES). Struts were well eliminated from the center of SB orifice (I). Stent struts crossing the SB orifice were highlighted by gray color in the en face view.
Table 4.
Independent predictors for smaller %open area* in no SB dilatation cases
| Factors | OR | 95% CI | P value |
|---|---|---|---|
| ZES (Resolute Integrity) | 4.85 | 1.23 – 19.1 | 0.02 |
| Orifice width < 1.84mm† | 7.52 | 1.84 – 30.8 | 0.01 |
| Orifice length < 1.1mm† | 1.64 | 0.40 – 6.80 | 0.49 |
| Small vessel < 2.5mm‡ | 3.06 | 0.59 – 15.8 | 0.18 |
| Small side branch < 2.5mm‡ | 0.33 | 0.07 – 1.57 | 0.16 |
Smaller %open area was defined as below median (85.5) in no side branch dilatation cases;
Measured on the cross sectional OCT image. The values 1.84 and 1.1 were the median of orifice width and length respectively;
Measured by quantitative coronary angiography;
SB: side branch; OR: odds ratio; CI: confidential interval
Comparison between single balloon and kissing balloon dilatation
Different treatment strategies for the treatment of bifurcation lesions are compared. The %open area of both single balloon dilatation (91.1% [83.7 – 93.6]) and kissing balloon dilatation (89.4% [87.9 – 91.0]) were significantly larger than no SB dilatation cases (85.5% [81.0 – 89.5], P=0.02, P=0.02, respectively). No statistically significant difference was seen in %open area between single balloon dilatation and kissing balloon dilatation (P=0.78).
Discussion
The main findings of this study are: 1) The quantitative evaluation of SB orifice by an en face projection processing of OCT images was validated using a bifurcation phantom model, and 2) In vivo application of the en face projection processing demonstrated differences in %open area among stent types and treatment strategies. In the no SB dilatation case, the Xience V stent provides a larger %open area compared to the Resolute Integrity stent. A significant difference in the %open area between with and without SB dilatation was demonstrated in the Resolute Integrity stent.
Risk of thrombotic complications after PCI for bifurcation lesions
Bifurcation lesions remain a challenging subset in PCI due to procedural complexity, and they are associated with worse clinical outcomes, including a higher incidence of stent thrombosis1. Several pathological studies4, 16 have demonstrated that ISA and uncovered struts crossing the SB orifice can serve as a nidus for stent thrombosis. Indeed, a recent OCT study demonstrated both ISA and NASB struts were not well covered by the neointima in the chronic phase, in contrast to well apposed struts5. The incidence of ISA and the percent of uncovered struts is significantly lower with current generation drug-eluting stent (DES) compared to the first generation DES17, 18. However, side branch orifice and NASB struts had not previously been well evaluated in vivo due to lack of diagnostic modality.
Assessment of stent struts around the side branch orifice in ex vivo models
Several ex vivo studies have attempted to quantitatively evaluate stent struts around the SB orifice. Ormiston et al. investigated the area of SB orifice obstructed by the residual struts after crush stenting using a micro-computed tomographic imaging of bench top model6. They demonstrated that the obstruction area was dependent on the stenting / ballooning strategy and the stent cell size. Mortier et al. evaluated the SB orifice after provisional stenting using a dedicated finite element simulation model7. They assessed the SB orifice en face and compared orifice area stenosis due to the remaining struts among different ballooning techniques and stent types. They reported the modified KBT (2-step dilatation) was better than simultaneous KBT (1-step dilatation) and the results were similar among three stent types [Integrity (Medtronic, Minneapolis, MN, USA), Omega (Boston Scientific, Natick, MA, USA), and Multi-link 8 (Abbott Vascular, Santa Clara, California)]. Another study also reported the characteristics of the stent platform affected the orifice area free from strut after SB dilatation19. These studies indicate that the quantitative evaluation of the SB orifice and NASB struts in vivo might give new insight into optimal strategies for bifurcation treatment.
In vivo assessment of side branch orifice area free from non-apposed strut
To quantitatively evaluate the SB orifice area after stenting in vivo, we developed an algorithm and reconstructed raw OCT digital data into en face projection images. We demonstrated the accuracy of area measurement on the en face projection images using a bifurcation phantom model, though important differences exist between a mimic side hole in a phantom model and a SB orifice in a human coronary artery (e.g. angulation between main vessel and side branch10, variation of shape, and border irregularity). Next, we applied the algorithm to actual cases of bifurcation lesion stenting to examine the impact of stent design and treatment strategies on the SB orifice area free from non-apposed struts in vivo, as the previous ex vivo studies examined. To the best of our knowledge, this is the first study quantitatively evaluating SB orifice after stenting in consecutive bifurcation cases.
Stent cell design and orifice area free from non-apposed strut
In this study, the %open area in cases with no SB dilatation was significantly larger in the EES group compared with the ZES group, which was confirmed by multivariate analysis. In the provisional approach for bifurcation lesions, if the angiographic results of both the main vessel and SB are satisfactory after initial stent placement, the SB orifice would not be further dilated20. Therefore, the SB orifice remains jailed by the struts of stent within the main vessel. In those cases, the number of crossing struts should depend on the stent cell design itself, and the relative position and size between stent struts and SB orifice. The Xience V is a cobalt-chromium-based, open-cell design DES with thinner strut thickness (81̅m) covered by a 7.8̅m fluoropolymer coating. The Resolute Integrity stent is the open-cell and cobalt-chromium-based DES with 91̅m strut thickness and 6̅m polymer. The platform is made with a continuous sinusoid technology with laser fusion of the adjacent crown21. We believe that the difference in %open area among the two stents in cases without SB dilatation was mainly caused by these differences in the mechanical performance.
Side branch dilatation and orifice area free from non-apposed strut
Unlike the results in no SB dilatation cases, the %open area was similar between the two stents after SB dilatation. In EES cases, the %open area after SB dilatation was not significantly different from cases without SB dilatation indicating that sufficient open area was already achieved without SB dilatation and additional balloon dilatation of the SB did not further improve the open area. On the other hand, the %open area was significantly larger in cases with SB dilatation than in those without SB dilatation in ZES group, whereas we need further study to evaluate the importance of SB dilatation in cases treated with ZES.
In contrast to our results, Burzotta et al. demonstrated with a virtual simulation model that the %open area after SB dilatation varied among stent types19. They reported the %open area after KBT was better with EES (90.6%) than with Cypher stents (Cordis, Johnson & Johnson, Warren, NJ, USA) (84.7%) but worse than with the TAXUS Liberte (Boston Scientific, Natick, MA, USA) (98.3%) and the Endeavor Resolute (Medtronic, Minneapolis, MN, USA) (97.2%). Therefore, the %open area after SB dilatation might also be affected by the stent design. However, they simulated only a single case for each stent type in a silicon phantom model without any varieties of vessel morphologies. Thus, their findings need to be confirmed in larger number of cases in vivo.
As for the dilatation strategies, we did not find an advantage of KBT over single balloon technique with respect to %open area. Foin et al. also demonstrated comparable results of SB orifice area free from strut between KBT and single ballooning22 in a micro CT computational model.
Clinical implication and future issues
To date, there has been no clinical study which demonstrates the impact of NASB on the long-term clinical outcomes. This is partly due to the lack of a diagnostic modality to quantitatively assess NASB despite the development in the field of intracoronary image reconstruction. The recent development of 3D OCT image reconstruction enables us to qualitatively evaluate the anatomical structure of bifurcation lesions and struts around the SB orifice after the implantation of bioresorbable vascular scaffolds or stents8–10. These might contribute to the optimization of PCI in bifurcation lesions11. In addition, a fusion of 3D angiography and OCT images using a dedicated software was reported to more accurately measure the SB orifice area by taking into account the angle between the main branch and SB23. However, these methods could not quantitatively evaluate stent struts around the SB orifice so far whereas those have given some new aspects in the field of bifurcation PCI. In the present study, we report the differences in open area within the SB orifice among different stent types and strategies using a novel approach. This method can be applied to standard intravascular OCT data and is therefore widely applicable with existing clinical OCT systems. We believe our method should provide information to better understand the mechanisms of restenosis and stent thrombosis at previously treated bifurcation lesions, and will therefore lead to better strategies with optimal stent design and techniques to prevent these complications.
Limitations
Several limitations should be noted. First, this is not a randomized study but rather a retrospective analysis with limited number of cases from a registry database. Therefore, some underlying differences in patient or procedural characteristics might have affected the results. Second, we didn’t take account of the SB angulation and plaque burden, since this algorithm is not 3D. In future development, these features may be incorporated. Third, we applied our method in only two cases of left main disease due to the limited utilization of OCT for them in our registry. Fourth, most EES cases with SB dilatation were dilated by KBT. Thus, the results of open area should be interpreted with caution. Finally, due to the small number of cases, we could not examine the impact of %open area on clinical outcomes. Further studies with larger cohorts will be needed to evaluate the clinical impact of our findings.
Conclusions
The accuracy of SB orifice measurements using an en face image processing algorithm was demonstrated. This novel approach could quantitatively evaluate the area of NASB obstructing the SB orifice after bifurcation stenting in vivo. The SB orifice %open area was larger in the EES group compared to the ZES group after no SB dilatation. However, after SB balloon dilatation the SB orifice %open area was significantly improved in the ZES group.
Supplementary Material
Acknowledgement
The authors thank all the investigators and supporting staff at all sites of MGH OCT Registry for their contributions. We also thank Iris A McNulty, RN, Shankha Mukhopadhyay, MS, and James Chan, MS for exceptional work in the core laboratory and Russell Joye, AS for his editorial assistance.
Source of funding
A. Aguirre received support from the American Heart Association (14FTF20380185). Z. Wang and J. Fujimoto are supported in part by NIH R01-CA075289-17. J. Fujimoto receives royalties for intellectual property owned by MIT and licensed to St. Jude Medical. I. Jang received Research grant from Medtronic Inc., St. Jude Medical and Boston Scientific. This study was also supported by Mr. and Mrs. Michael Park and Allan and Gillian Gray.
Disclosure
This study was partly funded by Medtronic Inc. However, the financial sponsor was not involved in the process of data collection, analysis, and manuscript preparation.
Footnotes
Supplementary material for online only publication
Supplement Fig. 1 Ex vivo phantom model
References
- 1.Iakovou I, Schmidt T, Bonizzoni E, Ge L, Sangiorgi GM, Stankovic G, et al. Incidence, predictors, and outcome of thrombosis after successful implantation of drug-eluting stents. JAMA. 2005;293:2126–2130. doi: 10.1001/jama.293.17.2126. [DOI] [PubMed] [Google Scholar]
- 2.Attizzani GF, Capodanno D, Ohno Y, Tamburino C. Mechanisms, pathophysiology, and clinical aspects of incomplete stent apposition. J Am Coll Cardiol. 2014;63:1355–1367. doi: 10.1016/j.jacc.2014.01.019. [DOI] [PubMed] [Google Scholar]
- 3.Viceconte N, Tyczynski P, Ferrante G, Foin N, Chan PH, Barrero EA, et al. Immediate results of bifurcational stenting assessed with optical coherence tomography. Catheter Cardiovasc Interv. 2013;81:519–528. doi: 10.1002/ccd.24337. [DOI] [PubMed] [Google Scholar]
- 4.Farb A, Burke AP, Kolodgie FD, Virmani R. Pathological mechanisms of fatal late coronary stent thrombosis in humans. Circulation. 2003;108:1701–1706. doi: 10.1161/01.CIR.0000091115.05480.B0. [DOI] [PubMed] [Google Scholar]
- 5.Gutierrez-Chico JL, Regar E, Nuesch E, Okamura T, Wykrzykowska J, di Mario C, et al. Delayed coverage in malapposed and side-branch struts with respect to well-apposed struts in drug-eluting stents: In vivo assessment with optical coherence tomography. Circulation. 2011;124:612–623. doi: 10.1161/CIRCULATIONAHA.110.014514. [DOI] [PubMed] [Google Scholar]
- 6.Ormiston JA, Webster MW, Webber B, Stewart JT, Ruygrok PN, Hatrick RI. The "crush" technique for coronary artery bifurcation stenting: Insights from micro-computed tomographic imaging of bench deployments. JACC Cardiovasc Interv. 2008;1:351–357. doi: 10.1016/j.jcin.2008.06.003. [DOI] [PubMed] [Google Scholar]
- 7.Mortier P, Hikichi Y, Foin N, De Santis G, Segers P, Verhegghe B, et al. Provisional stenting of coronary bifurcations: Insights into final kissing balloon post-dilation and stent design by computational modeling. JACC Cardiovasc Interv. 2014;7:325–333. doi: 10.1016/j.jcin.2013.09.012. [DOI] [PubMed] [Google Scholar]
- 8.Grundeken MJ, Kraak RP, de Bruin DM, Wykrzykowska JJ. Three-dimensional optical coherence tomography evaluation of a left main bifurcation lesion treated with absorb(r) bioresorbable vascular scaffold including fenestration and dilatation of the side branch. Int J Cardiol. 2013;168:e107–e108. doi: 10.1016/j.ijcard.2013.07.255. [DOI] [PubMed] [Google Scholar]
- 9.Okamura T, Onuma Y, Garcia-Garcia HM, Regar E, Wykrzykowska JJ, Koolen J, et al. 3-dimensional optical coherence tomography assessment of jailed side branches by bioresorbable vascular scaffolds: A proposal for classification. JACC Cardiovasc Interv. 2010;3:836–844. doi: 10.1016/j.jcin.2010.05.011. [DOI] [PubMed] [Google Scholar]
- 10.Farooq V, Serruys PW, Heo JH, Gogas BD, Okamura T, Gomez-Lara J, et al. New insights into the coronary artery bifurcation hypothesis-generating concepts utilizing 3-dimensional optical frequency domain imaging. JACC Cardiovasc Interv. 2011;4:921–931. doi: 10.1016/j.jcin.2011.06.004. [DOI] [PubMed] [Google Scholar]
- 11.Onuma Y, Okamura T, Muramatsu T, Uemura S, Serruys PW. New implication of three-dimensional optical coherence tomography in optimising bifurcation pci. EuroIntervention. 2015;11(Suppl V):V71–V74. doi: 10.4244/EIJV11SVA15. [DOI] [PubMed] [Google Scholar]
- 12.Okamura T, Onuma Y, Yamada J, Iqbal J, Tateishi H, Nao T, et al. 3d optical coherence tomography: New insights into the process of optimal rewiring of side branches during bifurcational stenting. EuroIntervention. 2014;10:907–915. doi: 10.4244/EIJV10I8A157. [DOI] [PubMed] [Google Scholar]
- 13.Wang Z, Chamie D, Bezerra HG, Yamamoto H, Kanovsky J, Wilson DL, et al. Volumetric quantification of fibrous caps using intravascular optical coherence tomography. Biomed Opt Express. 2012;3:1413–1426. doi: 10.1364/BOE.3.001413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Di Vito L, Yoon JH, Kato K, Yonetsu T, Vergallo R, Costa M, et al. Comprehensive overview of definitions for optical coherence tomography-based plaque and stent analyses. Coron Artery Dis. 2014;25:172–185. doi: 10.1097/MCA.0000000000000072. [DOI] [PubMed] [Google Scholar]
- 15.Wang Z, Jenkins M, Linderman G, Bezerra H, Fujino Y, Costa M, et al. 3-d stent detection in intravascular oct using a bayesian network and graph search. IEEE Trans Med Imaging. 2015;34:1549–1561. doi: 10.1109/TMI.2015.2405341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Joner M, Finn AV, Farb A, Mont EK, Kolodgie FD, Ladich E, et al. Pathology of drug-eluting stents in humans: Delayed healing and late thrombotic risk. J Am Coll Cardiol. 2006;48:193–202. doi: 10.1016/j.jacc.2006.03.042. [DOI] [PubMed] [Google Scholar]
- 17.Toledano Delgado FJ, Alvarez-Ossorio MP, de Lezo Cruz-Conde JS, Bellido FM, Romero Moreno MA, Fernandez-Aceytuno AM, et al. Optical coherence tomography evaluation of late strut coverage patterns between first-generation drug-eluting stents and everolimus-eluting stent. Catheter Cardiovasc Interv. 2014;84:720–726. doi: 10.1002/ccd.25235. [DOI] [PubMed] [Google Scholar]
- 18.Papayannis AC, Cipher D, Banerjee S, Brilakis ES. Optical coherence tomography evaluation of drug-eluting stents: A systematic review. Catheter Cardiovasc Interv. 2013;81:481–487. doi: 10.1002/ccd.24327. [DOI] [PubMed] [Google Scholar]
- 19.Burzotta F, Mortier P, Trani C. Characteristics of drug-eluting stent platforms potentially influencing bifurcated lesion provisional stenting procedure. EuroIntervention. 2014;10:124–132. doi: 10.4244/EIJV10I1A19. [DOI] [PubMed] [Google Scholar]
- 20.Latib A, Colombo A. Bifurcation disease: What do we know, what should we do? JACC Cardiovasc Interv. 2008;1:218–226. doi: 10.1016/j.jcin.2007.12.008. [DOI] [PubMed] [Google Scholar]
- 21.Turco MA. The integrity bare-metal stent made by continuous sinusoid technology. Expert Rev Med Devices. 2011;8:303–306. doi: 10.1586/erd.11.2. [DOI] [PubMed] [Google Scholar]
- 22.Foin N, Torii R, Mortier P, De Beule M, Viceconte N, Chan PH, et al. Kissing balloon or sequential dilation of the side branch and main vessel for provisional stenting of bifurcations: Lessons from micro-computed tomography and computational simulations. JACC Cardiovasc Interv. 2012;5:47–56. doi: 10.1016/j.jcin.2011.08.019. [DOI] [PubMed] [Google Scholar]
- 23.Karanasos A, Tu S, van Ditzhuijzen NS, Ligthart JM, Witberg K, Van Mieghem N, et al. A novel method to assess coronary artery bifurcations by oct: Cut-plane analysis for side-branch ostial assessment from a main-vessel pullback. Eur Heart J Cardiovasc Imaging. 2015;16:177–189. doi: 10.1093/ehjci/jeu176. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.