appendix - JACC: Cardiovascular Imaging

advertisement
Online Appendix for the following JACC: Cardiovascular Imaging paper
TITLE: Ex Vivo Assessment of Vascular Response to Coronary Stents in Humans by HighResolution Optical Frequency Domain Imaging
AUTHORS: Masataka Nakano, MD, Marc Vorpahl, MD, Fumiyuki Otsuka, MD, Masanori
Taniwaki, MD, Saami K. Yazdani, PhD, Aloke V. Finn, MD, Elena R. Ladich, MD, Frank D.
Kolodgie, PhD, Renu Virmani, MD
APPENDIX
Supplemental Methods
Stented Lesions
Unfixed or formalin-fixed hearts or vessels were radiographed by high-contrast film–based
X-ray to determine the presence and location of stent implants. For intact hearts, epicardial
coronary arteries with device implants were initially removed from the heart, and major side
branches were ligated to prevent the loss of pressure during artery interrogation. Before imaging,
the stented arterial segments were mounted on a custom frame by attaching the proximal and
distal ends to fixed luer connections. The proximal end included a Y-shaped connector for the
introduction of imaging catheters and the pressure perfusion. The stented segments were
perfused at 90 mm Hg with phosphate-buffered saline (PBS, pH 7.4) using a continuous pump
with an open circuitry system. In cases in which unfixed hearts were received (cases 3 and 9 in
Table 1 of the main paper), imaging was repeatedly performed before and after formalin fixation.
Histology Processing for Coronary Stents
Following artery interrogation, the stented vessels were removed from the perfusion
apparatus, dehydrated in a graded series of ethanol, and embedded in methylmethacrylate resin.
After polymerization, the stented artery was segmented at 2-mm intervals using a diamond
wafering blade. Histological sections were then prepared on a rotary microtome equipped with a
tungsten carbide knife, at four to six microns, mounted and stained with hematoxylin and eosin
(H&E) and modified Movat pentachrome (1). The confirmation of hypersensitivity was based on
recognition of eosinophils by Luna stain (2).
Difference of OFDI Signal from Unfixed and Formalin-fixed Tissue
In the limited cases (case numbers 3 and 9 in Table 1 of the main paper), optical frequency
domain imaging was performed before and after formalin fixation. Histological slides from those
cases including i) normal neointima >1 year (n = 6 sections) and ii) organized thrombus (n = 5)
were co-registered with corresponding OFDI images obtained in the unfixed and fixed status.
The OFDI signal intensity was calculated from luminal surface to 0.3-0.4 mm depth, and fitted to
an approximate exponential formula (y = A×exp−Bx), where index A represents a “Peak
intensity,” and index B reflects an “Attenuation rate” as described in the main paper. For all
sections, 3 regions of interest (ROI) were selected for each feature were examined and averaged.
Prior to this evaluation, all OFDI images were adjusted to the same default setting, and the
analysis was performed by Image J software (National Institutes of Health, Rockville, MD).
Statistical Analysis
The values were expressed as mean  SD for continuous valuables. Paired t-test was used to
calculate the significance of differences between OFDI signals from unfixed and formalin-fixed
tissues. A value of p < 0.05 was considered statistically significant.
2
3
Supplemental Results/Discussion
Results of OFDI signal analysis of the differences between unfixed and formalin-fixed tissue
are summarized in Online Table 1 and Online Figure 6. There were no en face differences in
OFDI appearance of unfixed and fixed tissues. Similarly, quantitative evaluation of OFDI
signals did not show absolute disparities, where signal Peak intensity of normal neointima >1
year in unfixed and fixed tissues were 180.9 ± 5.0 and 182.1 ± 4.4, respectively (p = 0.33), and
Peak intensity of organized thrombus were 136.1 ± 3.8 (unfixed) vs. 135.4 ± 3.8 (fixed) (p =
0.62). In addition, no difference was identified in the analysis of Attenuation rate: normal
neointima >1 year, 0.37 ± 0.17 (unfixed) vs. 0.37 ± 0.13 (fixed) (p = 0.92); organized thrombus,
0.40 ± 0.06 (unfixed) vs. 0.41 ± 0.07 (fixed) (p = 0.67).
Our observation in the limited cases showed no substantial differences in the OFDI signal
reflection regardless of tissue fixation. However, the results was acknowledged only in the
normal neointima >1 year and organized thrombus, and was not confirmed in other neointimal
features described in the main manuscript. Further, it is difficult to determine whether the
absence of significant difference in the quantitative signal analysis derived from the plausible
analogy or from the lack of statistical power. Therefore, further studies are required to confirm
the results in a larger cohort of cases.
4
Supplemental Table 1. Difference of OFDI Signal from Unfixed and Formalin-fixed Tissue
Unfixed
Fixed
P value
Intensity Index
180.9 ± 5.0
182.1 ± 4.4
0.33
Attenuation Index
0.37 ± 0.17
0.37 ± 0.13
0.92
Intensity Index
136.1 ± 3.8
135.4 ± 3.8
0.62
Attenuation Index
0.40 ± 0.06
0.41 ± 0.07
0.67
Normal neointima >1 year (n = 6)
Organized thrombus (n = 5)
5
Supplemental Figure 1. Human coronary histology with a stent implant with corresponding
IVUS and OFDI
A and B: histological sections of stented coronary artery segment (A: H&E; B: Movat,
magnification ×12.5). A stent was implanted over a lesion with necrotic core (NC). Stent struts
were observed on the luminal surface (black arrows). C: co-registered OFDI image. D: coregistered IVUS image.
6
Supplemental Figure 2. Example of strut identification in sequential frames in OFDI
A: histological section of stented coronary artery. B to G: sequential OFDI images. Individual
struts in the OFDI images were matched with the corresponding histological section.
7
Supplemental Figure 3. Atherosclerotic change in neointima with foam cell macrophage
infiltration
A: histological image shows lipid core (LC) formation deep inside neointima (Movat,
magnification ×20). B: magnified image of the inset in A (×100). Macrophages with foam cell
morphology are infiltrating the “shoulder” of a newly formed area of atherosclerosis in
neointima (yellow arrows) (asterisk: fixation artifact). C: corresponding OFDI to image A. LC in
histology appears dark without clear border. D: corresponding OFDI to image B. A narrow
8
signal-rich layer was observed in the area with macrophages in histology (white arrows).
9
Supplemental Figure 4. Thin cap fibroatheroma in neointima
A: histological image shows a relatively small eccentric necrotic core (NC) near the luminal
surface (Movat, magnification ×20). B: magnified image of the inset in A (×100). NC containing
cholesterol crystals (area enclosed by yellow arrowheads) is covered by a thin fibrous cap. C:
corresponding OFDI to image A. D: corresponding OFDI to image B. A thin, signal-rich layer
was observed near the luminal surface, sequentially followed by signal attenuation in deeper
intima.
10
Supplemental Figure 5. Other histological features identified in neointima
A: histological image with sheet of calcification within neointima matrix (black arrows) (Movat,
magnification ×20). B: corresponding OFDI to image A. The stent lumen was severely narrowed
with neointimal tissues, in which a black mass sharply delineated with clear borders was
identified. C: corresponding IVUS to image A. A strong signal with a trailing shadow was
observed with less clear margins in the calcified area. D: histological image shows deep
dissection in neointima. E: corresponding OFDI to image D. A discontinuation of neointimal
surface is clearly identified. F: corresponding IVUS image shows a neointimal disruption but
with poorer resolution compared to OFDI.
11
Supplemental Figure 6. Difference of OFDI appearance in unfixed and fixed tissue
A: OFDI image of coronary artery >1 year following stent implantation. Right: magnified image
of the inset in the left image demonstrating bright and homogeneous appearance of neointima
above the strut. B: OFDI image acquired following formalin fixation. The neointimal tissue
shows a similarity to the image in A. C: Histology corresponding to the images in A and B
(Movat, magnification ×20). Right: Magnified image of the inset in the left image showing
normal neointima consisting of smooth muscle cells and extracellular matrix such as
proteoglycans (Movat, ×200). D: OFDI image of stented coronary segment with severe luminal
narrowing. Right: magnified image of the inset in the left showing relatively dark, coarse tissue
with various sizes of black holes (white arrows) (* lumen area). E: OFDI image acquired
following formalin fixation. The tissue within the stent appears analogous to the image D, and
12
also has black holes (white arrows) (* lumen area). F: Histology corresponding to the images in
D and E (Movat, ×20). Right: Magnified image of the inset in the left image showing sparse
tissue rich in proteoglycans with various sizes of cavities (black arrows) representing
angiogenesis in the organized thrombus (Movat, ×400).
13
Supplemental References
1.
Finn AV, Joner M, Nakazawa G, et al. Pathological correlates of late drug-eluting stent
thrombosis: strut coverage as a marker of endothelialization. Circulation 2007;115:243541.
2.
Luna LG. Histologic Staining Methods of the Armed Forces Institute of Pathology. New
York: McGrow-Hill Book Co, 1968.
14
Download