|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 2004; 44:972-979, doi:10.1016/j.jacc.2004.05.066 © 2004 by the American College of Cardiology Foundation |
,||,
,,*
* Cardiology Division
Wellman Center for Photomedicine
Department of Pathology
Department of Radiology, Harvard Medical School, Boston, MassachusettsUSA
|| Department of Cardiology, Kinki University School of Medicine, Kinki, Japan
Manuscript received October 23, 2003; accepted May 31, 2004.
* Reprint requests and correspondence: Dr. Guillermo J. Tearney, Wellman Center for Photomedicine, Massachusetts General Hospital, BAR 703, Boston, Massachusetts 02114 (Email: gtearney{at}partners.org).
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
BACKGROUND: The relative significance of focal macrophage infiltration and generalized coronary inflammation for predicting acute coronary events is a currently a source of considerable controversy in cardiology. Lack of a high-resolution cross-sectional imaging modality has limited macrophage evaluation in vivo.
METHODS: Intracoronary OCT imaging was performed at culprit and non-culprit plaques in patients presenting with stable angina pectoris, unstable angina pectoris,and ST-segment elevation myocardial infarction. Macrophage densities were quantified from these images and analyzed with respect to the clinical presentations of the patients under investigation.
RESULTS: A significantly greater macrophage density was found in unstable patients, both for fibrous and lipid-rich plaques (p = 0.025 and p = 0.002, respectively). Within each patient, the macrophage densities at culprit and non-culprit lesions correlated significantly (r = 0.66, y = 0.88x + 0.43, p = 0.01). Sites of plaque rupture demonstrated a greater macrophage density than non-ruptured sites (6.95 ± 1.60%, 5.29 ± 1.17%; p = 0.002). Surface macrophage infiltration was a stronger predictor of unstable clinical presentation than subsurface infiltration for culprit lesions (p = 0.035) but not for remote lesions (p = 0.80).
CONCLUSIONS: Our results demonstrate that increases in both multi-focal and focal macrophage densities are highly correlated with symptom severity. By providing a means of detecting increases in plaque macrophage content before an acute event, this technique may aid in determining prognosis and guiding preventive therapy.
| |||||||||
An important goal of research in coronary atherosclerosis is to elucidate the degree of risk imparted by individual or focal plaque features versus multi-focal coronary inflammation. Measurement of macrophage densities within high-resolution cross-sectional images of plaques from patients with different clinical syndromes could shed light on this issue. Intravascular optical coherence tomography (OCT) is an optical analog of ultrasound that provides high-resolution (
10 µm) cross-sectional images of the arterial wall (15–17). Optical coherence tomography is a reliable and reproducible modality for plaque characterization (17,18). Recently, we demonstrated the ability of OCT to accurately quantify macrophage infiltration within atherosclerotic plaque ex vivo by establishing a high degree of correlation between OCT signal variance and the presence of CD68-positive cells (19). Extension of this method enables transformation of the conventional OCT image into a two-dimensional representation of macrophage content (Fig. 1a). The purpose of the present study was to utilize this novel technique to evaluate the distribution of coronary macrophages in living patients with ST-segment elevation myocardial infarction (STEMI), acute coronary syndromes (ACS), and stable angina pectoris (SAP).
|
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
1.8 mg/dl), coronary artery bypass graft intervention, and emergent procedures including primary angioplasty. The study was approved by the Partners Institutional Review Board, and all patients gave informed consent before participation. Clinical presentation. Using standard definitions, clinical presentation was defined as STEMI, ACS constituting non-STEMI and unstable angina pectoris, and SAP (20). The culprit lesion was determined using coronary angiography in conjunction with the patient's electrocardiogram and ventriculogram (21).
OCT imaging. The technique of intracoronary OCT imaging has previously been described (17). Briefly, before percutaneous coronary intervention, a 3.0-F OCT catheter was passed over a 0.014-inch angioplasty wire through a standard 7-F guide catheter. Under angiographic guidance, the OCT catheter was advanced to the culprit lesion. Images were acquired at 4 frames/s during intermittent saline injection (8 to 10 ml) and digitally recorded. We acquired images at the center of the culprit plaque and at its proximal and distal segments. Thus, up to three angiographically distinct images of the culprit plaque were obtained per patient. In addition, within the same vessel, we imaged angiographically mild or moderate lesions (30% to 70% stenosis) that were remote from the culprit site. Each OCT image was analyzed independently. Using previously established criteria, images of lipid-rich or fibrous plaques were selected for analysis by two observers who were blinded to the clinical presentation (18). Images with significant signal attenuation that precluded satisfactory evaluation of plaque morphology were excluded from the analysis.
Fibroatheroma cap segmentation. After the selection of images of lipid-rich plaques, the caps were outlined using automatic bimodal histogram segmentation (22,23). The threshold was set at the nadir of the bimodal histogram distribution computed from the pixel values within the plaque (22). The lateral boundaries of the cap were marked at the interface between the lipid pool and the adjacent fibrous tissue (Fig. 1c). For rupture sites, the interior boundary was marked 250 µm from the lateral-most portion of the intimal disruption. For each image, the fibrous cap was further segmented by morphologically eroding the cap with a 50 µm2 kernel into the most superficial 50 µm, which we defined as "surface," and the remainder of the cap, which we defined as "subsurface." This image processing provided two distinct cap segments, which allowed comparison between the surface (<50 µm from the lumen) and subsurface (>50 µm from the lumen) macrophage densities of the fibrous cap. Portions of the cap that were obscured by: 1) overlying thrombus, 2) intraluminal blood, 3) guide wire, or 4) arterial wall compression by the catheter sheath were excluded from analysis.
Quantitative macrophage analysis.
We performed measurement of macrophage content on raw OCT data within the region of interest using a technique previously validated in an ex vivo study (19). Median filtering was performed with a 3 x 3 square kernel to remove speckle noise (IPLab Spectrum 3.1, Scanalytics, Fairfax, Virginia) (23,24). We then measured the normalized standard deviation (NSD) for each pixel within each cap using a 125 µm2 window centered at the pixel location (Fig. 1d):
 | (1) |
where NSD(x,y) was the normalized standard deviation of the OCT signal at pixel location (x,y), Smax was the maximum OCT image value, and Smin was the minimum OCT image value. Pixels within the (125 x 125) µm2 window that did not overlap with the segmented cap were excluded. For each image, we assessed macrophage density by obtaining the average of the NSD values within the segmented cap (mean NSD). Macrophage densities within rupture sites, surface, and subsurface caps were calculated in the same manner. Because the segmentation of the caps and the computation of the NSD were conducted in an automated fashion, the determination of macrophage density was objective and not likely subject to investigator bias.
Statistical analysis.
Results are expressed as means ± SD unless otherwise stated. The NSD measurements between the groups were compared using analysis of variance. Significant variance was further analyzed using the least squares mean t test, which does not control for multiple comparisons. Correlation between continuous variables was estimated using Pearson's correlation coefficient. Comparison between the acute presentations (ACS + STEMI) and the stable presentations (SAP) were performed using the Student t test. Receiver operating characteristic (ROC) curves were constructed comparing the true positive rate (sensitivity) to the false positive rate of surface and subsurface macrophage densities for predicting patients with an acute coronary event. Results for ROC analysis are expressed as area ± standard error. Pairwise comparisons of the area under the ROC curve (AUC) were conducted using MedCalc (MedCalc Software, Mariakerke, Belgium) and the method described by Hanley and McNeil (25,26). All other analyses were performed using SAS software (SAS Institute Inc., Cary, North Carolina). A p value of 
0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
Multi-focal macrophage infiltration. In order to determine if macrophage content was increased in multiple locations in unstable patients, we further analyzed macrophage densities within lipid-rich plaques at both culprit and remote sites within the same vessel. For culprit lesions, the macrophage density was significantly higher in STEMI (5.66 ± 1.44%) and ACS (5.91 ± 2.06%) groups than in the SAP (4.21 ± 1.74%) group (p = 0.013 and p = 0.003, respectively). At remote sites, a significantly greater macrophage density was found in the unstable (STEMI + ACS) group than in the stable group (5.53 ± 1.71% vs. 4.02 ± 2.02%; p = 0.03). Moreover, when we compared culprit and remote macrophage densities within each clinical syndrome, we found that no significant difference existed between culprit and remote sites for each clinical group (STEMI: p = 0.52; ACS: p = 0.81; SAP: p = 0.82) (Fig. 2a). Furthermore, within the same patients (n = 15), macrophage density at remote sites correlated significantly with that of culprit sites (r = 0.67, y = 0.88x + 0.43, p = 0.006)(Fig. 2b).
Focal macrophage infiltration. Having established that multi-focal macrophage content was elevated in patients with unstable coronary syndromes, we asked whether focal elevations in macrophage content were independently related to symptom severity. In order to address this question, we analyzed culprit lipid-rich plaques within the STEMI and ACS groups that demonstrated clear OCT evidence of plaque rupture (n = 6; 2 STEMI, 4 ACS). For each rupture site we analyzed a 250 µm segment at the point of disruption and compared the macrophage density of the rupture location to that of the remainder of the plaque (Fig. 3a). We found a significantly higher macrophage density at the rupture site than of the adjacent non-ruptured cap within the same image (6.95 ± 1.6%, 5.29 ± 1.17%; p = 0.002) (Fig. 3b). Furthermore, macrophage density at rupture sites was significantly greater than that of all non-ruptured culprit sites in the combined STEMI and ACS groups (6.95 ± 1.6%, 5.75 ± 1.8%; p = 0.04).
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
In this study, we have extended the results of previous investigations (32,33) by obtaining the first measurements of macrophage distributions within high-resolution cross-sectional images of human coronary plaques in vivo. In so doing we observed a link between the structural morphology and biologic activity of coronary plaques. The value of this biologic measurement is reflected in the significant differences in cap macrophage densities found in patients with stable and unstable coronary syndromes.
Our results demonstrate that increased multi-focal inflammation is present in patients with severe manifestations of coronary atherosclerosis. First, a high degree of correlation was found between the macrophage densities at remote and culprit sites within the same patient. Second, no significant difference in macrophage content was detected between the remote and culprit sites within each clinical group. Finally, within remote sites, a significantly greater macrophage density was found in the acute compared with the stable clinical syndromes. These findings support recent studies that demonstrate inflammatory cell activation not only within the culprit artery but also within non-culprit arteries (6,7). Our data therefore provide further proof of the multi-focal nature of inflammation in symptomatic coronary atherosclerosis and demonstrate in vivo that this inflammation is manifested, at least in part, by increased macrophage density in the fibrous caps of both culprit and non-culprit lesions.
By demonstrating a significantly greater macrophage density at sites of plaque rupture, our results indicate that focal increases in macrophage content within individual plaques also affect plaque stability. Focal macrophage activation is thought to cause plaque instability by reducing collagen synthesis, increasing collagen degradation, and inducing smooth muscle cell apoptosis (28). The significant role of focal inflammatory plaque features is further supported by our observation that surface macrophage content was more predictive of clinical syndrome in culprit plaques, but not in remote lesions. The increased significance of macrophages at the surface of culprit lesions alone indicates that the spatial distribution of macrophages is intrinsically different in culprit lesions from that in non-culprit lesions within the same patient, even in the setting of elevated generalized inflammation. This finding may be explained by the results of recent studies that have shown that macrophages present at plaque surfaces may promote instability through endothelial erosion and tissue factor expression (34–37). This predilection of macrophages for the surface of culprit lesions may provide us with a new parameter for assessing individual plaque vulnerability.
An added advantage of this method is the potential to assess macrophage distribution in the context of high-resolution cross-sectional images of plaque morphology (Fig. 1). Previous studies have demonstrated that OCT is capable of characterizing plaque type with high sensitivity and specificity (18). In this study, we demonstrated that in acute patients inflammation is increased not only within the fibroatheromas typically implicated in plaque rupture, but also in fibrous plaques. Although fibrous plaques have traditionally been considered more stable, erosion of the superficial endothelial layer overlying fibrous plaques is found in a significant number of patients with an acute coronary event (3,4).
Other methods for detecting arterial inflammation. Atherosclerotic plaque temperature and thermal heterogeneity have previously been shown to correlate with plaque macrophage content (38). Clinical application of this technique demonstrated significantly greater intracoronary thermal heterogeneity in patients presenting with acute coronary events (39), findings that are consistent with our data. Similarly, magnetic resonance imaging of iron oxide-labeled macrophages has recently been shown to predict ruptured or rupture-prone carotid plaques in vivo (40). Both techniques are of great interest insofar as they provide a functional measure of macrophages within atherosclerotic plaque. Optical coherence tomography complements these modalities, as it detects macrophages in the context of other microstructural features of plaque instability, such as a thin fibrous cap, a lipid-rich core, or evidence of rupture.
Study limitations. Because blood attenuates light, it must be removed from the field of view in order to obtain high quality optical images. In the current study, blood was purged during imaging by flushing the artery with saline. This procedure permitted clear cross-sectional imaging at discrete locations but precluded comprehensive evaluation of the entire vessel. It is likely, therefore, that some rupture sites within some culprit lesions were not visualized. Inadequate saline purging resulted in exclusion of some images from analysis because of poor OCT image quality. Exclusion of these images may have introduced an inadvertent selection bias and, as a result, our findings should be considered preliminary. Forthcoming enhancements to OCT technology, including improvements in saline delivery and faster acquisition rates, will enable comprehensive screening of long coronary segments. Unfortunately, thrombus rapidly attenuates the OCT signal, prohibiting accurate measurement of macrophage content arising from the vessel wall underlying the thrombus. This attenuation would affect the macrophage density measurement, independent of the actual macrophage content. Although in principle it is possible to correct for the attenuation of the overlying thrombus, this correction method has not yet been validated against a gold standard. As a result, we chose to eliminate lesions containing a thrombus from the data set. In doing so, we may have underestimated the effect of macrophages at sites of endothelial erosion and rupture. Despite this limitation, our results achieved a high level of significance.
Imaging was performed at a single time point following symptom onset. As a result, the degree to which inflammation was a cause or a consequence of the clinical presentation could not be examined. Several studies have demonstrated elevation of acute-phase reactants or inflammatory markers before clinical presentation, indicating that inflammation precedes plaque disruption (41,42). Upcoming clinical trials will address this question by investigating the evolution of macrophage distributions in patients with a variety of symptoms.
The use of HMG-CoA reductase inhibitors (statins) is currently thought to improve cardiac mortality by a variety of mechanisms, one of which includes the inflammatory modulation properties of statins (28). Unfortunately, in this study information on statin use was not adequately recorded, and we were therefore unable to evaluate the effect of statins on macrophage content. Larger clinical studies utilizing OCT are planned that will investigate the relationship between macrophage distributions and patient-specific factors including statin use, age, gender, diabetes, hyperlipidemia, homocysteinemia, smoking, and high sensitivity C-reactive protein.
Conclusions. This study provides evidence supporting both the vulnerable plaque model and the hypothesis of multi-focal inflammatory risk, linked by the common thread of increased macrophage infiltration. Therefore, we hypothesize that elevated multi-focal coronary macrophage content, present in both culprit lesions and at remote sites, serves as a background for heightened risk. Superimposed on this inflammatory background, local increases in macrophage content, particularly at the cap surface and at areas at high risk for rupture, further promote the instability of individual lesions. Future studies will test these hypotheses by using this technology to evaluate the significance of both local and generalized macrophage content for predicting acute coronary events. As our understanding of coronary atherosclerosis advances, we anticipate that the relationship between these two hypotheses will become clearer and that optimal treatment strategies will combine pharmacologic therapy to control multi-focal intracoronary inflammation and directed local therapy to further diminish the likelihood of an acute coronary event.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  O C Raffel, T Akasaka, and I-K Jang Cardiac optical coherence tomography Heart, September 1, 2008; 94(9): 1200 - 1210. [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Honda and P. J. Fitzgerald Frontiers in Intravascular Imaging Technologies Circulation, April 15, 2008; 117(15): 2024 - 2037. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Di Mario and P. Barlis Optical Coherence Tomography: A New Tool to Detect Tissue Coverage in Drug-Eluting Stents J. Am. Coll. Cardiol. Intv., April 1, 2008; 1(2): 174 - 175. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gossl, L. O. Lerman, and A. Lerman Frontiers in Nephrology: Early Atherosclerosis A View Beyond the Lumen J. Am. Soc. Nephrol., November 1, 2007; 18(11): 2836 - 2842. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kubo, T. Imanishi, S. Takarada, A. Kuroi, S. Ueno, T. Yamano, T. Tanimoto, Y. Matsuo, T. Masho, H. Kitabata, et al. Assessment of Culprit Lesion Morphology in Acute Myocardial Infarction: Ability of Optical Coherence Tomography Compared With Intravascular Ultrasound and Coronary Angioscopy J. Am. Coll. Cardiol., September 4, 2007; 50(10): 933 - 939. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. W. Serruys, H. M. Garcia-Garcia, and E. Regar From Postmortem Characterization to the In Vivo Detection of Thin-Capped Fibroatheromas: The Missing Link Toward Percutaneous Treatment: What If Diogenes Would Have Found What He Was Looking For? J. Am. Coll. Cardiol., September 4, 2007; 50(10): 950 - 952. [Full Text] [PDF]
|
|
|
|
|
|
S. K. Nadkarni, M. C. Pierce, B. H. Park, J. F. de Boer, P. Whittaker, B. E. Bouma, J. E. Bressner, E. Halpern, S. L. Houser, and G. J. Tearney Measurement of Collagen and Smooth Muscle Cell Content in Atherosclerotic Plaques Using Polarization-Sensitive Optical Coherence Tomography J. Am. Coll. Cardiol., April 3, 2007; 49(13): 1474 - 1481. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Amirbekian, M. J. Lipinski, K. C. Briley-Saebo, S. Amirbekian, J. G. S. Aguinaldo, D. B. Weinreb, E. Vucic, J. C. Frias, F. Hyafil, V. Mani, et al. Detecting and assessing macrophages in vivo to evaluate atherosclerosis noninvasively using molecular MRI PNAS, January 16, 2007; 104(3): 961 - 966. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Waxman, F. Ishibashi, and J. E. Muller Detection and Treatment of Vulnerable Plaques and Vulnerable Patients: Novel Approaches to Prevention of Coronary Events Circulation, November 28, 2006; 114(22): 2390 - 2411. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Hosokawa, N. Kambara, M. Ohba, T. Mukai, M. Ogawa, H. Motomura, N. Kume, H. Saji, T. Kita, and R. Nohara A Catheter-Based Intravascular Radiation Detector of Vulnerable Plaques J. Nucl. Med., May 1, 2006; 47(5): 863 - 867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Stamper, N. J. Weissman, and M. Brezinski Plaque Characterization With Optical Coherence Tomography. J. Am. Coll. Cardiol., April 18, 2006; 47(8S): C69 - C79. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Sampson, M. S. Winterbone, J. C. Hughes, N. Dozio, and D. A. Hughes Monocyte Telomere Shortening and Oxidative DNA Damage in Type 2 Diabetes Diabetes Care, February 1, 2006; 29(2): 283 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Toutouzas, M. Drakopoulou, J. Mitropoulos, E. Tsiamis, S. Vaina, M. Vavuranakis, V. Markou, E. Bosinakou, and C. Stefanadis Elevated Plaque Temperature in Non-Culprit De Novo Atheromatous Lesions of Patients With Acute Coronary Syndromes J. Am. Coll. Cardiol., January 17, 2006; 47(2): 301 - 306. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Ginsburg, M. P. Donahue, and L. K. Newby Prospects for Personalized Cardiovascular Medicine: The Impact of Genomics J. Am. Coll. Cardiol., November 1, 2005; 46(9): 1615 - 1627. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Nadkarni, B. E. Bouma, T. Helg, R. Chan, E. Halpern, A. Chau, M. S. Minsky, J. T. Motz, S. L. Houser, and G. J. Tearney Characterization of Atherosclerotic Plaques by Laser Speckle Imaging Circulation, August 9, 2005; 112(6): 885 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P.S. Dunphy, A. Freiman, S. M. Larson, and H. W. Strauss Association of Vascular 18F-FDG Uptake with Vascular Calcification J. Nucl. Med., August 1, 2005; 46(8): 1278 - 1284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Muanza, A. P. Cotrim, M. McAuliffe, A. L. Sowers, B. J. Baum, J. A. Cook, F. Feldchtein, P. Amazeen, C. N. Coleman, and J. B. Mitchell Evaluation of Radiation-Induced Oral Mucositis by Optical Coherence Tomography Clin. Cancer Res., July 15, 2005; 11(14): 5121 - 5127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Priebe Perioperative myocardial infarction--aetiology and prevention Br. J. Anaesth., July 1, 2005; 95(1): 3 - 19. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
red) at a site of disruption (arrow) within a lipid-rich plaque (LP). The * represents guide wire shadow artifact. (b) Optical coherence tomography (OCT) image of an LP. The * represents guide wire shadow artifact. A 500 µm scale bar is shown in the top right-hand corner. (c) Outline (red) of the segmented fibrous cap of the OCT image depicted in panel b. (d) Normalized standard deviation image superimposed over a standard intensity image shows locations (blue 
