|
|
||||||||||
|
J Am Coll Cardiol, 2007; 49:2264-2271, doi:10.1016/j.jacc.2007.03.026
(Published online 24 May 2007). © 2007 by the American College of Cardiology Foundation |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
First Department of Cardiology, Hippokration Hospital, University of Athens, Greece.
Manuscript received November 20, 2006; revised manuscript received January 29, 2007, accepted March 6, 2007.
* Reprint requests and correspondence: Dr. Konstantinos Toutouzas, 26 Karaoli and Dimitriou str., Holargos, 15562 Athens, Greece. (Email: ktoutouz{at}otenet.gr).
| Abstract |
|---|
|
|
|---|
Background: Intravascular ultrasound (IVUS) provides morphologic assessment and intracoronary thermography (ICT) evaluates the local inflammatory activation of CL.
Methods: Eighty-one consecutive patients, 48 with ACS and 33 with CSA, were enrolled. Ratio of lesion to reference external elastic membrane area, indicated by IVUS, was defined as positive remodeling index (pRi) (
1) or negative remodeling index (nRi) (<1). We also investigated the existence of ruptured plaque (rp) in the CL. By ICT temperature difference (
T) between the CL and the proximal vessel wall was measured.
Results: Patients with ACS had greater remodeling index than patients with CSA (1.15 ± 0.18 vs. 0.90 ± 0.12; p < 0.01), as well as increased
T (0.08 ± 0.03°C vs. 0.04 ± 0.02°C; p < 0.01). Patients with pRi had higher
T than patients with nRi (0.07 ± 0.03°C vs. 0.04 ± 0.02°C; p < 0.001). In patients with nRi there was no difference in
T between ACS and CSA (p = 0.22). Patients with rp had increased
T compared with patients without rp (0.09 ± 0.03°C vs. 0.05 ± 0.02°C; p < 0.01). Multivariate analysis showed that
T was independently correlated with the presence of rp, pRi, and ACS.
Conclusions: The present study showed that culprit lesions with plaque rupture and positive arterial remodeling have increased thermal heterogeneity, although in certain patients a discrepancy between morphogic and functional characteristics was observed. A combination of morphologic and functional examination may offer additional diagnostic and prognostic information.
| |||||||||||||||
Certain risk factors have been related to vascular remodeling. Decreased adaptive or even constrictive remodeling has been related to hypertension (6,7), decreased high-density lipoprotein levels (7), age (8,9), and insulin-treated diabetes mellitus (10,11). In contrast, insulin resistance in nondiabetics is associated with expansive remodeling (12). Interestingly, a constrictive arterial wall remodeling is observed during plaque-stabilizing therapy with statin medications (13).
Studies of human autopsy specimens investigated the correlation between local inflammatory activation with arterial remodeling. Indeed, increased lipid core size, calcium, macrophage, and matrix metalloproteinases content were associated with expansive arterial remodeling (1416). Animal studies have confirmed this to be mediated through interleukin-6 (17), matrix metalloproteinases (18), and P-selectin (19,20). Furthermore, human systemic inflammatory mediators, such as C-reactive protein, interleukin-6, soluble vascular and intercellular adhesion molecules 1, are related to culprit artery plaque size and arterial remodeling (21,22). However, there are no in vivo human studies correlating the local temperature with the morphologic characteristics, including arterial remodeling and plaque rupture.
Both experimental and human studies demonstrated that plaque temperature is correlated with increased macrophage content (2325), as well as systemic inflammatory indexes (26,27). Thus, intracoronary thermography (ICT) provides in vivo evidence of local inflammatory activation in coronary lesions (28). The aim of the present study was to investigate a possible correlation between certain morphologic characteristics and observed temperature of culprit lesions in patients with symptomatic coronary artery disease (CAD) suffering from CSA or ACS.
| Methods |
|---|
|
|
|---|
20 mm in length and proximal reference vessel diameter
2.5 mm. Patients with multivessel CAD of inflammatory or neoplastic condition were excluded from the study, as were patients treated with corticosteroids or nonsteroidal antiinflammatory drugs, except for aspirin. The institutional ethics committee approved the study protocol, and each patient provided written informed consent. Chronic stable angina was defined as angina without change in frequency, duration, or intensity of symptoms over the last 6 weeks before the study. In patients with CSA, ischemia had to be documented with noninvasive techniques. Patients suffering from unstable angina, defined as new onset of severe angina, accelerated angina, or angina at rest, were included in the ACS group. Patients with recent myocardial infarction (<1 month) were not included in the study.
ICT catheter. The technical characteristics of the thermistor and the data acquisition and processing system have been previously described (2931). Briefly, the coronary thermography catheter contains 2 lumens: The first runs through the distal 20 cm of the device and is used for insertion of a guidewire (0.014-inch), which is advanced by a monorail system. The thermistor is positioned at the distal part of the thermography catheter. The ICT catheter used was 3-, 3.5-, or 4-F in diameter, depending on the size of the vessel.
IVUS imaging. The IVUS catheter (Volcano Corp., Rancho Cordova, California) has a synthetic aperture array electronic transducer design with a 20-MHz operating frequency. The images were stored on a compact disc for later review. Electrocardiographic tracings were continuously displayed on the console screen during the study.
Procedure. All patients underwent quantitative coronary angiography. The lesion of interest was well delineated in 2 views, on which the positioning of the catheters was based.
IVUS Examination
The IVUS imaging catheter was inserted and advanced along the guidewire as distally as possible, using fluoroscopic guidance. With automatic pullback (0.5 mm/s), ultrasound images were obtained and recorded for quantitative data analysis.
ICT Procedure
Five minutes after the last injection of contrast medium, the ICT catheter was advanced through the guidewire. Baseline temperature was obtained when the thermistor had just emerged from the tip of the guiding catheter without being in contact with the vessel wall. Thereafter, temperature was recorded at the proximal nondiseased segment and at the atherosclerotic lesions, being confirmed by IVUS (26,32,33). The fluoroscopic images of target lesion and proximal nondiseased segment were frozen during IVUS catheter pullback and guided the ICT procedure. Temperature difference (
T) was calculated by subtracting the mean temperature of proximal vessel wall from the maximal temperature at the lesion site. Afterward, percutaneous coronary intervention was performed.
IVUS analysis.
After the procedure, off-line analysis of the recorded images was performed blinded to the clinical characteristics and the temperature measurements by experienced investigators. Quantitative measurements were obtained from the frames at the end-diastolic phase. Cross-sectional ultrasound measurements were performed at the lesion site, which was the image slice with the smallest lumen area (LA); if there were several image slices with an equally small lumen, the one with the largest external elastic membrane area (EEMA) and plaque area (PA) was analyzed. Cross-sectional ultrasound images in proximal healthy reference segments within 10 mm but before any major side branch were also measured. The PA was calculated as EEMA minus LA. Stenosis (%) was calculated as the ratio of PA to EEMA. Remodeling index was defined as the ratio of EEMAL (lesion) to EEMAR (reference). Two patterns of Ri were recognized: positive remodeling index (pRi) and negative remodeling index (nRi) were defined as indices of
1 and <1, respectively.
We assigned plaques as ruptured when containing a cavity communicating with the lumen via an overlying residual fibrous cap fragment. A fissure without a cavity communicating with the true lumen was not included in the analysis. Rupture sites separated by a length of artery containing a smooth lumen contour and no cavity were considered to represent different plaque ruptures (4,5).
Statistical analysis.
Continuous variables are presented as mean ± standard deviation, and categoric variables are presented as absolute and relative frequencies. The presented figures show median values and interquartile range owing to the skewed distribution of
T. Comparisons of
T between groups of study were performed with nonparametric statistical methods (Mann-Whitney criterion) owing to the small sample and the skewed distribution of the investigated indices. Categoric variables were compared using the chi-square test and continuous variables using the Mann-Whitney test (independent comparison). Correlations between temperature and IVUS measurements were evaluated by the use of Spearman correlation coefficient. The effect of the interaction between Ri and clinical syndrome on log-transformed temperature differences was tested by the application of the analysis of covariance. The correlation of
T, as a continuous variable, with all variables was tested by linear multiple regression analysis. A stepwise backward elimination procedure evaluated the association between
T and age, gender, smoking, hypertension, hypercholesterolemia, diabetes mellitus, family history of coronary artery disease, plaque rupture, Ri, use of statins, and clinical syndrome. The cut-off for the p value for entering a variable in the model was set to 0.05, and the cut-off for removing a variable from the model was 0.10. Results from multiple regression analysis are presented as standardized beta coefficients. All reported p values are exact, based on 2-sided tests and compared with a significance level of 5%. Stata 6 software was used for the calculations (Stata Corp., College Station, Texas).
| Results |
|---|
|
|
|---|
|
|
|
|
|
T compared with CSA (0.08 ± 0.03°C vs. 0.04 ± 0.02°C; p < 0.01) (Fig. 2).
|
T compared with those with nRi (0.07 ± 0.03°C vs. 0.04 ± 0.02°C; p < 0.01) (Fig. 2). Among patients with ACS, pRi showed higher
T than nRi (p < 0.01), whereas no statistical significance was achieved among patients with CSA (p = 0.53) (Fig. 3, Table 5).
|
|
T than those with CSA (p < 0.01). In patients with nRi there was no difference in
T between ACS and CSA (p = 0.22) (Table 5). A good correlation was found between
T and Ri (p < 0.01; r = 0.59) (Fig. 4).
|
T compared with nonruptured (0.09 ± 0.03°C vs. 0.05 ± 0.02°C; p < 0.01). Among patients with CSA, those with ruptured plaque had higher
T than those without (p < 0.01). In the subgroups of patients with ruptured and nonruptured plaques,
T was increased in the presence of ACS compared to CSA (p < 0.01) (Table 6, Fig. 5).
|
|
T was significantly associated with the presence of ruptured plaque (ß = 0.48; p < 0.001), positive remodeling (ß = 0.18; p = 0.03), and presence of ACS (ß = 0.34; p < 0.001). Moreover, regarding the other variables entered in the initial model,
T was borderline associated with age (ß = 0.19; p = 0.03) and male gender (ß = 0.16; p = 0.08). | Discussion |
|---|
|
|
|---|
T. We also confirmed previously reported impact of clinical syndrome in local temperature measurements (30,34,35). Moreover, we observed that in patients suffering from CSA there is no difference in
T between lesions with positive and negative Ri. Several factors have been found to be associated with positive arterial remodeling, primarily, the clinical syndrome, because positive remodeling is found more frequently in patients with ACS and negative remodeling in patients with CSA (3).
Several variables affect the process of remodeling, including hypertension (6,7), hypercholesterolemia (7), age (8,9), diabetes mellitus (10,11), and statin treatment (13). In the present study, after stratification of the population by the clinical syndrome, all of the mentioned variables were similar between the compared groups.
A significant correlation between inflammation and arterial remodeling has been recognized in several autopsy and animal studies. Enhanced macrophage infiltration in human atheromatic plaques is associated with positive arterial remodeling (1416). Experimental animal studies showed this to be mediated through inflammatory indexes including interleukin-6 (17), matrix metalloproteinases (18), and P-selectin (19,20). In vivo human studies, however, have not investigated this correlation, except for recent studies exploring the linkage between systemic C-reactive protein and arterial remodeling at the culprit lesions of patients with acute myocardial infarction (21,22).
Local inflammatory activation was evaluated by ICT, because that is the only method currently available that can be used in vivo to measure heat generation from atheromatic plaques indicating local inflammation (28). Previous ex vivo and in vivo studies have shown that plaque temperature reflects the pathologic substrate of the lesion (2325,36,37) and moreover is correlated with systemic inflammatory indexes (26,30). Inflamed plaques, with greater percentage of macrophages, have higher thermal heterogeneity.
The results of the present study confirmed the correlation between increased functional characteristics, adaptive arterial compensatory enlargement, and distinct morphologic characteristics of the culprit lesion. This is clearly demonstrated, first, by the correlation of remodeling index with
T, second, by the increased
T of ruptured plaques, and, finally, by the results of the multivariate analysis, which indicated the independent correlation between the 2 latter variables and
T. In the multivariate analysis we did not confirm previous observations regarding the effect of statins on
T, possibly owing to the small number of patients receiving statins in the present study. We did not correlate other qualitative characteristics of the atheromatic plaques with
T, because the sensitivity of IVUS for tissue characterization, known to be related to plaque vulnerability, seems to be limited compared with other new invasive methods. The results of the present study justify the design of a study for the investigation of a possible correlation between temperature measurements and qualitative plaque characteristics evaluated by optical coherence tomography, which has an increased resolution even for quantification of macrophage concentration in the cap of atherosclerotic lesions (38).
It should be emphasized, however, that in certain patients a discrepancy was detected between the morphologic characteristics and plaque temperature. Interestingly, in patients with CSA
T was similar between patients with positive and negative Ri. This discrepancy was also found in patients with nRi, in which
T was similar between ACS and CSA groups. These observations suggest that local inflammatory activation, as detected by ICT, is not the only pathophysiologic substrate involved in ACS and positive remodeling, and the specific morphologic characteristics, as detected by IVUS, may not reveal all "high-risk" plaques. Therefore, the combined use of ICT measurements and pRi for the identification of vulnerable plaques warrants further investigation.
Clinical implications. Previous studies have shown that arterial remodeling (39) and thermal heterogeneity (35) are prognostic factors for long-term outcome after percutaneous coronary interventions. The combination of both morphologic and functional characteristics may provide more accurate risk stratification after percutaneous coronary intervention, especially in the era of drug-eluting stents. Patients with negative remodeling and lower thermal heterogeneity may have lower risk for an event. Especially for patients presenting with ACS, the role of plaque erosion and not rupture should be further investigated, because erosion is implicated in a minority of patients with ACS. In contrast, patients suffering from ACS with increased heat production and positive remodeling may require an aggressive interventional and pharmaceutic treatment. All these issues need to be clarified in a prospective study.
Study limitations.
We included only patients with 1-vessel disease and significant stenosis, as confirmed by IVUS, at the culprit lesions. A minimal lumen area of 5.7 mm2 was found in ACS, possibly owing to the positive remodeling observed in these patients, although the stenosis was similar to that in the CSA group. Significant lesions were studied to avoid the "cooling effect" of blood flow on plaque temperature measurements (31,40) and to ensure contact of the thermistor to the target lesion. Moreover, different-diameter coronary thermography catheters were used, depending on the diameter of the target coronary artery. We therefore cannot extrapolate the present results to intermediate lesions and patients with multivessel disease. Furthermore, there was no difference in stenosis between the ACS and CSA groups. Therefore, the difference observed in
T between the 2 groups confirmed previous observations (29,30) and cannot be attributed to the "cooling effect" of blood flow.
Earlier studies (5,41) have demonstrated that patients with ACS have diffuse vulnerability and multiple plaque ruptures even proximal to the target lesion. Although proximal ruptured sites would only underestimate the present results, because the temperature of the proximal vessel wall would be increased, we included patients with distinct lesions, and the proximal vessel wall was free of rupture as evaluated by IVUS.
Another known factor influencing thermal heterogeneity is the administration of statins (26,33,42). We did not find an effect of statins on
T, because, first, a limited number of patients were under statin therapy in the whole study group and, second, the percentage of treated patients was similar between the studied groups. Finally, these findings were observed only in culprit lesions producing significant stenosis. Whether this correlation between certain morphologic and functional characteristics is also found in nonculprit lesions needs further investigation.
| Conclusions |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
O. C. Raffel, F. M. Merchant, G. J. Tearney, S. Chia, D. D. Gauthier, E. Pomerantsev, K. Mizuno, B. E. Bouma, and I.-K. Jang In vivo association between positive coronary artery remodelling and coronary plaque characteristics assessed by intravascular optical coherence tomography Eur. Heart J., July 2, 2008; 29(14): 1721 - 1728. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. V. Finn, G. Nakazawa, J. Narula, and R. Virmani Culprit Plaque in Myocardial Infarction: Going Beyond Angiography J. Am. Coll. Cardiol., December 4, 2007; 50(23): 2204 - 2206. [Full Text] [PDF] |
||||
![]() |
P. Schoenhagen Plaque Temperature, Arterial Remodeling, and Inflammation: Understanding "Hot-Spots" in the Coronary Arteries J. Am. Coll. Cardiol., June 12, 2007; 49(23): 2272 - 2273. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |