|
|
| |||||||||||||||||||||||||||||||
Background Neointimal proliferation leading to restenosis frequently develops after coronary angioplasty. This process is associated with a change in vascular smooth-muscle cells from a contractile (quiescent) phenotype to a synthetic or proliferating (activated) one. We investigated whether the presence of activated smooth-muscle cells in coronary lesions at the time of coronary atherectomy predisposes patients to subsequent restenosis.
Methods We used in situ hybridization to study the expression of messenger RNA in coronary-atherectomy specimens from 20 patients. Plaque material was hybridized with a probe for the B isoform of human nonmuscle myosin heavy chain, a major nonmuscle myosin isoform in activated, but not quiescent, smooth-muscle cells. Angiographic follow-up data were obtained a mean (±SD) of 174 ±54 days after atherectomy in 16 of the 20 patients, and the extent of recurrent luminal narrowing was analyzed quantitatively. The presence of restenosis was assessed by exercise thallium scintigraphy in the other four patients.
Results Atherectomy specimens from 10 of the 20 patients showed hybridization with the probe, defined as the clustering of more than 20 silver grains per cell nucleus in more than 10 nuclei in five high-power fields (x250); specimens from the other 10 patients showed no such hybridization. At follow-up, restenosis had developed in 8 of the 10 patients with positive hybridization results, but was absent in 9 of the 10 patients with negative results (P = 0.007). The degree of late loss in luminal diameter was significantly higher in patients with positive hybridization results than in those with negative results (ratio of late loss to immediate gain after atherectomy, 0.76 ±0.3 vs. 0.36 ±0.3; P<0.001).
Conclusions We conclude that the expression of the B isoform of nonmuscle myosin heavy chain is increased in some coronary atherosclerotic plaques and that this increase in expression identifies a group of lesions at high risk for restenosis after atherectomy.
A number of studies have demonstrated that the hallmark of restenosis is the abundant proliferation of neointimal smooth-muscle cells3,4. Similar pathological findings have been described in coronary-artery bypass grafts5 and in small arteries in patients with longstanding hypertension6. Although there is some degree of smooth-muscle proliferation after virtually any injury (such as angioplasty), less than half of patients have sufficient proliferation to create the clinically important narrowing of the vessel known as restenosis. A number of factors may thus be operating simultaneously to explain the apparently random variability between patients. One such factor could be that the proliferative potential of various atherosclerotic plaques differs and that clinically inapparent differences in plaque composition at the time of angioplasty (or other coronary intervention) may influence the subsequent clinical course.
Smooth-muscle cells have two different phenotypes7. The contractile, or quiescent, phenotype is characterized by the presence of typical smooth-muscle contractile proteins (such as alpha-actin and smooth-muscle myosin), well-developed thick filaments, contraction in response to chemical and mechanical stimuli, and the inability to undergo cytokinesis. The synthetic, or activated, phenotype, on the other hand, is characterized by a loss of contractile function, decline in levels of contractile proteins, and greatly increased synthetic and proliferative capacity. Smooth-muscle cells found in restenotic lesions appear to have the synthetic phenotype8. We reasoned that the presence of these activated smooth-muscle cells in atherosclerotic lesions at the time of atherectomy might predispose patients to higher rates of subsequent growth and proliferation and hence to the more likely development of restenosis.
The advent of directional atherectomy9 has for the first time allowed direct access to pathologic intravascular material in vivo. Earlier results of histologic analysis of atherectomy tissues10 demonstrated the presence of sheets of proliferating smooth-muscle cells in one third of coronary plaques. The histologic appearance of these cells is similar to that of proliferative smooth-muscle cells found in restenotic lesions. Using in situ hybridization analysis of atherectomy specimens, we previously demonstrated that proliferating smooth-muscle cells in restenotic vascular lesions express an isoform of nonmuscle myosin heavy chain that does not appear to be present in normal vascular smooth muscle11. In a number of animal models, the same B isoform is expressed in the neointima after arterial injury8,12,13. These findings suggest that the B isoform of nonmuscle myosin may be intimately involved in the process of smooth-muscle proliferation and restenosis. We therefore undertook the present study to characterize further the cells in primary atherosclerotic plaques with regard to their expression of the messenger RNA (mRNA) for the B isoform of nonmuscle myosin, and to correlate its appearance with the subsequent development of restenosis after atherectomy.
Methods
Coronary Atherectomy
Coronary-atherectomy specimens were obtained in a prospective manner from 20 patients (chosen without any systematic selection criteria from all patients undergoing coronary atherectomy at Beth Israel Hospital between February 16, 1990, and December 9, 1991) with the Simpson AtheroCath (Devices for Vascular Interventions, Redwood, Calif.), as previously described10. The decision to perform atherectomy was based strictly on anatomical and technical factors and was not influenced by the study. Of the 20 atherectomy samples obtained in this study, 17 were from left anterior descending coronary arteries and 3 from saphenous-vein grafts. Seventeen lesions were primary (not previously treated; 15 from left anterior descending coronary arteries and 2 from saphenous-vein grafts), and 3 were restenotic (2 from left anterior descending arteries and 1 from a saphenous-vein graft). A single lesion from each patient was used for this study. Approximately 20 mg of tissue was removed with each procedure; 10 mg was used for in situ hybridization, and 10 mg for light microscopy. The study protocol was approved by the Clinical Studies Review Board of Beth Israel Hospital, and informed consent was obtained from all the patients.
In Situ Hybridization
The tissues and probes were prepared as previously described11. All slides were reviewed by two observers who were unaware of the results of clinical and angiographic follow-up and the nature of the specimens (primary or restenotic). An average of two slides of each tissue specimen were hybridized with antisense probes (which are complementary to the mRNA), and two slides with sense probes (which are not complementary to the mRNA). The relation of silver grains to the cell nucleus was used to judge the strength of the signal. A slide was said to show hybridization if more than 10 nuclei had overlying clusters of silver grains (more than 20 grains per nucleus) in five high-power fields (x250)11. This criterion was established prospectively, before the results of clinical and angiographic follow-up became available. A measure of hybridization was also obtained by counting the number of positive nuclei per five high-power fields on each slide and averaging this number for the two slides that were prepared for each patient. Any disagreement between the observers was resolved by a joint review of slides.
Light Microscopy
A portion of every specimen was fixed in 10 percent formalin for light-microscopical examination. The presence or absence of intimal hyperplasia, defined as increased cellularity, was assessed for each paraformaldehyde-fixed specimen.
Angiographic Analysis
Coronary angiography was performed immediately before and after atherectomy. To assess their severity, lesions were measured with electronic calipers (Digital Caliper System, Sandhill Scientific, Littleton, Colo.) on optically magnified images of the view showing the greatest degree of stenosis, with a 9-French guiding catheter as a reference. Such electronic measurements are reproducible and correlate well with the results of computerized digital analysis14. A follow-up angiographic study was performed in 16 of the 20 patients and was analyzed in a similar manner. The remaining four patients declined follow-up angiography and were evaluated with serial exercise thallium tests. For all the patients the diameter of the lesion and that of the reference segment were recorded before and after atherectomy and at a follow-up study. Restenosis was defined as stenosis of more than 50 percent at the atherectomy site at the time of follow-up angiography. The amount of immediate gain was defined as the difference in the diameter of the lesion before and after atherectomy. The amount of late loss was defined as the absolute change in luminal diameter between the measurement after atherectomy and the follow-up study. The amount of late loss was adjusted for the immediate gain according to the method of Kuntz et al15.
Clinical Data
The medical records of all 20 patients were reviewed with particular attention to the clinical presentations that had led to their referral for atherectomy. The type of angina (resting or effort-induced) was noted, and the presence of various coronary risk factors was recorded. All the patients received aspirin, dipyridamole, and a calcium-channel blocker. No attempt was made to control the use of other cardiac medications or to modify cardiac risk factors such as smoking and hypercholesterolemia.
Statistical Analysis
All data are expressed as means ±SD. Differences were assessed with the use of unpaired two-tailed Student's t-tests, chi-square analysis, and linear regression, as appropriate. Differences were considered significant if P was less than 0.05.
Results
Characteristics of the Patients at Base Line
Of the 20 patients in the study (Table 1), 18 were men and 2 women; the average age was 57.9 ±10.9 years (range, 43 to 74). Eleven patients presented with new-onset angina at rest, and nine had effort-induced angina. In all the patients, atherectomy was chosen as the treatment because of eccentricity or the very proximal location of the lesion. The study atherectomy was the first intracoronary procedure for 17 patients; the other 3 were undergoing atherectomy for restenosis at the site of earlier angioplasty.
|
Directional atherectomy was technically successful in all the patients (Table 1), leaving in each case less than 20 percent residual luminal narrowing. The mean immediate gain in these patients was 2.4 ±0.57 mm (range, 1.43 to 3.34). Angiographic follow-up data were obtained in 16 patients (80 percent) a mean of 174 ±54 days later (range, 95 to 306). Four patients refused to undergo angiography again. Three of them were asymptomatic six months after atherectomy and, because of serial negative exercise thallium tests, were considered to have no restenosis. One patient again had angina three months after atherectomy, had a positive exercise test with thallium imaging that showed ischemia in the territory supplied by the treated artery, and was considered to have restenosis.
Of the 16 patients who underwent repeated angiography, 8 had stenosis of more than 50 percent at the site of the atherectomy. The average ratio of late loss to immediate gain (the "loss index") was 0.92 ±0.18 in the patients with restenosis and 0.30 ±0.14 in the patients without restenosis (P<0.001). Seven of the eight patients with restenosis underwent additional revascularization procedures. The eighth patient had only mild (53 percent) stenosis on follow-up angiography and was treated conservatively, without further symptoms. The remaining 12 patients (8 with no evidence of restenosis on angiographic follow-up, 3 with negative noninvasive evaluations, and 1 with a positive noninvasive evaluation) did not undergo further revascularization procedures.
In Situ Hybridization
The results of in situ hybridization with the nonmuscle myosin heavy-chain probe were considered to be positive in 10 patients. Figure 1 shows typical results in the case of restenosis, a positive primary lesion, and a negative primary lesion. As demonstrated by the intense hybridization signal with the myosin probe, the restenotic lesion was composed predominantly of activated smooth-muscle cells. The positive primary lesion also had several clusters of granules around cell nuclei, which the negative primary lesion did not. Controls with sense probes indicated that the background distribution of the label was similar in all cases.
|
Increased cellularity on light-microscopical examination (performed by an independent observer without knowledge of the in situ hybridization results) was noted in equal numbers of patients with positive and negative hybridization results (5 of 10 in each case), suggesting no correlation between the two observations.
Follow-up Angiography
Of the nine patients with positive hybridization results who underwent repeated angiography, seven had narrowing of more than 50 percent (Table 1), and all seven required further revascularization procedures. Only two patients with positive hybridization results (one with a primary lesion in the left anterior descending coronary artery and the other with a lesion in a saphenous-vein graft) were free of restenosis at follow-up angiography. The 10th patient with positive hybridization results had recurrent symptoms and a positive exercise thallium test (presumed to reflect restenosis) but declined further evaluation. In contrast, only 1 of the 10 patients with negative hybridization results had restenosis (of 53 percent, which is regarded as borderline); the remaining 9 did not (6 had negative angiograms and 3 had negative exercise thallium tests). Patients with positive in situ hybridization results tended to be more symptomatic after the atherectomy and on average underwent repeat catheterization earlier than the patients with negative results (152 ±45 vs. 194 ±53 days, P = 0.07). The number of positive nuclei in the atherectomy specimen also correlated with angiographic follow-up data, with patients who had more strongly hybridizing specimens having greater restenosis at follow-up (Figure 2).
|
|
We examined whether the presence of activated smooth-muscle cells in coronary atherosclerotic plaques influences the subsequent development of restenosis after atherectomy. In doing so we took advantage of the availability of tissue specimens of coronary lesions removed by directional atherectomy. To identify activated smooth-muscle cells in these specimens, we used the technique of in situ hybridization, a tool that allows the analysis of mRNA in very small amounts of tissue.
Nonmuscle myosin is a ubiquitous contractile protein present in all eukaryotic cells. A growth-related change from smooth-muscle to nonmuscle myosin isoforms has been shown to occur in vascular smooth-muscle cells in culture16 and in vivo13. Although nonmuscle myosin constitutes 10 to 20 percent of the total myosin protein in isolated primary smooth-muscle cells,16 no direct measurement is available for smooth-muscle cells in the intact vessel wall. Nonmuscle myosin is thought to be involved in a number of cellular processes, including cytokinesis,17,18 receptor capping,19 and secretion20. The protein is composed of two heavy chains and two pairs of light chains. At least two isoforms of heavy chains are known to exist in humans21 and other vertebrates22. The isoforms are differentially expressed in various tissues, suggesting different functions, although expression of the A isoform almost always predominates over that of the B isoform. The B isoform appears to be identical to the embryonal isoform of smooth-muscle myosin heavy chain (SMemb) cloned by Kuro-o et al.13. This protein is expressed during embryogenesis but does not appear to be present in appreciable amounts in adult vascular smooth-muscle cells. In a previous study using in situ hybridization, we were unable to detect the B isoform in normal smooth-muscle cells from the internal mammary artery11. The B isoform is expressed after arterial injury in a number of animal models,8,13 as well as in human restenotic lesions,11 which suggests that this isoform may be involved in the restenotic process.
The key part played by nonmuscle myosin is underscored by the observation that inhibiting its expression with antisense oligonucleotides inhibits the proliferation of smooth-muscle cells in vitro23. The presence of the B isoform of nonmuscle myosin does not suggest that the cell is actively proliferating. Rather, it is simply an indication that the cell has been activated and converted to a synthetic phenotype. In cell culture, the expression of this isoform can be detected long after cells have ceased to proliferate, presumably reflecting the incomplete restoration of the contractile phenotype. Indeed, Gordon et al.24 used the presence of proliferating cell nuclear antigen as a marker of active cell proliferation and found that only a small minority of the smooth-muscle cells in primary atherosclerotic plaques are replicating. Instead, the expression of nonmuscle myosin suggests that the cells were once activated and perhaps went through several cycles of proliferation before incompletely shutting down. This is similar to the incomplete restoration of the contractile phenotype of smooth-muscle cells that has been observed after arterial injury8.
In the present study, atherectomy of atherosclerotic plaques containing cells that were positive for the B isoform was more likely to be complicated by subsequent restenosis than was the removal of plaques that did not contain positive cells. This finding could not be attributed to differences in the size or length of the lesions, anatomical or technical factors, or clinical differences between the patients. Of the 10 patients with positive lesions, 7 had symptomatic clinical restenosis. In contrast, only 1 of the 10 patients with negative lesions had restenosis, and it was only borderline.
We observed a strong correlation between the presence of activated smooth-muscle cells in the original atherectomy specimen and the subsequent development of restenosis, defined as a narrowing of the lumen by more than 50 percent at the site of the procedure. The frequency of restenosis seen in this study (47 percent) is in accord with that reported for lesions of the proximal left anterior descending artery treated by atherectomy (38 percent25) or angioplasty (45 percent26). Restenosis, however, is a continuous rather than a dichotomous process, a feature not adequately reflected in the rather arbitrary definitions in current clinical use. Analysis of our data suggests that there is a linear relation between the presence of activated smooth-muscle cells and the luminal diameter some time after atherectomy, with patients who have more strongly positive hybridization results also having more severe restenosis. This point is further emphasized when one considers the loss index,15 the ratio that adjusts the late loss of luminal diameter seen on follow-up angiography for the amount of immediate gain obtained during the initial procedure. As expected, patients with a greater number of activated smooth-muscle cells in their lesions had significantly greater loss indexes as a measure of their proclivity for proliferation.
In summary, the presence of greater numbers of activated smooth-muscle cells in atherosclerotic plaques defines lesions that have greater proliferative potential and are more likely to produce clinically important restenosis after atherectomy. Further investigations of the biology of smooth-muscle-cell activation and recruitment in neointimal proliferation should provide a better understanding of this process.
Supported in part by an award (91004240) from the American Heart Association to Dr. Simons and grants (HL-32747 and HL-40518) from the National Institutes of Health to Dr. Isner. Dr. Simons is a Clinician-Scientist of the American Heart Association, and Dr. Leclerc is a Research Fellow of the Canadian Heart and Stroke Foundation.
We are indebted to Stuart J. Schnitt for review of light-microscopical results, to Cynthia Senerchia for patient follow-up, and to Robert D. Rosenberg and Richard E. Kuntz for helpful discussions.
Source Information
From the Charles A. Dana Research Institute and the Harvard-Thorndike Laboratory, Department of Medicine (Cardiovascular Division), Beth Israel Hospital and Harvard Medical School (M.S., R.D.S., D.S.B.); and the Division of Cardiology, St. Elizabeth's Hospital (G.L., J.M.I., L.W.) -- all in Boston. Presented in part at the 64th Scientific Sessions of the American Heart Association, Anaheim, Calif., November 11-14, 1991.
Address reprint requests to Dr. Simons at the Cardiovascular Division, Beth Israel Hospital, 330 Brookline Ave., Boston, MA 02215.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
