Prior Cytomegalovirus Infection and the Risk of Restenosis After Percutaneous Transluminal Coronary Balloon Angioplasty
Background—Restenosis is a common problem after all revascularization procedures in atherosclerotic coronary arteries. Reactivated human cytomegalovirus (CMV) has been detected in tissues of restenotic vascular lesions and was hypothesized to be a contributing pathogenic factor. Recent data suggest an association of restenosis after optimal coronary atherectomy with CMV serostatus, and a possible role of antiviral therapy was discussed. We therefore tested the hypothesis that prior CMV infection might be a risk factor for restenosis after conventional coronary balloon angioplasty (PTCA).
Methods and Results—We analyzed 92 consecutive patients who had been admitted for control angiography after previous PTCA within a mean interval of 6 months. Anti-CMV antibodies were measured as an indicator of prior CMV infection and latency. The coronary angiograms before PTCA, directly after, and 6 months later were analyzed quantitatively. Sixty-five percent of the patients were CMV-positive. Before PTCA, the degree (mean±SD) of stenosis was 69±10% in CMV-positive and 68±8.3% in CMV-negative subjects. PTCA resulted in a residual stenosis of 39% in both groups. After 6 months, the late losses of luminal diameter in the CMV-positive and -negative groups were 11±13% and 12±15%, respectively (P=0.658). In an ANCOVA with 25 potential risk factors for restenosis, CMV serostatus was not significantly associated with restenosis development.
Conclusions—Our data indicate that prior CMV infection, in contrast to optimal atherectomy, is not associated with chronic restenosis after conventional coronary balloon angioplasty. The results do not support a possible benefit from antiviral therapy.
Percutaneous transluminal coronary balloon angioplasty (PTCA) is the most common interventional therapy of symptomatic coronary artery disease. However, restenosis occurs in 20% to 60% of patients within 6 months after angioplasty.1 2 3 Remodeling and smooth muscle cell hyperplasia are the main elements of restenosis development.4 Several complex interactions of cellular and extracellular factors that could contribute to restenosis have been identified in the past. Recently, as an additional possible pathogenetic factor, reactivated human cytomegalovirus (CMV) from smooth muscle cells cultured from restenotic lesions, was shown to inhibit cellular p53 by the viral immediate early antigen IE84,5 thus leading to reduced counterregulation of cellular proliferative factors. This possible viral effect to increase smooth muscle proliferation may consequently contribute to the development of vascular restenosis. It was suggested that mechanical irritation of stenotic lesions that latently harbor CMV by interventional procedures such as atherectomy or PTCA could lead to CMV reactivation.6 Indeed, a first clinical study showed that prior CMV infection (indicated by serum antibodies) of individuals undergoing coronary atherectomy was associated with a higher risk of restenosis.7 The patients for this study were recruited from the Optimal Atherectomy Restenosis Study (OARS) and had a mean residual stenosis of <10% directly after intervention. In comparison, conventional PTCA results in an average residual stenosis of ≈30% to 42%.8 9 10 Therefore, PTCA appears to be considerably less traumatic to coronary vessels and might be less likely to induce CMV reactivation. We conducted this study to evaluate a possible impact of prior CMV infection on the risk of restenosis after conventional percutaneous transluminal coronary balloon angioplasty.
From September to December 1996, serum samples and clinical data were collected from consecutive patients who were admitted to the cardiological unit in Düsseldorf. Patients who had control coronary angiography after previous PTCA and who fulfilled the inclusion and exclusion criteria were included in the study for angiographic evaluation. In the area of Düsseldorf, patients are assigned to one of several districts with a responsible cardiological center according to their place of residence. All patients undergoing PTCA or other invasive procedures are scheduled for follow-up angiography in the same center. In the case of uncomplicated PTCA, as in our study population, the follow-up interval is 6 months.
Inclusion criteria of the study population were interval between PTCA and control angiography ≥2 months and ≤12 months, minimal luminal diameter of stenotic vessel >0.2 mm, age 35 to 80 years, and no signs or symptoms of active infection. Exclusion criteria were total chronic occlusion of vessel; additional interventions such as atherectomy, laser ablation, stent implantation; or acute myocardial infarction. In patients with PTCA of 2 vessels in the same session (n=4), only 1 vessel (the larger one) was used in the analysis for statistical reasons.
To assess comparative data on the prevalence of prior CMV infection, blood samples were collected from all consecutive internal patients not included in the study who were admitted to the unit within the study period.
PTCA and Coronary Angiography
A computerized analysis of the cineangiograms (Cardio 500 system, Kontron Elektronik GmbH) was done without knowledge of CMV serostatus. All measurements were done twice before intervention, immediately after PTCA, and at follow-up after a mean of 6 months. The following parameters were measured: minimal luminal diameter, interpolated reference diameter, luminal area of stenotic region, interpolated reference area of stenotic region, and length and eccentricity of atheroma. Stenosis is defined as reference diameter minus minimal luminal diameter divided by the reference diameter expressed as percent, early gain as stenosis before PTCA minus stenosis immediately after PTCA, late loss as stenosis at follow-up minus stenosis directly after PTCA, effective gain as stenosis before PTCA minus stenosis at follow-up, and loss ratio as late loss divided by early gain expressed as percent.
Blood Samples and Laboratory Analysis
After informed consent had been obtained, blood samples were drawn from peripheral veins, centrifuged, frozen at −75°C, and collected until analysis in a parallel batch. Anti-CMV IgG was determined by an ELISA kit (Behring), with whole virus cell lysates as antigen. A serum was defined as positive for CMV antibodies if the optical density at 405 nm exceeded 0.2 after subtraction of the reactivity with the control antigen of uninfected cell lysates. Quantitative analysis was done by the α-method according to the manufacturer’s instructions. The analysis was performed blinded with a numerical sample identification code.
χ2 test and Fisher’s exact test in case of <5 counts per cell were used for frequency counts. Differences between means of continuous data were tested with Student’s t test. ANCOVA was performed on loss ratio with CMV serostatus and potential risk factors for restenosis. Parameter estimates of a linear model of CMV serostatus and 25 potential risk factors with probability value are reported. For correlation analysis, Pearson coefficients were calculated. All computations were done with the SAS/STAT Software (SAS Institute Inc).
Basic Patient Data and CMV Serostatus
Blood samples and complete clinical data from 124 patients who were admitted for control angiography after previous PTCA were obtained. Ninety-two patients met the inclusion and exclusion criteria. Four hundred seventy-two patients served as controls for CMV prevalence. CMV seropositivity was detected in 65% of both patient groups. The mean titer of the CMV-seropositive subjects was not statistically different in the study population and the control group (1:18 900 versus 1:15 700, respectively). In the study population, CMV-seropositive and CMV-seronegative subjects were well matched for basic patient data, cardiovascular history, risk factors for coronary artery disease, time interval between PTCA and control angiography, and concomitant cardiovascular medication. There was a difference of 4.2 years in mean age between the groups (Table 1⇓). In later analysis, however, there was no association of age with restenosis development.
Factors With Potential Influence on Restenosis
Control coronary angiography was performed after a mean interval of 6 months post PTCA. As shown in Figure 1⇓, there was no correlation between time interval and degree of restenosis (late loss: r=0.041, P=0.697). If the subgroups of patients with a time interval between 60 and 120 days and >120 days are analyzed separately, the proportion of patients with vessel renarrowing (75% and 78%, respectively) and the degree of renarrowing (late loss, 14.4% and 10.8%; loss ratio, 40.9% and 36.3%, respectively) is statistically not different.
The occurrence and location of former myocardial infarctions, the cumulative number of angioplasties in each patient or vessel, and the location of stenosis were statistically not different between the 2 groups. Also, prevalent risk factors for coronary artery disease and concomitant medication were evenly distributed (Table 1⇑).
There is a high prevalence of smoking (as a main risk factor of coronary artery disease) in the study population: 47% in CMV-seronegative and 43% in CMV-seropositive patients. A separate analysis of smoking, however, revealed no influence on the development of restenosis.
The angiographic findings and calculated parameters at baseline, immediately after PTCA, and at the 6-month control are listed in Table 2⇓. There were no differences between subjects with and those without prior CMV infection. Both groups exhibited an equal distribution of minimal luminal diameter at all time points of coronary angiography (Figure 2⇓). The dilatated region at the time of control angiography showed a late loss of >10% of the reference diameter in 16 CMV-negative (53%) and 30 CMV-positive (50%) subjects. Additional parameters, such as area, eccentricity, and length of atheroma, also did not differ between the groups. There was no difference between the subgroups with regard to the initial success of angioplasty (early gain), the late success (effective gain), and the renarrowing degree at control (late loss, loss ratio) by t test (Figure 3⇓). A power analysis was performed on the parameters of restenosis. If the degree of restenosis quantified by loss ratio is assumed to be twice as high in CMV-seropositive patients as in CMV-seronegative patients, as observed in a former study (67% versus 33%, respectively),7 the power is 0.988 with α=0.05 for the 2-sided test. If, because of a higher loss ratio in the CMV-seronegative population of our patients (41%), the effect of CMV seropositivity is assumed to be 33% less, the power is 0.847. For late loss as parameter of restenosis, the respective power values are 0.994 and 0.870.
Potential Risk Factors for Restenosis
To evaluate the impact of potential risk factors for restenosis development, an ANCOVA was performed (Table 3⇓). The analysis of a multiparameter model showed that CMV serostatus is not associated with the degree of vessel narrowing (loss ratio) after PTCA.
Two parameters were correlated with loss ratio: concomitant medication of nitrates negatively and ACE inhibitors positively. However, the frequency of taking nitrates was identical in CMV-seropositive and -negative subjects, whereas a trend of more frequent ACE inhibitor medication in CMV-seropositive subjects was observed. In the subgroup of CMV-seropositive patients, no significant correlation (r=−0.128, P=0.331) between degree of restenosis and CMV antibody titer could be detected (Figure 4⇓).
Restenosis after percutaneous transluminal coronary angioplasty is a common problem with high impact on secondary coronary morbidity. Despite considerable efforts, the results of various pharmacological approaches for prevention of restenosis after angioplasty are unsatisfactory.11
On the basis of clinical, pathological, and experimental data, it was suggested that reactivation of latent human CMV in blood vessels may contribute to development of coronary artery disease12 13 and restenosis after directional coronary atherectomy.6 7 Therefore, if CMV is a contributory factor for restenosis, systemic antiviral treatment could be a new therapeutic option for the reduction of late failure after coronary interventions.
Patients with prior CMV infection harbor latent virus in atherosclerotic coronary vessels.14 If CMV reactivation occurs and contributes to restenosis development as a consequence of mechanical irritation of vascular walls, CMV-seropositive patients should have a higher frequency or higher degree of renarrowing after vessel reopening procedures. Data from a previous study7 suggest that prior CMV infection is an independent risk factor for restenosis after optimal atherectomy. However, as presented in this study, prior CMV infection is no risk factor for restenosis after conventional balloon angioplasty. The most likely reasons for this discrepancy are the different methods used and the extent of vessel reopening achieved by each procedure. In our study, only patients with conventional balloon angioplasty were investigated. This procedure resulted in a residual stenosis of 39% immediately after angioplasty (mean immediate gain, 0.8 mm). In contrast, optimal atherectomy was performed by intensive mechanical plaque resection followed by additional balloon dilatation with a residual stenosis of 7% (mean immediate gain, 1.88 mm). Therefore, optimal atherectomy appears to produce considerably greater injury of vascular tissue and may induce more intensive cellular repair mechanisms, thus possibly leading to better conditions for CMV reactivation.6 The greater degree of vessel narrowing (late loss) in the atherectomy study compared with our patients (CMV-positive subjects, 37% versus 10%; CMV-negative subjects, 18% versus 12%, respectively) supports this concept. However, this hypothesis has to be supported by further experimental evidence.
The average time needed for full development of restenosis is ≈4 to 6 months. We therefore readmit all patients for control angiography after 4 to 8 months. Some patients who develop restenosis after initial successful PTCA may be readmitted for control angiography earlier than scheduled, mainly because of symptoms. We therefore also included patients with a time interval of 60 to 120 days in the study. The analysis showed that CMV status as well as proportion and degree of renarrowing did not differ between these subgroups and that the prevalence of CMV antibodies in the study group (65%) was equivalent to that in the control group (65%). Therefore, a bias with significant impact on the study results due to an incongruent study population seems unlikely in this setting.
The influence of CMV serostatus on the extent of restenosis was further investigated in a model of covariance. There was no association with CMV seropositivity. Furthermore, in the subgroup of patients with prior CMV infection, no correlation with the titer of anti-CMV antibodies could be detected.
In conclusion, in our study population, prior CMV infection is no risk factor for restenosis after PTCA. Therefore, in contrast to optimal atherectomy, the data do not support a possible benefit from specific antiviral therapy against human CMV to prevent chronic restenosis after conventional coronary balloon angioplasty.
This work is dedicated to Prof Dr Georg Strohmeyer on the occasion of his 70th birthday.
- Received June 1, 1998.
- Revision received November 30, 1998.
- Accepted December 18, 1998.
- Copyright © 1999 by American Heart Association
Leimgruber PP, Roubin GS, Hollman J, Cotsonis GA, Meier B, Douglas JS, King SB Jr, Gruentzig AR. Restenosis after successful coronary angioplasty in patients with single-vessel disease. Circulation. 1986;73:710–717.
Serruys PW, Luijten HE, Beatt KJ, Geuskens R, de Feyter PJ, van den Brand M, Reiber JH, ten Katen HJ, van Es GA, Hugenholtz PG. Incidence of restenosis after successful coronary angioplasty: a time-related phenomenon: a quantitative angiographic study in 342 consecutive patients at 1, 2, 3, and 4 months. Circulation. 1988;77:361–371.
Speir E, Modali R, Huang ES, Leon MB, Shawl F, Finkel T, Epstein SE. Potential role of human cytomegalovirus and p53 interaction in coronary restenosis. Science. 1994;265:391–394.
Epstein SE, Speir E, Zhou YF, Guetta E, Leon M, Finkel T. The role of infection in restenosis and atherosclerosis: focus on cytomegalovirus. Lancet. 1996;348:s13–s17.
Serruys PW, de Jaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, Emanuelsson H, Marco J, Legrand V, Materne P, Benestent Study Group. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med. 1994;331:489–495.
Melnick JL, Adam E, Debakey ME. Cytomegalovirus and atherosclerosis. Eur Heart J. 1993;14:30–38.