Coronary dilatation (CD) has an incidence of approximately 2% in the normal population and approximately 5% in the population with high atherosclerotic burden as demonstrated in the Coronary Artery Surgery Study registry (1). The incidence in heart transplant recipients is yet unknown. On the one hand, the term CD subsumes coronary aneurysms, which are defined as a localized dilatation that exceeds 1.5 times the diameter of the adjacent segment of an artery, and on the other hand, the term CD subsumes coronary ectasia, which is a diffuse dilatation involving more than 50% of the coronary artery. Types of CD are classified according to Markis et al. as diffuse dilatation with aneurysmal lesions in two vessels (type I), diffuse dilatation in one vessel and discrete dilatation in another vessel (type II), diffuse dilatation in one vessel (type III), and discrete dilatation in one vessel (type IV) (2, 3). The pathogenesis of the dilatation was believed to be closely associated with atherosclerosis. In the Coronary Artery Surgery Study registry, the coincidence of atherosclerosis and CD was found in up to 75% of all cases (2, 4). Yet, a number of authors did not interpret this as a relevant finding, because of the high incidence of atherosclerosis in the general population (2, 4, 5). Various facts support a different pathogenesis. The preferred localization in the right coronary artery (RCA), increased matrix metalloproteinase-3 concentration, and decreased endothelial-independent dilation suggest other causes (6), whereas increased occurrence in males and in particular of those with higher age and hyperlipidemia are consistent with a common pathogenesis (7, 8). The correlate to coronary artery disease (CAD) in native coronary arteries is cardiac allograft vasculopathy (CAV) in transplanted hearts. CAV is the main cause of morbidity and mortality beyond the first year after heart transplantation (HTx). The mechanisms of CAV and its predisposing factors are multifactorial and not well understood by now. The early immunologic and nonimmunologic (e.g., cytomegalovirus [CMV] status, immunosuppressive drugs) injury of the endothelium may initiate this process (9). In fact, epicardial and microvascular endothelial function is detectable early after transplantation and is of prognostic importance (9).
The impact of CD in transplant patients and its association with functional coronary defects is unknown. This study is the first to systematically analyze the incidence of CD after HTx. We also retrospectively correlated allograft vasomotor function, gender mismatch, and CMV status with CD to potentially identify transplantation-specific pathogenetic aspects.
CD was detectable in 3.8% (26 of 688; 95% confidence interval 2.5–5.5) at a median of 4.5 (range 2–18) years after HTx (Table 1). Regarding the classification of the ectasia, we found type I or type II in 63% of all CD patients (16 of 26). In particular, 50% (13 of 26) of the CD patients showed a diffuse dilatation with ectatic lesions in two vessels (type I) and 12.5% (3 of 26) showed a diffuse dilatation in one vessel and discrete dilatation in another (type II). Furthermore, we found a diffuse dilatation in only one vessel (type III) in 12.5% of the CD patients (3 of 26) and finally only a discrete dilation in one vessel (type IV) in 25% of the patients (7 of 26). Remarkably, the RCA was always affected.
CD was neither associated with donor-recipient gender mismatching nor with donor and recipient age (Table 1). Cumulative survival of the patients diagnosed with CD was comparable with the non-CD control collective using Cox regression with CD as time-dependent covariate. No significant differences in long-term survival were detected (P=0.96; Kaplan-Meier analysis, Fig. 1).
With reference to CMV constellation, we found a high-risk constellation (donor CMV positive, recipient CMV negative) for CD patients in 27% (6 of 22; for 4 patients, no data were available) and for CD-negative patients in 30% (31 of 104). In terms of CMV infections, we could only find CMV viremia in one CD patient (1 of 26, i.e., 6%) over the time period. For the CMV-tested patients without CD (n=102), viremia was detectable over time in 42 patients (41%). In all cases, the positive CMV viremia was treated with ganciclovir therapy.
We detected 12 (46%) rejection episodes in the CD-collective, which occurred between 1990 and 1996. During the same period, acute rejection episodes were found in 73% of the non-CD population (P<0.05).Coronary artery stenoses (defined as lumen reduction of ≥50%) were present in 27% of the CD segments in transplant patients. Coronary stenosis was not enhanced in CD patients compared with patients without CD (26%, not significant; Table 1; Fig. 2).
Epicardial Vasomotor Function
The analysis of the epicardial integrity of the transplanted patients showed a reduction of the endothelial- dependent coronary vasodilatation in both groups. For the CD group (n=5), we found a median change in coronary diameter of −12.1% (−19.5% to −3.7%) versus a median change of −6.5% (−57% to 32.15%) in the nonectatic group (n=172) in response to acetylcholine (P=0.3; Fig. 3).
In line, endothelial-independent coronary vasodilatation was not different between the ectatic (n=5) and the nonectatic group (n=172) 8.5% (−2.35% to 12.15%) versus 9% (−11% to 43.25%) in response to adenosine in the nonectatic group (P=0.2; Fig. 3).
Microvascular Vasomotor Function
We observed a general flow velocity reduction (−31%) in arteries with CD (median 20 cm/sec, range 13–27 cm/sec; n=5) compared with the nonectatic coronary arteries (28 cm/sec, range 13–110 cm/sec; n=172; P=0.03; Fig. 4).
Analysis of the microvascular integrity showed an insignificant increase of the coronary flow reserve (CFR) by 8.5% for the CD group under stimulation with acetylcholine (endothelial depending reaction) compared with the nonectatic group (mean fold increase 2.9 [1.3–3.8] vs. 2.5 [1–5.2]; P=0.45; Fig. 5). In response to adenosine (endothelial independent stimulation), we found a significantly reduced CFR by 29.6% (mean fold increase 1.9 [1–2.9] vs. 2.7 [1 to 6]; P=0.04) in patients with CD (Fig. 5).
Intimal thickening in the ectatic segments was not increased compared to nonectatic segments (Table 2) of the HTx patients.
We provide here the first systematic analysis of the incidence and clinical appearance of CD in heart transplant recipients. Furthermore, we analyzed functional changes associated with coronary ectasia. The incidence of CD in heart transplant recipients is not increased when compared with the normal population. CD develops at a median of 4.5 years after HTx. Compared with a nontransplant population (5, 10), CD after HTx had a higher grade of diffuse vascular involvement but a reduced coincidence with stenotic CAD. We only found a coincidence of CD and CAD in approximately 27% of our patients, whereas according to the literature, approximately 75% of patients with CD have concomitant stenosing CAD (5, 10). Therefore, our data support previous finding by others who have postulated that the coincidence of CD and CAD results more of the high incidence of the atherosclerosis in the general population than of a common pathogenesis (5). RCA has been described to be most frequently involved in the ectatic process. Correspondingly, we found a 100% involvement of the RCA, compared with 45% to 75% in published nontransplant collectives (10–12). Another finding was that, based on the classification of Markis et al. (3), the patients with CD after HTx had a much more distinctive form of dilation. Sixty-three percent of the patients had a I or II grade of dilatation, which means that two vessels exhibit local or diffuse dilatation.
One of the main coronary angiographic characteristics of CD is the impaired coronary blood flow (13). Our data confirm this fact in the transplanted vessels, as the coronary blood flow was approximately 30% lower in ectatic coronary arteries. Importantly, we measured the coronary flow distally to the ectatic segment, reflecting coronary microvascular disturbances. In fact, coronary microvascular response to adenosine was impaired in the CD group, demonstrating endothelial independent dysfunction. In contrast, endothelial-dependent vasodilatation was impaired in the ectatic and nonectatic group, reflecting diffuse endothelial injury after HTx. The pathogenesis of CD is controversially discussed (4, 9, 14). Several facts suggest that CD is not simply a variant of coronary atherosclerosis (4). There are various studies that support the hypothesis of a primary degenerative process of the lamina media as the first step in the pathogenesis of the CAD (4, 15). These findings do not depend on the degree of the intimal involvement and the local atheromatous burden (11, 16, 17). Our findings support this conclusion, because in our CD-HTx population, we only detected mild intimal hyperplasia in the dilated segments. Thus, a functional loss of the musculoelastic components of the coronary artery media seems to be the major cause in the pathogenesis of CD and can be tested by the intracoronary adenosine application and the measurement changes in epicardial diameter and microvascular flow velocity (4, 11). Our data support this pathogenetic aspect by measuring a reduced CFR of 30.7% after intracoronary adenosine application in vessels with CD.
The clinical course of CD in the nontransplant population seems to depend mainly on whether it is an isolated phenomenon or co-exists with CAD. In the CAAS study by Baman et al. (18) showed no significant reduction in the 5-year survival of patients with CD (29% of the population with a high atherosclerotic burden died during the 5-year follow-up). Patients with CD and CAD are comparable with patients with CAD and without CD with respect to outcome (4). In line, we did not detect an increased mortality in patients with CD after HTx. Therefore, patients presenting with CD have a good prognosis. However, these forms of CD are not completely harmless, as case studies have reported thrombotic events with consecutive cardiac infarction and also ruptures of the aneurysms (13) (14). We did not find any of these side effects in our study population. Every patient presenting with CD should receive a thrombocyte-aggregation inhibitor, for example, aspirin, and furthermore an individual treatment, depending on thrombotic events or ecstatic dimension (13). Therefore, covered stents or coronary artery surgery may be necessary in exceptional cases (14).
It is important to note that only 26 patients with CD could be identified in our transplant population (although we screened 688 patients for this study). Furthermore, only 5 of the patients with CD versus 172 without CD had invasive testing. Even if the difference between 26 of 688 and 5 of 177 is not statistically different (Fish-test; P=0.65), the number of patients with CD is low. Therefore, the invasive results only permit a hypothesis for the pathogenesis of CD in heart transplant recipients.
All CD patients in our study were male, reflecting the fact that approximately 80% of our transplant collective are male. Coincidentally, we had only three female donors. In addition, CMV infection was seen in only one CD patient. There is a well-documented relationship of CMV and transplant vasculopathy, and similarly, a general gender difference in CD has been reported earlier (7, 8). Because of our special population, it seems not to be opportune to convey gender and CMV-specific statements in context with CD in our heart transplanted collective. Because of the low prevalence of CD in heart transplant recipients and the low complication rate, further prospective studies and individual approaches are required to determine optimal treatment.
MATERIALS AND METHODS
We retrospectively analyzed the annually performed coronary angiographic results of 409 heart transplant recipients who were followed up in our outpatient department. In addition, we evaluated the coronary angiograms of 279 deceased patients for the description of CD.
All angiograms were initially performed within the first 6 months after transplantation with the rare exception when an angiogram was performed immediately before transplantation. After transplantation, additional annual angiograms were performed except in patients with specific contraindications (e.g., severely impaired renal function). Usually after 5 years, coronary angiograms were performed on a 2-year bases with stress echocardiographies in between. The latest angiogram on all 688 patients was reviewed, and if there was CD, then the previous angiograms were reviewed. The median of three angiograms (range 2–8) were looked at for the patients for whom CD was diagnosed. In total, we analyzed 770 angiograms of 688 patients during the time period 1988 to 2009.
Invasive functional testing and intravascular ultrasound (IVUS) were performed in a subgroup of 177 patients (of 688; 26%; Fig. 6). In addition, donor and recipient age and gender, time to first diagnosis of CD, CMVstatus, rejection episodes, and if available data of invasive functional testing and IVUS were registered.
Since 1994, CMV prophylaxis was initiated with ganciclovir in patients at high risk for CMV infection (donor CMV positive, recipient CMV negative). Patients presenting with other CMV constellations were scheduled for a preemptive approach with CMV testing and ganciclovir treatment only after proven CMV viremia. CMV testing was performed every 2 weeks. After the first year, CMV testing was only performed, when an infection was suspected. For the detection of virus activity, we used an antigen testing (CMV-Clonab-PP65) until 2001 and CMV gene expression (real-time PCR) thereafter.
Definition and Graduation of CD
CD was defined as more than or equal to 1.5-fold localized increased vessel diameter or diffuse dilatation involving more than 50% of the coronary artery (detected by quantitative coronary angiography). CD was classified according to the following definition of Markis et al. (3): type I: diffuse dilatation with ectatic lesions in two vessels; type II: diffuse dilatation in one vessel and discrete dilatation in another type; type III: diffuse dilatation in one vessel; and type IV: discrete dilatation in one vessel.
Study Protocol of Endothelial Dysfunction
Vasodilative drugs were discontinued at least 18 hr before cardiac catheterization. After completion of diagnostic catheterization, 5000 units of sodium heparin were administered, and an 8F-Judkins catheter was positioned in the left coronary artery ostium. An angioplasty guide wire (0.014 inch) was advanced to the distal segment of the left anterior descending (LAD) coronary artery, and a 3F infusion catheter was placed 0.5 to 1 cm distal to the origin of the first diagonal branch. The guide wire was removed. Subsequently, a series of sequential 3-min intracoronary perfusions of 5% dextrose and graded concentrations of acetylcholine (10–6, 10–5, and 10–4 mol/L) and nitroglycerine (40 μg/mL) were administered with a perfusion pump at a constant flow rate of 0.80 mL/min. A coronary angiogram was performed at the end of each perfusion.
Quantitative Coronary Angiography
Quantitative coronary angiography was performed to investigate the epicardial vasomotor response using a computerized automatic analysis system (HICOR, Siemens, Erlangen, Germany). Nonstenotic proximal and distal coronary segments identified between easily visualized branch points were selected for analysis in the LAD. Epicardial endothelial dysfunction was defined as luminal diameter constriction of more than 10% in response to acetylcholine. Epicardial smooth muscle cell dysfunction was defined as the luminal diameter dilatation of less than 10% (19) in response to adenosine. The prevalence of pathologic findings in the proximal and distal coronary parts was determined, and means were calculated. Intraobserver and interobserver variabilities showed high reproducibility (r2=0.84, P<0.01). CAD was defined as more than or equal to 50% stenosis by angiographic detection.
Coronary Flow Reserve
The CFR as a marker of microvascular integrity was determined as the ratio of the maximal coronary flow velocity (cm/sec) after pharmacologic stimulation to the basal flow velocity. Coronary flow is critically controlled at the level of the resistance vessels, provided no severe stenosis is present in the epicardial arteries. In this study, no epicardial constriction greater than 50% in response to acetylcholine was observed. Therefore, diameter changes were not included in the calculation of CFR. Coronary endothelial dysfunction was defined as CFR of less than 2 in response to acetylcholine (20, 21), and microvascular smooth muscle cell dysfunction was defined as CFR less than 2 in response to adenosine (22).
IVUS was performed as described previously in detail (20). Image analysis was performed offline using a digital system. Mean intimal area (10 randomly assigned segments), lumen area, and vessel area were calculated. The lumen area was defined as the area within the intimal border. The total vessel area was defined as the area within the media/adventitia boundary. Intimal (plaque) area was the space between vessel and lumen area. The mean intimal index was calculated as the quotient of the intimal area and the vessel area.
Our data are described with median and range. The Mann-Whitney U test signed for not normally distributed paired samples was applied. Because our study population was beyond 20, we indicate exact P values. For the statistical analyses of the nominal scaled unpaired clinical variables, we used the Pearson's chi-quadrate test and Fisher's exact Test. Differences between means were considered significant with P less than 0.05. SPSS (version 16, IBM, New York) was used for statistical analysis.
Overall survival was analyzed using Cox regression with CD as a time-dependent covariate. For graphical representation as Kaplan-Meier plots, we performed a landmark analysis using the median time to CD (4.5 years) as landmark.
The authors thank Nicole Kreißl, Silke Geserick, and Claudia Strohmeyer for their kindly support.
1.Swaye PS, Fisher LD, Litwin P, et al. Aneurysmal coronary artery disease. Circulation
1983; 67: 134.
2.Syed M, Lesch M. Coronary artery aneurysm: A review. Prog Cardiovasc Dis
1997; 40: 77.
3.Markis JE, Joffe CD, Cohn PF, et al. Clinical significance of coronary arterial ectasia. Am J Cardiol
1976; 37: 217.
4.Yetkin E, Waltenberger J. Novel insights into an old controversy: Is coronary artery ectasia a variant of coronary atherosclerosis? Clin Res Cardiol
2007; 96: 331.
5.Farto e Abreu P, Mesquita A, Silva JA, et al. [Coronary artery ectasia: Clinical and angiographic characteristics and prognosis]. Rev Port Cardiol
1993; 12: 305.
6.Newman KM, Ogata Y, Malon AM, et al. Identification of matrix metalloproteinases 3 (stromelysin-1) and 9 (gelatinase B) in abdominal aortic aneurysm. Arterioscler Thromb
1994; 14: 1315.
7.Pinar Bermudez E, Lopez Palop R, Lozano Martinez-Luengas I, et al. [Coronary ectasia
: Prevalence, and clinical and angiographic characteristics]. Rev Esp Cardiol
2003; 56: 473.
8.Giannoglou GD, Antoniadis AP, Chatzizisis YS, et al. Prevalence of ectasia in human coronary arteries in patients in northern Greece referred for coronary angiography. Am J Cardiol
2006; 98: 314.
9.Nickel T, Schlichting CL, Weis M. Drugs modulating endothelial function after transplantation. Transplantation
2006; 82(1 suppl): S41.
10.Demopoulos VP, Olympios CD, Fakiolas CN, et al. The natural history of aneurysmal coronary artery disease. Heart
1997; 78: 136.
11.Befeler B, Aranda MJ, Embi A, et al. Coronary artery aneurysms: Study of the etiology, clinical course and effect on left ventricular function and prognosis. Am J Med
1977; 62: 597.
12.Celik S, Erdogan T, Kasap H, et al. Carotid intima-media thickness in patients with isolated coronary artery ectasia. Atherosclerosis
2007; 190: 385.
13.Kruger D, Stierle U, Herrmann G, et al. Exercise-induced myocardial ischemia in isolated coronary artery ectasias and aneurysms (“dilated coronopathy”). J Am Coll Cardiol
1999; 34: 1461.
14.Haddad F, Perez M, Fleischmann D, et al. Giant coronary aneurysms in heart transplantation
: An unusual presentation of cardiac allograft vasculopathy. J Heart Lung Transplant
2006; 25: 1367.
15.LaMendola CL, Culliford AT, Harris LJ, et al. Multiple aneurysms of the coronary arteries in a patient with systemic aneurysmal disease. Ann Thorac Surg
1990; 49: 1009.
16.Rath S, Har-Zahav Y, Battler A, et al. Fate of nonobstructive aneurysmatic coronary artery disease: Angiographic and clinical follow-up report. Am Heart J
1985; 109: 785.
17.Chugh R, Perloff JK, Fishbein M, et al. Extramural coronary arteries in adults with cyanotic congenital heart disease. Am J Cardiol
2004; 94: 1355.
18.Baman TS, Cole JH, Devireddy CM, et al. Risk factors and outcomes in patients with coronary artery aneurysms. Am J Cardiol
2004; 93: 1549.
19.Kubrich M, Petrakopoulou P, Kofler S, et al. Impact of coronary endothelial dysfunction on adverse long-term outcome after heart transplantation
2008; 85: 1580.
20.Petrakopoulou P, Anthopoulou L, Muscholl M, et al. Coronary endothelial vasomotor function and vascular remodeling in heart transplant recipients randomized for tacrolimus or cyclosporine immunosuppression. J Am Coll Cardiol
2006; 47: 1622.
21.Hodgson JM, Marshall JJ. Direct vasoconstriction and endothelium-dependent vasodilation. Mechanisms of acetylcholine effects on coronary flow and arterial diameter in patients with nonstenotic coronary arteries. Circulation
1989; 79: 1043.
22.Petrakopoulou P, Kubrich M, Pehlivanli S, et al. Cytomegalovirus infection in heart transplant recipients is associated with impaired endothelial function. Circulation
2004; 110(11 suppl 1): II207.