Author + information
- Received May 11, 2005
- Revision received October 11, 2005
- Accepted October 31, 2005
- Published online April 18, 2006.
- Paraskevi Petrakopoulou, MD⁎,
- Lydia Anthopoulou, MD‡,
- Michael Muscholl, MD⁎,
- Volker Klauss, MD§,
- Wolfgang von Scheidt, MD¶,
- Peter Überfuhr, MD†,
- Bruno M. Meiser, MD†,
- Bruno Reichart, MD† and
- Michael Weis, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Michael Weis, Medizinische Klinik und Poliklinik I, Klinikum Grosshadern, University of Munich, Marchioninistrasse 15, 81377 Munich, Germany
Objectives This study aimed to compare changes in coronary endothelial function, systemic endothelin-1 (ET-1) levels, and vascular remodeling in heart transplant recipients randomized to cyclosporin A (CyA) or tacrolimus (Tac) immunosuppression.
Background Functional endothelial abnormalities and intimal thickening are sensitive measures of early cardiac allograft vasculopathy (CAV).
Methods The randomized, prospective study was performed in two groups of 22 patients, maintained on Tac or CyA and mycophenolate mofetil immunosuppression, 1 and 12 months after heart transplantation. We investigated epicardial luminal diameter, coronary blood flow velocity, and ET-1 plasma levels at 1 and 12 months after transplantation. Structural coronary alterations were determined using intravascular ultrasound.
Results Epicardial vasomotor function at baseline and during follow-up was comparable between the groups. Deterioration of microvascular endothelial function during follow-up was significantly enhanced in the CyA versus Tac group (p < 0.05). Circulating ET-1 concentration increased in the CyA group but significantly decreased over time in the Tac group (CyA +17% vs. Tac −25%; p < 0.05). The time-dependent increase in mean intimal area was significantly enhanced in the CyA versus Tac group, whereas the vessel area significantly increased during follow-up in the Tac compared with the CyA group.
Conclusions Epicardial endothelial function is comparable between CyA- and Tac-treated patients. Microvascular endothelial function deteriorates more in CyA-treated patients, a finding that correlates with enhanced ET-1 concentration and an increased intimal area during follow-up. The mean vessel area in the Tac group increased over time, indicating positive vascular remodeling. Tac is superior to CyA with respect to microvascular endothelial function, intimal thickening, and vascular remodeling.
Cardiac allograft vasculopathy (CAV) represents the major cause of death in long-term survivors after heart transplantation (1). The chronic disease is characterized by intimal proliferation during the early phase and ultimately manifests as a luminal stenosis of epicardial branches, occlusion of smaller vessels, and myocardial infarction (2). Most evidence indicates that CAV represents a combination of donor-transmitted atherosclerosis and de novo generated lesions after transplantation as a form of chronic allograft rejection. The endothelium, as an antigen-presenting cell layer between vessel wall and circulating immune cells, has an important role in development of the CAV. Coronary endothelial dysfunction is of major impact for the long-term follow-up in transplantation patients (3).
The progress of CAV is a very complex process. Initially, interest was focused in the severity of the intimal hyperplasia in the development of CAV, which is characterized by myointimal hyperplasia resulting in a luminal loss of the arteries. However, studies using intravascular ultrasound have shown that not only plaque-induced compensatory enlargement but also vessel constriction plays a role in the luminal obstruction.
Newer immunosuppressive regimens have strongly improved prophylaxis and treatment of acute rejection episodes, whereas the therapeutic options for the treatment of allograft vascular disease are limited. Cyclosporin A (CyA) or tacrolimus (Tac) in combination with azathioprine (AZA), mycophenolate mofetil (MMF), or rapamycin are considered standard immunosuppressive regimens in human heart transplantation.
Typical side effects of CyA and Tac are metabolic alterations such as nephrotoxicity, dyslipidemia, and glucose intolerance (4). Moreover, calcineurin inhibitors have been shown to interact with vasoactive metabolites, causing arterial hypertension and arteriosclerosis (5,6). Therefore, we may presume that the chronic exposure of the transplanted heart to calcineurin inhibitors augments the risk of chronic allograft failure because of acceleration of allograft endothelial dysfunction.
Endothelin-1 (ET-1) is a modulator of coronary vascular reactivity in the early stages of atherosclerosis and cardiac allograft vasculopathy (7). Plasma ET-1 levels have been shown to be elevated after human cardiac transplantation (8). Former studies by our group have shown enhanced concentrations of circulating and coronary ET-1 in transplantation patients with endothelial dysfunction (9). No clinical relationship between immunosuppressive therapy, development of coronary endothelial dysfunction, and endothelin activation has been shown thus far. The goal of this prospective, randomized study was to compare the impact of Tac and CyA on functional and structural coronary alterations, as well as on circulating ET-1 concentrations during the first year after cardiac transplantation. The primary goals were to compare changes in coronary endothelial function and intimal thickening between the two treatment groups. The secondary goals were to compare changes in coronary endothelium-independent function, vascular remodeling, and ET-1 concentrations between the groups.
Patients and methods
The study was formally approved by our local ethics committee. Each participant gave informed consent before recruitment. The trial was conducted in accordance with the Declaration of Helsinki.
After orthotopic heart transplantation, 60 patients entered the prospective, open-label, single-center, randomized trial between 1999 and 2001. Patients were randomized to treatment with Tac (n = 30) or CyA (n = 30) in combination with MMF and corticosteroids (for the first six months). Assignment to treatment was done according to an alternating 1:1 (Tac-to-CyA) ratio. Therapeutic drug monitoring was performed for Tac, CyA, and mycophenolic acid (MPA) concentrations throughout the study. Adult male or female patients (≥18 years old) suffering from terminal heart disease or heart failure were included in the study. Seventy-two patients were screened, but 12 patients were not enrolled because of various exclusion criteria. Reasons for exclusion were pregnancy or nursing, unwillingness or inability to use adequate contraception during the study, awaiting cardiac retransplantation, previous or multiple organ transplantations, human immunodeficiency virus-positive donor or recipient, significant renal impairment (creatinine >2.5 mg/dl) or elevated liver enzyme levels (serum glutamic oxaloacetic transaminase >1.5 times upper limit of reference value) for reasons other than liver congestion, or participation in any other investigational drug study within 28 days of entry into this study.
Coronary vasomotor testing and intravascular ultrasound at baseline and follow-up were performed in 44 patients (Tac n = 22; CyA n = 22). Sixteen patients (of 60) were excluded from coronary functional measurements based on renal insufficiency (n = 4), cerebral apoplexy (n = 1), unwillingness (n = 3), death (n = 2), stent implantation (n = 1), coronary vasospasm (n = 1), or technical difficulties (n = 4).
Each study group (CyA and Tac groups) consisted of 22 patients undergoing routine diagnostic coronary angiography 1 month (CyA 38 ± 14 days; Tac 44 ± 25 days) after orthotopic heart transplantation (Table 1).All patients studied were maintained on MMF- and prednisolone (for six months)-based immunosuppression protocols. Patients were followed up for 12 months (CyA 369 ± 20 days; Tac 356 ± 24 days), including coronary angiography, intravascular ultrasound, and coronary function testing at 1 and 12 months after transplantation.
Before listing a patient for heart transplantation, we routinely performed a panel reactive antibody test against a series of antigen-coated beads representing typical human leukocyte antigen (HLA) types. The result was expressed as a numerical percentage. All included patients, both in the Tac group as well as in the CyA group, showed a percentage of panel reactive antibody <10%. There was no difference concerning the HLA mismatching between the two patient groups (human leucocyte associated antigen A/B and D-related human leukocyte antigen). Moreover, the serological cytomegalovirus mismatches between transplant recipients and donors were comparable between the groups (Table 1).
Demographic data and primary diagnosis, as well as peri-operative data, were comparable between both treatment groups.
Study medication and therapeutic drug monitoring
All three compounds, Tac (Prograf; Fujisawa GmbH, Munich, Germany), CyA (Neoral; Novartis AG, Basel, Switzerland), and MMF (CellCept; Roche Pharmaceuticals, Basel, Switzerland), were administrated intravenously for the initial 2 to 3 days after transplantation, Tac at a dosage of 0.01 to 0.02 mg/kg/day and CyA at a dosage of 0.03 to 0.1 mg/kg/day. Both drugs were administrated as a continuous 24-h infusion. The MMF was administrated at a daily dosage of 2 g/day as a short-term infusion. Subsequent dosages were adjusted according to blood trough levels. Target blood trough levels for Tac were 13 to 16 ng/ml (Tacro II assay; Abbot, Wiesbaden, Germany) during the first 6 months and 10 to 12 ng/ml thereafter. The target blood trough levels for CyA ranged between 200 and 300 ng/ml (EMIT assay; Dade-Behring, Schwalbach, Germany) during the first 6 months and then decreased to 100 to 200 ng/ml. The MMF dose was adjusted according to MPA plasma concentrations (EMIT assay) of 2.5 to 4.0 ng/ml for the first 6 months and 1.5 to 2.5 ng/ml thereafter. Blood samples were collected daily for the first 3 weeks after transplantation and subsequently at biweekly, weekly, and monthly intervals. The dosage of each drug was adjusted to the respective target concentration. The MPA plasma concentrations were comparable between CyA- and Tac-treated patients at 1 and 12 months after transplantation as outlined in Table 1(p = 0.32 at baseline and p = 0.40 during follow-up).
All patients received an intraoperative bolus of intravenous methylprednisolone (500 mg) followed by 3 × 125 mg doses within the initial 24-h period after transplantation. Prednisolone therapy was subsequently administered, and the dosage was tapered from 1.0 to 0.1 mg/kg/day over a period of 4 weeks, and finally was completely removed in all patients as early as 6 months after transplantation.
Lipid-lowering agents and hypertension treatment
All patients received routinely a minimum dose of 5 mg simvastatin (Zocor; Merck Sharp & Dohme, Haar, Germany) per day, independent of the total or low-density-lipoprotein cholesterol levels. In cases of persistent hypercholesterolemia (fasting total cholesterol >200 mg/dl or low-density-lipoprotein cholesterol >140 mg/dl), the simvastatin dose was increased up to 20 mg/day. Simvastatin dosage was not different between the groups (CyA group 9.5 ± 5 mg/day vs. Tac group 8.6 ± 6 mg/day at 12 months; p = NS).
Patients received an antihypertensive treatment with calcium antagonists, angiotensin-converting enzyme inhibitors, or a combination of both drug groups. The usual daily dose of enalapril and diltiazem was 10 mg and 180 mg, respectively. In some cases, patients received an antihypertensive treatment with ramipril (usual daily dose, 5 mg) instead of enalapril. Angiotensin-converting enzyme inhibitor doses were comparable between CyA- and Tac-treated patients (CyA group 10 ± 4 mg/day vs. Tac group 12 ± 3 mg/day at 12 months; p = NS). The use of antihypertensive drugs was not significantly different between the groups (Table 1).
Left heart catheterization
Protocols for coronary angiography and coronary vasomotor testing have been described in detail (10,11). In brief, after the diagnostic procedure including left ventriculography and coronary angiography, a Cardiometrics Doppler Flow-wire (Endosonics Corp., Rancho Cordova, California) was placed in the proximal left anterior descending or circumflex artery, permitting measurement of coronary blood flow velocities (12,13). The blood flow velocity was recorded continuously during the administration of the study agents. First, adenosine (160 μg/min over 5 min; Adrekar; Sanofi Winthrop, Fuerstenfeldbruck, Germany) was infused into the left coronary system to achieve maximal endothelium-independent coronary flow. Secondly, acetylcholine (1 and 30 μg/min over 5 min each; Miochol; CIBA Visions Vertrieb GmbH, Grossostheim, Germany) was infused intracoronarily to investigate endothelium-dependent microvascular and epicardial endothelial vasomotor function. Finally, nifedipine (0.2 mg intracoronarily; Adalat, Bayer, Leverkusen, Germany) was administrated. In that way, we achieved maximal epicardial vasodilatation. At the end of each infusion, coronary angiography was performed with a biplane imaging system in a right and left oblique position with adequate cranial or caudal angulation for optimal analysis of the left coronary tree on end-diastolic frames. The position was kept constant during the protocol. Throughout each infusion, heart rate, arterial pressure, coronary flow velocity, and electrocardiogram were monitored and documented on SVHS videotape for additional offline analysis. The ultrasound catheter (Visions Five-64 F/X; Endosonics Corp.) was advanced to the distal left coronary descending and/or circumflex artery after intracoronary application of 5,000 IE heparin and 0.1 mg nitroglycerin. The catheter was advanced to the distal left coronary descending and/or circumflex artery. During the subsequent standardized pullback maneuver, images were documented on videotape for further off-line analysis.
Measurement of coronary flow velocity and reserve
Coronary flow measurement is an established functional parameter for assessing the integrity of the microcirculation. It was determined by the ratio of the maximal coronary flow velocity (in cm/s) after pharmacologic stimulation to the basal flow velocity (coronary flow velocity reserve [CFR] = Fp/Fb, where Fpis maximal and Fbis baseline flow velocity). The coronary flow is controlled at the level of the resistance vessels if no severe stenosis is present in the epicardial arteries. Coronary microvascular endothelial dysfunction was defined as CFR <2.0 in response to acetylcholine.
Quantitative coronary angiography
Quantitative coronary angiography was performed to investigate epicardial vasomotor response using a computerized automatic-analysis system (HICOR; Siemens, Erlangen, Germany). Nonstenotic proximal and distal coronary arterial segments identified between easily visualized branch points were selected for analysis in the left anterior descending artery. Responses of the coronary segments to the different stimuli were expressed as percent change versus control value. The diameter change for each vessel was calculated on the basis of the average of the two segments. Intraobserver and interobserver variability showed high reproducibility (r > 0.89, p < 0.001). Epicardial endothelial dysfunction has been defined as >10% diameter reduction in response to acetylcholine compared with baseline (10).
Intravascular ultrasound (IVUS)
Intravascular ultrasound was performed as described previously in detail (11,13). The IVUS was performed immediately after the Doppler flow measurement in both subgroups early (1 month) and late (12 months) after transplantation. Image analysis was performed offline with the help of 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 to the vessel area.
Measurement of ET-1 has been described in detail elsewhere (9). Plasma samples obtained before measurement of coronary vasomotor response were immediately placed on ice, centrifuged, and stored at −80°C until they were assayed. The ET-1 was extracted through Sep-Pak C18 cartridges (Waters, Eschborn, Germany) washed with 10 ml 0.1% trifluoroacetic acid and eluted with 5 ml 60% acetonitrile/0.1% trifluoroacetic acid. For radioimmunoassay, eluates were deep frozen, dried, and re-dissolved in radioimmunoassay buffer. The ET antibody 2428 (Medor, Herrsching, Germany) was added to samples and standards (binding characteristics: ET-1 100%; ET-2 73%; ET-3 31%; Big-ET 13%) for radioimmunoassay analysis.
For both treatment groups, values of donor and recipient demographic characteristics and intracoronary Doppler parameters are presented as proportions or means ± SD. For comparison of time-dependent changes in coronary function, intimal thickening, vascular remodeling, and ET-1 concentrations between the groups (CyA vs. Tac), we used an unpaired Student ttest. Examination of the association of MPA plasma levels and IVUS parameters were performed using linear regression analysis. A p value of 0.05 was considered significant.
The values of heart rate and mean arterial blood pressure after drug administration were not significantly altered compared with baseline (data not shown).
Resting coronary flow
Resting coronary blood flow velocity was not different between the groups at baseline and during follow-up (data not shown).
Microvascular endothelium-independent function
Coronary endothelial-independent dysfunction (defined as coronary flow velocity increase to adenosine <2) was detectable in 37% versus 18% (1st month vs. 12th month) of the CyA group, as well as in 33% versus 17% of the Tac group.
Coronary flow reserve in response to adenosine and nifedipine was not significantly different at baseline and during follow-up between the groups (Table 2).Adenosine-induced as well as nifedipine-induced CFR tended to increase during follow-up only in the Tac group (Table 2).
Microvascular endothelial function
Microvascular endothelial function at baseline was comparable between CyA- and Tac-treated transplant recipients (Table 2). Deterioration of microvascular endothelial function during follow-up was significantly enhanced in the CyA versus Tac group (p < 0.05; change in acetylcholine-induced CFR −30% in the CyA group vs. −10% in the Tac group) (Table 2). The vasomotor response in the microcirculation was not related to the number of treated rejected episodes (r = 0.16; p = NS), ischemic time (r = 0.06; p = NS), concomitant drug therapy (statin and antihypertensive therapy; r = 0.19; p = NS), cytomegalovirus mismatch (r = 0.28; p = NS), or hyperlipidemia (r = 0.23; p = NS).
Resting epicardial diameters
Resting proximal and distal epicardial diameters were not different between the groups at baseline and during follow-up (data not shown).
Epicardial endothelium-independent vasomotor function
Adenosine- and nifedipine-mediated epicardial vasodilatation was comparable between CyA- and Tac-treated patients at baseline and during follow-up (Table 3).Nifedipine-mediated vasodilatation tended to improve over time in the CyA-treated versus the Tac-treated group, without reaching statistical significance (Table 3).
Epicardial endothelial vasomotor function
Epicardial endothelial dysfunction (defined as a >10% diameter reduction to acetylcholine) was detectable in 56% versus 33% (1st month vs. 12th month) in the CyA group and in 33% versus 33% in the Tac group.
Acetylcholine-mediated epicardial responses were comparable between CyA- and Tac-treated patients at baseline and during follow-up (Table 3). Coronary epicardial endothelial function did not significantly improve over time in either group (absolute change +17% in the CyA group vs. +4% in the Tac group).
Coronary morphology and vascular remodeling
The time-dependent increase in the mean intimal area was significantly enhanced in the CyA versus Tac group (Fig. 1).The percent change in intimal area from 1 month to 12 months after transplantation amounted to 103% in the CyA group versus 60% in the Tac group (p < 0.05 comparing absolute changes) (Fig. 1). The lumen area did not expand significantly during the same period in either group (Tac +44% vs. CyA +25% increase in lumen area during follow-up), and it remained comparable in the Tac and CyA groups at baseline and during follow-up (6.4 to 10.0 mm2in the Tac vs. 6.8 to 8.2 mm2in the CyA group). However, vessel area significantly increased during follow-up in the Tac compared with the CyA group (p < 0.05) (Fig. 2).The mean intimal index increased not significantly over time in both patient groups. The mean intimal index at 12 months after heart transplantation was 27 ± 18% (19 ± 14% at baseline) in the CyA and 22 ± 15% (18 ± 11% at baseline) in the Tac group.
The indices of coronary morphology (as detected by IVUS) were not related to the number of treated rejection episodes during the first year after transplantation in either group (data not shown). Moreover, no correlation was detected between MPA levels and the intimal area (r = 0.20 in the CyA group and r = 0.32 in the Tac group, p = NS) and between MPA levels and the vessel area (r = 0.19 in the CyA group and r = 0.09 in the Tac group, p = NS).
Circulating ET-1 concentration
In the CyA group, the average circulating ET-1 plasma level during the first postoperative month amounted to 10.2 ± 2.3 pg/ml (Fig. 3).During follow-up, the ET-1 level increased to 11.9 ± 3.9 pg/ml (+17% increase). Contrary to these findings, the circulating ET-1 concentration in the Tac-treated group decreased during follow-up (11.5 ± 2.7 pg/ml at 1 month vs. 8.6 ± 1.7 pg/ml at 12 months; −25%). The absolute change in circulating ET-1 concentration over time was significantly different between the groups (p < 0.05) (Fig. 3).
The most salient findings of our prospective study are: 1) Epicardial and microvascular vasomotor function in response to adenosine is not significantly different between heart transplant recipients under CyA or Tac immunosuppression. 2) Nifedipine-mediated epicardial vasomotor response tended to improve in the CyA group during follow-up. 3) Epicardial and microvascular endothelial dysfunction is detectable in approximately 30% of the patients in the early post-transplantation phase. 4) During follow-up, deterioration of microvascular endothelial function was significantly enhanced in the CyA versus the Tac group, whereas epicardial responses tended to improve in both groups. 5) The mean intimal area significantly increased in the CyA versus the Tac group, whereas vessel area significantly increased in the Tac versus the CyA group. 6) During follow-up, circulating ET-1-concentration decreased in the Tac group, whereas it tended to increase in the CyA group.
Immunosuppression and acute/chronic rejection
Both Tac and CyA have proven to be effective immunosuppressive agents in clinical cardiac transplantation. Pham et al. (14) have shown that patient survival as well as major side effects, such as infection, renal toxicity, and hypertension, are comparable between Tac- and CyA-treated patients. However, they also observed that patients receiving Tac have fewer episodes of acute and refractory rejection, require less treatment for rejection, and need lower doses of maintenance steroids. Concerning cardiac allograft vasculopathy at four years after transplantation, defined as any luminal irregularity and any coronary stenosis seen on coronary angiography or any diffuse coronary disease at autopsy, they did not detect any significant difference between the two groups.
Meiser et al. (15) have recently shown that freedom from acute rejection is significantly higher and the incidence of acute rejection episodes per 100 patient days is significantly lower in heart transplant recipients treated with Tac + MMF than in those treated with CyA + MMF. Importantly, in the present analysis, the incidence of acute rejection episodes was not (independently) related to changes in coronary morphology or coronary function. We could not detect a significant association between the rejection score and coronary endothelial function or intimal thickening. These data suggest that the effect of calcineurin inhibitors on vascular structure and function occurs independently of their immunosuppressive properties.
Immunosuppression, endothelial dysfunction, and intimal thickening
The dysfunction of the coronary vascular endothelium, as assessed by functional vasomotor testing, is an early characteristic feature of both native and allograft vasculopathy. A decrease of the local bioavailability of endothelium-derived nitric oxide can contribute to initiation and progression of atherosclerosis.
Experimental models have shown that CyA has an impact on the endothelial and smooth muscle cell dysfunction. The CyA impaired the endothelium-dependent vasodilatation and increased the reactivity of the smooth muscle cells to certain vasoconstrictors, such as angiotensin II (16).
An in vivo comparative study (17) on the effects of CyA and Tac on brachial endothelial dysfunction in renal transplant recipients detected a more pronounced endothelial-dependent and -independent vasodilatation impairment in CyA- than in Tac-treated patients (6.5 ± 3.7% vs. 12.1 ± 5.1%; p < 0.001).
In line with the present study, calcineurin inhibition with CyA, but not with Tac, affected the microvascular endothelial vasomotor response over time. Whereas microvascular endothelial function significantly decreased over time in CyA-treated patients, the nifedipine-mediated vasodilatation tended to improve in the same group. In a recent study we have shown that epicardial and microvascular dilatation to nifedipine is compensatorily enhanced in the setting of coronary endothelial dysfunction (18). However, before recommending nifedipine as a standard therapy in CyA patients, a prospective study on the impact of oral nifedipine therapy on functional and structural coronary changes after transplantation should be available.
In a former nonrandomized in vivo study by our group, the immunosuppressive combinations of CyA + AZA and Tac + MMF seemed to be superior to Tac + AZA regarding preservation of early coronary vasomotor function within two months after transplantation, whereas intimal thickening was not different between the groups (19). In a prospective trial using volumetric intravascular ultrasound analysis, a trend toward a more pronounced progression of intimal thickening was noted in the Tac + AZA versus the CyA + AZA group during 12 months of follow-up (20). However, these functional and morphological coronary alterations in the early period after transplantation refer to comparative studies of immunosuppressive drug combinations other than in the present study (AZA vs. MMF) and do not consider the impact of immunosuppressive drugs on vascular remodeling.
Importantly, MMF dosage and MPA levels were comparable between Tac- and CyA-treated patients, and no association was detectable between MPA levels and IVUS parameters in the current study.
Immunosuppression and vascular remodeling
It is supposable that the same mechanisms that are responsible for the intimal proliferation have an impact on the remodeling process. The variation of remodeling from constriction to dilatation may reflect the influence of immune mechanisms over the vessel during the complex pathogenesis of CAV. Our study points out a significant increase in the mean intimal area without any statistically significant compensatory increase of the vessel and lumen area during the follow-up examination in the CyA group. In contrast, the patients in the Tac group tended to have an increased mean intimal area at follow-up associated with a significant enlargement of the vessel area over time, representing compensatory vascular remodeling. In parallel, we saw no progression of the microvascular endothelial dysfunction in the Tac group, suggesting that the adequate production and release of vasoactive substances from the microvascular tree (and the diminished systemic ET-1 bioactivity) contributes to the positive vascular remodeling.
Immunosuppression and ET-1
In an in vitro study (21) analyzing the prostacyclin (PGI2) and ET-1 release in microvascular capillaries, a significant difference in morphological and biochemical effects of the two common calcineurin inhibitors was found. The CyA had a pronounced injurious effect on the morphology of the in vitro capillaries, whereas Tac did not. The CyA also significantly increased ET-1 release by the capillaries, and Tac did not. As suggested in a subsequent in vitro study by the same group, ET-1, as a potent vasoconstrictive peptide and a proliferative factor of endothelial cells, is implicated in CyA-induced endothelial dysfunction (22). Our findings could verify these in vitro results. We detected that 1 year after heart transplantation, the circulating ET-1 concentration decreased significantly in the Tac-treated group, whereas this feature tended to be enhanced in the CyA-treated patients.
Rapamycin and allograft vasculopathy
Recently, some new combinations of immunosuppressive medications have been tested, among these everolimus and rapamycin. Experimental studies associate the combination of Tac/rapamycin and CyA/rapamycin with a reduction of stenosis, a reduced proliferation of smooth muscle cells, and a lower accumulation of extracellular matrix proteins. Both of these regimens can inhibit the early intimal proliferation, but only the combination of Tac/rapamycin can induce a marginal reduction of the intimal proliferation (23). The clinical importance of everolimus, a derivate of rapamycin, for the reduction of the cardiac allograft vasculopathy was shown in a clinical study by Eisen et al. (24) The patients were treated with everolimus or AZA in combination with CyA, corticoids, and statins. The incidence of cardiac allograft vasculopathy, as examined with intravascular ultrasound, after 12 months was significantly reduced in the everolimus group compared with the AZA group (30% vs. 53%). Clinical data on the impact of everolimus + Tac for the development of CAV have not yet been published.
The impact of rapamycin on endothelial release of vasoactive substances has been discussed controversially. The effect of sirolimus on PGI-2 and ET-1 release was first determined in cultured rabbit endothelial cells (25). Sirolimus significantly increased release of PGI-2 and ET-1 in TGF-beta-treated endothelial cells, whereas Tac caused a decrease in prostacyclin but an increase in ET-1 (25). We could show in human microvascular endothelial cells that rapamycin produced strong increases of ET-1, with only minor nitric oxide production (26). In a different study, CyA but neither Tac nor rapamycin significantly increased ET-1 release from human aortic endothelial cells (21). Differences between the studies could be explained by the different cell types and drug concentrations used.
A major limitation refers to the relatively small patient population studied in each group. However, to exclude potentially confounding factors, the patients were carefully selected from a consecutive series of transplant recipients according to predefined exclusion criteria. In addition, each individual patient served as his or her own control. Using post-hoc power analysis (G Power 2.0, Institute of Psychology, University of Trier, Trier, Germany), the power of our study amounted to 0.52, therefore false-negative results cannot be excluded.
In summary, our results indicate that the epicardial endothelial vasomotor function is not significantly different between heart transplant recipients under CyA or Tac immunosuppression. In contrast, the microvascular endothelial function significantly deteriorated in the CyA group, associated with an enhanced coronary ET-1 concentration during follow-up. The vessel area increased in the Tac group despite the fact that the mean intimal area tended to be increased, indicating positive vascular remodeling. Both Tac and CyA differ with respect to their impact on allograft coronary function and morphology, effects that occur independently of their immunosuppressive properties. Correspondingly, CyA immunosuppression with subsequent allograft endothelial dysfunction may predispose patients to more frequent cardiovascular complications.
The authors thank technician, C. Grimm, for measuring the ET-1 levels, as well as the nursing staff and physicians of the cardiac catheterization laboratory for their kind support.
This study was supported in part by the German research foundation Sachmittelstipendium We 2258/5-1 and by grants from Fujisawa GmbH (Munich, Germany), Novartis AG (Basel, Switzerland), and Roche Pharmaceuticals (Basel, Switzerland).
- Abbreviations and Acronyms
- cardiac allograft vasculopathy
- coronary flow velocity reserve
- cyclosporin A
- baseline flow velocity
- maximal flow velocity
- human leukocyte antigen
- intravascular ultrasound
- mycophenolate mofetil
- mycophenolic acid
- Received May 11, 2005.
- Revision received October 11, 2005.
- Accepted October 31, 2005.
- American College of Cardiology Foundation
- Hollerberg S.M.,
- Klein L.W.,
- Parillo J.E.,
- et al.
- Fitzgerald P.J.,
- St Goar F.G.,
- Connolly A.J.,
- et al.
- Haas G.J.,
- Wooding-Scott M.,
- Binkley P.F.,
- Mzerowity P.D.,
- Kellez R.,
- Cody R.S.
- Doucette J.W.,
- Corl P.D.,
- Payne H.M.,
- et al.
- Weis M.,
- Pehlivanli S.,
- Meiser B.,
- von Scheidt W.
- Meiser B.M.,
- Groetzner J.,
- Kaczmarek I.,
- et al.