Increased cardiac afterload caused by aortic stenosis and systemic arterial hypertension induces several molecular and cellular events leading to adaptative left ventricular hypertrophy (LVH).1 LVH affects 23% of men and 33% of women over the age of 59 years2 and is associated with adverse cardiovascular outcomes.3 Alterations of endothelium-dependent relaxations induced by acetylcholine, adenosine, and substance P have been demonstrated in LVH in clinical and experimental studies.4−6 The decrease of endothelium-dependent relaxations mediated by nitric oxide (NO•) in animal models of LVH has been attributed to decreased synthesis and/or bioavailability of NO• and to the concomitant release of endothelium-derived contracting factors, such as endothelin-1 and vasoconstrictor prostanoids.7 Oxidative stress, mainly by the action of superoxide anion (•O2−), also plays an important role in the development of endothelial dysfunction in LVH8 and in cardiomyocyte hypertrophy.9
We have previously characterized the endothelial dysfunction of epicardial coronary arteries in LVH secondary to aortic banding (AB) in a porcine model.10 Alterations of signaling mechanisms in endothelial cells mainly involve relaxations mediated through Gi proteins and, to a lesser degree, by Gq proteins. These observations, combined with a decreased basal level of cyclic guanosine-3′,5′-monophosphate (cGMP) in the coronary wall and a reduction in the plasma NO2−/NO3− ratio, reflect the decreased bioavailability of NO• in this model, due to a decreased production and/or increased destruction or scavenging.10 Moreover, in the same model, circulating levels of endothelin-1 are increased, associated with decreased endothelial ETB receptor density, which can explain, in part, the increased vasoconstriction.11 Finally, plasma concentrations of peroxynitrite (ONOO−) are increased in this model of LVH, confirming the state of increased oxidative stress.12 Acute treatment with the antioxidant enzymes superoxide dismutase and catalase has been shown to improve endothelial dysfunction associated with LVH.12 This implies that, in this model, endogenous antioxidant activity is insufficient and/or that reactive oxygen species levels such as •O2− and hydrogen peroxide are increased.
Probucol, a lipid-lowering agent possessing few similarities with other lipid-lowering agents, has been largely studied for its antirestenotic properties after angioplasty.13 Moreover, in a rat model of heart failure caused by ligation of the coronary artery, probucol improved numerous processes regulated by redox potential, including inhibition of cardiomyocyte hypertrophy, decrease in ventricular dilation, and improvement of left ventricular remodeling.14 Therefore, the aim of the present study was to assess the effect of the chain, breaking the antioxidant drug probucol on the initiation, progression, and reversal of the coronary endothelial dysfunction associated with LVH secondary to AB in a porcine model.
Twenty-five 8-week old Landrace male swine (Primiporc Inc., St-Gabriel de Brandon, Quebec, Canada), were randomly divided into 5 experimental groups. The animals from the control group (group 1) were submitted to the thoracotomy without an AB (n=3). The untreated aortic banded group (group 2) received a placebo (lactose; 400 mg/d per os) for 60 days, starting on the day of the surgery (n=3). The first actively treated group (group 3) was treated with high-dose probucol (1000 mg/d orally)13,15 from the day of banding for 60 days (n=5). The 2 other treated groups (groups 4 and 5) were started on probucol on days 30 and 60 postbanding, respectively, for a 30-day period (n=7 for both) (Table 1). Experiments were performed in compliance with recommendations of the guidelines on the care and use of laboratory animals issued by the Canadian Council on Animal Research and the guidelines of the Animal Care, and the experimental protocol was approved by a local committee.
Pigs were anesthetized by an intramuscular injection of a mixture of ketamine (20 mg/kg; Rogarsetic, Toronto, Ontario, Canada) and xylazine (2 mg/kg; Rompun, Cambridge, Ontario, Canada). Artificial ventilation with an oxygen/air mixture was provided throughout the surgical intervention to maintain an arterial oxygen saturation of 95%, and anesthesia was induced by isoflurane 1% vol/vol (Abbott Laboratories, Montreal, Quebec, Canada). Respiratory control was assured by frequent determinations of arterial blood gases and acidosis was balanced with 8.4% sodium bicarbonate (Abbott Laboratories, St-Laurent, Quebec, Canada). Hair was shaved in the operative field and the skin was disinfected with a surgical scrub and chlorexidine 0.5% solution. A catheter was placed in an auricular vein for administration of antibiotics at the beginning of surgery (Excenel 0.06 mL/kg; Pharmacia & Upjohn, Orangeville, Ontario, Canada) and for intravenous fluid infusion during surgery. Arterial cannulation was performed through the right femoral artery for blood pressure analysis and a rectal probe was used for monitoring the temperature.
The chest was entered through a left anterior thoracotomy in the third intercostal space and the AB was performed by gently tying an umbilical tape around the aorta, 3 cm above the coronary ostia, to obtain a systolic peak gradient of at least 15 mm Hg. The pericardium and chest were closed in multiple layers. An intramuscular injection of an analgesic (Buprenorphine, 0.6 mg/mL; Reckitt Benckiser Healthcare Limited, Dansom Lane, UK) was given at the end of the surgery and the animals were left to recover in temperature-controlled quarters. Pigs were treated with the antibiotic Excenel (0.06 mL/kg) for the 2 following days and were fed a standard piglet chow. For the following days, additional doses of analgesic (Buprenorphine, 0.3 mg/mL) were given to animals that had not completely recovered from surgery.
Pigs were sedated (see above) for transthoracic echocardiograms performed before the surgery, for baseline data, and at the end of the treatment, before being killed. An S3 probe (ranging from 1.6 to 3.2 MHz) with a standard echocardiographic system (Sonos 5500 Hewlett-Packard, Andover, MA) was used. Ascending aorta dimensions at the banding area, 5 mm both proximally and distally away from the banding were measured in the parasternal long-axis view. Continuous-wave Doppler was used in a modified left ventricular apical long-axis view to follow the peak velocity and gradient across the banding. A 2-dimensional guided M-mode study of the left ventricle and 2-dimensional apical 4-chamber and 2-chamber views were performed and recorded on videotape for off-line measurements. The thickness of the interventricular septum (IVS) and left ventricular posterior wall (LVPW) were measured on M-mode at end-diastole using the average of 3 cardiac cycles. Left ventricular mass was calculated using the following formula: left ventricular mass=1.05 [(D+LVPW+IVS)3–D3]−14 g, where D represents the left ventricular cavity end-diastolic diameter.
After removal of the heart and coronary harvesting, 6 surgical biopsies were taken from the free left ventricle and the septum. They were then fixed in 10% buffered formalin, embedded in paraffin and cut in 4-μm thick sections for histologic documentation of LVH. Each section was selected by an experienced cardiovascular pathologist blinded to the treatment. Cardiomyocyte hypertrophy (cardiomyocytes transversal area: approximately 600 cells/animal, in the septum and the LV-free wall) and interstitial fibrosis (collagen I content) were determined using hematoxilin-phloxin safran and sirius red stain, respectively. Images were taken with a Sony DKC-5000 camera mounted on a BX60 microscope and opened with the program Adobe Photoshop 5.02. Quantification was done with the program Scion Image version 1.63.
Killing and Coronary Harvesting
After their respective period of treatment, swine from the 5 groups were anesthetized using the same approach as described previously. They were killed by exsanguination through a median sternotomy. The heart was removed, weighed for heart-to-body mass ratio, and rapidly placed in a modified Krebs bicarbonate solution (composition in mmol/L: NaCl 118.3, KCl 4.7, CaCl2 2.5, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25, ethylenediaminetetraacetate (EDTA) 0.026, dextrose 11.1, control solution) for harvesting coronary arteries. The left anterior descending, the left circumflex, and the right epicardial coronary arteries were dissected free from adherent fat and connective tissue, and were divided into rings 4 mm in length.
Vascular Reactivity Studies
Rings were then placed in organ chambers filled with the Krebs bicarbonate solution (20 mL), maintained at 37°C, and oxygenated with a mixture of 95% O2/5% CO2. The rings were suspended between 2 metal stirrups, 1 connected to an isometric force transducer. Data were recorded on data acquisition software (IOS3, Emka Inc., Paris, France). After 30 minutes of stabilization, tension on rings was progressively increased to the optimal tension of its active length-tension curve (approximately 3.5 g), as determined by measuring the contraction to potassium chloride (KCl; 30 mmol/L) at different levels of stretch. A maximal contraction was determined with KCl (60 mmol/L) and the baths were then washed. Rings were excluded if they failed to contract with potassium chloride (exclusion rate of less than 5%). Afterward, all studies were performed in the presence of indomethacin (10−5 mol/L; to prevent the endogenous production of prostanoids), propranolol (10−7 mol/L; to prevent the activation of β-adrenergic receptors), and ketanserin (10−6 mol/L; incubated 45 min before the addition of serotonin to antagonize smooth muscle cell serotonin 5-HT2 receptors). After 45 minutes of stabilization, prostaglandin F2α (range 2×10−6 to 10−5 mol/L) was added to achieve contraction averaging 50% of the maximal contraction to KCl (60 mmol/L).
The NO mediated-relaxation pathway was studied by constructing concentration-response curves to serotonin (5-HT; 10−10 to 10−5 mol/L, an agonist binding to endothelial 5-HT1D receptors coupled to Gi proteins) and to bradykinin (10−12 to 10−6 mol/L, an agonist binding to endothelial B2 receptors coupled to Gq-proteins leading to the release of NO and endothelial-derived hyperpolarizing factor).
Assessment of Endothelial Function
Plasma Nitrite and Nitrate (NOX) Levels
NO production from endothelial nitric oxide synthase (eNOS) was assessed from NOX levels present in plasma samples of aortic banded treated and untreated groups and controls. Blood samples were drawn before killing from swine coronary sinus in EDTA to prevent coagulation and were then centrifuged at 4000 rpm for 15 minutes at 4°C. The isolated plasma was consequently frozen in liquid nitrogen and kept at −70°C until analysis. Nitrite (NO2−) was reduced to NO using iodide and acetic acid, whereas both nitrate (NO3−) and NO2− were reduced using vanadium (III) and hydrochloric acid. A Serviers 280 NO analyzer (Serviers Instruments Inc., Boulder, CO), combined with the NO analysis-liquid program, was used to obtain the NO2− and NO3− concentrations in the different samples.
Coronary Artery cGMP Levels
Basal cGMP levels in epicardial coronary arteries of controls and swine submitted to AB, treated and untreated, were measured to assess NO bioavailability. Following vessels harvesting, a few segments were frozen in liquid nitrogen and stored at −70°C until measurements of cGMP levels. The samples were subsequently pulverized, resuspended in trichloroacetic solution (6.25% wt/vol) to precipitate tissue membranes, and centrifuged at 3000 rpm for 15 minutes at 4°C. The supernatants were then washed with diethylether to preserve cGMP and eliminate trichloroacetic solution. Finally, the washed supernatants were heat dried with nitrogen to obtain purified cGMP. cGMP levels were measured using a nonacetylation enzyme immunoassay system based on rabbit anti-cGMP antibody (Amersham Pharmacia Biotech, Baie d'Urfé, Quebec, Canada). The levels were adjusted to the quantity of proteins measured in the tissue using the Bradford microassay technique (Bio-Rad Laboratories, Hercules, CA).
Assessment of Oxidative Stress
Plasma Angiotensin II Levels
Plasma samples were collected before killing from the coronary sinus and immediately centrifuged at 1000g for 15 minutes at 4°C. The reversed phase method with columns packed with phenylsilylsilica was performed for plasma sample extraction. Quantitative measurements of angiotensin II were assessed using an 125I-angiotensin II radioimmunoassay kit (Angiotensin II Radioimmunoassay, ALPCO Diagnostic, Windham, NH).
Plasma Lipid Hydroperoxide Levels
Plasma samples, drawn from coronary sinus before killing, were mixed 1:1 with the extract reagent saturated with methanol and vortexed. Cold chloroform (1 mL) was added to each tube and the solution was centrifuged at 1500g for 5 minutes at 0°C. The bottom chloroform layer was carefully removed and used for lipid peroxidation assessment with a commercial kit (Lipid Hydroperoxide Assay, Cayman Chemicals, Ann Arbor, MI).
Vascular Superoxide Dismutase
Epicardial coronary arteries were homogenized in HEPES buffer (20 mmol/L, pH 7.4, containing 1 mmol/L ethylene glycol tetra-acetic acid, 210 mmol/L mannitol, and 70 mmol/L sucrose) and centrifuged at 1500g for 5 minutes at 4°C. The supernatant, containing both cytosolic and mitochondrial Cu/Zn−, extracellular and Mn-vascular superoxide dismutase (SOD), was kept frozen at −80°C until analysis. The activity of the antioxidant enzyme was then assessed with a commercial kit (Superoxide Dismutase Assay, Cayman Chemicals, Ann Arbor, MI).
All solutions were prepared daily. Five-hydroxytryptamine creatinine sulfate (serotonin), bradykinin, chloroform, EDTA, indomethacin, ketanserin, probucol, propranolol, prostaglandin F2α, mannitol, and methanol were purchased from Sigma (Oakville, Ontario, Canada). Ethylene glycol tetra-acetic acid was purchased from MP Biochemicals (Aurora, OH) and sucrose from EMD Chemicals (Gibbstown, NJ). The placebo was purchased from ODAN Laboratoires Ltée (Montreal, Quebec, Canada).
Statistical Analysis and Data Interpretation
Relaxations are expressed as a percentage of the maximal contraction to prostaglandin F2α (for serotonin) and to serotonin (for bradykinin) for each group and presented as mean±standard error of the mean (±SEM). Repeated measure analysis of variance studies were performed to compare concentration-response curves. Analysis of variance studies followed by Dunnet pairwise comparisons, using the untreated aortic banded group as reference, were used for the comparison of cardiac fibrosis, myocyte hypertrophy, basal production of cGMP, plasma levels of NOX and angiotensin II, lipid peroxidation, and vascular activity of SOD. The EC50 and the Emax were measured from each individual concentration-response curve using a 5-parameter logistic function with SigmaPlot curve-fitting software. The pD2 value reported is the negative log of the EC50. For these parameters, differences between groups were evaluated with a 2-tailed unpaired Student t test. A P value less than 0.05 was considered statistically significant.
The ratio of aorta dimensions was assessed to determine whether the banding should induce the development of LVH, especially in the third group since the treatment was started on the day of the banding. Aortic diameter at the banding site divided by aorta diameter distal to banding represents a ratio of the decrease in aorta lumen dimension and, indirectly, the increase in afterload. In group 2, a 40% decrease in the aortic ratio induced the development of LVH 60 days after surgery whereas group 1 had no decrease in the ratio (P=0.001). All treated groups showed no statistically significant difference in the aortic ratio in comparison with group 2 (Table 2), suggesting that the development of LVH would have occurred without probucol treatment. Moreover, in the fourth group, the echocardiographic measures at day 30, before the beginning of the treatment, demonstrate that the left ventricle/body weight ratio was 3.99±0.44, a value that was not significantly different from the one obtained after 60 days of banding in the untreated group (4.13±0.21). In the fifth group, the echocardiographic data on day 60, before the beginning of the treatment, revealed a left ventricle/body weight ratio of 4.25±0.32 versus 4.13±0.21 for the untreated group, attesting to a similar degree of LVH severity. Finally, at the day of the killing, the ratios of left ventricle/body weight and heart/body weight were significantly increased in group 2 versus group 1, whereas they were significantly smaller in treated groups compared with group 2 (Table 2) (P<0.05).
Hearts from group 2 developed LVH, as assessed by cardiomyocyte transversal area, and treatment with probucol statistically decreased cardiomyocyte hypertrophy in the 3 actively treated groups (Table 3) (P<0.001). Interstitial fibrosis was approximately 5 times greater in group 2 in comparison with group 1 (ratio fibrosis/total area in 10 to 12 subendocardial fields/animal). In the 3 probucol groups, there was a significant decrease in the left ventricular fibrosis versus group 2 (Table 3) (P<0.001).
Vascular Reactivity Studies
Dose-response curves of coronary rings from group 2 demonstrated a statistically significant difference compared with group 1 (P<0.05), confirming the presence of an endothelial dysfunction (Figs. 1, 2). On the other hand, dose-response curves of the rings from the 3 treated groups showed a statistically significant increase of relaxations to serotonin (Fig. 1) and bradykinin (Fig. 2) (P < 0.05), compared with curves obtained with rings from group 2.
There was a statistically significant decrease in the maximal relaxation (Emax) to serotonin and bradykinin in group 2 compared with controls (Table 4), accompanied by a decrease in sensitivity, attested by the right shift of the relaxation-concentration curve in group 2 for both agonists (pD2). In probucol-treated groups, the treatment did not affect the sensitivity in comparison with the untreated group for both serotonin and bradykinin, but significantly improved the maximal relaxation (Table 4).
Assessment of Endothelial Function
Plasma Nitrite and Nitrate (NOX) Levels
Values obtained from the untreated AB group were significantly decreased in comparison with group 1, confirming a reduction in NO• production associated with LVH (P<0.001) (Fig. 3). Nitrites/nitrates ratio was statistically increased in groups 3 and 4 in comparison with group 2 (P≤0.001). The fifth group showed no significant difference with group 2.
Coronary Artery cGMP Levels
Bioavailability of NO, assessed by cGMP levels, was significantly decreased in group 2 in comparison with values obtained from group 1 (P<0.01). The 3 treated groups showed a significant increase in vascular cGMP content in comparison with group 2 (P<0.01) (Fig. 4).
Assessment of Oxidative Stress
Plasma Angiotensin II Levels
Plasma angiotensin II levels were increased in group 2 compared with the control group (P<0.001). Probucol significantly decreased plasma angiotensin II levels in the 3 treated groups in comparison with group 2 (P<0.001) (Fig. 5).
Plasma Lipid Hydroperoxide Levels
In group 2, the levels of hydroperoxides were significantly increased in comparison with group 1 (P=0.003) (Fig. 6). Plasma hydroperoxide levels from the treated groups were significantly decreased relative to group 2 (P≤0.008), independent of the cholesterol lipid-lowering property of probucol as no significant variations were detected in cholesterol content between treated and untreated groups (data not shown).
The activity of the antioxidant SOD had a trend toward an increase in group 2 compared with group 1 (P=0.07). There were no significant differences in SOD in probucol groups compared with group 2 (Fig. 7).
The major findings of the present study are that, in this porcine LVH model, the antioxidant probucol: (1) prevented LVH and interstitial fibrosis; (2) significantly improved endothelium-dependent relaxations of epicardial coronary arteries; (3) increased production and bioavailability of NO•; and (4) reduced levels of plasma angiotensin II, a strong activator of NAD(P)H oxidase.
Oxidative stress, mainly by the action of •O2−, plays an important role in the development of cardiomyocyte hypertrophy. In fact, in cultured rat hypertrophied cardiomyocytes, the hypertrophy is subsequent to an increased formation of •O2− and activity of NAD(P)H oxidase.9 Moreover, in a guinea pig model of pressure overload, the expression of multiple subunits of this pro-oxidant enzyme is increased in the myocardium of the left ventricle, generating more •O2−.16 The antioxidant effects of probucol may explain at least in part its ability to prevent LVH in this model. This scavenging action of probucol has been demonstrated in vitro, in cultured neonatal rat hypertrophied cardiomyocytes.9 Furthermore, NO• is known to have an antihypertrophic effect on the myocardium.17 In a condition of oxidative stress, NO• is scavenged by •O2− so that it is less available for cardiomyocytes. After antioxidant therapy with probucol, NO• is more accessible to cardiomyocytes to inhibit hypertrophy.
The presence of interstitial fibrosis in this porcine model of pressure-overload LVH was significantly decreased in the 3 groups treated with probucol. In a rat model of LVH, the genesis of fibrosis is subsequent to increased •O2− formation.9 This observation might explain in part the beneficial impact of probucol on the development of interstitial fibrosis in the myocardium. In fact, after left anterior descending artery ligation in a rat model, treatment of 80 days with probucol 1 mg/kg/d partially decreased fibrosis and improved left ventricular remodeling.14
As demonstrated in a previous study from our laboratory, no statistically significant difference was present in dose-response curves of rings between untreated and control groups for endothelium-independent relaxations (sodium nitroprusside, an NO• donor) and for endothelium-dependent receptor-independent relaxations (calcium ionophore A23187).10 Hence, these agonists were not tested on rings from the treated groups to assess the impact of probucol on vascular reactivity. On the other hand, epicardial coronary endothelium-dependent vasodilations are statistically decreased in the untreated group in comparison with the control group, confirming the presence of an endothelial dysfunction. The alterations in the endothelial cell signaling pathway preferentially involves the Gi protein-mediated relaxations, as demonstrated by concentration-response curves to serotonin, and moderately the Gq protein-mediated pathway, as demonstrated by concentration-response curves to bradykinin. Vascular reactivity is significantly improved in the 3 treated groups in comparison with the untreated group. The action of probucol on vascular reactivity has also been demonstrated in hypercholesterolemic rabbits, after stimulation with acetylcholine, whereas it has no effect on controls.18 This improvement in endothelium-dependent relaxations could be explained in vitro by the increase in NO• bioavailability for vascular smooth muscle cells (VSMCs), demonstrated by the increase in vascular cGMP content.
The rapid regression of LVH (attested by echocardiography as a significant reduction in left ventricle/body weight ratio between days 60 and 90) and of the associated endothelial dysfunction even when probucol was started 60 days after banding (group 5) may be linked to the neutralization of •O2− by its phenol groups. In fact, NO• production was not significantly increased in these animals compared with the untreated group whereas its bioavailability was restored, suggesting decreased scavenging of •O2− by NO•. Ramasamy et al19 have demonstrated that probucol positively modulates the expression of eNOS and its functionality, in part by decreasing ONOO− levels, and subsequently tetrahydrobiopterin (BH4) oxidation. As NO production levels have not been restored in the group beginning treatment once the endothelial dysfunction was established, this may be suggest that the duration of the treatment with probucol was not sufficient to modulate the possibly decreased expression of eNOS. On the other hand, the decreased NO synthesis by endothelial cells may not be the consequence of a failure of the action of probucol action on eNOS. This decrease may be explained by other factors including asymmetric dimethylarginine, an endogenous eNOS inhibitor of L-arginine, known to be increased in pathologic states including coronary artery disease and heart failure.20 VSMCs possess the same capacity to respond to NO• in epicardial coronary arteries of LVH and control groups, demonstrated by similar concentration-response curves to SNP.10 In this way, the restoration of NO• bioavailability was sufficient to obtain a comparable improvement in vascular reactivity in the 3 treated groups compared with the untreated group, as the downstream cascade is intact in VSMCs. On the other hand, most tissues contain endogenous peroxidases (eg, glutathione peroxidase), which decrease hydroperoxide concentrations and probucol, owing to its antioxidant effect on superoxide anion, inhibiting further lipid peroxide generation. These 2 factors combined may explain the regression of the oxidative load demonstrated by lipid hydroperoxide.
Cardiac hypertrophy has been shown to be mediated by •O2−.9 Our results show that oxidative stress is required not only for the induction but also for the maintenance of LVH. This raises the intriguing possibility that a powerful synthetic antioxidant may be used to induce regression of LVH. The reduction in interstitial fibrosis and cardiomyocyte transversal area was, however, slightly less marked when probucol was started 60 days after banding than when it was started earlier. The significance of these findings in patients with LVH of different etiologies will have to be determined in clinical trials.
In a previous study from our laboratory, Malo et al12 have demonstrated that an in vitro supplementation with methyltetrahydrobiopterin, an analog of the cofactor BH4, significantly improved epicardial coronary endothelium-dependent relaxations in this porcine model. In fact, ONOO− possesses the capacity to oxidize BH4, inducing a functional uncoupling of eNOS. This promotes oxidative stress instead of generation of NO•.21 Treatment with probucol may have induced a decrease in ONOO− production, increasing nonoxidized BH4 and restoring the functional coupling of the enzyme for an increased production of NO•. The increased NO• production in probucol-treated groups may also be the result of an increased expression of the enzyme eNOS. Indeed, probucol possesses the capacity to positively modulate eNOS expression, as demonstrated in a study using cultured bovine aortic endothelial cells.19
Angiotensin II is a strong activator of the NAD(P)H oxidase, even at a concentration under the threshold that augments blood pressure. Following AB in a rat model, the rennin-angiotensin system (RAS) is rapidly activated, acting on NAD(P)H oxidase and generating important levels of •O2−.22 In the untreated group of thepresent study, the RAS was activated, as demonstrated by the significant increase in angiotensin II levels. On the other hand, the treated groups showed a significant decrease in angiotensin II, suggesting an impact of probucol on RAS. In a dog model of pacing-induced heart failure, renin and angiotensin II levels were also decreased after probucol treatment.23
Left ventricular hypertrophy secondary to AB is characterized by decreased NO• bioavailability, increased oxidative stress, and alterations in endothelium-dependent relaxations of epicardial coronary arteries. In this porcine model, the antioxidant probucol prevented LVH and interstitial fibrosis, improved coronary artery endothelial function, increased NO• bioavailability, and reduced lipid hydroperoxide and angiotensin II levels. Considering the negative prognostic and pathophysiologic implications of LVH and endothelial dysfunction, clinical studies should be performed to assess the effects of probucol or similar agents in patients with established LVH or those at risk for the development of LVH and its deleterious consequences.
The authors thank Marie-Pierre Mathieu, Steve Laurendeau, Olivier Bouchot, Simon Maltais, and Émilie Rény-Nolin for their technical assistance, and Karine Tétrault for statistical analysis.
© 2006 Lippincott Williams & Wilkins, Inc.