The role of statins in the deterioration of native and bioprosthetic aortic valves

From the Boston University School of Medicine and the Department of Cardiothoracic Surgery, Boston Medical Center, Boston, Massachusetts

Progressive aortic valve calcification leading to severe aortic stenosis has been attributed to the natural aging process and the hemodynamic stress placed on anatomically abnormal valve leaflets. There is now evidence to suggest that the underlying pathophysiology of aortic valve calcification may be related to other factors, such as abnormal lipid metabolism, endothelial dysfunction, and inflammation.

This review will (1) document the evidence suggesting that native and prosthetic aortic valve calcification is similar to the tissue processes responsible for atherosclerotic vascular disease, and (2) discuss how HMG-CoA reductase inhibitor (statin) therapy may play an important role in the reversal and stabilization of this process.

Natural history of aortic valve calcification Faggiano and colleagues studied the natural history of aortic valve calcification in 400 patients using serial Doppler echocardiography over a 4-year period.1 They found that aortic valve calcification is a progressive disorder, with nearly a third of patients developing some degree of aortic valve stenosis. Further insight into the pathophysiology responsible for the natural progression of aortic valve calcification to stenotic lesions was provided by Palta and colleagues.2 They found that the annual rate of reduction in aortic valve area was 0.1 ± 0.27 cm2 in 170 patients followed up with serial echocardiograms. Furthermore, they noted that patients with serum cholesterol levels above 200 mg/dL had a rate of aortic valve area reduction nearly twice that of patients with lower cholesterol levels (0.14 ± 0.35 cm2 versus 0.07 ± 0.19 cm2; P = .04). Similar findings were noted by Aronow and colleagues in a retrospective analysis of 180 patients with mild aortic stenosis.3 Independent predictors of the progression of aortic stenosis were male sex, smoking, hypertension, diabetes, and a low-density lipoprotein (LDL) cholesterol level above 125 mg/dL. These factors are similar to those responsible for the progression of atherosclerotic vascular disease. In this study, the use statins was associated with a decreased progression of aortic stenosis. The relationship between the progression of aortic valve calcification and hypercholesterolemia was further defined by Pohle and colleagues in a retrospective study

of 104 patients using electron beam computed tomography.4 Patients with LDL cholesterol levels below 130 mg/dL had significantly less progression of both aortic valve and coronary artery calcification.

Pathophysiology of calcific aortic stenosis

The pathophysiologies of calcific aortic stenosis and vascular atherosclerotic disease have many similarities. Calcified aortic valve leaflets explanted at the time of surgery have an increased number of inflammatory cells, including macrophages, foam cells, and T lymphocytes.5 The macrophages have been shown to produce osteopontin, a protein involved in tissue calcification.6 They also contain increased amounts of LDL cholesterol and lipoprotein, suggestive of abnormal cholesterol metabolism.7 Calcified aortic valve leaflets also show evidence of an increased inflammatory response, similar to the atherosclerotic pro-

cess. In addition to accumulation of macrophages and T cells, they have been shown to have increased levels of C-reactive protein8 and angiotensin-converting enzyme.9

Aortic valve calcification

and adverse cardiovascular events

In addition to similarities in pathophysiology, there is now evidence to suggest that patients with aortic valve calcification may also have an increased incidence of adverse cardiovascular events similar to patients with vascular atherosclerotic disease. Chandra and colleagues followed 425 patients who went to the emergency department with chest pain, 49% of whom had evidence of aortic valve calcification without significant gradients.10 After 1 year, patients with calcified aortic valves had an increased incidence of mortality from all causes (18.7% versus 2.4%; P < .001), an increased cardiovascular mortality (14.7% versus 1.4%; P < .001), and an increase in the combined end point of cardiovascular death or myocardial in-farction (MI; 16.7% versus 7.1%;

P = .003). The incidence of coronary artery disease (CAD) was similar

for both groups of patients. Aortic valve calcification was also associated with increased levels of C-reactive protein.

Similar adverse events were shown in the Cardiovascular Health Study, which consisted of 5,000 patients with aortic valve calcifications followed up for 5 years.11 These patients had a 40% increase in the incidence of MI and a trend toward an increased risk of angina, congestive heart failure, and stroke, regardless of the presence or absence of CAD. The increase in adverse cardiovascular events in both studies was unrelated to progression of valvular gradients. An increased incidence of endothelial dysfunction, as suggested by reduced brachial artery flow-mediated dilatation,12 and significantly increased levels of serum LDL cholesterol and lipoprotein levels found in patients with aortic valve calcification13 suggest that the same factors responsible for increased ischemic events in atherosclerotic vascular patients may also be the culprit in patients with aortic valve calcification.

The role of statins in aortic valve calcification

In view of the evidence of an increased inflammatory response, endothelial dysfunction, and altered cholesterol metabolism associated with aortic valve calcification, is there a role for statins in the treatment of these patients? In addition to blocking cholesterol synthetic pathways at the level of the conversion of HMG-CoA to mevalonate, statins also reduce intermediates, such as geranyl pyrophosphate and farnesyl pyrophosphate. These isoprenoid compounds play an important role in the regulation of the inflammatory response and endothelial function through their effects on macrophage function. Recent studies have shown that statins reduce the secretion of matrix metalloproteinases, which lead to the degradation of extracellular matrix and contribute to the deposition of calcium in these injured tissues.14 Statins have been shown to retard calcification deposits in coronary vessels,15 protect against bone fractures,16 and stimulate bone protein production, resulting in increased bone formation.17

Could these non—cholesterol-lowering “pleiotropic” properties of statins also play a therapeutic role in patients with aortic valve calcification? Bellamy and colleagues studied 156 patients with mild aortic stenosis, 38 of whom received statin therapy, over 3.7 years.18 There was a significant progression of stenosis in the untreated group (mean gradient, 22 ± 12 mm Hg versus 39 ± 19 mm Hg; P < .001), and aortic valve area decreased from 1.20 ± 0.35 cm2 to 0.91 ± 0.33 cm2 (P < .001). In contrast, the progression of stenosis was significantly reduced in patients receiving statins (P < .04). There was no relationship between the progression of aortic stenosis and total cholesterol or LDL cholesterol levels. Hence, the beneficial effects of statins could not be explained by changes in serum lipids alone, suggesting a pleiotropic effect.

Novaro and colleagues retrospectively studied 174 patients with mild to moderate aortic stenosis, of which 57 patients (33%) received some type of statin therapy.19 Over a 21-month follow-up period, the nonstatin group showed the anticipated annual reduction in aortic valve area of 0.11 cm2. The statin-treated group had a 45% reduction in the annual stenosis rate (0.06 cm2; P = .03). Patients receiving statins also had a smaller increase in peak and mean valvular gradients. There was a significant relationship between changes in serum LDL cholesterol and peak and mean gradients. Multivariable analysis showed that statin use was a significant independent predictor of smaller decreases in valve area (P = .01) and peak gradients (P = .02). Bioprosthetic aortic valves Primary tissue failure of bioprosthetic aortic valves, similar to changes in native aortic valves, is related to calcification in tissue cusps. Histologic studies of stenotic explanted bioprostheses show areas of inflammation with cellular infiltration of macrophages and T lymphocytes, and lipoprotein deposition in areas of calcification.6

Farivar and Cohn recently examined whether risk factors for atherosclerosis could also be related to bioprosthetic valve calcification and dysfunction.20 Their study consisted of 144 patients who had a bioprosthetic aortic or mitral valve removed and a case-control analysis of a group of 66 patients with explanted tissue valves compared with a group of 66 patients with intact functioning protheses with similar duration of implantation. In the retrospective cohort study, they found that serum cholesterol levels (P = .035), younger age at implantation (P = .014), and presence of CAD (P = .017) were associated with valve calcification using a univariate analysis. In a multivariate regression analysis, only mean cholesterol levels were linked to calcification (P = .02). In the case-control analysis, mean serum cho-lesterol levels of the explanted valves were significantly higher (189 mg/dL versus 163 mg/dL; P < .001) than in the group whose valves did not require explantation. In patients whose serum cholesterol levels were above 200 mg/dL, the odds ratio for valve explantation and re-replacement was 3.9 times higher. In the case-control study, patients whose valves were not explanted had a higher incidence of statin use (18% versus 6%; P = .018).

Antonini-Canterin and colleagues also looked at the relationship of statins and echocardiographically derived parameters of bioprosthetic aortic valve degeneration in 167 patients after 4 years of implantation.21 Throughout the follow-up period, 22 patients (13.2%) received statins. Statin therapy significantly decreased the annual rate of increase of peak velocity (P < .001), the rate of increase of mean gradient (0.54 mm Hg versus 2.47 mm Hg; P < .001), the rate of decrease of effective orifice area (0.031 cm2 versus 0.100 cm2; P < .001), the rate of decrease of indexed effective orifice area (0.019 cm2/m2 ver-sus 0.056 cm2/m2; P < .001), and the worsening of aortic insufficiency

(P = .022). A combined parameter of prosthetic degeneration progression, defined as the existence of either an annual rate of increase in peak velocity of 0.3 m/sec per year or more or worsening of aortic insufficiency with greater than one third degrees was more frequently found in the group of patients who did

not receive statins. The only factor associated with progression of bioprosthetic aortic valve degeneration was the absence of statin therapy

as assessed by univariate analysis

(P < .001). Patients treated with statins exhibited a reduced progression of bioprosthetic degeneration despite higher serum cholesterol levels, suggesting that the benefits of statin therapy were due to their anti-inflammatory, pleiotropic effects.

Conclusions and future

directions

There is now sufficient evidence to suggest that the augmented inflammatory response seen in patients with atherosclerotic vascular decrease may also be responsible for the progression of calcification seen in native and bioprosthetic aortic valve disease. Furthermore, it appears that statins may play an important role in stabilizing the progression of valvular deterioration. However, Pearlman noted that several important questions remain.22 Are all statins equally effective in decreasing calcifications, and will higher doses be necessary to achieve these pleiotropic effects? Will statins be effective as monotherapy, or will their benefits be enhanced if combined with other anti-inflammatory agents or drugs that alter calcium metabolism? When is the optimal time to begin statin therapy? Will it be equally effective in cases of mild stenosis in contrast with more moderate and severe stenoses? And finally, will statin therapy significantly alter the natural progression of aortic stenosis and avoid valve replacement, enhance the durability of bioprostheses already implanted, and, most importantly, improve long-term survival?

The studies cited in this review are all retrospective and nonrandomized. The types of statins used, their dosages, and the duration of therapy varied considerably. A specific level of LDL cholesterol was not targeted. In studies involving bioprosthetic valves, it is not known which echocardiographic parameters of valve function correlate with the inflammatory response actually witnessed on an explanted tissue valve. Furthermore, most bioprosthetic valves show their greatest incidence of deterioration after 7 years of implantation. None of these studies exceeded a mean of 4 years of follow-up.

Prospective, randomized clinical trials using established doses of a specific statin targeted to achieve finite LDL cholesterol levels will be necessary to determine the role of statin therapy for these patients. In the meantime, the evidence suggests that patients with both native and bioprosthetic aortic valves are susceptible to the same inflammatory changes seen in patients with atherosclerotic vascular disease. Clinicians may wish to consider initiating some form of statin therapy in these patients when native aortic valve calcification is present or after the implantation of a new bioprosthesis.