|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 2007; 49:822-825, doi:10.1016/j.jacc.2006.11.025
(Published online 6 February 2007). © 2007 by the American College of Cardiology Foundation |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Cardiology, VA Medical Center and University of Minnesota, Minneapolis, Minnesota.
* Reprint requests and correspondence: Dr. Yellaprada Chandrashekhar, Division of Cardiology 111c, VA Medical Center/University of Minnesota, One Veterans Drive, Minneapolis, Minnesota 55417. (Email: shekh003{at}umn.edu).
The universe is full of magical things waiting for our wits to grow sharper.Eden Phillpotts (1)
Heart failure (HF) is characterized by adverse cardiac structural changes, including myocellular hypertrophy, interstitial disruption, and a reduction in number of cardiomyocytes (2–4). These changes, collectively called cardiac remodeling, cause chamber enlargement/dysfunction, mediate progression, and predict adverse prognosis in patients with HF. Traditionally, 2 major patterns have been described: concentric remodeling, related to pressure overload (PO), in which increased mass is associated with normal chamber volume, and eccentric remodeling, related to volume overload (VO), in which increased mass is associated with an increased chamber volume. Very often a mixed picture is present, and concentric remodeling might change to an eccentric one with worsening HF.
Despite controversy about these terms (5), they continue to have utility to practicing clinicians for describing morphology, identifying possible etiology, and predicting prognosis. However, clinicians have been less concerned about differences at a cellular level and more importantly, have not seen a need to adopt different treatment strategies based on remodeling patterns. Thus, vasodilator therapy, so effective in ischemic and dilated cardiomyopathy, has been tried in patients with VO remodeling after aortic regurgitation (AR) and mitral regurgitation (MR) with controversial and mostly negative results (6). The report by Ryan et al. (7) in this issue of the Journal helps to place some cellular level background to clinical therapeutic strategies. This group extends their previous work, showing inefficacy of angiotensin-converting enzyme inhibition (ACE-I) in VO remodeling in yet another condition (aortocaval fistula [ACF]). Their study had 2 important findings. First, early left ventricular (LV) dilatation after an ACF is caused by a rapid loss of interstitial collagen (caused by a bradykinin [BK]–matrix metalloproteinase [MMP] mediated degradation), without cardiomyocyte remodeling. A BK2 receptor antagonist (Hoe 140) largely attenuated this early LV dilatation. Second, ACE-I did not attenuate LV dilatation after ACF. This study has implications for therapy in VO heart failure. A brief overview of the pathogenesis of VO versus PO remodeling and some related issues might help to put this study into a clinical perspective.
 |
Components of Remodeling: Myocyte Versus Interstitium |
|---|
 Top Components of Remodeling:... Why Is ACE-I Less...ACE-I, BK, and MMP...Clinical PerspectiveReferences |
|---|
The true mystery of the world is the visible, not the invisible.Oscar Wilde (8)
Ryan et al. (7) found that LV dilatation in VO was caused by connective tissue dissolution and not myocyte remodeling. Does it matter what components make up the remodeled heart? All forms of remodeling respond with some change in cell volume, number, and collagen. The degree of change in cardiomyocyte versus the interstitium is, however, dynamic and varies with the stimulus (PO vs. VO) as well as its duration. In general, volume overload, with its diastolic wall stress, is initially geared to accommodating a larger volume; it accomplishes this through a lack of increase (or reduction) of interstitial collagen rather than a lengthening of the myocyte. Subsequently, the myocyte starts to hypertrophy. These models often do not show contractile dysfunction even with significant remodeling (9) until very late. This is in agreement with clinical experience in MR and AR. Pressure overload, with high systolic wall stress, is initially geared to accommodate high pressure; it accomplishes that with cellular hypertrophy and a somewhat later increase in interstitial fibrosis. Significant fibrosis often marks the onset of decompensation (transition from hypertrophy to failure) (9). In both VO and PO, length-to-width ratio is initially preserved. Clinical decompensation is often marked by a further increase in length without a proportionate increase in width (3). Many of the clinical heart failure syndromes have a mixed picture of cellular hypertrophy and increased fibrosis.
Is one component more important than the other? Although both of these elements are changed in most heart failure syndromes, one can change without the other (10). There is emerging doubt regarding whether cardiomyocyte remodeling is the most important defect in the remodeling heart. First, all cardiomyocyte remodeling is not bad and we can now differentiate and manipulate pathways to induce good vs. bad cardiomyocyte hypertrophy (11,12). Second, there is growing evidence, albeit controversial, that the remodeled cardiomyocyte does not necessarily contribute to whole-heart dysfunction (13,14). Finally, transgenic studies have shown that florid heart failure can coexist at a time when there is dissociation between cardiomyocyte hypertrophy and interstitial tissue changes; hearts can manifest fibrosis and contractile dysfunction without hypertrophy (10), although some forms of obvious hypertrophy can avoid heart failure (15). Similarly, one can selectively modulate contractility and hypertrophy independent of each other (12,16), and inhibiting hypertrophy is not essential to improving function in the failing heart.
Fibrosis seems to be detrimental in ventricular remodeling of nearly all etiologies, but this conviction may change with further incisive investigations. Myocardial injury generates fibrosis both at the site of and remote from the initial injury (17). A dynamic interaction between extracellular matrix formation, qualitative alterations (e.g., type I/type III), and degradation is an ongoing element of ventricular remodeling. Significant ventricular dilatation can occur even when collagen synthesis exceeds degradation, related to the type of collagen laid down (e.g., the more elastic type III), change in collagen cross-linking, or alterations in interaction with adhesion molecules or cardiomyocytes (9). Not surprisingly, decompensation is associated with increasing fibrosis. Conditions with significant remodeling that have predominant fibrosis (and little cardiomyocyte hypertrophy) respond to modulation of the fibrosis (18). Finally, the benefit seen with MMP inhibition strengthens the role of fibrosis in all forms of heart failure. The data from Ryan et al. (7), from MMP knock out studies, and from inhibiting endothelin 1 suggest that any deviation from normal collagen, either less or more, might have pathogenetic implications. It thus seems that the proportion of changes in cell volume, number, and collagen might also affect strategies to attenuate remodeling.
|
Why Is ACE-I Less Effective in VO Remodeling? |
|---|
|
TopComponents of Remodeling:...Why Is ACE-I Less...ACE-I, BK, and MMP...Clinical PerspectiveReferences |
|---|
We shall never cease from exploration and the end of all our exploring will be to arrive where we started and know the place for the first time.T. S. Eliot (8)
Ryan et al. (7) did not find a benefit of ACE-I in ACF. This raises a question about where ACE inhibitors act and whether there are any predictors that separate VO and PO in terms of therapy. A complete renin–angiotensin system exists in various cells of the heart (15). Locally produced angiotensin II (Ang-II) modulates cardiomyocytes and fibrous tissue at the site of myocardial injury as well as remote from it in both VO and PO. One of the main effects of Ang-II is augmenting fibrosis at the site of injury. These responses are transduced by AT-1 receptors, which are mainly expressed in cardiac fibroblasts. It might even be that the bulk of the Ang-II action is on modulating and sculpting the interstitium, maybe even more than myocyte remodeling. The ACE-I reduces cardiomyocyte remodeling and cardiac fibrosis in a number of cardiac injury models. So why did ACE-I therapy not benefit VO HF? There could be multiple reasons, in addition to the one postulated by Ryan et al. (7). A comparison of the degree of myocyte and chamber remodeling, collagen volume, and known effect of ACE-I is instructive. As seen in Figure 1, beneficial effects of ACE-I seem to be little correlated with myocyte size, myocyte function, and LV chamber size. However, ACE-I has consistently benefited conditions in which fibrosis is a strong component of remodeling, although ACE inhibitors have had little success in conditions in which fibrosis is not prominent. Consistent with the AT-1 receptor distribution data, one could thus postulate that ACE-I benefits are more predicated on there being a strong fibrous tissue response and less so on cardiomyocyte remodeling. Lacking significant fibrosis, ACE-I possibly had no major target to modulate in VO remodeling of ACF or MR. This construct needs further evaluation. This postulate in no way, however, negates the voluminous data about Ang II and ACE-I on myocyte hypertrophy, apoptosis, and other phenomena, but just establishes a preliminary correlation with the best predictor of ACE-I response.
|
|
ACE-I, BK, and MMP in the Remodeling Heart |
|---|
|
TopComponents of Remodeling:...Why Is ACE-I Less...ACE-I, BK, and MMP...Clinical PerspectiveReferences |
|---|
Two paradoxes are better than one; they may even suggest a solution.Edward Teller (19)
Finally, Ryan et al. (7) also showed that ventricular remodeling in the VO heart was mediated by the BK–MMP axis, and this was postulated to be the reason that ACE-I was ineffective despite the fact that the renin–angiotensin axis is very active in the VO heart. Once again it is instructive to place BK in the context of all forms of ventricular remodeling.
Bradykinin, because of its potent hypotrophic effects on the myocyte and interstitium via BK2 receptors, can strongly attenuate ventricular remodeling (20,21); ACE has a great proclivity to inactivate bradykinin, and ACE-I increases tissue bradykinin levels. Not surprisingly, a significant amount of the ACE-I effectiveness is contingent on BK increase (20). Ryan et al. (7) suggest that a significant amount of the ACE-I ineffectiveness in VO may also be contingent on BK increase. Unfortunately, they do not have data to show that long-term BK2 inhibition alone, or better yet given with ACE-I, would improve remodeling. This leaves the story incomplete. Second, although ACE-I did not benefit remodeling in ACF, it did not significantly worsen it either (even with the additional possibility that ACE-I concomitantly could also have increased another potent antifibrotic compound N-acetyl-seryl-aspartyl-lysyl-proline [22]). Presumably, if BK is responsible for adverse remodeling (by fibrous tissue dissolution, etc.), more BK after ACE-I theoretically should have made remodeling worse. This did not happen. Either the additional excess of BK is not needed for further remodeling (ceiling effect?) or other mechanisms are also at play.
Alternatively, because BK (either spontaneously or after ACE-I) is antiproliferative and antifibrotic, BK2 benefits (and thus ACE-I benefits) might depend on the degree of underlying fibrosis or myocyte hypertrophy. Although there are few comprehensive data, BK seems to have a somewhat greater effect on fibrous tissue compared with cardiomyocytes (18). It is also interesting that the AT2 receptors (which have potent antigrowth activities) seem to predominantly modulate fibrosis (via a BK2 pathway) and not myocyte hypertrophy after Ang II infusion or after myocardial infarction (23). Thus, one could logically postulate that the BK2 effect (or ACE-I–induced BK effect) might be most prominent and beneficial in conditions in which there is significant fibrous tissue proliferation, such as after myocardial infarction, in aortic banding, in renovascular hypertension, or in diabetic cardiomyopathy, in which attenuation of fibrous tissue might be most important. In VO, with little proliferation or even a net loss of fibrous tissue, BK augmentation might have a lesser effect. This fits with the known difference in the magnitude of benefits of ACE-I in PO versus VO remodeling (Fig. 1). The availability of serum markers for interstitial collagen turnover and newer imaging techniques might help us to clarify whether fibrous tissue quantitation will allow us to identify responders versus nonresponders to ACE-I in various forms of ventricular remodeling.
Are there other explanations for the ACE-I–BK–MMP theory in ACF? There remains the possibility that Hoe 140 increased blood pressure, and this might have reduced the "threatened blood pressure" response with ACF (24). Because vasodilation after ACF has a prominent role in salt and water retention and thus diastolic wall stress, any pressor agent could have reduced the stimulus to remodel and thus falsely show benefit. The investigators rightly used other pressor controls. For a similar degree of pressor response, Ang-II increased collagen volume (less than with Hoe 140), but caused myocyte hypertrophy while reducing remodeling. This lends credence to the collagen dissolution theory, but also suggests that interventions inducing controlled myocyte hypertrophy might also be beneficial (reduced wall stress theory). Second, even subpressor doses of ACE-I can directly reduce MMP generation and influence remodeling (25); effects on MMP generation via BK2 might be attenuated by those directly on MMP itself. Thus, ACE-I did not benefit or worsen the situation in ACF. The MMP level and activity in the chronic ACE-I group might have been instructive. Finally, endothelin is increased in ACF and has a similar effect on interstitial dissolution via mast cells and MMP. An endothelin antagonist might have benefits similar to those of Hoe 140. This would be an interesting area to explore (26).
|
Clinical Perspective |
|---|
|
TopComponents of Remodeling:...Why Is ACE-I Less...ACE-I, BK, and MMP...Clinical PerspectiveReferences |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopComponents of Remodeling:...Why Is ACE-I Less...ACE-I, BK, and MMP...Clinical PerspectiveReferences |
|---|
1-Adrenergic receptors prevent a maladaptive cardiac response to pressure overload J Clin Invest 2006;116:1005-1015.[CrossRef][ISI][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
