What happens to cirrhotic cardiomyopathy after liver transplantation?

Authors


  • Potential conflict of interest: Nothing to report.

Torregrosa M, Aguade S, Dos L, Segura R, Gonzalez A, Evangelista A, et al. Cardiac alterations in cirrhosis: reversibility after liver transplantation. J Hepatol 2005;42:68-74. http://www.sciencedirect.com/science/journal/01688278 (Reprinted by permission of the European Association for the Study of the Liver.)

Abstract

BACKGROUND/AIMS

Liver cirrhosis induces cardiac alterations. We aimed to define these alterations and assess their reversibility after transplantation.

METHODS

Cirrhotic patients (n=40) and controls (n=15) underwent echocardiography and stress ventriculography. Fifteen cirrhotics were reevaluated 6-12 months after transplantation.

RESULTS

Cirrhotics had higher left ventricular wall thickness (9.6 ± 1.2 vs. 8.8 ± 1.2mm; p< 0.05) and ejection fraction (73 ± 6 vs. 65 ± 4%,p< 0.001) than controls. Basal diastolic function was similar. During stress, patients with cirrhosis presented lower increases of heart rate, left ventricular ejection fraction, stroke volume and cardiac index (p< 0.05 for all), and diastolic dysfunction with lower ventricular peak filling rate (p= 0.001). Exercise capacity was reduced (48 ± 21 vs. 76 ± 24W;p< 0.001). Ascitic patients exhibited more diastolic dysfunction at rest and during stress compared with non-ascitic patients. Liver transplantation caused regression of ventricular wall thickness (10.2 ± 1.3 vs. 9.5 ± 1.2mm;p< 0.05), improvement of diastolic function, and normalization of systolic response and exercise capacity during stress (significant increases in heart rate, ventricular ejection fraction, stroke volume, and cardiac index;p< 0.05 for all).

CONCLUSIONS

Cardiac alterations in cirrhosis present with mild increases in ventricular wall thickness, diastolic dysfunction that worsens with ascites and physical stress, and abnormal systolic response to stress limiting exercise capacity. Liver transplantation reverses these alterations.

“A chance to cut is a chance to cure.”surgical proverb

In patients with cirrhosis, myocardial contractile responsiveness is impaired under stressful stimuli such as exercise or pharmacological challenge, a condition termed cirrhotic cardiomyopathy.1 This syndrome has remained an academic curiosity until very recently, initially because it was misdiagnosed as alcoholic cardiomyopathy, then later because it appeared to have little clinical relevance. Because overt severe heart failure is rare in patients with cirrhosis, some even doubt the existence of cirrhotic cardiomyopathy. Skeptics contend that all changes of this syndrome can be explained by a cardiac adaptation to the hyperdynamic circulation, or ventricular hypertrophy due to chronic volume overload.

However, the past decade has been edifying to those of us who believe in the syndrome. Recent research strongly suggests that cardiodepression in the face of sepsis contributes to the pathogenesis of hepatorenal syndrome.2, 3 In addition, the advent of significant cardiac stresses such as TIPS insertion, or major surgery has precipitated cardiac dysfunction, even overt left ventricular failure in some patients.4–7 Of course, the most intense cardiovascular stress is induced by the acme of all surgeries, orthotopic liver transplantation. This operation has underscored the clinical significance of ventricular dysfunction in several ways. Before the idea of cirrhotic cardiomyopathy was raised, transplant specialists were puzzled by the apparently inexplicable appearance of sometimes severe cardiac dysfunction in the postoperative period, often in patients with no previous history or risk factors for heart disease. The myriad cardiac complications of liver transplantation have recently been reviewed in detail.8, 9 However, a few points bear mention. Cardiac complications including hypotension due to postreperfusion syndrome, myocardial infarction, significant arrhythmias, heart failure, and sudden cardiac death afflict a significant percentage of patients in the immediate and delayed postoperative period. Moreover, the degree of cardiac reserve function may directly affect the survival chances of the transplanted patient. Those with higher cardiac output both preoperatively and after transplantation show better survival rates compared with those with lower cardiac output.10

Cardiac contractile function appears to undergo a biphasic or even triphasic response after transplantation. The immediate-early stresses of the transplantation surgery impose a significantly increased workload on the heart. Predominant among these is the rapid increase in arterial pressure and systemic vascular resistance, which abruptly increases the ventricular afterload. In conjunction with the typical volume challenges attendant with the surgery, these factors can precipitate acute left ventricular failure, manifesting as pulmonary edema. Clinical or radiographic evidence of pulmonary edema is surprisingly frequent, observed in 12% to 56% of patients during their hospitalization, usually in the first postoperative week.8, 9 These episodes almost always resolve with appropriate treatment. Although most episodes are mild, a minority present with overt or severe dilated cardiomyopathy. A severe but reversible cardiomyopathy was observed in 7 of 754 (1%) patients after liver transplantation in one series.11

Over the next few weeks, hypertension may be further worsened by calcineurin inhibitors. Whether this significantly increased afterload actually induces further cardiac dysfunction remains unsettled. Therapondos and colleagues12 sequentially studied cardiac structure and function at several time points up to 3 months after transplantation. Compared with the immediate postoperative period, diastolic function further deteriorated, and the degree of ventricular hypertrophy slightly increased. Other markers of cardiac distress such as serum B-type or brain natriuretic peptide (BNP) levels also increased at the 3-month mark compared with pretransplantation. These authors speculated that the changes reflected cardiotoxicity of calcineurin inhibitors, particularly tacrolimus.12 However, whether these drugs are even cardiotoxic remains open to doubt. Deterioration of cirrhotic cardiomyopathy remains a viable alternative explanation for those results, as the authors admitted.

With this background, the results of Torregrosa and colleagues13 from Barcelona show a final phase of improvement in cardiac parameters at a later time point. Using both echocardiography and radionuclide angiography, they carefully documented complete reversal of almost all abnormal cardiovascular parameters in a subset of 15 of their 40 patients. Indices of both systolic and diastolic function reverted to normal when studied an average of 9 months after transplantation. They also documented improved cardiac workload and exercise capacity in their cohort. This study is thorough and well executed with very few weaknesses.

Some findings of this study elicit no controversy, as they confirm or support previous work. First, patients with alcoholic versus. nonalcoholic cirrhosis were hemodynamically indistinguishable. Second, diastolic dysfunction was more pronounced in those with ascites versus nonascitic patients, supporting previous observations that the degree of cirrhotic cardiomyopathy tends to correlate with extent of liver failure.4–7 Third, the patients with prolonged QT interval normalized this electrocardiographic abnormality after transplantation. In this regard, a rate-corrected QT interval greater than the accepted upper limit of normal (440 msec) is found in approximately 30% to 60% of patients with cirrhosis, and generally improves after liver transplantation.14, 15

However, the main finding of this study apparently contradicts an earlier Spanish study. The only previous description of cardiac function pretransplantation and posttransplantation was reported by Acosta and colleagues16 from Murcia. These authors performed baseline pre-transplantation echocardiography, then repeated these measurements a mean of 21 months after transplantation (range, 13-40 months) in 30 patients. Left ventricular sizes did not change significantly, but diastolic function deteriorated from baseline values. The E/A ratio, defined as the ratio of early ventricular diastolic filling wave velocity divided by the late (or atrial) filling velocity, is accepted as an index of ventricular diastolic compliance. In other words, a stiff, noncompliant ventricle displays a low E/A ratio. In the Murcia study, E/A ratio decreased significantly from 1.32 to 1.01 (normal, >1.0), and peak filling rate, another index of diastolic function, also worsened. Moreover, whereas 20% of the patients showed abnormal E/A values preoperatively, 60% had abnormal E/A on the second examination.16 How can we reconcile these contradictory studies? The answer is unlikely to be different study populations, because the two cities are only 500 km apart on the Mediterranean coast of Eastern Spain.

We believe that differences in the protocols to examine cardiac function underlie the discrepant results. Specifically, the Murcia group did not stress the heart to elicit evidence of attenuated responses, whereas the Barcelona investigators examined responses both at rest and with exercise stress. Indeed, the two datasets may not be that discrepant after all. The resting E/A ratio in the Barcelona study actually showed an insignificant decline on the second examination, from 1.1 to 1.0. Only with exercise stress did the altered ventricular function become manifest. Because a hallmark of cirrhotic cardiomyopathy is the subnormal response to stressful stimuli, the fact that studies that do not stress the ventricle fail to detect the presence of this syndrome is not surprising.

A weakness of the Barcelona study is the lack of sequential repeated measurements. Presumably, patients progressed from possible worsening of hemodynamics during the first 3 months to regression of these changes, and finally to complete normalization by a mean of 9 months after transplantation. In the lack of several repeat measurements, the exact time sequence of such changes remains conjectural. Is it plausible that the further deterioration of cardiac structure and function as described by Therapondos and colleagues in the first 3 months reverses and then improves all the way back to complete normalcy in only 6 more months? The answer seems to be yes. Hepatologists accustomed to the glacial pace of the progression of liver fibrosis in most diseases, and the even slower rate of fibrosis regression with removal of the injurious factor, may be surprised to discover that cardiac remodeling, befitting an organ that contracts 100,000 times daily, is lightning-quick by comparison. For example, relief from aortic stenosis by valve replacement results in decreased ventricular wall thickness within a few weeks, with almost complete remodeling by 6 months.17 Even more pertinent is the recent study by Pozzi and colleagues,18 who administered the aldosterone receptor antagonist potassium canrenoate for 6 months to patients with cirrhotic cardiomyopathy. This treatment significantly reduced indices of ventricular hypertrophy such as the left ventricular posterior wall thickness and end-diastolic volume, although measures of diastolic function such as the E/A ratio remained unaltered.18

The Barcelona study was unable to suggest a mechanism by which improvement occurs. Our experimental animal studies, mostly in the 4-week bile duct–ligated (BDL) rat, have suggested several possible cellular mechanisms underlying cirrhotic cardiomyopathy.4, 7, 8 Although the intense jaundice of the BDL rat directly inhibits cardiomyocyte adenylate cyclase,19 thus contributing to contractile dysfunction, that other nonjaundiced rat models of cirrhosis show similar cardiodepression is reassuring.19 Nevertheless, direct extrapolation of results in this and other animal models of cirrhosis to the human condition should be done cautiously. That said, mechanistic abnormalities in these models include changes in the cardiomyocyte plasma membrane beta-adrenergic receptor system, cholesterol content of the lipid bilayer leading to decreased membrane fluidity, and increased activity of cardiodepressant systems such as nitric oxide, carbon monoxide, and endogenous cannabinoids.4, 7, 8, 20 Cardiotropic hormones such as the renin-angiotensin- aldosterone system also may play a role in ventricular hypertrophy.18 Relatively little is known about improvement in all these altered systems after liver transplantation; this should be addressed in future work.

In summary, cirrhotic cardiomyopathy joins the list of cardiovascular and other extrahepatic complications of cirrhosis that may be corrected after liver transplantation. Even though cardiac function may worsen transiently in the first few days to weeks, this study demonstrates normalization of structure and function by 9 to 12 postoperative months. The exact time sequence of this correction, as well as the underlying cellular mechanisms, remains to be clarified in future studies. Torregrosa and colleagues have contributed a crucial step in improving our understanding of cardiac function in cirrhosis after liver transplantation.

Ancillary