Journal of Hepatology
Volume 44, Issue 5 , Pages 994-1002, May 2006

Cardiac electrophysiological abnormalities in patients with cirrhosis

Dipartimento di Medicina Interna, Cardioangiologia ed Epatologia, Alma Mater Studiorum—Università di Bologna, Semeiotica Medica Policlinico S. Orsola-Malpighi Via Albertoni, 15 40138 Bologna, Italy

published online 23 January 2006.

Article Outline

 

Back to Article Outline

1. Introduction 

Abnormalities in cardiac electrophysiology are well documented in patients with liver cirrhosis. The mechanisms underlying their occurrence are not fully understood, although it is likely that some of these defects have common pathophysiogical alterations at molecular level.

In order to attribute the electrophysiological abnormalities found in a patient to cirrhosis, it is important to exclude other causes of cardiac disease, such as ischemic, valvular, restrictive or congenital heart disease, as well as conduction abnormalities, arterial hypertension, lung disease or the use of anti-arrhythmic drugs. Patients who are actively drinking alcohol may have cardiac abnormalities due to alcohol abuse, and, as with cardiac abnormalities in patients with hemochromatosis, it may be difficult, if not impossible, to attribute an electrophysiologial abnormality to cirrhosis per se in these contexts.

This review will deal with the three abnormalities of cardiac electrophysiology most frequently encountered in cirrhosis: (1) chronotropic incompetence; (2) electromechanical uncoupling, and (3) electrocardiographic QT interval prolongation.

Back to Article Outline

2. Chronotropic incompetence 

Chronotropic incompetence consists of a defective cardiac response to physiological and pharmacological stimuli able to increase heart rate, and has long been recognized in cirrhosis.

Although resting tachycardia is a feature of the hyperdynamic circulatory syndrome [1], manoeuvres which either activate the sympathetic nervous system (SNS) or alter the sympathetic/vagal balance in favour of the SNS, or both, such as Valsava, ice-cold skin stimulation, tilting, isometric and isotonic physical exercise, do not evoke an adequate acceleration of heart rate [2], [3], [4], [5], [6]. Moreover, heart rate circadian variability is disrupted in patients with cirrhosis [7]. Impaired chronotropic responses may be exhibited by patients with cirrhosis of any cause, and their prevalence increases with the disease severity [4].

Chronotropic incompetence is a common feature of autonomic neuropathy [8] and impaired autonomic function is often found in cirrhosis [9]. However, as vagal impairment usually predominates [9], [10], and the sympathetic response to tilting or exercise is preserved or even enhanced [3], [4] the cause of this abnormality likely resides at the receptor and/or post-receptor level. Accordingly, the dose of β-agonists, such as isoproterenol, required for the heart rate to increase by 25bpm was threefold with respect to healthy subjects [11]. Likewise, in rats with biliary cirrhosis, the isoproterenol doses required for the heart rate to increase by 50bpm were four times higher than those required in sham-operated animals [12].

Impaired chronotropic responses may be diagnosed by heart rate monitoring during tests that activate the SNS activity or decrease vagal tone [13]. One problem that has arisen in the literature is that many studies have used healthy controls as a reference group, but there is no universal agreement on a reference range of values to define chronotropic competence in response to either drugs or manoeuvres.

Whether and to what extent abnormal chronotropic responses have a clinical impact in patients with cirrhosis is unknown. Do patients with impaired chronotropic responses have a shorter survival or a worse outcome of therapeutic procedures, such as TIPS insertion or major surgery including liver transplantation? It seems likely that impaired chronotropic responses would enhance the death risk in cirrhotic patients, but this has not been investigated. Recent studies on the hemodynamic and cardiac events associated with paracentesis-induced circulatory dysfunction [14], renal failure precipitated by spontaneous bacterial peritonitis [15] and hepatorenal syndrome [16] suggest that chronotropic incompetence may play an important role in the pathophysiology of such complications. All these conditions are due to a worsening in effective volemia, as witnessed by sharp increases in plasma renin activity and plasma norepinephrine, which are indirect markers of arterial circulation fullness. Contrary to what was generally thought, renal failure precipitated by spontaneous bacterial peritonitis and hepatorenal syndrome were not associated with a further drop in peripheral vascular resistance, but with a reduction in cardiac output. There may be several causes of this, such as impaired heart contractility and/or reduced cardiac pre-load. What is certain is that heart rate failed to increase, even in the face of a dramatic activation of the SNS [15], [16]. A similar hemodynamic pattern was seen in paracentesis-induced circulatory dysfunction. In fact, even though a further reduction in peripheral vascular resistance actually occurred, a striking feature was that cardiac index did not increase at all. Again, the main reason for this was that heart rate did not accelerate [14]. Possibly, these results are the first evidence on how chronotropic incompetence can assume clinical relevance in patients with cirrhosis (Fig. 1).

  • View full-size image.
  • Fig. 1. 

    Events associated with the occurrence of paracentesis-induced circulatory dysfunction (panel A) and renal failure induced by spontaneaous bacterial peritonitis (panel B). Both conditions are characterized by a worsening in effective volemia, as witnessed by the striking increases in plasma renin activity (PRA) and plasma norepinephrine concentration (NorE) and the reduction in mean arterial pressure (MAP). A drop in systemic vascular resistance (SVR) only occurred with paracentesis-induced circulatory dysfunction (panel A). In both conditions, an inadequate increase (panel A) or even a reduction (panel B) in cardiac index (CI) or output (CO) was seen. This can be attributed to several causes, such as impaired heart contractility and/or reduced cardiac pre-load. However, the failure of heart rate (HR) to increase (chronotropic incompetence) certainly played an important role. *= statistically significant change. Data derived from Ref. 14 (panel A) and 15 (panel B). In panel A, the value of the increase in PRA has to be multiplied by 10.

Back to Article Outline

3. Electromechanical uncoupling 

Electromechanical uncoupling consists of a disruption of the process leading to the contraction of cardiac myocytes following electrical excitation.

The measurement of systolic time intervals can non-invasively evaluate cardiac performance and allows an indirect assessment of electromechanical coupling (Fig. 2A). The reliability of this technique has been validated by cardiac catheterisation studies [17]. Studies evaluating cardiac function under rest and after isometric exercise showed a prolongation and defective shortening of some systolic time intervals (STI) under basal conditions as after exercise, respectively [4] (Fig. 3). The pattern of such changes suggests an abnormal electromechanical coupling, rather than mechanical failure in the ventricle. In fact, the abnormalities found involved electromechanical delay and the pre-ejection period, but not mechanical systole or the left ventricular ejection time, which are typically impaired in patients with heart failure. Dys-synchrony between electrical and mechanical systole in patients with cirrhosis has been confirmed by a different experimental approach. In fact, electromechanical coupling can also be evaluated by the simultaneous measurement of mechanical and electrical systole duration as assessed, respectively, by the aortic pressure curve, which provides an accurate appraisal of the mechanical components of the cardiac cycle, and the ECG tracing (Fig. 2B). This experimental approach, showed that the duration of mechanical systole in cirrhotic patients was normal, but its relationship with the duration of the electrical systole was clearly altered [18].

  • View full-size image.
  • Fig. 2. 

    Panel A. Systolic time intervals (STI) are derived from the simultaneous tracings of ECG, carotid artery pulse and phonocardiogram (PCG). The length of the total electromechanical systole is represented by QS2 interval, which begins at the onset of QRS complex and ends at the first high frequency vibrations of the aortic component of the second heart sound (S2). QS2 includes intervals identifying the mechanical systole, such as left ventricular ejection time (LVET; from the beginning upstroke to the though of the incisura of the carotid artery pulse) and mechanical systole (S1S2; from the first heart sound [S1] to the beginning of the aortic component of S2), and those influenced by electromechanical coupling (see Fig. 7), which are derived from the former intervals. These include electromechanical delay (QS1), isometric contraction time (ICT) and pre-ejection period (PEP). Alterations in LVET and S1S2 typically occur in the presence of impaired cardiac contractility, as in patients with heart failure. Isolated changes in PEP, QS1 and ICT suggest electromechanical uncoupling. Panel B. The duration of electrical and mechanical events associated with cardiac systole can be evaluated by the simultaneous reading of ECG tracing and aortic pressure curve. The measurement of QT interval gives an estimate of the duration of electrical systole. The mechanical components of the cardiac cycle (time to peak pressure [tP]; systolic time [tS]; diastolic time [tD]) can be measured on the aortic pressure curve. (TRR: time of one heart cycle). The assessment of the aortic pressure curve through catheterisation allows a more direct estimate of left ventricular pressure than the evaluation of the carotid artery pulse by pressure transducer. Hence, a better estimate of the duration of the mechanical components of the cardiac cycle can be achieved.

  • View full-size image.
  • Fig. 3. 

    Systolic time intervals in resting healthy controls and patients with cirrhosis. The total electromechanical systole (QS2) was prolonged in cirrhotic patients. However, this was not due to the prolongation of the mechanical components (mechanical systole: S1S2; left ventricular ejection time: LVET) as it happens with impaired contractility, but to the lengthening of systolic time intervals influenced by electromechanical coupling such as electromechanical delay (QS1) and pre-ejection period (PEP). This defect could be related to a reduced response to the adrenergic drive, which is known to shorten all systolic time intervals except mechanical systole and left ventricular ejection time. ICT: isometric contraction time. Data derived from Ref. 4.

Electromechanical uncoupling occurs irrespective of the etiology of cirrhosis, and is more pronounced in patients with advanced liver disease compared to those with well-compensated cirrhosis [4]. Since the activation of the SNS shortens all STI except mechanical systole and left ventricular ejection time [17], the above abnormalities may be secondary to an impaired response to the adrenergic drive. This seems to be cardiac in origin, since activation of the SNS was normal or even increased under baseline conditions and was increased further by exercise [4].

As regards the clinical relevance, it is not known whether electromechanical uncoupling per se influences the clinical outcome or prognosis of patients with cirrhosis.

Back to Article Outline

4. Prolongation of the QT interval 

Prolongation of the QT interval is the most widely recognized electrophysiological abnormality of cirrhosis. The QT interval represents the duration of the ventricular electric systole, and its prolongation predisposes to the development of ventricular arrhythmias [19]. QT interval duration can be determined by conventional ECG recording from the onset of the QRS complex to the end of the T wave (Fig. 4). The presence of a prominent U wave in ECG tracing, the meaning of which is debated but is likely correlated with ventricular after depolarizations [20], may render difficult the QT interval measurement. In such cases, most authorities agree that the end of QT interval corresponds to the nadir of the curve between the T and U waves [21] (Fig. 4).

  • View full-size image.
  • Fig. 4. 

    Normal ECG tracing. The various intervals are illustrated. The QT interval is measured from the onset of the QRS complex to the end of the T wave, defined as the return to T–P baseline. If a U wave is present, the QT interval is measured from the onset of the QRS complex to the nadir of the curve between the T and U wave. The duration of QT is calculated by measuring three consecutive intervals in each of the 12 ECG leads and averaged; alternatively, the maximal averaged value of the QT interval in any of the 12 leads is recognized (QTmax). The QT interval varies with heart rate, and its direct measurement should be corrected to avoid such an influence. The most frequently used formula was proposed by Bazzett [22]: QTc (QT corrected for heart rate) =QT/square root RR. In order to overcome possible pitfalls of Bazzett's formula, other ways to correct QT for heart rate have been suggested, such as: QTcub =QT/cubic root RR [25]; QTquadratic = QT/quadratic root RR [24].

QT interval varies with gender, being longer in females, and, chiefly, heart rate, as it shortens when heart rate accelerates. The most frequently employed formula to correct QT for heart rate (QTc) has been proposed by Bazzett (Fig. 4), who studied the relationship between QT and heart rate under resting and exercise [22]. Its reliability has been questioned as it overestimates QT length at high heart rates [23], as clearly shown with atrial pacing or administration of drugs such as atropine or isoproterenol [24]. A trend to overestimation was also seen in alcoholic patients with liver disease [25]. For this reason, other ways to correct QT for heart rate have been suggested [24], [25] (Fig. 4). However, almost all studies that addressed the issue of QT interval prolongation in cirrhosis are based on QTc. Moreover, the prevalence of a prolonged QT interval in patients with cirrhosis of different etiology, as evaluated by both QTc and QTcub, was almost the same [26].

Although the normal value of QTc is still open to question, 440ms is usually identified as the cut-off value. Modern 12 lead electrocardiographs usually provide the computed measurement of the QTc, which can also be measured manually in an ECG tracing where all 12 ECG leads are simultaneously recorded at 50mm/s (Fig. 4). In our experience [26], [27], the inter-observer variation co-efficient of the manual measurement of QT interval is less than 5%.

QT duration normally varies in the 12 ECG leads (QT dispersion). To quantify the extent of such heterogeneity in ventricular repolarisation, various indices have been used. One is based on the difference between the longest and the shortest interval measured in the 12 leads, the second is the relative dispersion of QT or QTc calculated as: standard deviation of QT/mean QT×100 [21].

An elevated prevalence of QT interval prolongation was first shown in patients with alcoholic liver disease [25]. Subsequent studies demonstrated that this abnormality is also common in patients with cirrhosis of different etiology [26], [28]. Thus, in one study [26], QT interval was longer than 440ms in 47% of cases in a large population of patients with cirrhosis, while 5% of control subjects showed this abnormality (Fig. 5). Interestingly, the impact of liver disease was such that the physiological difference in QT length between genders was lost. Several investigations have now shown that QT interval prolongation increases with the severity of liver disease, but can also occur in patients with well-compensated cirrhosis [26], [29]. In fact, its prevalence was 25% in class A of Child–Pugh classification [30], 51% in class B, and 60% in class C [26] (Fig. 5). The relationship between QT prolongation and severity of cirrhosis was further confirmed by following patients through liver disease progression, as a worsening of Child–Pugh score was associated with a further prolongation of the QT interval [26].

  • View full-size image.
  • Fig. 5. 

    Prevalence of prolonged (above 440ms) QTc interval in patients with cirrhosis belonging to Child–Pugh classes A, B and C, and in healthy controls. Such a prevalence is exceedingly high in patients, and increases in parallel with the severity of cirrhosis. Data derived from Ref. 26.

The mechanisms leading to QT interval prolongation in cirrhosis are unknown. In one study [26], multi-variate analysis showed that the independent factors predicting this abnormality were Child–Pugh score and plasma norepinephrine concentration, a marker of SNS activity. Increased activity of the SNS is common in advanced cirrhosis, and its impact on cardiac function may be magnified by vagal denervation, which is also common in this context [9]. Interestingly, acute β-blockade shortens the QT interval and abolishes its correlation with plasma norepinephrine concentration in cirrhosis [31]. Ongoing studies demonstrate that chronic β-blockade also shortens the QT interval (Bernardi M, unpublished observation) (Fig. 6). As a whole, these findings suggest that the chronic activation of the SNS that occurs in cirrhosis likely plays an important role in the genesis of QT interval prolongation.

  • View full-size image.
  • Fig. 6. 

    Effect of chronic β-blockade on the duration of QTc interval in patients with cirrhosis. A substantial shortening was seen in most subjects, the most striking reductions generally occurring in patients showing the highest baseline value of QT interval (correlation between baseline QTc interval duration and degree of shortening: r=0.70; P<0.001). QTcmax = frequency adjusted QT interval (Bernardi, M. Unpublished data).

Other pathogenetic mechanisms may contribute to QT prolongation in cirrhosis. Since it occurs in up to 80% of patients with non-cirrhotic portal hypertension, and worsens after TIPS placement in cirrhotic patients [27], shunting of splanch-nic blood, which may spread several ‘cardiotoxic’ substances into the systemic circulation, likely plays a role.

The improvement of the QT interval following liver transplantation [28] suggests that there is a functional defect(s) underlying QT interval prolongation. However, the recent finding that QT interval prolongation correlates with circulating concentrations of brain natriuretic peptide, the gene of which is predominantly expressed in myocytes of failing left ventricles, suggests that a sub-clinical cardiomyopathy may be responsible for its development [32].

The clinical relevance of QT prolongation in cirrhosis is still uncertain. The survival of patients with this abnormality is shorter than in those with a normal QT interval [26]. At least in part, however, this finding is due to patients with prolonged QT usually presenting a more advanced disease. In fact, the prolonged QT did not significantly influence mortality in Child–Pugh classes B and C, while it was associated with a shorter survival rate in class A. This suggests that this abnormality may be an important risk factor for death in the early stages of cirrhosis; later on, its potential prognostic impact is somewhat effaced by many other risk factors affecting patient survival.

Although the incidence of sudden death is higher in alcoholic patients with chronic liver disease and prolonged QT interval [25], sudden death is uncommon in cirrhotic patients. This likely reflects the fact that inter-lead QT interval dispersion, which facilitates the mechanisms of re-entry and identifies those subjects who are at high risk for ventricular arrhythmias [33], is usually normal in cirrhotic patients [27], [28]. However, an abnormal cardiac repolarisation may affect patient survival during complications of liver disease, such as gastro-intestinal bleeding, infections, or major surgery, such as liver transplant. Severe and fatal ventricular arrhythmias have been reported in patients treated with vasopressin during variceal bleeding, or undergoing plasma exchange [34], [35], [36]. Moreover, the QT interval in patients dying after liver transplantation was reported to be longer than that of survivors [28]. Given this background, it is advisable to be prudent when administering drugs that will affect heart repolarisation in patients with cirrhosis, such as anti-arrhythmics, i.e. disopyramide, procainamide, quinidine, sotalol and amiodarone, or other drugs which cause QT prolongation, such as clarithromycin, erythromycin and domperidone, since all of these have the potential to precipitate torsade de pointes [37].

Back to Article Outline

5. Common molecular defects underlying electrophysiological abnormalities 

The sympathetic drive influences heart rate and electromechanical coupling. Indeed, the pathways leading heart rate to increase and coupling of myocyte depolarisation and contraction have several steps in common [38] (Fig. 7): norepinephrine binding with β-receptors, receptor-stimulatory G protein interaction, consequent adenylcyclase stimulation, activation of cAMP-dependent phosphokinase A, and channel phosphorylation. Phosphorylation of Na+ channels (INa-B—inward Na+ background leak current) favours the inward pacemaker current, thus enhancing depolarisation of action potential (AP) phase 4; as a result heart rate accelerates. Phosphorylation of L-type Ca2+ channels and ryanodine receptors RyR2 favours calcium entry from the extracellular compartment and Ca2+ release from the sarcoplasmic reticulum, respectively. The resultant troponin C–Ca2+ complex initiates cross-bridge cycling between actin and myosin, which represent the molecular background for contraction.

  • View full-size image.
  • Fig. 7. 

    Receptor and post-receptor pathways following β1-adrenergic stimulation in the cardiomyocyte. Norepinephrine binding with β1-receptors leads to receptor-stimulatory G protein interaction, consequent adenylcyclase stimulation, activation of cAMP-dependent phosphokinase A, and channel phosphorylation. Phosphorylation of L-type Ca2+ channels and ryanodine receptors located in the sarcoplasmic reticulum favours calcium entry from the extracellular compartment (ICa-L: slowly decaying inward Ca2+ current) and Ca2+ release from the sarcoplasmic reticulum, respectively. The resultant troponin C–Ca2+ complex initiates cross-bridge cycling between actin and myosin, which represent the molecular background for contraction (electromechanical coupling). Phosphorylation of Na+ channels (INa-B: inward Na+ background leak current) favours the inward pacemaker current, thus enhancing depolarisation of action potential phase 4; as a result heart rate accelerates. Several receptor and post-receptor defects have been described in cirrhosis, such as β-adrenoceptor density reduction, altered G protein and adenylcyclase functions, altered physical properties of myocyte plasma membrane, which may lead to receptor and ion flux abnormalities, and reduced density and functional depression of L-type Ca2+ channels. These defects can account for chronotropic incompetence and abnormalities in electromechanical coupling.

Several receptor and post-receptor defects have been described in cirrhotic patients and animal models of cirrhosis. First, β-adrenoceptor density and sensitivity are reduced [12], [39]. Second, altered G protein and adenylcyclase functions have been found in different experimental models of cirrhosis and portal hypertension [40], [41]. Third, altered physical properties of myocyte plasma membrane may lead to receptor and ion flux abnormalities [39]. At last, reduced density and functional depression of L-type Ca2+ channels have been described [42], [43]. In conclusion, all these defects could account for both impaired chronotropic responses and electromechanical uncoupling.

The sympathetic drive also influences AP duration in myocytes and in Purkinje fibres (Fig. 8). The molecular mechanisms underlying the effects of the adrenergic system on AP duration are complex and not fully understood as yet. The activation of β1-adrenoceptors in ventricular myocytes leads to phosphokinase A-dependent phosphorylation of both L-type Ca2+ and K+ channels. This enhances, respectively, the slowly decaying inward Ca2+ current (ICa-L), which prolongs the AP plateau phase 2, and the transient outward (ITO) and delayed rectifier K+ currents (IK), which promote the repolarisation in phase 2 and 3, thus shortening AP duration [19]. There is evidence that one effect prevails on the other depending on the intensity of adrenergic stimulation [44], [45].

  • View full-size image.
  • Fig. 8. 

    The cardiac action potential is primed by the inflow of Na+ and Ca2+ ions (INa: inward fast Na+ current; ICa-T: inward T-type Ca2+ current) leading to the abrupt depolarisation in phase 0. Thereafter, the outflow of K+ (ITO-K: transient outward K+ current) initiates repolarisation (phases 1 and 2), a process which is counterbalanced by Ca2+ and Na+ influx (INaCa: electrogenic Na+- Ca2+ exchange current; ICa-L: slowly decaying inward Ca2+ current). K+ extrusion continues during phase 3 (IK: delayed rectifier K+ current), which restores the resting potential of phase 4. Several conditions, known to prolong the QT interval and listed in the figure, counteract K+ efflux. The sympathoadrenergic drive leads to events which favour both repolarisation, by enhancing K+ efflux, and depolarisation, enhancing Ca2+ entry in phase 2. In the presence of altered K+ fluxes, adrenergic stimulation leads to the prolongation of the repolarisation phases, and, hence QT interval. Ward and co-workers [46] have demonstrated altered K+ currents in ventricular myocytes of cirrhotic rats.

It is likely that ion channel abnormalities also underlie QT interval prolongation in cirrhotic patients. Interestingly, in ventricular myocytes from rats with chronic bile duct ligation, the action potential was found to be prolonged and whole-cell patch-clamp studies showed an impaired function of K+ channels responsible for ITO and IK currents [46]. As a result, it is conceivable that, in the presence of potassium currents defects, chronic adrenergic stimulation leads to the prolongation of action potential (and QT interval), possibly due to a prevalent effect on the L-type Ca2+-dependent slow inward current. This contention is supported by the observation that hereditary long QT interval syndromes, most of which are due to altered potassium channel expression [19], are clearly related to adrenergic stimulation or enhancement of sympathetic nervous system tone: typically, bizarre and dynamic long QT segments are observed during periods of β1-adrenergic stimulation [47]. This is the background for treating these patients with β-blockers or left-sympathectomy.

Back to Article Outline

6. Potential ‘cardiotoxins’ responsible for electrophysiological abnormalities 

The picture emerging from the studies cited above is that electrophysiological abnormalities in cirrhosis are due to β-adrenoceptor and post-receptor pathway defects and a generalized ion channel dysfunction which affects the different phases of AP. In addition to chronic β-adrenergic stimulation, which can reduce receptor density and sensitivity, these defects likely arise from myocyte exposure to cardioactive substances, many of which of splanchnic origin and gaining access to the systemic circulation through portasystemic shunts induced by portal hypertension. Elevated plasma levels of endotoxins [48] and cytokines, such as interleukin-1, interleukin-6, tumor necrosis factor-α [49] have been found in cirrhosis. Endotoxins reduce inward Ca2+ current through L-type Ca2+ channels [50], and widen the QRS complex in experimental settings [51]. Cytokines also interfere with heart electrophysiology, as they alter electromechanical coupling contributing to the reversible myocardial depression and β-adrenergic desensitisation observed in diverse clinical conditions [52]. The plasma of cirrhotic patients is also rich in bile salts [53], which may contribute to the altered membrane fluidity of myocytes [54] seen in experimental biliary cirrhosis [39]. Changes in plasma membrane fluidity have several consequences, by affecting the proper function of β-adrenoceptors, G protein and ion channels [39], [55], [56]. A further pathogenetic factor may be represented by hyperinsulinism, which is common in cirrhosis [57]. In fact, insulin increases the transmembrane potential of different cells, including cardiac myocytes [58]. Indeed, hyperinsulinemia is responsible for QT interval prolongation in healthy subjects undergoing euglycemic insulin clamp [59] and insulin-resistant individuals [60].

Heart electrophysiology in cirrhosis is likely influenced by autacoids, some of which mediate the effects of some of the substances cited above. Endothelin-1, a potent vasoconstrictor whose plasma concentration is increased in cirrhosis [61] due to reduced hepatic clearance [62], has been shown to induce changes in ion currents and inhomogeneous prolongation of AP in various mammalian cardiac preparations [63]. These effects are due to the suppression of several cAMP-dependent ionic currents, such as ICa and IK [64]. Proposed mechanisms for the arrhythmogenicity of endothelin-1 are prolongation or increased dispersion of AP, QT prolongation and development of early after depolarizations [65]. Surprisingly, no data are available on the relationship between endothelin-1 and heart electrophysiological abnormalities in cirrhosis.

An established feature of cirrhosis is the increased production of endothelial nitric oxide, which is involved in the genesis of the hyperdynamic circulatory syndrome [66]. In addition to its vasodilating property, nitric oxide also influences cardiac function. In fact, by activating the soluble guanylate cyclase, it inhibits the L-type Ca2+ channels [67] and the sarcoplasmic ryanodine receptors [68]. In bile duct ligated rat ventricles, both soluble cGMP levels and inducible nitric oxide synthase mRNA transcription and protein expression are increased [69]. The effects of the increased nitric oxide production in the heart can also follow cGMP-independent mechanisms, such as nitration/nitrosylation of cardiac proteins [70], [71]. Indeed, the nitration of cardiac proteins, mainly mitochondrial proteins involved in ion homeostasis, is markedly increased in rats with biliary cirrhosis and is prevented by the administration of N-acetylcysteine and l-Nitro-arginine methyl ester (l-NAME) [72]. Interestingly, increased heart content in nitric oxide could account for the resistance against epinephrine-induced arrhythmia seen in bile duct ligated rats, which was restored by l-NAME administration [73]. This may be a mechanism through which cirrhotic patients, despite frequently showing a severe QT interval prolongation, rarely develop ventricular arrhythmias.

Back to Article Outline

7. Conclusion 

Heart electrophysiological abnormalities frequently occur in cirrhosis, are roughly related to the severity of the disease and appear to be independent of its etiology. Liver transplantation is followed by their normalisation or attenuation, suggesting that they are functional in nature, although, at present, an organic component (cirrhotic cardiomyopathy) cannot be excluded. Although many potential mechanisms at molecular level have been identified, mostly consisting of ion channel dysfunctions, the pathophysiology of heart electrophysiological abnormalities still awaits an exact definition. Namely, the pathogenetic role of putative ‘cardiotoxins’, such as endotoxins, cytokines, bile acids, insulin, nitric oxide and endothelins needs to be assessed in both experimental and human cirrhosis.

Even more needed are clinical studies aimed at establishing the clinical relevance of these abnormalities, which remains elusive at present. In particular, the association between a given electrophysiological abnormality and either patient survival or the outcome of therapeutic manoeuvres such as TIPS placement or major surgery, including liver transplantation, should be established. Monitoring of cardiac performance during acute complications, such as bleeding and infection would also be warranted. This may allow us to understand whether an association exists between the presence (or appearance or worsening) of electrophysiological abnormalities and outcome, with particular reference to the occurrence of cardiovascular events, such as shock and arrhythmias, endangering patient survival.

Back to Article Outline

References 

  1. Kowalski HJ, Abelmann WH. The cardiac output at rest in Laennec's cirrhosis. J Clin Invest. 1953;32:1025–1033
  2. Lunzer MR, Manghani KK, Newman SP, Sherlock S, Bernard AG, Ginsburg J. Impaired cardiovascular responsiveness in liver disease. Lancet. 1975;2:382–385
  3. Bernardi M, Trevisani F, Santini C, Zoli G, Baraldini M, Ligabue A, et al. Plasma norepinephrine, weak neurotransmitters and renin activity during active tilting in liver cirrhosis. Relationship with cardiovascular homeostasis and renal function. Hepatology. 1983;3:56–64
  4. Bernardi M, Rubboli A, Trevisani F, Cancellieri C, Ligabue A, Baraldini M, et al. Reduced cardiovascular responsiveness to exercise-induced sympatho-adrenergic stimulation in patients with cirrhosis. J Hepatol. 1991;12:207–216
  5. Grose RD, Nolan J, Dillon JE, Errington M, Hannan WJ, Bouchier IA, et al. Exercise induced left ventricular dysfunction in alcoholic and non-alcoholic cirrhosis. J Hepatol. 1995;22:326–332
  6. Wong F, Girgrah N, Graba J, Allidina Y, Liu P, Blendis L. The cardiac response to exercise in cirrhosis. Gut. 2001;49:268–275
  7. Bernardi M, Trevisani F, De Palma R, Ligabue A, Capani F, Baraldini M, et al. Chronobiological evaluation of sympathoadrenergic function in cirrhosis. Relationship with arterial pressure and heart rate. Gastroenterology. 1987;93:1178–1186
  8. Freeman R. Autonomic peripheral neuropathy. Lancet. 2005;365:1259–1270
  9. Thuluvath PJ, Triger DR. Autonomic neuropathy and chronic liver disease. Q J Med. 1989;72:737–747
  10. Trevisani F, Sica G, Mainquà P, Santese G, De Notariis S, Caraceni P, et al. Autonomic dysfunction and hyperdynamic circulation in cirrhosis with ascites. Hepatology. 1999;30:1387–1392
  11. Ramond MJ, Comoy E, Lebrec D. Alterations in isoprenaline sensitivity in patients with cirrhosis: evidence of abnormality of the sympathetic nervous system activity. Br J Pharmacol. 1986;21:191–196
  12. Lee SS, Marty J, Mantz J, Samain E, Braillon A, Lebrec D. Desensitisation of myocardial β-adrenergic receptors in cirrhotic rats. Hepatology. 1990;12:481–485
  13. Fletcher GF, Balady G, Froelicher VF, Hartley LH, Haskell WL, Pollock ML. Exercise standards. A statement for healthcare professionals from the American Heart Association. Writing group. Circulation. 1995;91:580–615
  14. Ruiz-del-Arbol L, Monescillo A, Jimenez W, Garcia-Plaza A, Arroyo V, Rodes J. Paracentesis-induced circulatory dysfunction: mechanism and effect on hepatic hemodynamics in cirrhosis. Gastroenterology. 1997;113:579–586
  15. Ruiz-del-Arbol L, Urman J, Fernandez J, Gonzalez M, Navasa M, Monescillo A, et al. Systemic, renal, and hepatic hemodynamic derangement in cirrhotic patients with spontaneous bacterial peritonitis. Hepatology. 2003;38:1210–1218
  16. Ruiz-Del-Arbol L, Monescillo A, Arocena C, Valer P, Gines P, Moreira V, et al. Circulatory function and hepatorenal syndrome in cirrhosis. Hepatology. 2005;42:439–447
  17. Weissler AM. Current concepts in cardiology. Systolic-time intervals. N Engl J Med. 1977;296:321–324
  18. Henriksen JH, Fuglsang S, Bendsten F, Christensen E, Møller S. Dyssynchronous electrical and mechanical systole in patients with cirrhosis. J Hepatol. 2002;36:513–520
  19. Kass RS, Moss AJ, Long QT. syndrome: novel insights into the mechanisms of cardiac arrhythmias. J Clin Invest. 2003;112:810–815
  20. Di Bernardo D, Murray A. Origin on the electrocardiogram of U-waves and abnormal U-wave inversion. Cardiovasc Res. 2002;53:202–208
  21. Castellanos A, Kessler KM, Myerburg RG. The resting electrocardiogram. In:  Alexander RW,  Schlant RC,  Fuster V editor. The heart. 9th ed. New York: McGraw-Hill; 1998;p. 351–385
  22. Bazzett HC. An analysis of the time-relations of electrocardiogram. Heart. 1920;7:353–370
  23. Ward DE. Prolongation of QT interval as an indicator of risk of a cardiac event. Eur Heart J. 1988;9:139–144
  24. Kawataki M, Kashima T, Toda H, Tanaka H. Relation between QT interval and heart rate. Applications and limitations of Bazzett's formula. J Electrocardiol. 1984;17:371–376
  25. Day PC, James FWO, Butler JT, Campbell RWF. Q–T prolongation and sudden cardiac death in patients with alcoholic liver disease. Lancet. 1993;341:1423–1428
  26. Bernardi M, Calandra S, Colantoni A, Trevisani F, Raimondo ML, Sica G, et al. Q–T interval prolongation in cirrhosis: prevalence, relationship with severity, and etiology of the disease and possible pathogenetic factors. Hepatology. 1998;27:28–34
  27. Trevisani F, Merli M, Savelli F, Valeriano V, Zambruni A, Riggio O, et al. QT interval in patients with noncirrhotic portal hypertension and in cirrhotic patients treated with transjugular intrahepatic porto-systemic shunt. J Hepatol. 2003;38:461–467
  28. Mohamed R, Forsey PR, Davies MK, Neuberger JM. Effect of liver transplantation on QT interval prolongation and autonomic dysfunction in end-stage liver disease. Hepatology. 1996;23:1128–1134
  29. Bal JS, Thuluvath PJ. Prolongation of QTc interval: relationship with etiology and severity of liver disease, mortality and liver transplantation. Liver Int. 2003;23:243–248
  30. Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg. 1973;60:646–649
  31. Henriksen JH, Bendtsen F, Hansen EF, Moller S. Acute non-selective β-adrenergic blockade reduces prolonged frequency-adjusted Q–T interval in patients with cirrhosis. J Hepatol. 2004;40:234–239
  32. Henriksen JH, Gotze JP, Fuglsang S, Christensen E, Bendtsen F, Moller S. Increased circulating pro-brain natriuretic peptide (proBNP) and brain natriuretic peptide (BNP) in patients with cirrhosis: relation to cardiovascular dysfunction and severity of disease. Gut. 2003;52:1511–1517
  33. Day CP, McComb JM, Campbell RWF. QT dispersion: an indication of arrhythmia risk in patients with long QT intervals. Br Heart J. 1990;63:342–344
  34. Stein LB, Dabezies MA, Silverman M, Brozena SC. Fatal torsade de pointes occurring in a patient receiving intravenous vasopressin and nitroglycerin. J Clin Gastroenterol. 1992;15:171–174
  35. Faigel DO, Metz DC, Kochman ML. Torsade de pointes complicating the treatment of bleeding esophageal varices: association with neuroleptics, vasopressin and electrolyte imbalance. Am J Gastroenterol. 1995;90:822–824
  36. Nakasone H, Sugama R, Sakugawa H, Matayoshi R, Miyagi T, Maeshiro T, et al. Alcoholic liver cirrhosis complicated with torsade de pointes during plasma exchange and hemodiafiltration. J Gastroenterol. 2001;36:564–568
  37. Wood AJJ. Drug-induced prolongation of the QT interval. N Engl J Med. 2004;350:1013–1022
  38. Schlant RC, Sonnenblick EH, Katz AM. Normal physiology of the cardiovascular system. In: Alexander RW, Schlant RC, Fuster V, editors. The heart, 9th ed, New York: McGraw-Hill. p. 81–124.
  39. Ma Z, Meddings JB, Lee SS. Membrane physical properties determine β-adrenergic receptor function in cirrhotic rats. Am J Physiol. 1994;267:G87–G93
  40. Ma Z, Miyamoto A, Lee SS. Role of altered β-adrenergic receptor signal transduction in the pathogenesis of cirrhotic cardiomyopathy in rats. Gastroenterology. 1996;110:1191–1198
  41. Zavecz JH, Battarbee HD, O'Donnell JM. Cardiac beta-adrenoceptor-effector coupling in portal vein-stenosed rats. Am J Physiol. 1995;268:G410–G415
  42. Zavecz JH, Bueno O, Maloney RE, O'Donnell JM, Roerig SC, Battarbee HD. Cardiac excitation–contraction coupling in the portal hypertensive rat. Am J Physiol. 2000;279:G28–G39
  43. Ward CA, Liu H, Lee SS. Altered cellular calcium regulatory systems in a rat model of cirrhotic cardiomyopathy. Gastroenterology. 2001;121:1209–1218
  44. Kass RS, Wiegers SE. The ionic basis of concentration-related effects of noradrenaline on the action potential of calf cardiac purkinje fibres. J Physiol. 1982;322:541–558
  45. Priori SG, Corr PB. Mechanisms underlying early and delayed afterdepolarizations induced by catecholamines. Am J Physiol. 1990;258:H1796–H1805
  46. Ward CA, Ma Z, Lee SS, Giles WR. Potassium currents in atrial and ventricular myocytes from a rat model of cirrhosis. Am J Physiol. 1997;273:G537–G544
  47. Tan HL, Hou CJ, Lauer MR, Sung RJ. Electrophysiologic mechanisms of the long QT interval syndromes and torsade de pointes. Ann Intern Med. 1995;122:701–714
  48. Lin RS, Lee FY, Lee SD, Tsai YT, Lin HC, Lu RH, et al. Endotoxemia in patients with chronic liver diseases: relationship to severity of liver diseases, presence of esophageal varices, and hyperdynamic circulation. J Hepatol. 1995;22:165–172
  49. Tilg H, Wilmer A, Vogel W, Herold M, Nolchen B, Judmaier G, et al. Serum levels of cytokines in chronic liver diseases. Gastroenterology. 1992;103:264–274
  50. Zhong J, Hwang TC, Adams HR, Rubin LJ. Reduced L-type calcium current in ventricular myocytes from endotoxemic guinea pigs. Am J Physiol. 1997;273:H2312–H2324
  51. Lolov R, Velkov Z. Disorders of ventricular contractility and electrogenesis in the early stage of endotoxin shocked rabbits. Acta Physiol Pharmacol Bulg. 1994;20:95–102
  52. Haque R, Kan H, Finkel MS. Effects of cytokines and nitric oxide on myocardial E–C coupling. Basic Res Cardiol. 1998;93:86–94
  53. LaRusso NF, Hoffman NE, Hofmann AF, Korman MG. Validity and sensitivity of an intravenous bile acid tolerance test in patients with liver disease. N Engl J Med. 1975;292:1209–1214
  54. Gazawi H, Ljubuncic P, Cogan U, Hochgraff E, Ben-Shachar D, Bomzon A. The effects of bile acids on beta-adrenoceptors, fluidity, and the extent of lipid peroxidation in rat cardiac membranes. Biochem Pharmacol. 2000;59:1623–1628
  55. Casey PJ. Lipid modifications of G proteins. Curr Opin Cell Biol. 1994;6:219–225
  56. Bastiaanse EM, Jongsma HJ, van der Laarse A, Takens-Kwak BR. Heptanol-induced decrease in cardiac gap junctional conductance is mediated by a decrease in the fluidity of membranous cholesterol-rich domains. J Membr Biol. 1993;136:135–145
  57. Marchesini G, Bianchi GP, Forlani G, Rusticali AG, Patrono D, Capelli M, et al. Insulin resistance is the main determinant of impaired glucose tolerance in patients with liver cirrhosis. Dig Dis Sci. 1987;32:1118–1124
  58. Zierler K, Wu FS. Insulin acts on Na, K, and Ca currents. Trans Assoc Am Physicians. 1988;101:320–325
  59. Gastaldelli A, Emdin M, Conforti F, Camastra S, Ferrannini E. Insulin prolongs the QTc interval in humans. Am J Physiol Regul Integr Comp Physiol. 2000;279:R2022–R2025
  60. Frank S, Colliver JA, Frank A. The electrocardiogram in obesity: statistical analysis of 1029 patients. J Am Coll Cardiol. 1986;7:295–299
  61. Bernardi M, Gulberg V, Colantoni A, Trevisani F, Gasbarrini A, Gerbes AL. Plasma endothelin-1 and -3 in cirrhosis: relationship with systemic hemodynamics, renal function and neurohumoral systems. J Hepatol. 1996;24:161–168
  62. Tran Duc AT, Schwab AJ, Simard A, Villeneuve L, Dupuis J. Reduction in hepatic endothelin-1 clearance in cirrhosis. Clin Sci. 2003;105:227–234
  63. Magyar J, Iost N, Kortvely A, Banyasz T, Virag L, Szigligeti P, et al. Effects of endothelin-1 on calcium and potassium currents in undiseased human ventricular myocytes. Pflugers Arch. 2000;441:144–149
  64. Mohacsi A, Magyar J, Tamas B, Nanasi PP. Effects of endothelins on cardiac and vascular cells: new therapeutic target for the future?. Curr Vasc Pharmacol. 2004;2:53–63
  65. Duru F, Barton M, Luscher TF, Candinas R. Endothelin and cardiac arrhythmias: do endothelin antagonists have a therapeutic potential as antiarrhythmic drugs?. Cardiovasc Res. 2001;49:272–280
  66. Bomzon A, Blendis LM. The nitric oxide hypothesis and the hyperdynamic circulation in cirrhosis. Hepatology. 1994;20:1343–1350
  67. Mery PF, Pavoine C, Belhassen L, Pecker F, Fischmeister R. Nitric oxide regulates cardiac Ca2+ current. Involvement of cGMP-inhibited and cGMP-stimulated phosphodiesterases through guanylyl cyclase activation. J Biol Chem. 1993;268:26286–26295
  68. Eu JP, Xu L, Stamler JS, Meissner G. Regulation of ryanodine receptors by reactive nitrogen species. Biochem Pharmacol. 1999;57:1079–1084
  69. Liu H, Ma Z, Lee SS. Contribution of nitric oxide to the pathogenesis of cirrhotic cardiomyopathy in bile duct-ligated rats. Gastroenterology. 2000;118:937–944
  70. Foster MW, McMahon TJ, Stamler JS. S-nitrosylation in health and disease. Trends Mol Med. 2003;9:160–168
  71. Schopfer FJ, Baker PR, Freeman BA. NO-dependent protein nitration: a cell signaling event or an oxidative inflammatory response?. Trends Biochem Sci. 2003;28:646–654
  72. Mani AR, Fabjan JS, Ollosson R, Ippolito S, Freeman BA, Moore KP. Increased nitration of critical mitochondrial proteins in biliary cirrhosis. J Hepatol. 2004;40:70
  73. Hajrasouliha AR, Tavakoli S, Jabehdar-Maralani P, Shafaroodi H, Borhani AA, Houshmand G, et al. Resistance of cholestatic rats against epinephrine-induced arrhythmia: the role of nitric oxide and endogenous opioids. Eur J Pharmacol. 2004;499:307–313

PII: S0168-8278(06)00004-3

doi:10.1016/j.jhep.2005.10.034

Journal of Hepatology
Volume 44, Issue 5 , Pages 994-1002, May 2006

Article Tools

* Image Usage & Resolution