Diastolic cardiac dysfunction is a predictor of dismal prognosis in patients with liver cirrhosis

Hepatol Int (2014) 8:588 594 DOI 10.1007/s12072-014-9544-6 ORIGINAL ARTICLE Diastolic cardiac dysfunction is a predictor of dismal prognosis in patients with liver cirrhosis Dimitrios S. Karagiannakis Jiannis Vlachogiannakos Georgios Anastasiadis Irini Vafiadis-Zouboulis Spiros D. Ladas Received: 26 February 2014 / Accepted: 13 May 2014 / Published online: 8 June 2014 Ó Asian Pacific Association for the Study of the Liver 2014 Abstract Background and purpose Left ventricular diastolic dysfunction (LVDD) constitutes the prominent characteristic of cirrhotic cardiomyopathy, but its relevance on the clinical course of cirrhotic patients has not been clearly defined. The aim of the study was to evaluate the relationship of LVDD with the severity and etiology of liver disease and to investigate whether it affects the outcome of cirrhotic patients. Methods Cardiac function of 45 cirrhotics was studied by a tissue Doppler imaging echocardiography. Diagnosis of LVDD was made according to the latest guidelines of the American Society of Echocardiography. All patients were followed up for a period of 2 years. Death or liver transplantation was the endpoint of the study. Results LVDD was found in 17 (38 %) of 45 patients. Its presence was not found to be associated with the etiology and stage of cirrhosis, but its severity was directly D. S. Karagiannakis (&) J. Vlachogiannakos I. Vafiadis-Zouboulis S. D. Ladas Academic Department of Gastroenterology, Medical School of Athens, Laikon General Hospital, 17 Aghiou Thoma Street, 11527 Athens, Greece e-mail: d_karagiannakis@hotmail.com J. Vlachogiannakos e-mail: jvlachog@hotmail.com I. Vafiadis-Zouboulis e-mail: izoubvaf@med.uoa.gr S. D. Ladas e-mail: sdladas@otenet.gr G. Anastasiadis Department of Cardiology, Laikon General Hospital, 17 Aghiou Thoma Street, 11527 Athens, Greece e-mail: fayk@otenet.gr correlated with the Child-Pugh score. At the end of followup, 14 patients had died; 9 had LVDD (9/17, 53 %) and 5 had not (5/28, 18 %). Patients who died at the beginning of observation period had a higher Child-Pugh and MELD score, higher BNP, lower albumin and more prolonged QTc. On Kaplan-Meier analysis, patients with LVDD had statistically significantly worse prognosis compared to those without (p = 0.013, log rank: 5.495). Low albumin values (p = 0.003) and presence of LVDD (p = 0.017) were independent predictive factors of mortality. Conclusions LVDD is a common complication of cirrhosis. As its development seems to be related to a worse prognosis, patients with LVDD must be under a strict follow-up. Keywords Cardiomyopathy Diastolic dysfunction Cirrhosis Survival Abbreviations LVDD Left ventricular diastolic dysfunction HRS Hepatorenal syndrome CC Cirrhotic cardiomyopathy SBP Spontaneous bacterial peritonitis BNP Brain natriuretic peptide ECG Electrocardiograph QTc Corrected QT interval PW Pulse wave TDI Tissue Doppler imaging HR Heart rate MAP Mean arterial pressure LVEDD Left ventricular end diastolic diameter LAvol in. Left atrium volume index LVEF % Left ventricular ejection fraction % CI Cardiac index SVRI Systemic vascular resistance index

Hepatol Int (2014) 8:588 594 589 DT IVRT PASP MELD Introduction Deceleration time Isovolumetric relaxation time Pulmonary artery systolic pressure Model of end-stage liver disease Cirrhotic patients exhibit a circulatory dysfunction, which is characterized by splanchnic arterial vasodilation, low systemic vascular resistance, low blood pressure and reduced central blood volume with central or effective hypovolemia, all secondary to portal hypertension [1]. At the initial phases, homeostasis is maintained by the activation of the sympathetic nervous system, which increases the heart rate and output. In advanced stages of liver disease, the splanchnic vasodilation is getting more distinct and cannot be compensated by further increase in cardiac output. Thus, more potent vasoconstricting systems are stimulated, such as the renin-angiotensin-aldosterone axis and non-osmotic release of vasopressin [2]. These systems help to maintain the effective arterial blood volume and pressure, but in case of extreme activation, they lead to the development of ascites, as well as a markedly renal vasoconstriction with a reduced glomerular filtration rate, a condition known as hepatorenal syndrome (HRS) [1, 3, 4]. Recent studies indicate that in these stages of cirrhosis a blunted cardiac function precedes the HRS and contributes to its development by aggravating the circulatory dysfunction further [5]. This clinical entity is called cirrhotic cardiomyopathy (CC) and is characterized by altered diastolic relaxation, electrophysiological abnormalities, such as prolongation of QT interval, and impaired contractility under physiological or pharmacological stress, all occurring in the absence of other known cardiac disease [6 8]. Left ventricular diastolic dysfunction (LVDD) seems to be the first manifestation of cardiac disease, but in later phases features of systolic dysfunction are also present [9, 10]. The clinical significance of LVDD in cirrhosis remains unclear, as symptoms of heart failure are absent because of the coexistence of splachnic vasodilation and reduced afterload. The aim of this study was to investigate the frequency and characteristics of LVDD and to assess its potential impact on the outcome of cirrhotic patients. Methods Patients Over a period of 12 months, all cirrhotic patients aged from 18 to 70 years old who attended our department were considered eligible for inclusion in the study, regardless of the etiology and severity of their liver disease. The diagnosis of cirrhosis was based on clinical and laboratory findings, endoscopy, imaging studies or liver biopsy. Every cirrhotic patient with a Child-Pugh score C7 and/or signs of hepatic encephalopathy, history of variceal bleeding or the presence of ascites was defined as decompensated. A routine abdominal ultrasound was performed in all patients with the purpose to detect even minimal ascites. Exclusion criteria were history of arterial hypertension, chronic cardiac, pulmonary or renal disease, diabetes mellitus, active bacterial infection, recent gastrointestinal bleeding (\3 months), hepatocellular carcinoma, recent or active ethanol abuse (\6 months) [11], and treatment with drugs that could affect cardiac function or circulatory parameters, such as vasoactive drugs or nitrates. Active bacterial infection was ruled out by history, clinical examination, blood tests, culture of urine, and chest radiograph, and in ascitic patients by culture and the white cell count of ascitic fluid. None of our patients had a history of spontaneous bacterial peritonitis (SBP) in the past. Large volume paracentesis was not performed in our ascitic patients during the last month before their enrollment to the study. Patients with alcoholic cirrhosis had prolonged periods of abstinence confirmed by detailed history, discussion with relatives and non-scheduled plasma alcohol determinations during their visits. Fourteen patients who were taking betablockers for the prevention of variceal bleeding were asked to stop them 7 days prior the estimation of their cardiac function. Subsequently, all patients underwent repeated sessions of band ligation until the eradication of varices. Patients who finally participated in the study were followed up for 24 months. During this period of time, every patient had been subjected to clinical examination, an abdominal ultrasound and laboratory tests month by month. None of the patients was receiving beta-blockers during the followup period. Bacterial infections, variceal bleeding, presence or worsening of ascites, episodes of hepatic encephalopathy, presence of hepatocellular carcinoma, changes in urinary function tests or other laboratory tests had been recorded and managed by standard therapies. None of our patients was transplanted during this follow-up period. The study protocol conformed to the principles outlined in the Declaration of Helsinki and was approved by the Ethics Committee of the Laikon General Hospital. All participants signed their informed consent. Clinical data Demographic and clinical data (such as age, gender, BMI, cause of liver disease, heart rate, blood pressure) as well as laboratory parameters (including biochemical and clotting profile) were recorded on admission. The Child-Pugh score

590 Hepatol Int (2014) 8:588 594 and MELD score were calculated. The plasma levels of brain natriuretic peptide (BNP) were also estimated (Alere Triage MeterPro CLIA test; Biosite Incorporated, San Diego, CA, USA, lower limit of detection 5 pg/ml). Electrocardiography and echocardiography protocol Electrocardiograph (ECG) was recorded by a conventional electrocardiogram (Cardioline ar 600, Italy). QT intervals were corrected (QTc) with Bazett s formula. A pulsedwave (PW) Doppler echocardiography with tissue Doppler imaging (TDI) (General Electric Vivid 3, USA) was used to assess the dimensions of cardiac cavities and the following cardiac parameters: heart rate (HR), mean arterial pressure (MAP), left ventricular end diastolic diameter (LVEDD), left atrium volume index (LAvol in.), left ventricular ejection fraction at rest (LVEF %), cardiac index (CI), systemic vascular resistance index (SVRI), peak early filling velocity during early ventricle diastole (E wave), late diastolic filling velocity during atrial systole (A wave), E/A ratio (E/A), deceleration time of E wave (DT), isovolumetric relaxation time (IVRT), early diastolic mitral annular velocity from the septal side (éseptal), early diastolic mitral annular velocity from the lateral side (élateral), average early diastolic mitral annular velocity (éav), E/éav. ratio (E/éav.) and pulmonary artery systolic pressure (PASP). The diagnosis of LVDD was made if a patient had a combination of éseptal \ 8 cm/s and élateral \ 10 cm/s, according to the recent guidelines of the American Society of Echocardiography [12]. In patients with LVDD, the E/éav. was used to define its severity. All echocardiography examinations were performed by an experienced cardiologist. Echocardiograms were stored digitally and analyzed offline twice in different periods of time. Differences were rarely found between the two measurements. If this happened, the average values were obtained. Statistical analysis For quantitative data with normal distribution, summary statistics are presented as mean ± SD, and analyses were performed by using Student s t test. For quantitative data with indications of non-normality, summary statistics are presented as median (range), and comparisons among groups were performed by using Mann-Whitney test. Comparisons involving categorical data were made by chisquare test with Yates correction or Fisher s exact test, as appropriate. Survival rates were evaluated by Kaplan-Meier curves, and groups were compared by the long-rank test. Multivariate logistic regression analysis models were used to identify independent, significant, predictive factors of a poor outcome. A p value of less than 0.05 was considered statistically significant. All statistical analyses were performed by using SPSS 16.0 for Windows (SPSS Inc. Chicago, IL, USA) and GraphPad Prism version 5.01. Results Demographic and clinical data Over the study period, 62 cirrhotic patients who attended our Liver Clinic fulfilled the inclusion criteria. Seventeen out of 62 were excluded for the following reasons: 3 had developed liver cancer, 2 had diabetes, 5 had arterial hypertension and/or coronary artery disease, 2 had valvular heart disease, 3 were active alcohol drinkers, and 2 patients refused to provide written informed consent. Therefore, 45 patients were finally included in our study. The demographics of these patients are listed in Table 1. The mean Child-Pugh score was 6.43 ± 1.9 (range 5 14). All patients had some of the features that are indicative of portal hypertension (varices, ascites, hepatic encephalopathy, splenomegaly, portal vein dilatation, etc.). Twenty-two Table 1 Baseline characteristics of patients Patients (n = 45) M/F 33 (73 %) 12 (27 %) Age (years) 57.2 ± 12.4 Etiology HBV 9 (20 %) HCV 10 (22 %) Alcohol 22 (49 %) PBC 2 (4.6 %) NASH 1 (2.2 %) Unknown 1 (2.2 %) Bilirubin (mg/dl) 1.2 (0.3 12) INR 1.3 (1 2.7) Creatinine (mg/dl) 1.1 (0.4 3.9) Child-Pugh grade A 26 (59 %) B 15 (34 %) C 3 (7 %) MELD score 11.5 ± 4.2 Ascites Yes/no 19 (42 %)/26 (58 %) History of encephalopathy Yes/no 13 (29 %)/32 (71 %) Varices Yes/no 30 (67 %)/15 (33 %) History of variceal bleeding Yes/no 9 (20 %)/36 (80 %) Variables are expressed as mean ± SD or median (range) values

Hepatol Int (2014) 8:588 594 591 Table 2 Laboratory characteristics and echocardiographic data among patients with normal diastolic cardiac function and patients with LVDD Normal (n = 28) LVDD (n = 17) p value Age (years) 53.8 ± 13 62.8 ± 9 0.016 Child score 6.5 ± 2.1 6.4 ± 1.6 0.828 MELD score 11.7 ± 4.6 11 ±.3.5 0.605 Bilirubin (mg/dl) 1.3 (0.3 12) 1 (0.6 2.3) 0.320 Albumin (mg/dl) 3.82 ± 0.67 3.66 ± 0.52 0.411 INR 1.3 (1 2.7) 1.3 (1 1.5) 0.455 Creatinine (mg/dl) 0.9 (0.4 2.1) 1 (0.5 3.9) 0.963 BNP (pg/ml) 20 (5 286) 46 (5 369) 0.160 MAP (mmhg) 94 ± 11 95 ± 9 0.956 12.28 % 10.80 % HR (bpm) 78 ± 12 75? 10 0.389 15.80 % 14.10 % SVRI (dyn/s/m 2 /cm -5 ) 2,320 ± 633 2,400 ± 609 0.683 27.26 % 25.38 % QTc (ms) 422 ± 23 438 ± 32 0.068 6.68 % 7.06 % LAvol in. (ml/m 2 ) 23 ± 5.5 26.5 ± 8 0.088 23.94 % 29.56 % E/e av ratio 7 ± 2 8.7 ± 2 0.008 28.52 % 23.23 % E/A ratio 1.2 ± 0.3 0.9 ± 0.2 0.002 30.41 % 26.37 % IVRT (ms) 102 ± 16 110 ± 13 0.091 15.61 % 11.64 % DT (ms) 211 ± 38 233 ± 27 0.047 17.81 % 11.91 % LVEDD (mm) 51.5 ± 4.6 49 ± 4.8 0.091 8.95 % 10.07 % LVEF % 66.4 ± 4.8 64.3 ± 4.5 0.163 7.29 % 7.04 % CI (l/min/m 2 ) 3.1 ± 0.8 3 ±.6 0.510 24.51 % 19.48 % PASP (mmhg) 32.5 ± 5.9 33.6 ± 5.6 0.544 18.09 % 16.81 % Variables are expressed as mean ± SD or median (range) values Underneath, the coefficients of variations are expressed BNP brain natriuretic peptide, MAP mean arterial pressure, HR heart rate, SVRI systemic vascular resistance index, Qtc QT corrected interval in ECG, LA vol in. left atrium volume index, E peak early filling velocity during early ventricle diastole, A late diastolic filling velocity during atrial systole, éav the average early diastolic mitral annular velocity, IVRT isovolumetric relaxation time, DT deceleration time of E wave, LVEDD left ventricular end diastolic diameter, LVEF % left ventricular ejection fraction at rest, CI cardiac index at rest, PASP pulmonary artery systolic pressure Table 3 Differences between patients with normal diastolic cardiac function and patients with LVDD stage II Normal (n = 28) LVDD II (n = 8) p value Age (years) 53.8 ± 13 63.9 ± 9.6 0.016 Child score 6.5 ± 2.1 7.3 ± 1.4 0.332 MELD score 11.7 ± 4.6 12 ±.4 0.880 BNP (pg/ml) 20 (5 286) 62 (7 369) 0.007 QTc (ms) 422 ± 23 444 ± 35 0.049 LAvol in. (ml/m 2 ) 23 ± 5.5 29 ± 8 0.013 Variables are expressed as mean ± SD or median (range) values BNP brain natriuretic peptide, MAP mean arterial pressure, Qtc QT corrected interval in ECG, LA vol in. left atrium volume index patients (48.9 %) had decompensated liver cirrhosis at the time of inclusion in the study, and 23 patients (51.1 %) had compensated liver cirrhosis. Features of left ventricular diastolic dysfunction (LVDD) Seventeen (38 %) out of 45 patients had LVDD (20 % with LVDD grade I and 18 % with grade II). Among these 17 patients, 6 (35.3 %) had liver cirrhosis due to alcohol abuse in the past. LVDD was less common in patients with alcoholic cirrhosis (6/21, 28.5 %) compared to patients with cirrhosis of other etiologies (11/24, 45.8 %), but the difference was not statistically significant (p = 0.208). LVDD was diagnosed not only in patients with decompensated cirrhosis (9/22, 40.9 %), but also in patients with compensated liver disease (8/23, 34.8 %). Patients with LVDD compared to those without LVDD Echocardiographic and biochemical data of patients classified according to the presence or not of LVDD are shown in Table 2. By definition, patients with LVDD had higher E/éav. (p = 0.008), lower E/A (p = 0.002) and longer DT (p = 0.047), whereas they were older in age (p = 0.016). They also had a trend forward a longer QTc (p = 0.068), larger LA vol in. (p = 0.088) and longer IVRT (p = 0.091). No statistically significant differences were found between these groups of patients regarding their biochemical data, Child-Pugh score, MELD score and hemodynamic parameters (HR, MAP, SVRI) as well as their LVEF % and CI, which represent markers of left ventricular systolic function. When the comparison was made only between patients with severe LVDD (stage II) and those without LVDD, it was found that the former group of patients was older in

592 Hepatol Int (2014) 8:588 594 Table 4 Characteristics of patients at the beginning of follow-up according to whether they were alive or dead after 24 months age (p = 0.016) and characterized by a longer QTc interval (p = 0.049), larger LA vol in. (p = 0.013) and higher BNP levels (p = 0.007) (Table 3). In the whole series of patients, the Child-Pugh score was found to have a positive correlation with the E/éav. (p = 0.007, r = 0.399), BNP (p = 0.017, r = 0.3585) and QTc (p = 0.031, r = 0.329). Furthermore E/éav., which defines the severity of diastolic dysfunction, was directly correlated with albumin (p = 0.0002, r =-0.537), LA vol in. (p = 0.007, r = 0.399) and BNP levels (p = 0.0001, r = 0.543). Follow-up Alive (n = 31) Dead (n = 14) p value Age (years) 56.1 ± 13.1 59.5 ± 10.9 0.401 Child score 5.9 ± 1.4 7.6 ± 2.4 0.005 MELD score 10.6 ± 3.5 13.5 ± 5 0.033 Bilirubin (mg/dl) 1.1 (0.3 2.5) 1.3 (0.6 18.4) 0.181 INR 1.2 (0.9 1.8) 1.4 (1 2.7) 0.014 Creatinine (mg/dl) 1(0.6 2.1) 0.8 (0.5 3.9) 0.173 Albumin (mg/dl) 4 ± 0.5 3.3 ± 0.5 \0.0001 BNP (pg/ml) 21 (5 369) 76 (8 357) 0.046 QTc (ms) 434 ± 30 464 ± 26 0.003 6.88 % 5.59 % LAvol in. (ml/m 2 ) 24 ± 6.5 25.2 ± 7.4 0.502 27.27 % 29.20 % E/e av ratio 7.1 ± 1.9 8.7 ± 2.3 0.018 26.53 % 26.86 % E/A ratio 1.1 ± 0.3 1 ± 0.3 0.372 31.32 % 35.17 % DT (ms) 212 ± 33 237 ± 37 0.027 15.40 % 15.77 % LVEDD (mm) 50.3 ± 4.9 51 ± 5 0.727 9.63 % 9.83 % LVEF % 65.7 ± 5 65.8 ± 4 0.954 7.77 % 6.12 % CI (l/min/m 2 ) 3.1 ± 0.8 3.1 ± 0.5 0.862 25.05 % 17.52 % Variables are expressed as mean ± SD or median (range) values. Underneath, the coefficients of variations are expressed BNP brain natriuretic peptide, Qtc QT corrected interval in ECG, LA vol in. left atrium volume index, E peak early filling velocity during early ventricle diastole, A late diastolic filling velocity during atrial systole, éav the average early diastolic mitral annular velocity, DT deceleration time of E wave, LVEDD left ventricular end diastolic diameter, LVEF % left ventricular ejection fraction at rest, CI cardiac index at rest All the 45 patients were followed up for 24 months. Fourteen of them died in that period of time; 9 had LVDD Table 5 Variables independently associated with poor outcome p value Exp (B) 95 % CI for Exp (B) Lower Upper Albumin 0.003 0.030 0.003 0.306 LVDD 0.017 0.086 0.012 0.642 Fig. 1 Differences in survival of patients according to the status of their cardiac diastolic function (9/17, 53 %), and 5 did not (5/28, 17.8 %). Table 4 shows the characteristics of patients who died and those who survived. The two groups had no statistically significant differences in age (p = 0.401), but in Child-Pugh score (p = 0.005), MELD score (p = 0.033), albumin (p \ 0.0001), INR (p = 0.014), QTc (p = 0.003), DT (p = 0.027), E/éav. (p = 0.018) and BNP (p = 0.046). In multivariate analysis, albumin (p = 0.003) and the presence of LVDD (p = 0.017) were found to significantly predict the 2-year mortality (Table 5). The Kaplan-Meier analysis of 2-year probability of patients survival, classified by their diastolic function, is marked in Fig. 1. The outcome of patients with LVDD was statistically significantly worse compared to those without (p = 0.013, log rank: 6.128). More precisely, the survival rates of patients with and without LVDD were 88.2 and 96.4 % at 6 months, 70.6 and 89.3 % at 12 months, 53 and 85.7 % at 18 months, and 47 and 82.1 % at 24 months, respectively. Causes of death were hepatic failure (8/15, 53.33 %), hepatorenal syndrome type I (4/15, 26.67 %), hepatocellular carcinoma (2/15, 13.33 %) and hemorrhage of the upper gastrointestinal tract (1/15, 6.67 %). In respect to patients with LVDD, the leading cause of death was hepatic failure (5/9, 55.56 %), followed by hepatorenal syndrome type I (3/9, 33.33 %) and hepatocellular carcinoma (1/9, 11.11 %). When patients were classified according to their E/éav. ratio, the survival rates were significantly higher in patients

Hepatol Int (2014) 8:588 594 593 Fig. 2 Survival of patients according to E/éav. value with E/éav. \ 9 cm/s compared to patients with E/éav. C 9 cm/s (p = 0.019, log rank: 5.495, Fig. 2). A value of E/éav. C 9 cm/s, as it is known by the latest guidelines, differentiates between patients with severe LVDD stage II and those with LVDD stage I. Discussion In this study, we investigated 45 cirrhotic patients and found that a substantial proportion of them had developed LVDD. After a 2-year follow-up, we clearly showed that the presence of LVDD is associated with poor outcome. In contrast to older studies, which used the abnormal E/A ratio and the prolongation of DT for the diagnosis of LVDD, which are neither sensitive nor specific markers as they can be influenced by changes in the preload and afterload, we used tissue Doppler measurements, which are less load dependent and more indicative of cardiac muscle changes and for that reasons more reliable for the estimation of myocardial dysfunction [13 15]. Thus, according to our measurements, it was found that 40 % of cirrhotic patients had features of LVDD. Between patients with LVDD and those without, no differences were found in their LVEF % and CI, confirming the results from other studies showing that the diastolic dysfunction precedes the systolic one [16 18]. The presence of LVDD did not seem to be correlated with the etiology of liver disease, as no statistically significant differences were established between patients with alcoholic cirrhosis and those with cirrhosis of other etiologies. Except for that, it was also shown that the presence of LVDD was not associated with the stage of liver failure as it was developed equally among patients with compensated and decompensated cirrhosis. However, the severity of LVDD seems to be related to the degree of liver failure as the E/éav, which reflects the severity of LVDD, was found to have a positive direct correlation with the Child- Pugh score and a negative direct correlation with the albumin levels. Furthermore, higher BNP and longer QTc, which are both identified in the more severe stages of LVDD, were found to be correlated with higher Child- Pugh score. This correlation between the severity of LVDD and the Child-Pugh score was not highlighted in Table 3, and this discrepancy could be explained by the fact that in our study the number of patients with a high Child-Pugh score (grade C) was small. The clinical significance of LVDD on the prognosis of cirrhotic patients who underwent liver transplantation or TIPS insertion is already documented [19 22]. Nevertheless, the clinical impact of LVDD on the survival of cirrhotics who have not undergone any invasive intervention has not been clearly defined yet. In our study, we showed that patients with LVDD have worse prognosis, regardless of the etiology of liver disease. Furthermore, the presence of LVDD and low albumin values were recognized as independent prognostic factors of survival. Our results are in accordance with those recently published by Ruiz-del- Arbol et al. [23]. However, in our study, we had a longer follow-up period, and we further displayed for the first time that the difference in survival became more significant after the first year of follow-up. This finding could explain why previous studies were not able to elucidate a relationship between LVDD and survival as most of them had a shorter follow-up period [24, 25]. Regarding the study of Alexopoulou et al. [26], which had a longer follow-up period, it must be emphasized that despite its inability to show a worse outcome of patients with LVDD, the trend forward to this was absolutely noteworthy. Previous studies have underlined high BNP levels and the prolongation of QTc as predictors of poor prognosis in cirrhotic patients [27 30]. In our study, we established the importance of another factor, the E/éav, which seems to have a good prognostic accuracy. Specifically, we verified that patients with a value of E/éav. C 9 cm/s, which defines stage II LVDD, had a worse outcome. This marker could be used to identify those cirrhotic patients who should be attended to more strictly as they are at higher risk of mortality. In conclusion, we have shown that LVDD is a frequent complication of liver cirrhosis, and its presence is not correlated with the etiology and stage of liver disease, but its severity is related to the deterioration of liver function. As LVDD seems to potentially affect the prognosis of cirrhotic patients, particularly after the first year, it is becoming obvious that a longer and more careful follow-up is required in all patients with diastolic dysfunction, mostly in those with low albumin levels and a high E/éav. ratio. Acknowledgements Guarantor of the study: Prof. S.D. Ladas. D.Karagiannakis planned and designed the study, recruited the pool of patients, analyzed the data and wrote the report. G. Anastasiadis was

594 Hepatol Int (2014) 8:588 594 responsible for the cardiovascular assessment that was performed on patients. Assistant Prof. J. Vlachogiannakos participated in the study design and interpretation of the data, and he was responsible for the statistical analysis. Prof. I. Vafiadis-Zouboulis participated in the interpretation of the data. Compliance with ethical requirements and Conflict of interest All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008. Informed consent was obtained from all patients for being included in the study. Dimitrios Karagiannakis, Jiannis Vlachogiannakos, Georgios Anastasiadis, Irini Vafiadis-Zouboulis and Spiros D. Ladas declare that they have no conflict of interest. References 1. Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rode s J. Peripheral artery vasodilatation hypothesis: a proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 1988;5:1151 1157 2. Iwakiri Y, Groszmann RJ. The hyperdynamic circulation of chronic liver diseases: from the patient to the molecule. Hepatology 2006;43:S121 S131 3. Salermo F, Cazzaniga M, Gobbo G. Pharmacological treatment of hepatorenal syndrome: a note of optimism. J Hepatol 2007; 47:729 731 4. Moller S, Henriksen JH. The systemic circulation in cirrhosis. In Gines P, Arroyo V, Rodes J, Schrier RW, editors. Ascites and Renal Dysfunction in Liver Disease. 2nd ed. Malden: Blackwell; 2005. 139 155 5. Krag A, Bendtsen F, Burroughs AK, Moller S. The cardiorenal link in advanced cirrhosis. Med Hypotheses 2012;79:53 55 6. Timoh T, Protano MA, Wagman G, Bloom M, Vittorio TJ. A perspective on cirrhotic cardiomyopathy. Transplant Proc 2011;43:1649 1653 7. Moller S, Hove JD, Dixen U, Bendtsen F. New insights into cirrhotic cardiomyopathy. Int J Cardiol 2013;167:1101 1108 8. Bernardi M, Maggioli C, Dibra V, Zaccherini G. QT interval prolongation in liver cirrhosis: innocent bystander or serious threat? Expert Rev Gastroenterol Hepatol 2012;6:57 66 9. Wiese S, Hove JD, Bendtsen F, Moller S. Cirrhotic cardiomyopathy: pathogenesis and clinical relevance. Nat Rev Gastroenterol Hepatol 2013;11:177 186 10. Pozzi M, Redaelli E, Ratti L, Poli G, Guidi C, Milanese M, et al. Time-course of diastolic dysfunction in different stages of chronic HCV related liver diseases. Minerva Gastroenterol Dietol 2005;51:179 186 11. Guilo P, Mansourati J, Maheu B, Etienne Y, Provost K, Simon O, et al. Long-term prognosis in patients with alcoholic cardiomyopathy and severe heart failure after total abstinence. Am J Cardiol 1997;79(9):1276 1278 12. Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. J Am Soc Echocardiogr 2009;22:107 133 13. Klein AL, Burstow DJ, Talik AJ, Zachariak PK, Bailey KR, Seward JB. Effects of age on left ventricular dimensions and filling dynamics in 117 normal persons. Mayo Clin Proc 1994;69: 212 224 14. Kazankov K, Holland-Fischer P, Andersen NH, Torp P, Sloth E, Aagaard NK, et al. Resting myocardial dysfunction in cirrhosis quantified by tissue Doppler imaging. Liver Int 2011;31:534 540 15. Sampaio F, Pimenta J, Bettencourt N, Fontes-Carvalho R, Silva AP, Valente J, et al. Systolic and diastolic dysfunction in cirrhosis: a tissue-doppler and speckle tracking echocardiography study. Liver int 2013;33:1158 1165 16. Lee RF, Glenn TK, Lee SS. Cardiac dysfunction in cirrhosis. Best Pract Res Clin Gastroenterol 2007;21:125 140 17. Mandell MS, Lindenfeld J, Tsou MY, Zimmerman M. Cardiac evaluation of liver transplant candidates. World J Gastroenterol 2008;14:3445 3451 18. Moller S, Bernardi M. Interactions of the heart and the liver. Eur Heart J 2013;34:2804 2811 19. Cazzaniga M, Salerno F, Pagnozzi G, Dionigi E, Visentin S, Cirello I, et al. Diastolic dysfunction is associated with poor survival in cirrhotic patients with transjugular intrahepatic portosystemic shunt. Gut 2007;56:869 875 20. Rabie RN, Cazzaniga M, Salerno F, Wong F. The use of E/A ratio as a predictor of outcome in cirrhotic patients treated with transjugular intrahepatic portosystemic shunt. Am J Gastroenterol 2009;104:2458 2466 21. Ripoll C, Catalina MV, Yotti R, Olmedilla L, Pérez-Peña J, Iacono O, et al. Cardiac dysfunction during liver transplantation: incidence and preoperative predictors. Transplantation 2008;85: 1766 1772 22. Holt EW, Woo G, Trilesskaya M, Haeusslein EA, Shaw RE, Frederick RT. Diastolic dysfunction defined by E/A ratio \ 1 on 2-D echo is an independent predictor of liver transplantation or death in patient with cirrhosis. J Hepatol 2011;54:S245 S246 23. Ruiz-del-Arbol L, Achecar L, Serradilla R, Rodríguez-Gandía MÁ, Rivero M, Garrido E, et al. Diastolic dysfunction is a predictor of poor outcomes in patients with cirrhosis, portal hypertension and I normal creatinine. Hepatology 2013;58:1732 1741 24. Nazar A, Guevara M, Sitges M, Terra C, Solà E, Guigou C, et al. Left ventricular function assessed by echocardiography in cirrhosis: Relationship to systemic hemodynamics and renal dysfunction. J Hepatol 2013;58:51 57 25. Merli M, Calicchia A, Ruffa A, Pellicori P, Riggio O, Giusto M, et al. Cardiac dysfunction in cirrhosis is not associated with the severity of liver disease. Euro J Int Med 2013;24:172 176 26. Alexopoulou A, Papatheodoridis G, Pouriki S, Chrysohoou C, Raftopoulos L, Stefanadis C, et al. Diastolic myocardial dysfunction does not affect survival in patients with cirrhosis. Transpl Int 2012;25:1174 1181 27. Henriksen JH, Goetze 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 1157 28. Pimenta J, Paulo C, Gomes A, Silva S, Rocha-Goncalves F, Bettencourt P. B-type natriuretic peptide is related to cardiac function and prognosis in hospitalized patients with decompensated cirrhosis. Liver Int 2010;30:1059 1066 29. Bal JS, Thuluvath PJ. Prolongation of QTc inteval: relationship with etiology and severity of liver disease, mortality and liver transplantation. Liver int 2003;23:243 248 30. Trevisani F, Di Micoli A, Zambruni A, Biselli M, Santi V, Erroi V, et al. QT interval prolongation by acute gastrointestinal bleeding in patients with cirrhosis. Liver Int 2012;32:1510 1515