Cirrhosis is associated with a specific subclinical

Transcription

1 LIVER FAILURE/CIRRHOSIS/PORTAL HYPERTENSION Diastolic Dysfunction Is a Predictor of Poor Outcomes in Patients With Cirrhosis, Portal Hypertension, and a Normal Creatinine Luıs Ruız-del-Arbol, 1 Linette Achecar, 1 Regina Serradilla, 1 Miguel A. Rodrıguez-Gandıa, 1 Miguel Rivero, 1 Elena Garrido, 1 and Jose J. Natcher 2 We investigated left ventricular diastolic dysfunction (LVDD) and its relationship with circulatory function and prognosis in cirrhosis with portal hypertension and normal creatinine. Conventional and tissue Doppler (TDI) echocardiography, systemic and hepatic hemodynamics, and the activity of endogenous vasoactive systems (AEVS) were measured prospectively in 80 patients. Plasma renin activity (PRA; >4 ng/ml/hour) was used as a surrogate of effective arterial blood volume. Patients were followed up for 12 months. Thirty-seven patients had LVDD (19 with grade 1 and 18 with grade 2). Left ventricular hypertrophy, left atrial volume, AEVS, and natriuretic peptide levels were significantly greater in patients with LVDD than without LVDD. Patients with grade 2 LVDD, compared to grade 1 LVDD and without LVDD, had significantly lower mean arterial pressure and higher Model for End-Stage Liver Disease (MELD) score, E-wave transmitral/early diastolic mitral annular velocity (E/e ratio), cardiopulmonary pressures, PRA, and natriuretic peptide levels. Systolic and cardiac chronotropic function were significantly lower in patients with grade 2 LVDD than without LVDD. LVDD was more frequent in patients with ascites and increased PRA than patients without ascites or with ascites but normal PRA. Fourteen patients with LVDD developed hepatorenal syndrome (HRS) type 1 on follow-up. Survival was different according to degree of LVDD (without LVDD: 95%; grade 1 LVDD: 79%; grade 2 LVDD: 39%; P < 0.001). Independent predictive factors of mortality were MELD score and E/e ratio. Conclusion: LVDD occurs simultaneously with other changes in cardiac structure and function and is associated with an impairment of effective arterial blood volume. LVDD is a sensitive marker of advanced cirrhosis, type 1 HRS development, and mortality. (HEPATOLOGY 2013;58: ) Cirrhosis is associated with a specific subclinical cardiomyopathy 1-4 characterized by diminished systolic responsiveness to stress stimuli, 5,6 impaired diastolic relaxation, 7,8 electrophysiological abnormalities, 9 and enlargement and hypertrophy of cardiac chambers, 10,11 all in the absence of known cardiac disease. However, evidence suggests that patients with cirrhosis display primarily left ventricular diastolic dysfunction (LVDD) with normal systolic function at rest. 7 Abbreviations: 2D, two-dimensional; A, atrial wave-filling peak; ALDO, plasma concentrations of aldosterone; ANF, atrial natriuretic factor; ASE, the American Society of Echocardiography; AUC, area under the curve; BNP, brain natriuretic peptide; CI, confidence interval; CO, cardiac output; DT, deceleration time; E, E-wave transmitral peak early filling; e 0, early diastolic mitral annular velocity; E/A, early diastolic mitral inflow velocity/late diastolic; E/e 0 ratio, E-wave transmitral/early diastolic mitral annular velocity; FHVP, free hepatic venous pressure; GI, gastrointestinal; HE, hepatic encephalopathy; HR, heart rate; HRS, hepatorenal syndrome; HVPG, hepatic venous pressure gradient; IVRT, isovolumetric relaxation time; LAVI, left atrial volume index; LT, liver transplantation; LV, left ventricular; LVDD, left ventricular diastolic dysfunction; LVEF, left ventricular ejection fraction; LVH, left ventricular hypertrophy; LVMI, left ventricular mass indexed; MAP, mean arterial pressure; MELD, Model for End-Stage Liver Disease; NE, plasma norepinephrine; PAP, pulmonary artery pressure; PCWP, pulmonary capillary wedged pressure; PH, portal hypertension; PRA, plasma renin activity; RAP, right atrial pressure; RR, relative risk; SBP, spontaneous bacterial peritonitis; SD, standard deviation; TDI, tissue Doppler imaging; TIPS, transjugular intrahepatic portosystemic shunt; TTE, transthoracic echocardiography; USG, ultrasonography; WHVP, wedged hepatic venous pressures. From the 1 Hepatic Hemodynamic Unit, Gastroenterology Department, Hospital Ramon y Cajal, University of Alcala, Madrid, Spain; 2 Cardiology Department, Hospital Ramon y Cajal, University of Alcala, Madrid, Spain. Received August 9, 2012; accepted April 27, This work was supported, in part, by grants from Fondo de Investigaciones Sanitarias (FIS 06/1082). 1732

2 HEPATOLOGY, Vol. 58, No. 5, 2013 RUIZ-DEL-ARBOL ET AL Doppler examination of mitral inflow has been the most common technique used in the evaluation of left ventricular (LV) diastolic function in cirrhosis. 7,8,10,11 However, conventional Doppler echocardiographic indices (E/A ratio) have clear limitations (age and load conditions) and rarely allow the accurate differentiation between normal from pseudonormal pattern. Currently, the most sensitive and reproducible echocardiographic technique for the assessment of LV filling dynamics is tissue Doppler imaging (TDI), because TDI can overcome some of these factors. TDI is an ultrasound modality that applies the Doppler principle to record velocities within the myocardium. TDI velocities have demonstrated a significant correlation with invasive indices of LV relaxation and minimal effect of preload in the setting of impaired relaxation. 12 Recently, the E-wave transmitral/early diastolic mitral annular velocity (E/e 0 ) index was considered to be the most important LV diastolic function parameter. 13 Some studies have suggested that LVDD plays a role in the pathogenesis of hepatorenal syndrome (HRS) precipitated by spontaneous bacterial peritonitis (SBP) 14,15 and the abnormal cardiac response after insertion of a transjugular intrahepatic portosystemic shunt (TIPS) 16 and liver transplantation (LT). 17,18 The relationship between LVDD and other alterations in cardiac function in cirrhosis is unknown. Finally, the potential role of LVDD in the pathogenesis of circulatory dysfunction and in the clinical course of cirrhosis remains unclear. We performed a prospective study assessing LV function, cardiac chronotropic response to the endogenous sympathetic nervous activity, and effective arterial blood volume in a large series of patients with cirrhosis, portal hypertension (PH), and normal serum creatinine. The aim of the study was to analyze the frequency, characteristics of LVDD, and its potential role in the impairment in circulatory function and clinical course of these patients. Patients and Methods Study Population. A total of 220 patients with complications of cirrhosis were admitted in hospital (Department of Gastroenterology, Hospital Ramon y Cajal, University of Alcala, Madrid, Spain) between November 2007 and November 2009 were evaluated. Inclusion criteria were (1) age range of years, (2) cirrhosis as diagnosed by histology or clinical, laboratory, and ultrasonography (USG) findings, and (3) presence of PH and normal serum creatinine concentration (<1.2 mg/dl). Patients were excluded if they had age >60 years (n 5 46), cardiac disease (n 5 5), arterial hypertension (n 5 7), obesity (n 5 3), diabetes mellitus (n 5 20), respiratory (n 5 9) or renal disease (n 5 10), portal vein thrombosis (n 5 2), TIPS insertion (n 5 5), hepatocellular carcinoma (n 5 23), or taking medications that could potentially affect cardiac function (n 5 10). No patient was receiving b-blockers because they had contraindication for the treatment or they were being treated with band ligation. Ten alcoholic patients were active drinkers at inclusion, and all of them had compensated cirrhosis. Patients with infection, encephalopathy grade III-IV, tense ascites, or gastrointestinal (GI) hemorrhage were considered after 1 month of recovery of these complications. All study subjects gave informed consent to participate in the study, which was approved by the clinical investigation and ethics committee of Ramon y Cajal Hospital of Madrid. A baseline study was performed after at least 4 days on a mmol/day sodium diet and without diuretics. Complete history and physical examination, chest and abdominal X-rays, electrocardiogram, abdominal USG, laboratory tests, and blood and ascitic fluid cultures were performed. At 8:00 a.m. of the fifth day after overnight fasting and after 1 hour of bed rest, samples were obtained to measure liver and renal function tests, plasma renin activity (PRA), plasma concentrations of aldosterone (ALDO), norepinephrine (NE), atrial natriuretic factor (ANF), and brain natriuretic peptide (BNP). Urine was collected for 24 hours. Patients were subsequently transferred to the Hepatic Hemodynamics Unit, and hemodynamics measurements were obtained. Subsequent to 2 hours after hemodynamics measurements, all study subjects underwent transthoracic echocardiography (TTE) to assess cardiac structure and systolic and diastolic function. Patients were discharged from the hospital with diuretics, norfloxacin, lactulose, or band ligation to prevent recurrence of ascites, SBP, hepatic encephalopathy Address reprint requests to: Luis Ruiz-del- Arbol, M.D., Hepatic Hemodynamic Unit, Gastroenterology Department, Hospital Ramon y Cajal, Ctra. de Colmenar Viejo Km 9.1, Madrid, Spain. lruizarbol@gmail.com; fax: Copyright VC 2013 by the American Association for the Study of Liver Diseases. View this article online at wileyonlinelibrary.com. DOI /hep Potential conflict of interest: Nothing to report. Additional Supporting Information may be found in the online version of this article.

3 1734 RUIZ-DEL-ARBOL ET AL. HEPATOLOGY, November 2013 (HE), and variceal bleeding, respectively. After discharge from the hospital, patients were followed up for at least 1 year in the outpatient clinic. During followup, we performed an evaluation of all bacterial infections, variceal bleeding, HE, and type 1 HRS 19 occurring in the patients included in the study. These patients were managed with standard therapy (Supporting Materials). Patients transplanted during followup were considered as censored at the time of transplantation. Hemodynamic and Neurohormonal Measurements. Under fluoroscopic control, a Swan-Ganz catheter (Abbott Labs, Abbott Park, IL) was advanced into the pulmonary artery for measurement of cardiopulmonary pressures (right atrial pressure [RAP], pulmonary artery pressure [PAP], and pulmonary capillary wedged pressure [PCWP]) and cardiac output (CO). A 7-F balloon-tipped catheter (MediTech Cooper Scientific Corp., Watertown, MA) was advanced into the main right hepatic vein to measure wedged and free hepatic venous pressures (WHVP and FHVP, respectively). Hepatic venous pressure gradient (HVPG) was calculated as the difference between WHVP and FHVP. All measurements were performed in triplicate and the average taken. 20 Heart rate and mean arterial pressure (MAP) were measured with an automatic sphygmomanometer. Systemic vascular resistance was calculated as follows: MAP (mmhg) 2 RAP (mmhg)/co (L/min 21 ) Left ventricular stroke work was calculated as follows: (stroke volume 3 [MAP 2 PCWP] ) (g m-m). PRA, ALDO, NE, and ANF were determined as previously described. 20 BNP was measured using a chemiluminometric immunoassay run on the ADVIA Centaur Immunochemistry analyzer (Siemens Healthcare Diagnostics, Tarrytown, NY). Values in healthy subjects on a low-sodium diet were as follows: ng/ml/hour, ng/dl, pg/ml, fmol/ml, and pg/ml, respectively. Echocardiography. TTE was performed using commercially available instruments operating in a MHz transducer in standard parasternal and apical views according to the recommendations of the American Society of Echocardiography (ASE). 21 Calculations of different cardiac dimension and volumes were assessed by M-mode cursor. Left ventricular ejection fraction (LVEF) was obtained by a modified version of Simpson s method. 22 Left ventricular hypertrophy (LVH) was defined as aleft ventricular mass indexed to body surface area (LVMI) that exceeded 143 or 102 g for men and women, respectively. 23 To obtain mitral inflow pattern, pulsed-wave Doppler echocardiography recordings were obtained from a sample volume positioned at the tips of the mitral valves parallel to inflow during diastole at end-expiration. The following parameters were measured: isovolumetric relaxation time (IVRT); peak early filling (E) and its deceleration time (DT); atrial filling peak (A); and the early diastolic mitral inflow velocity/late diastolic (E/A) ratio. E/A ratio was corrected for age. Recordings of mitral inflow with Valsalva maneuver were generally not performed. TDI measurements were sampled at the level of the mitral annulus over the septal wall. Peak early diastolic annular velocity (e 0 ) was measured at the septal and lateral mitral annular sites. 21 Values of e 0 measured at both sites were averaged. The combined E/e ratio was also calculated. All recordings were performed at a sweep speed of mm/sec and averaged over three consecutive cardiac cycles. All echocardiograms were interpreted by J.N., who had no knowledge of the clinical and laboratory data. Definitions. LVDD was defined and classified according to ASE guidelines. 21 LVDD included the following categories: grade 1: e <8 cm/sec, E/e ratio <8, E/A ratio <0.8, and DT >200 ms; grade 2: e <8 cm/sec, E/e ratio 9-15, E/A ratio , and DT ms; and grade 3: e <8 cm/sec, E/e ratio >15, E/A ratio >2, and DT <160 ms. Normal ventricular function at rest was defined by an LVEF >50% and without LVDD (e 8 cm/s, E/e ratio <8, and E/A ratio >1). Effective arterial blood volume was assessed by measuring plasma concentration of PRA. The criteria used to define decreased arterial blood volume were derived from those used in previous studies as an increase in PRA to a level >4 ng/ml/hour. 20 Statistical Analysis. Results are reported as frequencies or means 6 standard deviation (SD) plus 95% confidence interval (CI) of the mean. The Student t, Mann-Witney s, or chi-squared tests were used to compare continuous or categorical variables. For comparisons of multiple independent groups, Kruskal- Wallis test was used, followed by Mann-Withney s test. Univariate analyses were used to identify variables associated with development of type 1 HRS as well as with survival. Cox s proportional hazards method was used to assess the prognostic value of these variables. Accuracy of each independent predictive factor of survival was assessed by receiver operating characteristic curves. Kaplan-Meier s analysis was used to estimate survival, and probability curves were compared by logrank test. A P value <0.025 was considered statistically

4 HEPATOLOGY, Vol. 58, No. 5, 2013 RUIZ-DEL-ARBOL ET AL Table 1. Demographic, Clinical, Biochemical, Echocardiographic, Hemodynamic, and Neurohormonal Data of Patients With Cirrhosis at Time of Inclusion Classified According to Degree of LVDD Characteristics Grade 0 LVDD (n 5 43) Grade 1 LVDD (n 5 19) Grade 2 LVDD (n 5 18) Age, years (46 52) (48 55) (41 51) Sex, male/female 35/8 17/2 15/3 Etiology of cirrhosis, n (alcohol/viral/other) 19/18/6 10/5/4 6/9/3 Presence of ascites, n (%) 23 (53) 14 (74) 17 (94)** Presence of HE, grade I-II n (%) 2 (5) 2 (10) 7 (39)** Liver and renal function test Child-Pugh class (A/B/C), n 11/16/16 1/8/10 0/6/12 Child-Pugh score (7 9) (9 10) ** (9 11) MELD score (13 17) (14 19) *** (18 24) Serum creatinine, mg/dl ( ) ( ) ( ) Serum sodium, meq/l ( ) ( ) ( ) Echocardiographic data LVMI, g/m ( ) ( ) **** ( ) Diastolic function e, cm/sec ( ) ( ) **** ( ) E/e ratio ( ) ( ) **** ( ) LAVI, ml/m (17-22) (20 27) **** (29 40) IVRT, ms (83 96) ( ) ( ) E/A ratio ( ) ( ) ( ) DT, ms ( ) ( ) ( ) Systolic function LVEF, % (73 78) (70 77) *** (65 72) LVEDV, ml (92 124) ( ) ( ) LVESV, ml (16 23) (22 40) *** (27 49) Hemodynamic assesment and portal hepatic measurements MAP, mmhg (85 90) (79 87) **** (69 80) HR, bpm (75 84) (72 83) (68 81) RAP, mmhg ( ) ( ) ** ( ) PAP, mmhg ( ) ( ) *** ( ) PCWP, mmhg ( ) ( ) **** ( ) CO, L/min ( ) ( ) ** ( ) LV stroke work, gm-m (86 100) (69 93) **** (55 74) SVR, dyne/sec/cm 25 1, (1,002 1,211) 1, (965 1,318) 1, (887 1,139) WHVP, mmhg ( ) (26 34) (27 34) FHVP, mmhg ( ) ( ) ( ) HVPG, mmhg ( ) ( ) ( ) Neurohormonal systems Plasma renin activity, ng/ml/hour ( ) ( ) **** ( ) Aldosterone, ng/dl (32 55) (44 114) **** (63 145) Norepinephrine, pg/ml ( ) ( ) **** ( ) Natriuretic peptides B-type natriuretic peptide, pg/ml (22 30) (45 70) **** ( ) Atrial natriuretic factor, fmol/ml ( ) (15 31) **** (35 52) Data are presented as mean 6 SD (range: 95% CI of the mean). *P < 0.025; ** P < 0.01; *** P < 0.005; **** P < 0.001; LVDD grade 2 with respect to values of LVDD grade 0. P < 0.025; P < 0.01; P < 0.005; P < 0.001; LVDD grade 2 with respect to values of LVDD grade 1. P < 0.025; P < 0.01; P < 0.005; P < 0.001; LVDD grade 1 with respect to values of LVDD grade 0. Abbreviations: LVEDV, left ventricular end diastolic volume; LVESV, left ventricular end systolic volume; SVR, systemic vascular resistance. significant for comparisons of multiple groups. All statistical analyses were performed using SPSS 15.0 software (SPSS, Inc., Chicago, IL). Results The investigation included 80 patients. At rest, all had a normal ejection fraction (>50%). Forty-three patients had normal LV diastolic function, 19 had grade 1 LVDD, and 18 had grade 2 LVDD. There was no patient with LVDD grade 3. All patients with cirrhosis with active alcoholism at inclusion of the study had no LVDD. Characteristics of Patients. Table 1 shows the demographic, clinical, biochemical, and echocardiographic data of all patients included in the study. There were no significant differences among subgroups according to age, sex, etiology of liver disease, and

5 1736 RUIZ-DEL-ARBOL ET AL. HEPATOLOGY, November 2013 Fig. 1. Box-plot depiction of relationship between heart/plasma norepinephrine in patients with cirrhosis according to the presence of effective arterial blood volume: no ascites, ascites with normal plasma rennin activity and ascites with increased plasma rennin activity. The box defines the interquartile range, with the median indicated by the crossbar. Error bars indicate 5th and 95th percentiles. renal function tests. Patients from the grade 2 LVDD subgroup had greater frequency of HE, ascites, and higher Model for End-Stage Liver Disease (MELD) score, compared to patients of the other subgroups. There were no significant differences in these parameters between patients with grade 1 and grade 0 LVDD. LVDD: Relationship With Cardiopulmonary Pressures and Natriuretic Peptides. Echocardiographic data for patients classified according to grade of LVDD are shown in Table 1. By definition, patients with LVDD had a reduced e 0. As expected, E/e ratio significantly increased with advanced LVDD. Patients with grade 1 and grade 2 LVDD had ventricular hypertrophy (LVMI) and higher left atrial volume index (LAVI) than in patients with grade 0 LVDD, and this was associated with a significant progressive increase of plasma levels of ANF and BNP (Table 1). As compared to patients with grade 1 LVDD, patients with grade 2 LVDD showed significantly higher LAVI, cardiopulmonary pressures (RAP, PAP, and PCWP), and circulating plasma levels of natriuretic peptides (ANF and BNP; Table 1). In the whole series of patients, E/ e ratio correlated directly with PCWP (r ; P < 0.001) and BNP (r ; P < 0.001). Left Ventricular Systolic Function, Cardiac Chronotropic Function, Hemodynamics, and Neurohormonal Systems: Relationship to LVDD. Left ventricular systolic function, as estimated by CO, left ventricular stroke work and LVEF (Table 1), and cardiac chronotropic function (heart rate [HR]/plasma norepinephrine; Fig. 1) were significantly reduced in patients with grade 2 LVDD, as compared to grade 0 LVDD. No significant differences in these parameters were found when patients with grade 1 and grade 0 LVDD were compared. Patients with grade 2 LVDD showed significantly lower MAP, as compared to patients in the other two subgroups. Patients with grade 1 and grade 2 LVDD showed a higher, and significant, progressive stimulation of PRA as well as plasma levels of ALDO and NE than patients with grade 0 LVDD. There were no differences among the subgroups in peripheral vascular resistance and WHVP, FHVP, and HVPG. Systemic Circulatory Dysfunction: Relationship With Left Ventricular and Cardiac Chronotropic Function. According to the presence of ascites and degree of impairment in effective arterial blood volume assessed by measuring plasma concentration of PRA, patients were divided into three groups (Table 2). Group 1 included patients with compensated cirrhosis (patients who had never had ascites; n 5 26). All had normal PRA. Group 2 included patients with ascites, but normal PRA (<4 ng/ml/hour). Group 3 included patients with ascites and increased PRA. PRA is used as a surrogate of effective arterial blood volume. Grade 1 and grade 2 LVDD were significantly more frequent in patients with ascites and increased PRA than in the other two groups. No significant difference was observed when patients with ascites and normal PRA were compared to patients without ascites. Liver function was more deteriorated in patients with ascites than patients with compensated cirrhosis. MAP significantly decreased from patients without ascites to patients with ascites and increased PRA. Left ventricular systolic function (left ventricular stroke work) and cardiac chronotropic function (Fig. 1) were significantly reduced and plasma concentration of ALDO, NE, ANF, and BNP were significantly increased in patients with ascites and high PRA, as compared to the other two groups. Type 1 HRS Development During Follow-up. Twenty-seven of eighty patients (34%) developed at least one episode of complications of cirrhosis during follow-up. Variceal bleeding developed in 4 cases, HE in 12, and bacterial infections in 27. Fourteen patients developed type 1 HRS. Table 3 compares patients with moderate ascites who did and did not develop type 1 HRS during follow-up. Only patients with moderate ascites were compared, because no patients with minimal or without ascites developed the syndrome. Patients from group A (patients developing type 1 HRS) showed reduced LV

6 HEPATOLOGY, Vol. 58, No. 5, 2013 RUIZ-DEL-ARBOL ET AL Table 2. Relationship Between Degree of Circulatory and LVDD in Patients Divided According to Presence of Compensated and Decompensated Cirrhosis Variables Compensated Cirrhosis Patients Without Ascites (n 5 26) Patients With Ascites but Normal Plasma Renin Activity (n 5 18) Decompensated Cirrhosis Patients With Ascites and Increased Plasma Renin Activity (n 5 36) Child-Pugh Score *** MELD score *** Serum creatinine, mg/dl Echocardiographic data LVDD (%) Grade 0 20 (77) 16 (89) 7 (19.5) Grade 1 5 (19) 2 (11) 12 (33.3)* Grade 2 1 (4) 0 (0) 17 (47.2)**** e, cm/sec *** E/e ratio **** LVEF, % Hemodynamic assesment and systemic circulatory function MAP, mmhg **** HR, bpm RAP, mmhg PAP, mmhg PCWP, mmhg CO, L/min *** LV stroke work, gm-m **** SVR, dyne/sec/cm 25 1, , , Plasma renin activity, ng/ml/hour **** Aldosterone, ng/dl **** Norepinephrine, pg/ml **** Natriuretic peptides B-type natriuretic peptide, pg/ml **** Atrial natriuretic factor, fmol/ml * Data are presented as mean 6 SD. * P < 0.025; ** P < 0.01; *** P < 0.005; **** P < 0.001; patients with ascites and increased plasma renin activity with respect to values of patients without ascites. P < 0.025; P < 0.01; P < 0.005; P < 0.001; patients with ascites and increased plasma renin activity with respect to values of patients with ascites but normal plasma renin activity. P < 0.025; P < 0.01; P < 0.005; P < 0.001; patients with ascites but normal plasma renin activity with respect to values of patients without ascites. Abbreviation: SVR, systemic vascular resistance. diastolic function and significantly lower MAP and higher levels of PRA as well as plasma concentration of ALDO, NE, BNP, and ANF, compared to patients from group B. No significant differences were observed in liver function and hepatic hemodynamics. Of the variables showing significant differences between groups, only PRA (relative risk [RR]: 1.24; 95% CI: ; P ) and E/e ratio (RR, 1.55; 95% CI: ; P ) were independently associated with development of HRS type 1 according to a multivariate analysis. Survival. At the end of follow-up, 56 (70%) of the 80 patients were alive, 17 (21%) had died, and 7 (9%) had received a transplant. Table 4 shows the comparison between patients who died and those who survived. Significant differences were found in Child- Pugh and MELD scores, LV diastolic function (e and E/e ratio), MAP, PRA, and plasma levels of ALDO, NE, BNP, and ANF. In multivariate analysis, only E/e ratio and MELD score were significant for predicting 1-year mortality (area under the curve [AUC] [range, ] and [range, ], respectively). The accuracy of the E/e ratio alone in the prediction of survival was not modified by the contribution of liver failure, as estimated by MELD >15 points (E/e 0 alone RR: 2.10; 95% CI: ; P < 0.001; MELD plus E/e 0 RR: 1.99; 95% CI: ; P < 0.001). The value of the E/e ratio with higher sensitivity and specificity to predict 12-month survival was 10 (Fig. 2). Survival was significantly greater in E/e <10, compared to the E/e 10 group (91% and 29% [P < ], respectively). The relationship between the E/e ratio and 12-month probability of survival is shown in Fig. 3. Figure 4 shows 1-year probability of survival curves of patients classified according to diastolic function.

7 1738 RUIZ-DEL-ARBOL ET AL. HEPATOLOGY, November 2013 Table 3. Baseline Measurements in Patients With Cirrhosis and Moderate Ascites Who Presented Type 1 HRS (Group A) and in Patients Who Did Not Develop Type 1 HRS During Follow-up (Group B) Characteristics Group A (n 5 14) Group B (n 5 40) P Value Child-Pugh score MELD score Serum creatinine, mg/dl Serum sodium, meq/l Echocardiographic data LVMI, g/m Diastolic function e, cm/sec E/e ratio LAVI, ml/m IVRT, ms E/A ratio DT, ms Systolic function LVEF, % LVEDV, ml LVESV, ml Hemodynamic assesment and portal hepatic measurements MAP, mmhg HR, beats/min RAP, mmhg PAP, mmhg PCWP, mmhg CO, L/min LV stroke work, gm-m SVR, dyne/sec/cm 25 1, , HVPG, mmhg Neurohormonal systems Plasma renin activity, ng/ml/hour Aldosterone, ng/dl Norepinephrine, pg/ml Natriuretic peptides B-type natriuretic peptide, pg/ml Atrial natriuretic f actor, fmol/ml Fig. 2. Probability of survival of patients with cirrhosis and LVDD and without LVDD classified according to the E/e ratio obtained at inclusion of the study. Figures under curves are the patients at risk at different time points. relationship to clinical course and prognosis. In most studies thus far performed in cirrhosis, diagnosis of LVDD has been based on E/A ratio <1 using twodimensional (2D) Doppler echocardiography. However, the E/A ratio is strongly dependent on preload. 21,24 TDI is superior to conventional 2D Doppler echocardiography for diagnosing LVDD. Unlike transmitral valve Doppler flow, TDI directly measures the velocity of myocardial displacement as the LV expands in the diastole and therefore is independent of volume Data are presented as mean 6 SD. Abbreviations: LVEDV, left ventricular end diastolic volume; LVESV, left ventricular end systolic volume; SVR, systemic vascular resistance. Probability of survival was significantly different among the three groups. Patients without LVDD had the longest and those with grade 2 LVDD the shortest probability of survival (95% versus 39% [P < 0.01], respectively). Causes of death were hepatic failure/cirrhosis (n 5 2), HRS type 1 (n 5 5), multiorgan failure (n 5 5), infections (n 5 2), and GI hemorrhage (n 5 3). Discussion The current study reports on a prospective investigation of LVDD in patients with cirrhosis with PH and normal creatinine and provides information on the mechanisms of cardiocirculatory dysfunction and its Fig. 3. Relationship between the E/e ratio and probability of survival in patients with cirrhosis and LVDD and without LVDD included in the current study.

8 HEPATOLOGY, Vol. 58, No. 5, 2013 RUIZ-DEL-ARBOL ET AL Fig. 4. Twelve-month probability of survival in patients with cirrhosis categorized in three different groups according to severity of LV diastolic function: LVDD, grade 1, 2, and 3 (P < 0.001). Figures under curves are patients at risk at different time points. status and left atrial pressure. The tissue velocity measured at the basal part of the lateral and septal LV wall during early filling (e ) is primarily determined by the relaxation of the LV. TDI velocities continuously decline from normal to LVDD and have high feasibility and reproducibility. As a consequence, the ASE has included TDI parameters in the definition of LVDD. In our study, the diagnosis of LVDD was based on this technique, although we also explored our patients with conventional echocardiography for additional measurements. We excluded patients with several cobormidities to avoid confounding their effects on LV diastolic function. We did not include patients older than 60 years because it has been reported that LVDD is very frequent in healthy subjects above this age. 21 This represents a limitation of our study in the assessment of the prevalence of this condition in cirrhosis. LV systolic function was estimated by the CO, LV stroke volume was measured by standard hemodynamic techniques, and LVEF was estimated by conventional echocardiography. Cardiac inotropic function was estimated as the HR/plasma noradrenaline ratio, because plasma noradrenaline concentration is a surrogate of the effective arterial hypovolemia and secondary to activation of sympathetic nervous activity. Therefore, the HR/plasma noradrenaline ratio estimates cardiac chronotropic response to systemic circulatory dysfunction. The backward increase in cardiopulmonary pressures induced by LV dysfunction was estimated by measuring LAVI, RAP, PAP, and PWCP as well as the plasma concentration of brain and atrial natriuretic peptides. The cardiac production of these hormones increases in response to stretching of the ventricular wall and volume overload. 25 Peripheral vascular resistance is reduced in patients with cirrhosis as a consequence of splanchnic arterial vasodilation. 26 However, it is a poor index of the progression of this abnormality because it does not take into account the homeostatic response of the endogenous vasoconstrictor systems (the renin-angiotensin system, sympathetic nervous system, and antidiuretic hormone) that induces vasoconstriction in the extrasplanchnic vascular territories. 27,28 Therefore, effective arterial blood volume (which is the result of the interaction of CO and peripheral vascular resistance) in cirrhosis is generally estimated by the degree of activity Table 4. Comparison of Biochemical, Echocardiographic, Hemodynamic, and Neurohormonal Characteristics of Patients Included in the Study Divided According to Whether They Were Alive or Dead at 12 Months Characteristics Alive (n 5 63) Dead (n 5 17) P Value Child-Pugh Score MELD score Echocardiographic data LVMI, g/m Diastolic function e, cm/s <0.001 E/e ratio <0.001 LAVI, ml/m <0.001 IVRT, ms E/A ratio DT, ms Systolic function LVEF, % LVEDV, ml LVESV, ml Hemodynamic assesment and portal hepatic measurements MAP, mmhg <0.001 HR, beats/min RAP, mmhg PAP, mmhg PCWP, mmhg CO, L/min SVR, dyne/sec/cm 25 1, , HVPG, mmhg Neurohormonal systems Plasma renin activity, ng/ml/hour <0.001 Aldosterone, ng/dl <0.001 Norepinephrine, pg/ml <0.001 Natriuretic peptides B-type natriuretic peptide, pg/ml <0.001 Atrial natriuretic factor, fmol/ml <0.001 Data are presented as mean 6 SD. Abbreviations: LVEDV, left ventricular end diastolic volume; LVESV, left ventricular end systolic volume; SVR, systemic vascular resistance.

9 1740 RUIZ-DEL-ARBOL ET AL. HEPATOLOGY, November 2013 of these endogenous vasoconstrictor systems. 29 In our study, we used the PRA as a surrogate of effective arterial blood volume. LVDD was found in 37 patients. In 53% of these patients, LVDD was of grade 1 and in 47% of grade 2. Patients with LVDD grade 1 and 2 showed significantly greater LAVI than patients with grade 0. Cardiopulmonary pressures were significantly higher in patients with grade 2 LVDD than in patients with grade 0 and grade 1.There was a relationship between values of PCWP with BNP concentration and E/e ratio. This explains the higher plasma levels of both ANF and BNP found in LVDD patients. Although the increase in PCWP was less pronounced in patients with cirrhosis 30 than in patients with LVDD and cardiac disease, their values exclude the possibility that patients with LVDD did not have sufficient circulatory volume. In addition, LVDD was associated with changes in cardiac structure. Seventy-five percent of the patients with LVDD had increased LVMI, indicating the existence of LVH. In heart diseases, LVDD usually precedes LV systolic dysfunction. However, although systolic function was normal in all cases, patients with grade 2 LVDD had a significantly lower resting CO, LV stroke volume, and LVEF than those without LVDD. Recent studies have also reported that CO decreases during the course of the liver disease. 31 Therefore, LVDD may compromise the LV systolic function at rest. Finally, there was not any tachycardia in patients with LVDD, despite a significant increase of plasma norepinephrine levels. Consequently, the HR/norepinephrine ratio was reduced in these patients, indicating there is an impaired cardiac chronotropic function toward effective arterial blood volume. 5,6,32,33 These findings suggest there was a relationship between the severity of LVDD and other types of cardiac function abnormalities. We investigated the relationship between cardiac dysfunction and degree of impairment in effective arterial blood volume after classifying patients into three groups: (1) patients without effective arterial hypovolemia (compensated cirrhosis); (2) patients with ascites and normal PRA, representing a subgroup of patients with early decompensated cirrhosis. The mechanism of sodium retention and ascites in these patients is unknown, although it has been suggested that it could be related to a slight decrease in effective arterial blood volume not detectable by current markers 34 ; and (3) patients with ascites and increased PRA, representing a group with significant effective arterial hypovolemia. As expected, patients of this latter group also showed significantly increased plasma aldosterone and norepinephrine concentration. There are studies indicating that they have increased vascular resistances in nonsplanchnic vascular territories, such as the brain, kidneys, muscle, and skin. 27,28 The percentage of patients with grade 1 and 2 LVDD was more significant in the group of patients with ascites and increased PRA than those with ascites, but normal PRA, or without ascites. Patients with cirrhosis with significant effective arterial blood volume showed arterial hypotension and reduced left ventricular systolic function (CO and LV stroke work) and cardiac chronotropic function, as compared to the other two groups. These data indicate that cardiac dysfunction plays an important role in the pathogenesis of the impairment of effective arterial blood volume in cirrhosis. Our results indicate that nonazotemic patients with cirrhosis with LVDD are specially predisposed to develop HRS. During follow-up, HRS type 1 was diagnosed in 14 of 80 (17%) patients. Patients who went on to develop HRS type 1 were clearly at a more advanced stage of LVDD and lower effective arterial blood volume than patients who did not develop type 1 HRS. The categorization of patients in three groups according to diastolic function has prognostic relevance. Patients without LVDD had the longest and those with grade 2 LVDD the shortest probability of survival. Cardiac dysfunction were related to degree of liver failure, as indicated by a higher prevalence of ascites and HE and higher Child-Pugh and MELD scores in patients with grade 2 LVDD than in those with grade 0 LVDD. This observation has also been reported by other studies. 2,4,10 However, our data indicate that the E/e 0 ratio is an independent prognostic factor of survival. In addition, the accuracy of the E/e ratio in the prediction of survival was not modified by severity of liver function estimated by MELD >15. Previous studies have also demonstrated that an E/A ratio 1 was an independent predictor of death in patients with cirrhosis who are treated with TIPS. 35,36 These data indicate that the increased mortality risk in LVDD could be related to a more deteriorated cardiocirculatory function that occurs simultaneously and in parallel with the progression of liver failure. The current investigation shows that cardiac dysfunction in cirrhosis is the result of several primary cardiac disorders, which correlated with the impairment of effective arterial blood volume and degree of liver failure. Our results suggest that LVDD and its severity is a sensitive marker of advanced cirrhosis, type 1 HRS development, and mortality. Therefore, the examination of any patient with cirrhosis and PH and

10 HEPATOLOGY, Vol. 58, No. 5, 2013 RUIZ-DEL-ARBOL ET AL normal creatinine should include an echocardiographic study with TDI, and when an E/e ratio >10 is diagnosed, it should be necessary to carry out a careful follow-up of patients who are candidates for LT. Acknowledgment: The authors thank Wladimiro Jimenez (Biochemistry and Molecular Genetics, Hospital Clınico, Barcelona, Spain) for his collaboration in the hormonal study and Alfonso Muriel) Department of Bioestadistica, Hospital Ramon y Cajal, Madrid, Spain). References 1. Lee SS. Cardiac abnormalities in liver cirrhosis. West J Med 1989;151: Alqahtani SA, Fouad TR, Lee SS. Cirrhotic cardiomyopathy. Sem Liver Dis 2008;28: Mïller S, Henriksen JH. Cardiovascular complications of cirrhosis. Gut 2008;57: Wong F. Cirrhotic cardiomyopathy. Hepatol Int 2009;3: Bernardi M, Rubboli A, Trevisani F, Cancellieri C, Ligabue A, Baraldini M, et al. Reduced cardiovascular responsiveness to exercise-induced sympathoadrenergic stimulation in patients with cirrhosis. J Hepatol 1991;12: Wong F, Girgrah N, Graba J, Allidina Y, Liu P, Blendis L. The effect of cirrhotic cardiomyopathy on the cardiac response to exercise in cirrhosis. Gut 2001;49: Valeriano V, Funaro S, Lionetti R, Riggio O, Pulcinelli G, Fiore P, et al. Modification of cardiac function in cirrhotic patients with and without ascites. Am J Gastroenterol 2000;95: Finucci G, Desideri A, Sacerdoti D, Bolognesi M, Merkel C, Angeli P, et al. Left ventricular diastolic dysfunction in liver cirrhosis. Scan J Gastroenterol 1996;31: Zambruni A, Trevisani F, Caraceni P, Bernardi M. Cardiac electrophysiological abnormalities in patients with cirrhosis. J Hepatol 2006;44: Pozzi M, Carugo S, Boari G, Pecci V, de Ceglia S, Maggiolini S, et al. Functional and structural cardiac abnormalities in cirrhotic patients with and without ascites. HEPATOLOGY 1997;26: Wong F, Liu P, Lilly L, Bomzon A, Blendis L. Role or cardiac structural and functional abnormalities in the pathogenesis of hyperdynamic circulation and renal sodium retention in cirrhosis. Clin Sci (Lond) 1999;97: Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol 1997;30: Paulus WJ, Tschope C, Sanderson JE, Rusconi C, Flachskampf FA, Rademakers FE, et al. How to diagnose diastolic heart failure: a consensus statement on the diagnosis of heart failure with normal left ventricular ejection fraction by the Heart Failure and Echocardiography Associations of the European Society of Cardiology. Eur Heart J 2007;28: 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: Lee SS. Cardiac dysfunction in spontaneous bacterial peritonitis: a manifestation of cirrhotic cardiomyopathy? HEPATOLOGY 2003;38: Gines P, Uriz J, Calahorra B, Garcia-Tsao G, Kamath PS, Ruiz-del- Arbol L, et al. Transjugular intrahepatic portosystemic shunting versus paracentesis plus albumin for refractory ascites in cirrhosis. Gastroenterology 2002;123: Myers RP, Lee SS. Cirrhotic cardiomyopathy and liver transplantation. Liver Transpl 2000;6:S44-S Zardi EM, Abbate A, Zardi DM, Sobrina A, Margiotta D, Van Tassel BW, et al. Cirrhotic cardiomyopathy. J Am Coll Cardiol 2010;56: Salerno F, Gerbes A, Gines P, Wong F, Arroyo V. Diagnosis, prevention, and treatment of the hepatorenal syndrome in cirrhosis. A consensus workshop of the International Ascites Club. Gut 2007;56: Ruiz-del-Arbol L, Monescillo A, Jimenez W, Garcıa-Plaza A, Arroyo V, Rodes J. Paracentesis-induced circulatory disfunction. Mechanism and effect on hepatic hemodynamic in cirrhosis. Gastroenterology 1997;113: 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: Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, et al.; for the Chamber Quantification Writing Group. Recommendations for chamber quantification. J Am Soc Echocardiogr. 2005;18: Levy D, Savage DD, Garrison RJ, Anderson KM, Kannel WB, Castelli WP. Echocardiographic criteria for left ventricular hypertrophy: The Framingham Heart Study. Am J Cardiol 1987;59: Salerno F, Cazzaniga M, Pagnozzi G, Cirello I, Nicolini A, Meregaglia D, Burdick L. Humoral and cardiac effects of TIPS in cirrhotic patients with different effective blood volume. HEPATOLOGY 2003;38: Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998;339: Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rodes J. Peripheral arterial vasodilation hypothesis: a proposal for the initiation of renal sodium and water retention in cirrhosis. HEPATOLOGY 1988;8: Sacerdoti D, Bolognesi M, Merkel C, Angeli P, Gatta A. Renal vasoconstriction in cirrhosis evaluated by duplex Doppler ultrasonography. HEPATOLOGY 1993;17: Maroto A, Gines P, Arroyo V, Gines A, Salo J, Claria J, et al. Brachial and femoral artery blood flow in cirrhosis: relationship to kidney dysfunction. HEPATOLOGY 1993;17: Arroyo V, Bosch J, Mauri M, Ribera F, Navarro-Lopez F, Rodes J. Effect of angiotensin-ii blockade on systemic and hepatic haemodynamics and on the renin-angiotensin-aldosterone system in cirrhosis with ascites. Eur J Clin Invest 1981;11: Henriksen JH, Bendtsen F, Sorensen TI, Stadeager C, Ring-Larsen H. Reduced central blood volume in cirrhosis. Gastroenterology 1989;97: Krag A, Bendtsen F, Henriksen JH, Moller S. Low cardiac output predicts development of hepatorenal syndrome and survival in patients with cirrhosis and ascites. Gut 2010;59: Ruiz-del-Arbol L, Monescillo A, Arocena C, Valer P, Ginès P, Moreira V, et al. Circulatory function and hepatorenal syndrome in cirrhosis. HEPATOLOGY 2005;42: Ma Z, Miyamoto A, Lee SS. Role of altered beta-adrenoceptor signal transduction in the pathogenesis of cirrhotic cardiomyopathy in rats. Gastroenterology 1996;110: Arroyo V, Guevara M, Gines P. Hepatorenal syndrome in cirrhosis: pathogenesis and treatment. Gastroenterology 2002;122: Cazzaniga M, Salerno F, Pagnozzi G, Dionigi E, Visentin S, Cirello I, et al. Diastolic dysfunction is associated with poor survival in patients with cirrhosis with transjugular intrahepatic portosystemic shunt. Gut 2007;56: 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: