Advertisement

Abnormal Liver Function in Relation to Hemodynamic Profile in Heart Failure Patients

Published:September 27, 2009DOI:https://doi.org/10.1016/j.cardfail.2009.08.002

      Abstract

      Background

      We studied the relation between liver function abnormalities and hemodynamic profile in patients with heart failure (HF).

      Methods and Results

      In 323 HF patients, liver function was determined by aspartate and alanine aminotransferase (AST, ALT), alkaline phosphatase, γ-glutamyl transpeptidase (GGT), lactate dehydrogenase, and direct and total bilirubin (Bili dir, Bili tot). Central venous pressure (CVP) and cardiac index (CI) were determined invasively. Follow-up consisted of time to all-cause mortality. Mean age was 53 ± 15 years, and 60% were male. In multivariable analysis, all liver function tests related to CVP, but higher CVP was predominantly related to GGT (r = 0.336, P < .001) and Bili dir (r = 0.370, P < .001). Only elevated AST (r =−0.177, P < .01), ALT (r = −0.130, P < .05), and Bili tot (r = −0.158, P < .01) were associated with both low CI and elevated CVP. The prognostic value of abnormal liver function tests was related to their interaction with CI and CVP.

      Conclusions

      Elevated liver function tests mainly indicate higher CVP, whereas only the presence of elevated AST, ALT, or Bili dir may indicate a low CI. The absence of prognostic information in the presence of invasive hemodynamic measurements suggests that abnormal liver function tests in HF reflect a poor hemodynamic status.

      Key Words

      Despite initiation of new therapies, both the short- and long-term mortality rate of patients with heart failure is still high.
      • Dickstein K.
      • Cohen-Solal A.
      • Filippatos G.
      • et al.
      ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM).
      • Jaarsma T.
      • van der Wal M.H.
      • Lesman-Leegte I.
      • et al.
      Effect of moderate or intensive disease management program on outcome in patients with heart failure: Coordinating Study Evaluating Outcomes of Advising and Counseling in Heart Failure (COACH).
      This may be at least partly attributable to frequently present comorbidities.
      • Groenveld H.F.
      • Januzzi J.L.
      • Damman K.
      • et al.
      Anemia and mortality in heart failure patients a systematic review and meta-analysis.
      • Hillege H.L.
      • Nitsch D.
      • Pfeffer M.A.
      • et al.
      Renal function as a predictor of outcome in a broad spectrum of patients with heart failure.
      • Macchia A.
      • Monte S.
      • Romero M.
      • D'Ettorre A.
      • Tognoni G.
      The prognostic influence of chronic obstructive pulmonary disease in patients hospitalised for chronic heart failure.
      Heart failure itself is characterized by impaired organ perfusion resulting from both forward failure and increased central venous pressure (backward failure). We recently showed that both forward and backward failure are the most important determinants of renal dysfunction in heart failure.
      • Damman K.
      • Navis G.
      • Smilde T.D.
      • et al.
      Decreased cardiac output, venous congestion and the association with renal impairment in patients with cardiac dysfunction.
      • Damman K.
      • van Deursen V.M.
      • Navis G.
      • Voors A.A.
      • van Veldhuisen D.J.
      • Hillege H.L.
      Increased central venous pressure is associated with impaired renal function and mortality in a broad spectrum of patients with cardiovascular disease.
      Liver function abnormalities are frequently found in patients with heart failure and are related to a poor outcome.
      • Allen L.A.
      • Felker G.M.
      • Pocock S.
      • et al.
      Liver function abnormalities and outcome in patients with chronic heart failure: data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program.
      • Batin P.
      • Wickens M.
      • McEntegart D.
      • Fullwood L.
      • Cowley A.J.
      The importance of abnormalities of liver function tests in predicting mortality in chronic heart failure.
      • Felder L.
      • Mund A.
      • Parker J.G.
      Liver function tests in chronic congestive heart failure.
      • Jolliffe N.
      Liver function in congestive heart failure.
      • Myers J.D.
      • Hickam J.B.
      An estimation of the hepatic blood flow and splanchnic oxygen consumption in heart failure.
      • Richman S.M.
      • Delman A.J.
      • Grob D.
      Alterations in indices of liver function in congestive heart failure with particular reference to serum enzymes.
      • Sherlock S.
      The liver in heart failure; relation of anatomical, functional, and circulatory changes.
      • Shinagawa H.
      • Inomata T.
      • Koitabashi T.
      • et al.
      Prognostic significance of increased serum bilirubin levels coincident with cardiac decompensation in chronic heart failure.
      Individual small reports have highlighted the importance of either high central venous pressure or reduced hepatic perfusion.
      • Myers J.D.
      • Hickam J.B.
      An estimation of the hepatic blood flow and splanchnic oxygen consumption in heart failure.
      • Henrion J.
      • Schapira M.
      • Luwaert R.
      • Colin L.
      • Delannoy A.
      • Heller F.R.
      Hypoxic hepatitis: clinical and hemodynamic study in 142 consecutive cases.
      • Kubo S.H.
      • Walter B.A.
      • John D.H.
      • Clark M.
      • Cody R.J.
      Liver function abnormalities in chronic heart failure. Influence of systemic hemodynamics.
      • Lau G.T.
      • Tan H.C.
      • Kritharides L.
      Type of liver dysfunction in heart failure and its relation to the severity of tricuspid regurgitation.
      However, the relative contribution of reduced perfusion (forward failure) or venous congestion (backward failure) in causing alterations in specific markers of liver function has not been established.
      We studied the relation between liver function abnormalities and forward and backward failure in patients with heart failure.

      Methods

      Retrospective chart review was done to analyze characteristics of all patients that underwent right heart catheterization between January 1, 1989, and December 31, 2006, at the University Medical Center Groningen, The Netherlands. For each patient, date of birth, gender, weight, and height were collected. Comorbid conditions, medical history, laboratory values including serum creatinine and hemoglobin levels, and use of medication were also collected. Left ventricular ejection fraction (LVEF) measured within a 6-month interval before or after catheterization was recorded. For the present analysis, all patients with a diagnosis of clinical heart failure at the time of right heart catheterization were included.

      Heart Catheterization

      Hemodynamic variables obtained during catheterization included systolic blood pressure (SBP, mm Hg), diastolic blood pressure (DBP, mm Hg), cardiac output (thermodilution, L/min), and right atrial pressure as indicator of central venous pressure (CVP, mm Hg). Cardiac index (L·min·m2) was determined as cardiac output divided by the body surface area. Body surface area was calculated as 0.007184·weight0.425 ·length0.725. Body mass index was calculated as weight/length2. Measurements obtained from cardiac catheterization were obtained from the patient during a resting state.

      Liver Function Testing

      Laboratory measurements were extracted from samples drawn within 3 days before catheterization. Liver function tests that were extracted included aspartate aminotransferase (AST, upper limit of normal [ULN] 40 U/L), alanine aminotransferase (ALT, ULN 30 U/L), alkaline phosphatase (ALP, ULN 120 U/L), γ-glutamyl transpeptidase (GGT, ULN 65 U/L), lactate dehydrogenase (LDH, ULN 235 U/L), direct bilirubin (Bili dir, ULN 5 μmol/L), and total bilirubin (Bili tot, ULN 26 μmol/L). Abnormal liver function tests were defined as values above the upper limit of normal. To account for confounding by drug-induced liver injury, we investigated medication use of all patients that showed either a hepatocellular profile (ALT >3 × ULN), a cholestatic profile (ALP >2 × ULN, ALT/ALP <2) or a mixed profile (ALP and ALT >ULN) of liver injury, according to Chang et al.
      • Chang C.Y.
      • Schiano T.D.
      Review article: drug hepatotoxicity.
      In addition, we also screened patients with liver function tests values higher than 5 times ULN. After exclusion of subjects on possible hepatotoxic medication, a history of hepatitis or substance abuse, and missing laboratory samples within 3 days before right heart catheterization, 323 heart failure patients were available for the present analysis.

      Mortality Data

      Survival status was determined using the electronic patient registration database of the University Medical Center Groningen. Follow up started directly after right heart catheterization. The end point of interest was death from any cause.

      Statistical Analysis

      Data are given as mean ± standard deviation when normally distributed, as median and interquartile range when skewed distributed, and as frequencies and percentages for categorical variables. Differences between baseline variables were evaluated by means Student t-test, the Kruskal-Wallis test, and χ2 or Fisher exact tests, when appropriate. CI and CVP were ranked to arbitrary high versus low values (CI </≥ 2.5 L·min·m2 and CVP ≤ 8/> 8 mm Hg), corresponding to the lower limit of normal values of CI and higher upper limit of normal values of CVP, corresponding to hemodynamic profiles in heart failure.
      • Stevenson L.W.
      • Perloff J.K.
      The limited reliability of physical signs for estimating hemodynamics in chronic heart failure.
      Accordingly, we stratified patients to 4 different hemodynamic profiles.
      • 1.
        Patients with a normal/high cardiac index (>2.5 L·min·m2) and normal/low CVP (≤8 mm Hg)
      • 2.
        Patients with a normal/high cardiac index (>2.5 L·min·m2) and a high CVP (>8 mm Hg)
      • 3.
        Patients with a low cardiac index (≤2.5 L·min·m2) and a normal/low CVP (≤8 mm Hg)
      • 4.
        Patients with a low cardiac index (≤2.5 L·min·m2) and a high CVP (>8 mm Hg)
      Initial linear regression analyses for liver function parameters were performed using partial correlation coefficients, adjusted for age and gender. Afterwards, multivariable regression analysis included CI, CVP, age, and gender into the model. Interactions between CI and CVP were modeled and a P value < .1 was deemed as a significant interaction. To account for possible confounding by body mass index, a history of diabetes, SBP, and DBP, these factors were introduced as covariates in a secondary multivariable model. We used a Cox proportional hazards model to estimate hazard ratios for all-cause mortality with 95% CI. Multivariable models were constructed in the following way: liver function tests with noninvasive variable, liver function test with invasive measurements (CVP and CI), and a stepwise multivariable model. A P value < .05 was considered statistically significant. Statistical analyses were performed using SPSS, Chicago, version 12.0, and STATA, College Station, TX, version 9.0.

      Results

      Baseline characteristics of the patient population are shown in Table 1. Age was 53 ± 15 years, and 60% were male. Direct bilirubin (62%) and LDH (65%) were often abnormal in this heart failure patient cohort, whereas AST (18%) or ALT (43%) abnormalities were much less common (Fig. 1). When stratified for different hemodynamic profiles, there were marked differences in these 4 groups with respect to liver function tests. In general, patients with lower CI tended to have significantly higher values of liver function parameters, whereas a similar, but less pronounced pattern was observed in patients with higher CVP, but relatively normal CI levels. In patients with both high CVP and low CI, the highest values of most liver function parameters were observed.
      Table 1Baseline Characteristic According to Different Hemodynamic Profiles
      TotalHigh CI

      Low CVP
      Low CI

      Low CVP
      High CI

      High CVP
      Low CI

      High CVP
      n (%)323171 (53)71 (22)29 (9)52 (16)
      Age (y)53±1553±1554±1556±1150±16
      Gender (% male)6059614868
      SBP (mm Hg)121±28124±25114±29
      P < .05 vs. high CI, low CVP.
      118±27105±21
      P < .01 vs. high CI, low CVP.
      P < .01 vs. low CI, low CVP.
      DBP (mm Hg)67±1268±1265±1065±1171±12
      P < .05 vs. low CI, low CVP.
      BMI (kg/m2)25±425±425±427±4
      P < .05 vs. low CI, low CVP.
      25±4
      CVP (mm Hg)6±64±24±212±315±6
      CI (L·min·m2)2.7±0.83.2±0.62.1±0.33.2±0.61.9±0.4
      CO (L/min)5.2±1.66.2±1.43.9±0.76.3±1.53.6±0.9
      LVEF (%)28±1331±1326±12
      P < .05 vs. high CI, low CVP.
      32±1322±13
      P < .01 vs. high CI, low CVP.
      P < .01 vs. low CI, low CVP.
      P < .01 vs. High CI, High CVP.
      Hb (g/dL)13.5±1.913.6±2.014.1±1.512.1±2.6
      P < .01 vs. high CI, low CVP.
      P < .05 vs. high CI, high CVP.
      13.7±1.9
      P < .01 vs. low CI, low CVP.
      Creatinine (mg/dL)1.18 (1.00–1.41)1.08 (0.94–1.29)1.16 (1.01–1.35)1.30 (0.99–1.71)
      P < .05 vs. high CI, low CVP.
      1.32 (1.13–1.55)
      P < .01 vs. high CI, low CVP.
      P < .05 vs. low CI, low CVP.
      eGFR (mL·min·1.73 m2)63±2168±2163±1854±21
      P < .01 vs. high CI, low CVP.
      P < .05 vs. low CI, low CVP.
      59±20
      P < .05 vs. high CI, low CVP.
      Liver function tests
       GGT (U/L)48 (26–100)30 (20–56)46 (30–87)
      P < .01 vs. high CI, low CVP.
      82 (42–245)
      P < .01 vs. high CI, low CVP.
      P < .05 vs. low CI, low CVP.
      102 (54–147)
      P < .01 vs. high CI, low CVP.
      P < .05 vs. high CI, high CVP.
       ALP (U/L)87 (66–106)79 (62–94)88 (70–112)
      P < .01 vs. high CI, low CVP.
      95 (74–133)
      P < .05 vs. high CI, low CVP.
      96 (77–117)
      P < .01 vs. high CI, low CVP.
       Bili Tot (μmol/L)16 (10–26)13 (8–16)20 (13–26)
      P < .01 vs. high CI, low CVP.
      19 (14–27)
      P < .01 vs. high CI, low CVP.
      30 (18–46)
      P < .01 vs. high CI, low CVP.
      P < .05 vs. high CI, high CVP.
      P < .01 vs. low CI, low CVP.
       Bili Dir (μmol/L)6 (4–12)5 (3–8)7 (4–11)
      P < .05 vs. high CI, low CVP.
      9 (7–14)
      P < .01 vs. high CI, low CVP.
      P < .05 vs. low CI, low CVP.
      16 (7–25)
      P < .01 vs. high CI, low CVP.
      P < .05 vs. high CI, high CVP.
       AST (U/L)28 (23–38)25 (22–33)31 (23–42)
      P < .01 vs. high CI, low CVP.
      29 (22–39)36 (26–56)
      P < .01 vs. high CI, low CVP.
      P < .01 vs. low CI, low CVP.
       ALT (U/L)29 (19–43)24 (17–39)30 (21–44)
      P < .05 vs. high CI, low CVP.
      24 (16–33)29 (21–78)
      P < .01 vs. high CI, low CVP.
      P < .01 vs. low CI, low CVP.
       LDH (U/L)256 (220–321)238 (207–287)278 (234–343)
      P < .01 vs. high CI, low CVP.
      317 (257–429)
      P < .01 vs. high CI, low CVP.
      284 (239–362)
      P < .01 vs. high CI, low CVP.
      Medication (%)
       Diuretics614767
      P < .01 vs. high CI, low CVP.
      72
      P < .05 vs. high CI, low CVP.
      86
      P < .01 vs. high CI, low CVP.
      P < .05 vs. low CI, low CVP.
       β-blocker3327363647
      P < .05 vs. high CI, low CVP.
       ACEi or ARB5851576463
       Aldosterone-antagonist191230
      P < .01 vs. high CI, low CVP.
      40
      P < .01 vs. high CI, low CVP.
      26
      P < .05 vs. high CI, low CVP.
      ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; BMI, body mass index; CI, cardiac index; CO, cardiac output; CVP, central venous pressure; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; SBP, systolic blood pressure; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; Bili dir, direct bilirubin; Bili tot, total bilirubin; NS, nonsignificant; LVEF, left ventricular ejection fraction; Hb, hemoglobin.
      Values presented as mean±SD, median (25-75%), or total n and percentages.
      P < .05 vs. high CI, low CVP.
      P < .01 vs. high CI, low CVP.
      P < .05 vs. low CI, low CVP.
      || P < .01 vs. low CI, low CVP.
      § P < .05 vs. high CI, high CVP.
      P < .01 vs. High CI, High CVP.
      Figure thumbnail gr1
      Fig. 1Percentages of abnormal liver function tests in patients with heart failure. ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; bili dir, direct bilirubin; bili tot: total bilirubin.

      Liver Function Tests and Hemodynamic Parameters

      Table 2 summarizes linear regression analysis, adjusted for age and gender and for all liver function tests, highlighting the association with CVP and CI. All liver function tests showed a significant relationship with CVP and an inverse relationship with CI. Comparing CVP and CI, the relationship with CVP was stronger for all liver function tests compared with CI. We found significant interactions between CVP and CI on the relationship with total bilirubin, AST, and ALT, but with not direct bilirubin, LDH, ALP, or GGT (Table 3). Figure 2 shows the percentages of abnormal liver function tests when both CVP and CI were ranked to high versus low, visualizing the significant interaction between CVP and CI on these liver function tests. Most liver function tests showed an increase in percentages of abnormal values with decreasing CI and increasing CVP. AST and ALT were particularly increased when CI was low, and GGT was particularly increased when CVP was high. Bilirubin tests were increased both with elevated CVP and low CI. In multivariable analysis, CVP remained significantly associated with all liver function tests (Table 3). In contrast, CI remained only significantly associated with AST, ALT, and total bilirubin. In a second multivariable analysis, which included body mass index, SBP, DBP, and a history of diabetes as covariates, the associations between CI, CVP, and liver function tests were unaffected (coefficients not shown).
      Table 2Univariate Regression Analysis for Different Liver Function Tests in 323 Heart Failure Patients
      GGT
      Log-transformed.
      ALP
      Log-transformed.
      Bili tot
      Log-transformed.
      Bili dir
      Log-transformed.
      AST
      Log-transformed.
      ALT
      Log-transformed.
      LDH
      Log-transformed.
      CI−0.229
      P < .01.
      −0.154
      P < .05.
      −0.295
      P < .001.
      −0.190
      P < .05.
      −0.310
      P < .001.
      −0.236
      P < .001.
      −0.108
      CVP0.367
      P < .001.
      0.193
      P < .01.
      0.423
      P < .001.
      0.424
      P < .001.
      0.319
      P < .001.
      0.257
      P < .001.
      0.223
      P < .001.
      CI, cardiac index; CVP, central venous pressure; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; Bili dir, direct bilirubin; Bili tot, total bilirubin.
      Partial correlation coefficients are shown adjusted for age and gender.
      P < .05.
      P < .01.
      P < .001.
      § Log-transformed.
      Table 3Multivariable regression Analysis for Different Liver Function Tests in 323 Heart Failure Patients
      GGT
      Log-transformed.
      ALP
      Log-transformed.
      Bili tot
      Log-transformed.
      Bili dir
      Log-transformed.
      AST
      Log-transformed.
      ALT
      Log-transformed.
      LDH
      Log-transformed.
      CI−0.094−0.084−0.158
      P < .01.
      −0.040−0.177
      P < .01.
      −0.130
      P < .05.
      −0.026
      CVP0.318
      P < .001.
      0.161
      P < .01.
      0.357
      P < .001.
      0.350
      P < .001.
      0.156
      P < .01.
      0.120
      P < .05.
      0.199
      P < .01.
      P value for interactionNSNS.017NS.011.010NS
      Adjusted r20.1840.0640.1980.1450.1530.1620.044
      CI, cardiac index; CVP, central venous pressure; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; Bili dir, direct bilirubin; Bili tot, total bilirubin.
      Partial correlation coefficients are shown adjusted for age and gender.
      P < .05.
      P < .01.
      P < .001.
      § Log-transformed.
      Figure thumbnail gr2
      Fig. 2Prevalence of abnormal liver function tests in subgroups of CI and CVP: interaction between CI and CVP. (A) Total bilirubin, (B) AST, (C) ALT. CI, cardiac index; CVP, central venous pressure; AST, aspartate aminotransferase; ALT, alanine aminotransferase.

      Liver Function Tests and Prognosis

      During a median follow-up time of 7.9 (4.1–11.8) years, a total of 122 (36%) patients died. Table 4 summarizes the relationship between the individual liver function tests and prognosis. In univariate analysis, GGT, ALP, AST, and LDH levels were significant predictors of all-cause mortality. In addition, both CVP (HR 1.03 per mm Hg [1.00–1.06], P < .05) and CI (HR 0.61 per L·min·m2 [0.45–0.83], P = .0016) significantly determined prognosis. After adjustment for noninvasive covariates, AST and LDH remained significantly associated with impaired survival, whereas ALT showed a trend with survival (P = .070). However, after adjustment for hemodynamics (CI and CVP), none of the liver function tests remained associated with impaired survival, which was attributable to inclusion of both parameters (CI and CVP) into the model.
      Table 4Cox Proportional Hazard Analysis for Different Parameters of Liver Function in 323 Heart Failure Patients
      UnivariateMultivariable
      Adjusted for age; gender; eGFR; left ventricular ejection fraction; hemoglobin; history of diabetes mellitus, coronary heart disease, and hypertension; systolic blood pressure; diastolic blood pressure; medication use.
      Multivariable
      Adjusted for CI and CVP.
      Hazard RatioP ValueHazard RatioP ValueHazard RatioP Value
      GGT
      Log-transformed.
      1.81 (1.07–3.08).0281.05 (0.40–2.77)NS1.37 (0.69–2.72)NS
      ALP
      Log-transformed.
      4.12 (1.33–12.78).0141.39 (0.14–14.24)NS2.68 (0.72–10.1)NS
      Bili Tot
      Log-transformed.
      1.44 (0.68–3.04)NS3.07 (0.59–15.92)NS0.73 (0.29–1.84)NS
      Bili Dir
      Log-transformed.
      1.77 (0.99–3.16)NS2.04 (0.59–7.05)NS1.25 (0.62–2.57)NS
      AST
      Log-transformed.
      1.77 (1.05–2.97).0333.44 (1.22–9.75).0201.16 (0.60–2.21)NS
      ALT
      Log-transformed.
      1.09 (0.63–1.89)NS2.32 (0.93–5.79).0700.85 (0.48–1.50)NS
      LDH
      Log-transformed.
      2.26 (1.07–4.76).0335.80 (1.30–26.0).0221.47 (0.62–3.49)NS
      CI, cardiac index; CVP, central venous pressure; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; Bili dir, direct bilirubin; Bili tot, total bilirubin.
      Log-transformed.
      (1) Adjusted for age; gender; eGFR; left ventricular ejection fraction; hemoglobin; history of diabetes mellitus, coronary heart disease, and hypertension; systolic blood pressure; diastolic blood pressure; medication use.
      (2) Adjusted for CI and CVP.

      Discussion

      In the present study, we show that liver function abnormalities are frequently observed in patients with heart failure—in particular, high levels of direct bilirubin, LDH, and GGT. Most parameters of liver function were predominantly related to CVP, whereas only AST, ALT, and total bilirubin were also related to a reduced CI. CVP and CI showed a significant interaction on the association with liver function parameters, indicating a mutually mediating role of both parameters on the relationship of each individual parameter with liver function. Finally, levels of GGT, ALP, and especially AST and LDH were predictors of all-cause mortality, but not independent of CI and CVP.

      Prevalence and Pathophysiology of Liver Function Abnormalities in Heart Failure

      The presence of liver function abnormalities in heart failure has long been recognized.
      • Jolliffe N.
      Liver function in congestive heart failure.
      • Sherlock S.
      The liver in heart failure; relation of anatomical, functional, and circulatory changes.
      In our analysis, the prevalence of liver function abnormalities ranged from as little as 15% to as much as 65%, depending on definition and type of liver function abnormality. We observed similar percentages of abnormal levels GGT, bilirubin levels, and a higher percentage of abnormal levels of ALT and AST compared with a study by Lau et al.
      • Lau G.T.
      • Tan H.C.
      • Kritharides L.
      Type of liver dysfunction in heart failure and its relation to the severity of tricuspid regurgitation.
      In comparison with a recent substudy of the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) trials we found similar percentages of abnormal levels of ALP, but much higher percentages of abnormal levels of other liver function tests.
      • Allen L.A.
      • Felker G.M.
      • Pocock S.
      • et al.
      Liver function abnormalities and outcome in patients with chronic heart failure: data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program.
      The pathophysiology of liver failure in heart failure patients has been attributed to either venous congestion leading to hepatic congestion or reduced cardiac output leading to hepatic hypoxic injury.
      • Myers J.D.
      • Hickam J.B.
      An estimation of the hepatic blood flow and splanchnic oxygen consumption in heart failure.
      • Sherlock S.
      The liver in heart failure; relation of anatomical, functional, and circulatory changes.
      • Lau G.T.
      • Tan H.C.
      • Kritharides L.
      Type of liver dysfunction in heart failure and its relation to the severity of tricuspid regurgitation.
      Liver failure in heart failure includes necrosis in the central zone of hepatic lobules resulting from direct compression and congestion.
      • Richman S.M.
      • Delman A.J.
      • Grob D.
      Alterations in indices of liver function in congestive heart failure with particular reference to serum enzymes.
      Interestingly, low perfusion seems to be less important because oxygen consumption can easily be increased when hepatic blood flow is decreased,
      • Myers J.D.
      • Hickam J.B.
      An estimation of the hepatic blood flow and splanchnic oxygen consumption in heart failure.
      because 70% of the blood supply of the liver is dependent of the portal system, whereas only 30% is delivered by the hepatic artery.
      • Naschitz J.E.
      • Slobodin G.
      • Lewis R.J.
      • Zuckerman E.
      • Yeshurun D.
      Heart diseases affecting the liver and liver diseases affecting the heart.
      A nonhemodynamic factor that may be related to liver failure is poor nutrition state of heart failure patients, leading to fatty liver and fibrosis.
      • White T.J.
      • Leevy C.M.L.
      • Brusca A.M.
      • Gnassi A.M.
      The liver in congestive heart failure.
      The present study offers more insight in the association of a combination of congestion and reduced perfusion with the presence of liver function abnormalities. Historically, increased levels of AST and ALT in heart failure have been attributed to hepatocellular damage from decreased perfusion, whereas especially increased bilirubin levels, high ALP levels, and low ALT/ALP ratio in heart failure have been associated with cholestatic liver injury from an increased CVP. Kubo et al have shown that with increasingly severe heart failure, the occurrence of liver function abnormalities increases.
      • Kubo S.H.
      • Walter B.A.
      • John D.H.
      • Clark M.
      • Cody R.J.
      Liver function abnormalities in chronic heart failure. Influence of systemic hemodynamics.
      This seemed to be parallel to an increase in CVP and decrease in cardiac output, although no interaction or multivariable analysis was performed. Our present univariate analyses are consistent with their findings, resulting in moderate relationships between CVP and liver function tests, and even weaker relationships with CI. Interestingly however, to the best of our knowledge, no other data are available in heart failure patients to support this generally accepted concept. We found significant interactions between CVP and CI on the relationship with liver function tests. Essentially, all levels of liver function parameters increased with decreasing CI and increasing CVP, except for ALT and AST. Both ALT and AST only showed elevations when CI was reduced. Because these abnormalities often coexist, this finding seems to be a reflection of the pathophysiology of liver function abnormalities in heart failure being a combination of congestion and reduced cardiac output. A similar analysis for total bilirubin levels was carried out by Shinagawa et al in 183 acute heart failure patients.
      • Shinagawa H.
      • Inomata T.
      • Koitabashi T.
      • et al.
      Prognostic significance of increased serum bilirubin levels coincident with cardiac decompensation in chronic heart failure.
      Their analysis showed that especially the combination of congestion and reduced perfusion led to elevated total bilirubin levels, consistent with our present findings. Unfortunately, no data on other parameters of liver dysfunction were shown. Lau et al showed that, especially, GGT and bilirubin levels were increased in patients with elevated CVP.
      • Lau G.T.
      • Tan H.C.
      • Kritharides L.
      Type of liver dysfunction in heart failure and its relation to the severity of tricuspid regurgitation.
      Together with our findings, this suggest that elevations in GGT, ALP, LDH, and bilirubin levels are dependent of both reduced perfusion and congestion, whereas AST and ALT levels are mainly determined by reduced hepatic perfusion. This suggests that low hepatic perfusion in heart failure mainly predisposes to hepatocellular liver injury (high AST and ALT), whereas cholestatic liver injury is primarily observed when CVP is particularly high (high bilirubins, GGT, ALP). Additionally, our findings may help to identify patients in different hemodynamic profiles, according to abnormal values, or combination of abnormal values of different liver function test (Table 5).
      Table 5Different Hemodynamic Profiles and Expected Abnormal Levels of Liver Function Tests
      Low CVPHigh CVP
      High CIGGT =GGT ↑↑
      ALP =ALP ↑
      Bili tot =Bili tot ↑
      Bili dir =Bili dir ↑
      AST =AST ↑
      ALT =ALT ↑
      LDH =LDH ↑
      Low CIGGT =GGT ↑↑
      ALP =ALP ↑
      Bili tot ↑Bili tot ↑↑
      Bili dir =Bili dir ↑↑
      AST ↑↑AST ↑↑
      ALT ↑ALT ↑↑
      LDH =LDH ↑
      CI, cardiac index; CVP, central venous pressure; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase GGT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; Bili dir, direct bilirubin; Bili tot, total bilirubin.

      Prognosis

      Several reports have addressed the prognostic importance of liver function abnormalities in the setting of heart failure. Batin et al showed that individual abnormal liver function tests were related to increased mortality, including bilirubin and AST.
      • Batin P.
      • Wickens M.
      • McEntegart D.
      • Fullwood L.
      • Cowley A.J.
      The importance of abnormalities of liver function tests in predicting mortality in chronic heart failure.
      Shinagawa also emphasized the importance of elevated total bilirubin levels, whereas direct bilirubin, ALP, and GGT were also related to cardiac events.
      • Shinagawa H.
      • Inomata T.
      • Koitabashi T.
      • et al.
      Prognostic significance of increased serum bilirubin levels coincident with cardiac decompensation in chronic heart failure.
      In a recent report, Allen et al reported independent prognostic information of bilirubin in the CHARM study population, but in our present study, we were unable to find a relationship between bilirubin levels and mortality.
      • Allen L.A.
      • Felker G.M.
      • Pocock S.
      • et al.
      Liver function abnormalities and outcome in patients with chronic heart failure: data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program.
      We did, however, find that GGT, ALP, LDH, and AST were significantly related to mortality in univariate analysis. Even after correction for easily obtainable, noninvasive covariates, AST and LDH remained significant predictors of prognosis. Interestingly, after adjustment for hemodynamic derangement (CVP and CI), none of these liver function tests remained associated with reduced survival.
      This observation seems to indicate that the severity of liver function abnormalities is related to the type of liver function test and the relative contributions of CVP and CI. The prognostic importance of these abnormalities seems to be more a reflection of the poor hemodynamic status of these patients, instead of underlying secondary hepatic injury.

      Limitations

      This is a retrospective analysis of a selected patient population, which includes patients who were identified in the early 1990s. Considering the improvement in medical therapy in patients with heart failure in recent years, our heart failure population may have a higher prevalence of liver function abnormalities compared with the general heart failure population now. We carried out multiple comparisons in a limited set of patients, which increases the change of finding significant relationships. Our results should therefore be confirmed in other studies. Furthermore, changes in liver function abnormalities were not assessed and we cannot assess whether the observed abnormalities are transient or permanent. We did not evaluate neurohormonal activation, echocardiographic assessment of tricuspid regurgitation, or measured hepatic blood flow, all of which could potentially influenced the results. In our analysis, we have tried to exclude patients who were on hepatotoxic medication, but we cannot account for over-the-counter products with possible harmful effects on liver function, such as acetaminophen. Finally, we could not assess other reasons for liver test abnormalities in heart failures, such as hemochromatosis, because ferritin measurements were not routinely performed, and were therefore not available for this analysis.

      Conclusion

      Liver function abnormalities are frequently observed in patients with heart failure. Elevated CVP showed the strongest relationship with liver function abnormalities. Despite interactions between CI and CVP, only AST, ALT, and total bilirubin were related to both CVP and CI, whereas both GGT and direct bilirubin were the most prominent parameters related to CVP, in the absence of reduced CI. Finally, the prognostic information of liver function abnormalities in heart failure was blunted by elevated CVP and low CI, suggesting that liver function abnormalities are a reflection of poor hemodynamic status in these patients.

      References

        • Dickstein K.
        • Cohen-Solal A.
        • Filippatos G.
        • et al.
        ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008: the Task Force for the diagnosis and treatment of acute and chronic heart failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM).
        Eur J Heart Fail. 2008; 10: 933-989
        • Jaarsma T.
        • van der Wal M.H.
        • Lesman-Leegte I.
        • et al.
        Effect of moderate or intensive disease management program on outcome in patients with heart failure: Coordinating Study Evaluating Outcomes of Advising and Counseling in Heart Failure (COACH).
        Arch Intern Med. 2008; 168: 316-324
        • Groenveld H.F.
        • Januzzi J.L.
        • Damman K.
        • et al.
        Anemia and mortality in heart failure patients a systematic review and meta-analysis.
        J Am Coll Cardiol. 2008; 52: 818-827
        • Hillege H.L.
        • Nitsch D.
        • Pfeffer M.A.
        • et al.
        Renal function as a predictor of outcome in a broad spectrum of patients with heart failure.
        Circulation. 2006; 113: 671-678
        • Macchia A.
        • Monte S.
        • Romero M.
        • D'Ettorre A.
        • Tognoni G.
        The prognostic influence of chronic obstructive pulmonary disease in patients hospitalised for chronic heart failure.
        Eur J Heart Fail. 2007; 9: 942-948
        • Damman K.
        • Navis G.
        • Smilde T.D.
        • et al.
        Decreased cardiac output, venous congestion and the association with renal impairment in patients with cardiac dysfunction.
        Eur J Heart Fail. 2007; 9: 872-878
        • Damman K.
        • van Deursen V.M.
        • Navis G.
        • Voors A.A.
        • van Veldhuisen D.J.
        • Hillege H.L.
        Increased central venous pressure is associated with impaired renal function and mortality in a broad spectrum of patients with cardiovascular disease.
        J Am Coll Cardiol. 2009; 53: 582-588
        • Allen L.A.
        • Felker G.M.
        • Pocock S.
        • et al.
        Liver function abnormalities and outcome in patients with chronic heart failure: data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program.
        Eur J Heart Fail. 2009; 11: 170-177
        • Batin P.
        • Wickens M.
        • McEntegart D.
        • Fullwood L.
        • Cowley A.J.
        The importance of abnormalities of liver function tests in predicting mortality in chronic heart failure.
        Eur Heart J. 1995; 16: 1613-1618
        • Felder L.
        • Mund A.
        • Parker J.G.
        Liver function tests in chronic congestive heart failure.
        Circulation. 1950; 2: 286-297
        • Jolliffe N.
        Liver function in congestive heart failure.
        J Clin Invest. 1930; 8: 419-433
        • Myers J.D.
        • Hickam J.B.
        An estimation of the hepatic blood flow and splanchnic oxygen consumption in heart failure.
        J Clin Invest. 1948; 27: 620-627
        • Richman S.M.
        • Delman A.J.
        • Grob D.
        Alterations in indices of liver function in congestive heart failure with particular reference to serum enzymes.
        Am J Med. 1961; 30: 211-225
        • Sherlock S.
        The liver in heart failure; relation of anatomical, functional, and circulatory changes.
        Br Heart J. 1951; 13: 273-293
        • Shinagawa H.
        • Inomata T.
        • Koitabashi T.
        • et al.
        Prognostic significance of increased serum bilirubin levels coincident with cardiac decompensation in chronic heart failure.
        Circ J. 2008; 72: 364-369
        • Henrion J.
        • Schapira M.
        • Luwaert R.
        • Colin L.
        • Delannoy A.
        • Heller F.R.
        Hypoxic hepatitis: clinical and hemodynamic study in 142 consecutive cases.
        Medicine (Baltimore). 2003; 82: 392-406
        • Kubo S.H.
        • Walter B.A.
        • John D.H.
        • Clark M.
        • Cody R.J.
        Liver function abnormalities in chronic heart failure. Influence of systemic hemodynamics.
        Arch Intern Med. 1987; 147: 1227-1230
        • Lau G.T.
        • Tan H.C.
        • Kritharides L.
        Type of liver dysfunction in heart failure and its relation to the severity of tricuspid regurgitation.
        Am J Cardiol. 2002; 90: 1405-1409
        • Chang C.Y.
        • Schiano T.D.
        Review article: drug hepatotoxicity.
        Aliment Pharmacol Ther. 2007; 25: 1135-1151
        • Stevenson L.W.
        • Perloff J.K.
        The limited reliability of physical signs for estimating hemodynamics in chronic heart failure.
        JAMA. 1989; 261: 884-888
        • Naschitz J.E.
        • Slobodin G.
        • Lewis R.J.
        • Zuckerman E.
        • Yeshurun D.
        Heart diseases affecting the liver and liver diseases affecting the heart.
        Am Heart J. 2000; 140: 111-120
        • White T.J.
        • Leevy C.M.L.
        • Brusca A.M.
        • Gnassi A.M.
        The liver in congestive heart failure.
        Am Heart J. 1955; 49: 250-257