Increased B-Type Natriuretic Peptide During Liver Transplantation: Relationship to Invasively Measured Hemodynamic Parameters

Increased B-Type Natriuretic Peptide During Liver Transplantation: Relationship to Invasively Measured Hemodynamic Parameters I.Y. Huh, Y.K. Kim, W.J. Shin, S.E. Park, J.Y. Bang, and G.S. Hwang ABSTRACT Background. The role of B-type natriuretic peptide (BNP) concentration in predicting cardiac dysfunction has been extensively investigated in many clinical conditions. Little is known, however, about its relationships with hemodynamic parameters from right heart catheterization in patients undergoing liver transplant surgery. Methods. We retrospectively evaluated 525 patients who underwent liver transplantation. Hemodynamic variables from a Swan-Ganz catheter and BNP concentrations were measured 1 hour after induction of general anesthesia. Patients were stratified by quintiles of BNP concentrations. Univariate and multivariate logistic regression analysis were used to identify hemodynamic parameters associated with BNP ⱖ 135 pg/mL, a cutoff point for the 5th quintile. Results. Univariate analysis showed that factors significantly associated with BNP ⱖ 135 pg/mL included model for end-stage liver disease (MELD) score, diastolic blood pressure, mean pulmonary artery pressure, pulmonary capillary wedge pressure (PCWP), cardiac index, right ventricular end-diastolic volume index (RVEDVI), systemic vascular resistance index, pulmonary vascular resistance index (PVRI), and right ventricular stroke work index. Multivariate analysis revealed that MELD score (odds ratio [OR] ⫽ 1.059, P ⬍ .001), PCWP (OR ⫽ 1.116, P ⫽ .026), RVEDVI (OR ⫽ 1.010, P ⫽ .009), and PVRI (OR ⫽ 1.009, P ⫽ .002) were independent determinants of BNP ⱖ 135 pg/mL. Conclusions. Severity of liver disease, preload dependent hemodynamic parameters, and pulmonary vascular resistance were found to be significantly associated with increased BNP concentration, reinforcing the utility of BNP as a marker of cardiac strain and ventricular volume overload in liver failure patients undergoing liver transplant surgery. RAIN NATRIURETIC PEPTIDE (BNP) is a cardiac neurohormone secreted from the ventricular myocardium in response to volume expansion or pressure overload.1 In patients with heart failure, BNP concentration is correlated with the degree of cardiac dysfunction and is proportional to congestion.2– 4 BNP concentration is also elevated in most patients with pulmonary hypertension or pulmonary embolism resulting in right ventricular overload.5 In addition, higher circulating BNP concentration is a strong independent risk factor for major cardiovascular complications and mortality.6 – 8 Cirrhotic cardiomyopathy is a condition involving impaired cardiac contractility, systolic and diastolic dysfunction, and electromechanical abnormalities in cirrhotic patients without known cardiac diseases.9 BNP concentration has been shown to be associated with cirrhotic cardiomyopathy, the severity of cirrhosis,

B

the degree of cardiac dysfunction, and myocardial hypertrophy.9 –11 Although the ability of BNP concentration to predict cardiac dysfunction has been extensively investigated in many clinical conditions, less is known about the association

From the Department of Anesthesiology and Pain Medicine, Asan Medical Center (Y.-K.K., W.-J.S., J.-Y.B., G.-S.H.), University of Ulsan College of Medicine, Seoul, Korea, and Ulsan University Hospital (I.-Y.H., S.-E.P.), University of Ulsan College of Medicine, Ulsan, Korea. Address reprint requests to Gyu-Sam Hwang, MD, PhD, Professor, Department of Anesthesiology and Pain Medicine, Asan Medical Center, University of Ulsan College of Medicine, 388-1, Pungnap 2-dong, Songpa-gu, Seoul, 138-736, Korea. E-mail: [email protected]

0041-1345/12/$–see front matter http://dx.doi.org/10.1016/j.transproceed.2012.01.140

© 2012 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710

1318

Transplantation Proceedings, 44, 1318 –1322 (2012)

INCREASED B-TYPE NATRIURETIC PEPTIDE

between BNP concentration and invasively measured hemodynamic parameters during liver transplantation. We therefore evaluated the hemodynamic determinants associated with increased BNP concentration in patients undergoing liver transplantation. METHODS We retrospectively investigated 525 patients who underwent liver transplantation between January 2007 and February 2009; this patient cohort has been described previously.8 BNP concentration and hemodynamic parameters from right heart catheterization were measured concurrently in all recipients. The study population consisted of 391 males and 134 females, of mean age 50.2 ⫾ 8.6 years and mean body mass index 24.0 ⫾ 3.2 kg/m2. Of these recipients, 51.6% had hepatitis B virus-related cirrhosis, 31.2% had combined hepatocellular carcinoma and cirrhosis, 5.1% had alcoholic cirrhosis, 4% had fulminant hepatic failure, 1.7% had cryptogenic cirrhosis, 1.7% were undergoing retransplantation, 1.5% had autoimmune hepatitis, 1.0% had Wilson disease, 0.8% had Budd-Chiari syndrome, 0.8% had primary biliary cirrhosis, and 0.6% had primary sclerosing cholangitis. Disease severity was assessed by Model for End-stage Liver Disease (MELD) score.12 The study protocol was approved by the Institutional Review Board of the Asan Medical Center. All patients were anesthetized in the same manner. Briefly, anesthesia was induced with intravenous thiopental, fentanyl, and vecuronium and was maintained with 1% isoflurane, a 50% O2/air mixture, and continuous infusion of fentanyl and vecuronium. Five-lead electrocardiography and invasive radial arterial pressure were measured. A 7.5-French pulmonary artery catheter (SwanGanz CCOmbo V CCO/SvO2/CEDV, Edwards Lifesciences LLC, Calif, USA) connected to a Vigilance device (Vigilance II, Edwards Lifesciences) was inserted via a 9-French introducer sheath into the internal jugular vein and was advanced under guidance of the pressure curve to a wedged position to monitor invasive hemodynamic variables. One hour after induction of general anesthesia, and before surgical incision, hemodynamic variables were measured, including heart rate (HR), systolic radial arterial pressure (SRAP), diastolic radial arterial pressure (DRAP), cardiac index (CI), mean pulmonary artery pressure (MPAP), pulmonary capillary wedge pressure (PCWP), right ventricular end systolic volume index (RVESVI), right ventricular end diastolic volume index (RVEDVI), systemic vascular resistance index (SVRI), pulmonary vascular resistance index (PVRI), right ventricular stroke work index (RVSWI), left ventricular stroke work index (LVSWI), and right ventricular ejection fraction (RVEF). Plasma BNP concentration was measured simultaneously using the ADVIA Centaur CP Immunoassay System (Siemens Medical Solution Diagnostics, NY, USA). Following abdominal incision, the volume of ascites was measured directly from the suction bottle.

Statistics Normality of distributions was assessed using the Shapiro-Wilk test. Normally distributed continuous data were expressed as mean and standard deviation, and skewed data were expressed as median with interquartile range. Categorical data were expressed as absolute values and percentages. Quintiles of BNP concentration were compared using the ␹2 test for trend (for categorical data) and the Jonckheere-Terpstra test (for ordered continuous data). To assess the hemodynamic determinants associated with increased BNP, the recipients were dichotomized into two groups, those with BNP ⬍

1319 135 pg/mL and ⱖ 135 pg/mL by the cutoff point of the 5th quintile (ie, the highest 20% of BNP concentrations). Univariate logistic regression analysis was used to identify hemodynamic parameters associated with BNP ⱖ 135 pg/mL. Factors with a P value ⬍ .05 in univariate analysis were included in a multivariate logistic regression model using forward stepwise analysis. All statistical analyses were performed using SPSS version 12.0 software (SPSS Inc, Chicago, Ill, USA). P ⬍ .05 was considered statistically significant.

RESULTS

Median BNP concentration in all 525 patients was 60 pg/mL (range, 3–13,586 pg/mL), and cutoff values for the 5 BNP quintiles were 24, 47, 78, and 135 pg/mL, respectively. The relationships between intraoperative BNP quintiles and invasively measured hemodynamic variables are shown in Table 1. We found that, as BNP concentration increased, MPAP, PCWP, CI, PVRI, RVEDVI, RVSWI, and RVEF increased significantly, whereas DRAP and SVRI decreased significantly (P ⬍ .05 for trends; Table 1). Univariate logistic regression analysis revealed that MELD, ascites ⬎ 1 L, DRAP, MPAP, PCWP, CI, SVRI, PVRI, RVEDVI, RVSWI and RVEF were significant predictors of intraoperative BNP ⱖ 135 pg/mL, whereas age, sex, body mass index, SRAP, RVESVI, and LVSWI were not. In multivariate logistic regression analysis, MELD score (odds ratio [OR] ⫽ 1.059, P ⬍ .001), PCWP (OR ⫽ 1.116, P ⫽ .026), PVRI (OR ⫽ 1.009, P ⫽ .002), and RVEDVI (OR ⫽ 1.010, P ⫽ .009; Table 2) were independently associated with BNP ⱖ 135 pg/mL. DISCUSSION

We have shown here that elevated BNP concentration is independently associated with MELD score, PCWP, RVEDVI, and PVRI, indicating that the severity of liver disease, preload-dependent hemodynamic parameters, and pulmonary vascular resistance were significant predictors of intraoperatively elevated BNP during liver transplantation. Cardiovascular changes in cirrhotic patients vary from subtle to severe, thus a cardiac dysfunction termed “cirrhotic cardiomyopathy”.13,14 Previous studies have shown that elevated circulating concentrations of N-terminal-pro B-type natriuretic peptide and BNP in cirrhosis patients likely reflect increased cardiac ventricular generation of these peptides, thus indicating the presence of cardiac dysfunction, rather than being caused by the hyperdynamic circulatory changes (such as CI and SVRI) found in those patients.10,13 Our multivariate analysis showed similar results, demonstrating no relationship between increased BNP and CO or SVRI. Rather, increased BNP was proportional to volume index. Although cirrhotic patients had no signs of central overfilling, as central blood volume was either normal or decreased,10,13 an early report showed a direct relationship between end diastolic volume and central blood volume.15 In this study, in which patients showed evidence of congestion such as ascites, we found that BNP concentration was significantly related to central hypervolemia, as shown by PCWP and RVEDVI, which represent left and right ven-

1320

Table 1. Intraoperative BNP Quintiles and Demographic and Hemodynamic Data Q 1 (ⱕ 24 pg/mL)

Patient characteristics Sex (M/F) Age (y) BMI (kg/m2) MELD Diabetes Ascites (ⱖ1 L) Causes of liver disease Viruse-related HCC and cirrhosis Alcohol related FHF Other Laboratory variables BNP (pg/dL) Total bilirubin (mg/dL) Prothrombin time (INR) Creatinine (mg/dL) Hemodynamic variables SRAP (mm Hg) DRAP (mm Hg) MPAP (mm Hg) PCWP (mm Hg) CI (L/min/m2) SVRI (dyne · s/sm5/m2) PVRI (dyne · s/sm5/m2) RVDEVI (mL/m2) RVESVI (mL/m2) RVSWI (g · m/m2) LVSWI (g · m/m2) RVEF (%)

109 (84/25) 48.1 (46.6–50.0) 24.3 ⫾ 2.9 13.5 (12.0–15.1) 16 (14.7%) 19 (17.4%) 55 (50.5%) 36 (33.0%) 4 (3.7%) 5 (4.6%) 9 (8.3%)

Q 2 (25–46 pg/mL)

Q 3 (47–77 pg/mL)

Q 4 (78–134 pg/mL)

Q 5 (ⱖ 135 pg/mL)

101 (76/25) 50.6 (49.1–52.1) 23.9 ⫾ 3.2 17.2 (15.4–19.1) 16 (15.8%) 31 (30.7%)

105 (79/26) 51.9 (50.5–53.3) 24.2 ⫾ 3.5 18.9 (17.0–20.8) 21 (20%) 38 (36.2%)

106 (75/31) 50.8 (49.0–52.6) 24.0 ⫾ 3.3 20.5 (18.6–22.4) 22 (20.8%) 40 (37.7%)

104 (76/28) 50.0 (48.1–51.8) 23.6 ⫾ 2.9 26.4 (24.1–28.7) 17 (16.0%) 43 (41.3%)

56 (52.8%) 33 (31.1%) 6 (5.7%) 5 (4.7%) 6 (5.7%)

54 (51.9%) 30 (28.8%) 6 (5.8%) 5 (4.8%) 9 (8.7%)

51 (50.0%) 32 (31.7%) 5 (5.0%) 3 (3.0%) 10 (9.9%)

55 (52.4%) 33 (31.4%) 6 (5.7%) 3 (2.9%) 8 (7.6%)

P Value for Trend

.860 .041 .252 ⬍.001 .714 .002

17.0 (12.0–20.0) 2.0 (1.5–3.2) 1.4 (1.2–1.6) 0.7 (0.6–0.9)

35.0 (29.0–40.0) 2.5 (1.9–4.2) 1.6 (1.5–1.8) 0.7 (0.6–0.9)

60.0 (53.0–70.0) 3.4 (2.1–7.5) 1.7 (1.4–1.9) 0.7 (0.6–0.9)

100.5 (86.0–114.3) 3.4 (2.2–11.9) 1.7 (1.5–1.9) 0.7 (0.6–1.0)

245.0 (179.5–377.8) 8.6 (3.1–23.9) 1.9 (1.6–2.2) 0.9 (0.6–1.4)

⬍.001 ⬍.001 ⬍.001 .008

110.7 ⫾ 14.7 59.6 ⫾ 10.7 13.4 (12.7–14.0) 9.1 (8.6–9.6) 3.8 (3.7–4.0) 1617.5 ⫾ 505.3 99.0 ⫾ 43.4 148.3 (142.5–154.1) 103.1 ⫾ 36.7 4.7 (4.1–5.2) 46.8 ⫾ 11.3 33.9 ⫾ 6.4

109.2 ⫾ 16.5 57.2 ⫾ 11.4 14.1 (13.3–14.8) 9.5 (8.8–10.1) 3.9 (3.8–4.1) 1543.3 ⫾ 535.5 101.2 ⫾ 41.9 150.4 (144.6–156.2) 99.5 ⫾ 25.2 5.0 (4.4–5.6) 47.4 ⫾ 12.2 34.6 ⫾ 7.4

110.0 ⫾ 15.5 57.1 ⫾ 10.0 14.7 (14.0–15.3) 10.1 (9.5–10.7) 4.0 (3.8–4.2) 1441.4 ⫾ 438.6 105.2 ⫾ 41.9 155.3 (148.4–162.3) 102.6 ⫾ 28.3 5.5 (4.9–6.1) 49.2 ⫾ 11.8 35.2 ⫾ 6.1

112.1 ⫾ 14.6 57.4 ⫾ 9.3 16.4 (15.7–17.1) 10.8 (10.2–11.4) 4.0 (3.8–4.2) 1439.7 ⫾ 419.3 110.6 ⫾ 44.3 151.2 (145.4–157.1) 99.7 ⫾ 30.0 6.0 (5.3–6.7) 48.4 ⫾ 12.5 35.5 ⫾ 7.8

112.1 ⫾ 15.5 55.5 ⫾ 11.3 19.0 (18.8–20.2) 13.0 (12.1–14.0) 4.5 (4.2–4.8) 1263.3 ⫾ 447.8 119.2 ⫾ 47.0 165.3 (157.4–173.3) 105.6 ⫾ 30.7 7.4 (6.4–8.3) 47.6 ⫾ 12.4 36.0 ⫾ 7.4

.178 .015 ⬍.001 ⬍.001 .001 ⬍.001 .003 .003 .454 ⬍.001 .637 .015

HUH, KIM, SHIN ET AL

Data are expressed as mean ⫾ standard deviation, median (interquartile range), or number of patients, as appropriate. Q, quintile; BMI, body mass index; MELD, Model for End-stage Liver Disease; HCC, hepatocelluar cancer; FHF, fulminant hepatic failure; BNP, brain natriuretic peptide; INR, international normalized ratio; SRAP, systolic radial artery pressure; DRAP, diastolic radial artery pressure; MPAP, mean pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; CI, cardiac index; SVRI, systemic vascular resistance index; PVRI, pulmonary vascular resistance index; RVEDVI, right ventricular end diastolic volume index; RVESVI, right ventricular end systolic volume index; RVSWI, right ventricular stroke work index; LVSWI, left ventricular stroke work index; RVEF, right ventricular ejection fraction.

INCREASED B-TYPE NATRIURETIC PEPTIDE

1321

Table 2. Logistic Regression Analysis of Factors Predicting BNP > 135 pg/mL in Liver Transplant Recipients Univariate Analysis

Patient characteristics Sex (M/F) Age BMI MELD Diabetes Ascites (ⱖ 1L) Hemodynamic variables SRAP DRAP MPAP PCWP CI SVRI PVRI RVEDVI RVESVI RVSWI LVSWI RVEF

Multivariate Analysis

Odds Ratio (95% CI)

P Value

0.952 (0.583–1.556) 0.993 (0.968–1.017) 0.952 (0.887–1.022) 1.074 (1.053–1.095) 0.775 (0.425–1.413) 1.595 (1.022–2.490)

.846 .555 .173 ⬍.001 .405 .040

1.006 (0.992–1.020) 0.978 (0.957–0.998) 1.225 (1.161–1.293) 1.254 (1.176–1.338) 1.504 (1.243–1.820) 0.999 (0.998–0.999) 1.008 (1.003–1.013) 1.012 (1.006–1.018) 1.004 (0.998–1.011) 1.143 (1.082–1.207) 0.999 (0.981–1.017) 1.029 (0.998–1.061)

.415 .038 ⬍.001 ⬍.001 ⬍.001 ⬍.001 .002 ⬍.001 .209 ⬍.001 .880 .071

Odds Ratio (95% CI)

P Value

1.059 (1.036–1.084)

⬍.001

1.116 (1.013–1.228)

.026

1.009 (1.003–1.015) 1.010 (1.002–1.017)

.002 .009

BNP, brain natriuretic peptide; BMI, body mass index; MELD, Model for End-stage Liver Disease; SRAP, systolic radial artery pressure; DRAP, diastolic radial artery pressure; MPAP, mean pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; CI, cardiac index; SVRI, systemic vascular resistance index; PVRI, pulmonary vascular resistance index; RVEDVI, right ventricular end diastolic volume index; RVESVI, right ventricular end systolic volume index; RVSWI, right ventricular stroke work index; LVSWI, left ventricular stroke work index; RVEF, right ventricular ejection fraction.

tricular volume index, respectively. Therefore, our results demonstrate that elevated BNP is directly related to central hypervolemia in patients undergoing liver transplant surgery. Furthermore, we previously showed that elevated BNP (⬎136 pg/mL) was a significant independent predictor of 1-year all-cause mortality after liver transplantation.8 Since BNP is secreted in response to ventricular volume overload in heart failure patients,3 our findings emphasize the utility of BNP as a marker of cardiac strain in liver failure patients. In addition to being significantly elevated in pathological conditions that affect the left ventricle, BNP concentrations are increased in clinical conditions that lead to isolated acute or chronic right ventricular overload.5 We also showed that increased PVRI was related to increased BNP, reinforcing the importance of pulmonary circulation and right ventricular pressure and volume overload in patients undergoing liver transplantation surgery. Our study has the limitation that retrospective data from the operating room database was used, thus further study using prospective design will be needed. Additionally, we measured hemodynamic parameters under general anesthesia, thus our results probably cannot be generalized to all clinical conditions, since vascular tone and myocardial contractility may be sensitive to general anesthetics. In conclusion, using invasively measured hemodynamic parameters associated with right heart catheterization, we have shown that preload-dependent hemodynamic parameters such as PCWP, RVEDVI, and increased PVRI are significantly associated with increased plasma BNP concen-

tration. This finding reinforces the utility of BNP as a marker of cardiac strain and ventricular volume overload in liver failure patients undergoing liver transplant surgery. REFERENCES 1. Luchner A, Stevens TL, Borgeson DD, et al: Differential atrial and ventricular expression of myocardial BNP during evolution of heart failure. Am J Physiol 274:H1684, 1998 2. Baughman KL: B-type natriuretic peptide—a window to the heart. N Engl J Med 347:158, 2002 3. Dickstein K: Natriuretic peptides in detection of heart failure. Lancet 351:4, 1998 4. Levin ER, Gardner DG, Samson WK: Natriuretic peptides. N Engl J Med 339:321, 1998 5. Pruszczyk P: N-terminal pro-brain natriuretic peptide as an Indicator of right ventricular dysfunction. J Card Fail 11:S65, 2005 6. Richards AM, Nicholls MG, Espiner EA, et al: B-type natriuretic peptides and ejection fraction for prognosis after myocardial infarction. Circulation 107:2786, 2003 7. Scott PA, Barry J, Roberts PR, et al: Brain natriuretic peptide for the prediction of sudden cardiac death and ventricular arrhythmias: a meta-analysis. Eur J Heart Fail 11:958, 2009 8. Kim YK, Shin WJ, Song JG, et al: Evaluation of intraoperative brain natriuretic peptide as a predictor of 1-year mortality after liver transplantation. Transplant Proc 43:1684, 2011 9. Moller S, Henriksen JH: Cirrhotic cardiomyopathy: a pathophysiological review of circulatory dysfunction in liver disease. Heart 87:9, 2002 10. Henriksen JH, Gotze JP, Fuglsang S, et al: 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 52:1511, 2003 11. Wong F, Siu S, Liu P, et al: Brain natriuretic peptide: is it a predictor of cardiomyopathy in cirrhosis? Clin Sci (Lond) 101:621, 2001

1322 12. Hwang WJ, Jeon JP, Kang SH, et al: Sluggish decline in a post-transplant model for end-stage liver disease score is a predictor of mortality in living donor liver transplantation. Korean J Anesthesiol 59:160, 2010 13. Moller S, Henriksen JH: Cardiovascular complications of cirrhosis. Gut 57:268, 2008

HUH, KIM, SHIN ET AL 14. Ma Z, Lee SS: Cirrhotic cardiomyopathy: getting to the heart of the matter. Hepatology 24:451, 1996 15. Moller S, Sondergaard L, Mogelvang J, et al: Decreased right heart blood volume determined by magnetic resonance imaging: evidence of central underfilling in cirrhosis. Hepatology 22:472, 1995