- Email: [email protected]
Pretransplant Resting Heart Rate and Its Association With All-Cause Mortality in Liver Transplant Recipients
Pretransplant Resting Heart Rate and Its Association With All-Cause Mortality in Liver Transplant Recipients H.-M. Kwon, I.-G. Jun*, K.-W. Jung, Y.-J. Moon, W.-J. Shin, J.-G. Song, and G.-S. Hwang Department of Anesthesiology and Pain Medicine, Laboratory for Cardiovascular Dynamics, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
ABSTRACT Background. The importance of heart rate (HR) measurement as a prognostic factor has been recognized in many clinical conditions, such as hypertension, coronary artery disease, or heart failure. Patients with liver cirrhosis tend to have increased resting HR as consequence of hyperdynamic circulation. In the current study, we examined whether pretransplant resting increased HR is associated with overall mortality in cirrhotic patients following liver transplantation (LT). Patients and Methods. We retrospectively collected and analyzed the data of 881 liver recipients who underwent LT surgery between October 2009 and September 2012. Patients were categorized into 3 groups by tertile of resting HR as follows: tertile 1 group, HR 65 beats per minute (bpm); tertile 2 group, HR 66 to 80 bpm; and tertile 3 group, HR > 80 bpm. Results. Kaplan-Meier analysis showed that the all-cause mortality rate was significantly different according to tertiles of HR (P ¼ .016, log-rank test). The multivariate Cox regression analysis showed that tertile 3 group was significantly associated with higher risk for all-cause mortality (hazard ratio 1.83, 95% confidence interval, 1.10e3.07; P ¼ .021) compared with tertile 1 group, after adjusting for clinically significant variables in univariate analysis. Conclusions. Our results demonstrate that pretransplant resting tachycardia can identify patients at high risk of death in cirrhotic patients following LT, suggesting that further study will be need to clarify relationship between HR burden and sympathetic cardiac neuropathy.
H
EART rate (HR) is simply and routinely measured in hospitalized patients; it presents abundant information regarding the patient’s cardiovascular status. Recent studies demonstrated that preoperative resting tachycardia is a prognostic risk factor for perioperative myocardial injury, heart failure, and cardiac-related mortality across different patient populations such as patients with hypertension, coronary artery disease, or heart failure [1e4]. A widespread acknowledged explanation is that myocardial oxygen supplyedemand imbalance caused by tachycardia may provoke perioperative myocardial injury, which may contribute to increased morbidity and mortality [5e7]. Patients with advanced liver cirrhosis develop hyperdynamic circulation, which is characterized by increased 0041-1345/17 http://dx.doi.org/10.1016/j.transproceed.2017.03.043
1092
HR, activated sympathetic nervous system, and high cardiac output combined with low vascular resistance [8,9]. In this regard, understanding the basal cardiovascular status of patients is crucial to perform successful liver transplantation (LT). Furthermore, with advances in surgical technique, pretransplant cardiovascular disease is an important leading cause of morbidity and mortality after LT in the current
*Address correspondence to In-Gu Jun, MD, PhD, Department of Anesthesiology and Pain Medicine, Laboratory for Cardiovascular Dynamics, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea. E-mail: [email protected] ª 2017 Elsevier Inc. All rights reserved. 230 Park Avenue, New York, NY 10169
Transplantation Proceedings, 49, 1092e1096 (2017)
PRETRANSPLANT RESTING HEART RATE AND MORTALITY
era [6,10]. However, studies regarding relationship between resting HR and long-term mortality in liver recipients are limited. The aim of this study were to assess the usefulness of preoperative increased HR in predicting all-cause mortality after LT surgery, thereby addressing whether HR measurement provides additional prognostic information to show latent cardiovascular disease status of end-stage liver disease patients. PATIENTS AND METHODS In this retrospective study, we analyzed the database of 1048 consecutive patients who underwent LT surgery at Asan Medical Center from October 2009 to September 2012. All patients included were followed for 56 20 months, from the operational day to June 2016. The exclusion criteria were acute fulminant hepatic failure without chronic liver disease (n ¼ 39) or hepatic failure due to graft rejection following LT (n ¼ 32). We also excluded patients who had more than moderate valvular disease (n ¼ 4), significant obstructive coronary artery disease (n ¼ 6), previous history of cardiac surgery (n ¼ 3), significant arrhythmia such as atrial fibrillation (n ¼ 4), chronic kidney disease (n ¼ 14), and incomplete data (n ¼ 97). After approval of the institutional review board, we collected and analyzed data from computerized collecting registry (ABLE, Asan Biomedical Research Program, South Korea), which was performed in accordance with the ethical standards in the 1964 Declaration of Helsinki and its amendments. Because HR is quite variable, resting HR was defined as averaged value of HR when patients were admitted and the HR measured quietly before induction of
1093 anesthesia. Primary end point was 1-year overall all-cause mortality and mortality observed until June 2016. Patients were categorized into 3 groups by tertile analysis of resting HR as follows: tertile 1, HR 65 beats per minute (bpm) (median: 59 bpm, range 41e65 bpm); tertile 2, HR 66 to 80 bpm (median: 72 bpm, 66e80 bpm); and tertile 3, HR > 80 bpm (median: 88 bpm, 81e130 bpm). Laboratory data were collected the day before surgery. Categorical and continuous variables are shown as number (percentage), mean standard deviation if normally distributed, or median with interquartile range if skewed, appropriately. Differences between patients who survived or died were analyzed with a t test or Mann-Whitney U test and a c2 test or Fisher exact test. Variables according to HR tertiles were analyzed by a linear-bylinear c2 test for trend (for categorical data) and the JonckheereTerpstra test (for ordered continuous data). Survival rates were compared using the Kaplan-Meier survival curve analysis, and differences were compared using the log-rank test. To assess hazard ratios of resting HR on all-cause mortality, univariate and multiple logistic regression analyses using Cox proportional hazards regression models were performed. Variables with P value less than .1 in univariate analysis were included in the multivariable analysis. For the main statistical analysis, SPSS version 22 (IBM, New York, NY, USA) or R software version 2.10.1 were used.
RESULTS
A total of 881 recipients were included in the final analysis. Among LT recipients, 77% were men and the mean age was 53 years, ranging from 48 to 57. During a mean follow-up
Table 1. Demographic Characteristics, Laboratory Variables and Outcomes According to Tertile Analysis of Heart Rate Tertile 1 (65 bpm)
Demographics Male sex Age (y) Body mass index (kg/m2) MELD score Diabetes Hypertension b-blocker usage Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Pulse pressure (mm Hg) Laboratory variables Hemoglobin (g/dL) Platelet (103/mL) Total bilirubin (mg/dL) Albumin (g/dL) Creatinine (mg/dL) Troponin I (ng/mL) Sodium (mmol/L) Prothrombin time (INR) BNP (pg/mL) Outcome variables 6-mo mortality 1-y mortality Overall mortality
240 53 25 12 65 46 108 107 67 40 12 60 1.8 3.2 0.80 0.006 139 1.34 43
(84.5) (50e58) (22e27) (9e17) (22.9) (16.2) (38.0) (98e118) (61e75) (34e46) (10e13) (42e87) (1.1e3.5) (2.7e3.7) (0.70e0.90) (0.001e0.006) (136e141) (1.16e1.60) (21e95)
5 (1.8) 11 (3.9) 25 (8.8)
Tertile 2 (66e80 bpm)
248 53 23 13 73 42 48 110 69 40 11 56 2.2 3.2 0.75 0.006 138 1.40 40
(78.5) (48e57) (21e25) (9e19) (23.1) (13.3) (15.2) (99e121) (63e76) (34e48) (9e12) (37e84) (1.2e5.7) (2.8e3.6) (0.60e0.98) (0.002e0.006) (134e140) (1.20e1.76) (17e106)
17 (5.4) 26 (8.2) 40 (12.7)
Tertile 3 (>80 bpm)
193 52 24 18 63 34 18 113 72 41 10 58 4.7 3.0 0.82 0.006 136 1.62 53
P
(68.7) (46e57) (21e26) (12e29) (22.4) (12.1) (6.4) (103e124) (63e80) (35e50)
<.001 .003 <.001 <.001 .980 .347 <.001 <.001 <.001 .161
(9e11) (39e82) (2.0e20.9) (2.7e3.4) (0.60e1.20) (0.004e0.006) (132e139) (1.35e2.25) (21e142)
<.001 .191 <.001 .022 .018 .003 <.001 <.001 .089
27 (9.6) 39 (13.9) 46 (16.4)
Data are presented as number (percentage) or median (25the75th percentiles). Abbreviations: BNP, B-type natriuretic peptide; BPM, beats per minute; INR, international normalized ratio; MELD, Model for End-stage Liver Disease.
<.001 <.001 .025
1094
KWON, JUN, JUNG ET AL
Table 2. Demographic Characteristics and Laboratory Variables in Patients Who Survived and Patients Who Died After Liver Transplantation Survived (n ¼ 770)
Demographics Male sex Age (y) Body mass index (kg/m2) MELD score Diabetes Hypertension b-blocker usage Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Pulse pressure (mm Hg) Laboratory variables Hemoglobin (g/dL) Platelet (103/mL) Total bilirubin (mg/dL) Albumin (g/dL) Creatinine (mg/dL) Troponin I (ng/mL) Sodium (mmol/L) Prothrombin time (INR) BNP (pg/mL)
601 52 24 13.5 176 103 155 110 70 40 11.0 58 2.3 3.1 0.79 0.006 138 1.43 44
(78) (47e57) (22e26) (9.0e21.0) (22.9) (13.4) (20.1) (100e121) (62e77) (34e47) (9.0e12.0) (39e84) (1.3e6.5) (2.7e3.6) (0.60e0.97) (0.002e0.006) (134e140) (1.21e1.82) (19e103)
Died (n ¼ 111)
80 55 23 17.0 25 19 19 112 68 45 10.0 52 3.2 3.2 0.90 0.006 136 1.51 65
Total (n ¼ 881)
(72) (50e62) (21e26) (11.0e28.0) (22.5) (17.1) (17.1) (100e128) (62e74) (35e53)
681 53 23.8 14.0 201 122 174 110 69 40
(8.5e12.0) (38e88) (1.5e18.4) (2.8e3.5) (0.70e1.31) (0.006e0.008) (134e139) (1.27e2.02) (20e168)
11.0 58 2.4 3.1 0.80 0.006 138 1.45 46
P
(77) (48e57) (21.6e26.2) (10.0e21.0) (22.8) (13.8) (19.8) (100e121) (62e77) (34e48)
.199 <.001 .173 .002 >.999 .358 .537 .239 .099 .004
(9.0e12.0) (39e84) (1.3e6.8) (2.7e3.6) (0.60e1.00) (0.002e0.006) (134e140) (1.21e1.85) (19e110)
.002 .416 .088 .465 .001 <.001 .059 .098 .035
Data are presented as counts (percentages) or median (25the75th percentiles). Abbreviations: BNP, B-type natriuretic peptide; INR, international normalized ratio; MELD, Model for End-stage Liver Disease.
time of 56 20 months, 111 patients (12.6%) died, of whom 76 (8.6%) had died by 1 year after LT. Baseline characteristics and laboratory and outcome variables among the tertile groups of HR are summarized in Table 1. Patients with higher resting HR showed increases of Model for End-stage Liver Disease (MELD) score, total bilirubin, prothrombin time (all P < .001). The 1-year mortality rate was 3.9%, 8.2%, and 13.9% (P < .001) and overall mortality rate was 8.8%, 12.7%, and 16.4% (P ¼ .025) from lower to higher HR tertiles, respectively. Table 2 presents baseline characteristics and laboratory variables between the groups of patients who survived and patients who died. Patients who died were older (P < .001) and had higher MELD score (P ¼ .002), higher pulse pressure (P ¼ .004), lower hemoglobin level (P ¼ .031), and higher creatinine level (P ¼ .001). In the univariate Cox regression analysis, tertile 3 group showed a significantly higher overall mortality rate compared with tertile 1 group (hazard ratio 2.01; 95% confidence interval, 1.23e3.27; P ¼ .005) (Table 3). After adjusting for clinically significant variables in univariate analysis such as age, MELD score, hemoglobin, B-type natriuretic peptide, and pulse pressure, the multivariable Cox regression analysis showed that increased resting HR (tertile group 3) was significantly associated with higher risk for all-cause mortality (hazard ratio 1.83; 95% confidence interval, 1.10e3.07; P ¼ .021) compared with low resting HR group (tertile group 1) (Table 3). Kaplan-Meier analysis is shown on Fig 1 and revealed that the all-cause mortality rate was significantly higher for patients with increased HR (P ¼ .016, log-rank test).
DISCUSSION
In this retrospective study, the principle finding of our results is that increased preoperative HR is a prognostic risk factor for overall mortality in end-stage liver disease patients who underwent LT. Higher 1-year mortality rate
Table 3. Univariate and Multivariable Analysis of Risk Factors Associated With All-Cause Mortality After Liver Transplantation Univariate HR (95% CI)
Age Sex Body mass index Hypertension Diabetes b-blocker use BNP (log) Hemoglobin Platelet Creatinine MELD score Heart rate Tertile 1 Tertile 2 Tertile 3 Pulse pressure
1.07 1.35 0.97 1.27 0.98 0.82 1.56 0.87 1.00 1.33 1.04
Multivariable P
HR (95% CI)
P
(1.04e1.10) <.001 1.07 (1.04e1.10) <.001 (0.89e2.04) .159 (0.92e1.03) .328 (0.78e2.08) .344 (0.63e1.52) .923 (0.50e1.35) .441 (1.14e2.15) .006 (0.79e0.95) .002 (0.97e1.00) .921 (1.17e1.51) <.001 (1.02e1.05) <.001 1.03 (1.01e1.05) .006
1.43 (0.86e2.36) .169 1.46 (0.88e2.42) 2.01 (1.23e3.27) .005 1.83 (1.10e3.07) 1.03 (1.01e1.05) <.001 1.02 (1.00e1.03)
.147 .021 .060
Adjusted for age, MELD score, and laboratory variables including log-BNP, hemoglobin, and pulse pressure. Abbreviations: BNP, B-type natriuretic peptide; CI, confidence interval; HR, hazard ratio; MELD, Model for End-stage Liver Disease; tertile 1 group, heart rate 65 beats per minutes (bpm); tertile 2 group, 65 bpm < heart rate 80 bpm; tertile 3 group, heart rate >80 bpm.
PRETRANSPLANT RESTING HEART RATE AND MORTALITY
1095
Fig 1. Kaplan-Meier survival analysis of overall survival rate after liver transplantation in tertile groups of heart rate. Abbreviation: bpm, beats per minute.
(3.9% vs 8.2% vs 13.5%, from low to high HR, P < .001) and increased overall mortality rate during a mean followup time of 56 20 months (8.8% vs 13% vs 15.8%, from low to high HR, P ¼ .025) were observed. The relationship between HR and mortality persisted after adjusting for clinically significant factors such as MELD score, age, and B-type natriuretic peptide. Several hypothesized pathophysiological mechanisms had been proposed for the explanation of the relationship between increased HR and mortality. One of most acknowledged mechanism is myocardial oxygen supplyedemand imbalance. Tachycardia promotes higher oxygen consumption, which exposes the patient to the higher risk of perioperative myocardial injury, which greatly attributes to higher mortality [6,11]. Abbott et al have shown association between preoperative HR and myocardial injury, myocardial infarct, and overall mortality after noncardiac surgery [11]. Another proposed theory is that tachycardia could be a marker of elevated circulating levels of inflammatory markers or sympathetic autonomic dysfunction caused by underlying chronic disease states. Hyperdynamic circulation is a typical presentation in patients with liver cirrhosis, which is characterized by low overall systemic vascular resistance, increased cardiac output, and increased HR. It is caused by increased blood volume (increased preload), the presence of arteriovenous communications, and increased sympathetic nervous activity, which depends on severity of liver cirrhosis [8]. Henriksen et al have shown that increased circulating catecholamines is directly related to the severity of the disease in patients with cirrhosis, and Floras et al have shown a close association between circulating levels of catecholamine and increased burst frequency of skeletal muscle, which leads
to the conclusion that patients with liver cirrhosis exhibit enhanced activity of the sympathetic nervous system [6,8,12,13]. Therefore, a plausible mechanism by which tachycardia may be a marker of severity of a patient’s preoperative sympathetic cardiac neuropathy could be offered, which may affect the mortality as observed in our study. As far as we know, this is the first study to identify HR as a prognostic factor for mortality in end-stage liver disease patients after LT. Average preoperative HR in our study population was 73 bpm, ranging from 43 to 130 bpm. After categorizing the patients into groups of tertile proportionally, the median HR of highest tertile was 88 bpm, ranging from 81 to 130 bpm. Compared with the other study populations regarding the association between mortality and HR, our study populations tended to have higher resting HR, which may be caused by hyperactive sympathetic activation found in cirrhotic patients. This study has several limitations. Influence of other factors such as patient’s medications (eg, b-blockers) or emotional stress could have affected HR, which is well established. The effect of b-blockers on mortality has been actively studied but remains controversial. As for our study, the P value for univariate analysis for b-blocker was insignificant, so the factor was not included in multivariate analysis. Further studies are needed to clarify the effect of b-blockers on mortality in cirrhotic patients. We did not measure other sympathetic variables, such as blood levels of epinephrine and norepinephrine. In conclusion, pretransplant increased resting HR is associated with all-cause mortality in cirrhotic patients. This result allows us to identify end-stage liver disease patients at high risk of mortality following LT. Further research is
1096
needed to clarify the relationship between HR burden and sympathetic cardiac neuropathy in cirrhotic patients.
REFERENCES [1] Jensen MT, Suadicani P, Hein HO, Gyntelberg F. Elevated resting heart rate, physical fitness and all-cause mortality: a 16-year follow-up in the Copenhagen Male Study. Heart 2013;99:882e7. [2] Greenland P, Daviglus ML, Dyer AR, Liu K, Huang CF, Goldberger JJ, et al. Resting heart rate is a risk factor for cardiovascular and noncardiovascular mortality: the Chicago Heart Association Detection Project in Industry. Am J Epidemiol 1999;149:853e62. [3] Fox K, Ford I, Steg PG, Tendera M, Robertson M, Ferrari R, et al. Heart rate as a prognostic risk factor in patients with coronary artery disease and left-ventricular systolic dysfunction (BEAUTIFUL): a subgroup analysis of a randomised controlled trial. Lancet 2008;372:817e21. [4] Jensen MT, Marott JL, Allin KH, Nordestgaard BG, Jensen GB. Resting heart rate is associated with cardiovascular and all-cause mortality after adjusting for inflammatory markers: the Copenhagen City Heart Study. Eur J Prev Cardiol 2012;19:102e8. [5] Dyer AR, Persky V, Stamler J, Paul O, Shekelle RB, Berkson DM, et al. Heart rate as a prognostic factor for coronary heart disease and mortality: findings in three Chicago epidemiologic studies. Am J Epidemiol 1980;112:736e49.
KWON, JUN, JUNG ET AL [6] Landesberg G, Beattie WS, Mosseri M, Jaffe AS, Alpert JS. Perioperative myocardial infarction. Circulation 2009;119:2936e44. [7] Landesberg G, Mosseri M, Zahger D, Wolf Y, Perouansky M, Anner H, et al. Myocardial infarction after vascular surgery: the role of prolonged stress-induced, ST depression-type ischemia. J Am Coll Cardiol 2001;37:1839e45. [8] Trevisani F, Sica G, Mainqua P, Santese G, De Notariis S, Caraceni P, et al. Autonomic dysfunction and hyperdynamic circulation in cirrhosis with ascites. Hepatology 1999;30:1387e92. [9] Groszmann RJ. Hyperdynamic circulation of liver disease 40 years later: pathophysiology and clinical consequences. Hepatology 1994;20:1359e63. [10] Watt KD, Pedersen RA, Kremers WK, Heimbach JK, Charlton MR. Evolution of causes and risk factors for mortality post-liver transplant: results of the NIDDK long-term follow-up study. Am J Transplant 2010;10:1420e7. [11] Abbott TE, Ackland GL, Archbold RA, Wragg A, Kam E, Ahmad T, et al. Preoperative heart rate and myocardial injury after non-cardiac surgery: results of a predefined secondary analysis of the VISION study. Br J Anaesth 2016;117:172e81. [12] Henriksen JH, Ring-Larsen H, Christensen NJ. Sympathetic nervous activity in cirrhosis. A survey of plasma catecholamine studies. J Hepatol 1985;1:55e65. [13] Floras JS, Legault L, Morali GA, Hara K, Blendis LM. Increased sympathetic outflow in cirrhosis and ascites: direct evidence from intraneural recordings. Ann Intern Med 1991;114: 373e80.
ABSTRACT Background. The importance of heart rate (HR) measurement as a prognostic factor has been recognized in many clinical conditions, such as hypertension, coronary artery disease, or heart failure. Patients with liver cirrhosis tend to have increased resting HR as consequence of hyperdynamic circulation. In the current study, we examined whether pretransplant resting increased HR is associated with overall mortality in cirrhotic patients following liver transplantation (LT). Patients and Methods. We retrospectively collected and analyzed the data of 881 liver recipients who underwent LT surgery between October 2009 and September 2012. Patients were categorized into 3 groups by tertile of resting HR as follows: tertile 1 group, HR 65 beats per minute (bpm); tertile 2 group, HR 66 to 80 bpm; and tertile 3 group, HR > 80 bpm. Results. Kaplan-Meier analysis showed that the all-cause mortality rate was significantly different according to tertiles of HR (P ¼ .016, log-rank test). The multivariate Cox regression analysis showed that tertile 3 group was significantly associated with higher risk for all-cause mortality (hazard ratio 1.83, 95% confidence interval, 1.10e3.07; P ¼ .021) compared with tertile 1 group, after adjusting for clinically significant variables in univariate analysis. Conclusions. Our results demonstrate that pretransplant resting tachycardia can identify patients at high risk of death in cirrhotic patients following LT, suggesting that further study will be need to clarify relationship between HR burden and sympathetic cardiac neuropathy.
H
EART rate (HR) is simply and routinely measured in hospitalized patients; it presents abundant information regarding the patient’s cardiovascular status. Recent studies demonstrated that preoperative resting tachycardia is a prognostic risk factor for perioperative myocardial injury, heart failure, and cardiac-related mortality across different patient populations such as patients with hypertension, coronary artery disease, or heart failure [1e4]. A widespread acknowledged explanation is that myocardial oxygen supplyedemand imbalance caused by tachycardia may provoke perioperative myocardial injury, which may contribute to increased morbidity and mortality [5e7]. Patients with advanced liver cirrhosis develop hyperdynamic circulation, which is characterized by increased 0041-1345/17 http://dx.doi.org/10.1016/j.transproceed.2017.03.043
1092
HR, activated sympathetic nervous system, and high cardiac output combined with low vascular resistance [8,9]. In this regard, understanding the basal cardiovascular status of patients is crucial to perform successful liver transplantation (LT). Furthermore, with advances in surgical technique, pretransplant cardiovascular disease is an important leading cause of morbidity and mortality after LT in the current
*Address correspondence to In-Gu Jun, MD, PhD, Department of Anesthesiology and Pain Medicine, Laboratory for Cardiovascular Dynamics, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea. E-mail: [email protected] ª 2017 Elsevier Inc. All rights reserved. 230 Park Avenue, New York, NY 10169
Transplantation Proceedings, 49, 1092e1096 (2017)
PRETRANSPLANT RESTING HEART RATE AND MORTALITY
era [6,10]. However, studies regarding relationship between resting HR and long-term mortality in liver recipients are limited. The aim of this study were to assess the usefulness of preoperative increased HR in predicting all-cause mortality after LT surgery, thereby addressing whether HR measurement provides additional prognostic information to show latent cardiovascular disease status of end-stage liver disease patients. PATIENTS AND METHODS In this retrospective study, we analyzed the database of 1048 consecutive patients who underwent LT surgery at Asan Medical Center from October 2009 to September 2012. All patients included were followed for 56 20 months, from the operational day to June 2016. The exclusion criteria were acute fulminant hepatic failure without chronic liver disease (n ¼ 39) or hepatic failure due to graft rejection following LT (n ¼ 32). We also excluded patients who had more than moderate valvular disease (n ¼ 4), significant obstructive coronary artery disease (n ¼ 6), previous history of cardiac surgery (n ¼ 3), significant arrhythmia such as atrial fibrillation (n ¼ 4), chronic kidney disease (n ¼ 14), and incomplete data (n ¼ 97). After approval of the institutional review board, we collected and analyzed data from computerized collecting registry (ABLE, Asan Biomedical Research Program, South Korea), which was performed in accordance with the ethical standards in the 1964 Declaration of Helsinki and its amendments. Because HR is quite variable, resting HR was defined as averaged value of HR when patients were admitted and the HR measured quietly before induction of
1093 anesthesia. Primary end point was 1-year overall all-cause mortality and mortality observed until June 2016. Patients were categorized into 3 groups by tertile analysis of resting HR as follows: tertile 1, HR 65 beats per minute (bpm) (median: 59 bpm, range 41e65 bpm); tertile 2, HR 66 to 80 bpm (median: 72 bpm, 66e80 bpm); and tertile 3, HR > 80 bpm (median: 88 bpm, 81e130 bpm). Laboratory data were collected the day before surgery. Categorical and continuous variables are shown as number (percentage), mean standard deviation if normally distributed, or median with interquartile range if skewed, appropriately. Differences between patients who survived or died were analyzed with a t test or Mann-Whitney U test and a c2 test or Fisher exact test. Variables according to HR tertiles were analyzed by a linear-bylinear c2 test for trend (for categorical data) and the JonckheereTerpstra test (for ordered continuous data). Survival rates were compared using the Kaplan-Meier survival curve analysis, and differences were compared using the log-rank test. To assess hazard ratios of resting HR on all-cause mortality, univariate and multiple logistic regression analyses using Cox proportional hazards regression models were performed. Variables with P value less than .1 in univariate analysis were included in the multivariable analysis. For the main statistical analysis, SPSS version 22 (IBM, New York, NY, USA) or R software version 2.10.1 were used.
RESULTS
A total of 881 recipients were included in the final analysis. Among LT recipients, 77% were men and the mean age was 53 years, ranging from 48 to 57. During a mean follow-up
Table 1. Demographic Characteristics, Laboratory Variables and Outcomes According to Tertile Analysis of Heart Rate Tertile 1 (65 bpm)
Demographics Male sex Age (y) Body mass index (kg/m2) MELD score Diabetes Hypertension b-blocker usage Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Pulse pressure (mm Hg) Laboratory variables Hemoglobin (g/dL) Platelet (103/mL) Total bilirubin (mg/dL) Albumin (g/dL) Creatinine (mg/dL) Troponin I (ng/mL) Sodium (mmol/L) Prothrombin time (INR) BNP (pg/mL) Outcome variables 6-mo mortality 1-y mortality Overall mortality
240 53 25 12 65 46 108 107 67 40 12 60 1.8 3.2 0.80 0.006 139 1.34 43
(84.5) (50e58) (22e27) (9e17) (22.9) (16.2) (38.0) (98e118) (61e75) (34e46) (10e13) (42e87) (1.1e3.5) (2.7e3.7) (0.70e0.90) (0.001e0.006) (136e141) (1.16e1.60) (21e95)
5 (1.8) 11 (3.9) 25 (8.8)
Tertile 2 (66e80 bpm)
248 53 23 13 73 42 48 110 69 40 11 56 2.2 3.2 0.75 0.006 138 1.40 40
(78.5) (48e57) (21e25) (9e19) (23.1) (13.3) (15.2) (99e121) (63e76) (34e48) (9e12) (37e84) (1.2e5.7) (2.8e3.6) (0.60e0.98) (0.002e0.006) (134e140) (1.20e1.76) (17e106)
17 (5.4) 26 (8.2) 40 (12.7)
Tertile 3 (>80 bpm)
193 52 24 18 63 34 18 113 72 41 10 58 4.7 3.0 0.82 0.006 136 1.62 53
P
(68.7) (46e57) (21e26) (12e29) (22.4) (12.1) (6.4) (103e124) (63e80) (35e50)
<.001 .003 <.001 <.001 .980 .347 <.001 <.001 <.001 .161
(9e11) (39e82) (2.0e20.9) (2.7e3.4) (0.60e1.20) (0.004e0.006) (132e139) (1.35e2.25) (21e142)
<.001 .191 <.001 .022 .018 .003 <.001 <.001 .089
27 (9.6) 39 (13.9) 46 (16.4)
Data are presented as number (percentage) or median (25the75th percentiles). Abbreviations: BNP, B-type natriuretic peptide; BPM, beats per minute; INR, international normalized ratio; MELD, Model for End-stage Liver Disease.
<.001 <.001 .025
1094
KWON, JUN, JUNG ET AL
Table 2. Demographic Characteristics and Laboratory Variables in Patients Who Survived and Patients Who Died After Liver Transplantation Survived (n ¼ 770)
Demographics Male sex Age (y) Body mass index (kg/m2) MELD score Diabetes Hypertension b-blocker usage Systolic pressure (mm Hg) Diastolic pressure (mm Hg) Pulse pressure (mm Hg) Laboratory variables Hemoglobin (g/dL) Platelet (103/mL) Total bilirubin (mg/dL) Albumin (g/dL) Creatinine (mg/dL) Troponin I (ng/mL) Sodium (mmol/L) Prothrombin time (INR) BNP (pg/mL)
601 52 24 13.5 176 103 155 110 70 40 11.0 58 2.3 3.1 0.79 0.006 138 1.43 44
(78) (47e57) (22e26) (9.0e21.0) (22.9) (13.4) (20.1) (100e121) (62e77) (34e47) (9.0e12.0) (39e84) (1.3e6.5) (2.7e3.6) (0.60e0.97) (0.002e0.006) (134e140) (1.21e1.82) (19e103)
Died (n ¼ 111)
80 55 23 17.0 25 19 19 112 68 45 10.0 52 3.2 3.2 0.90 0.006 136 1.51 65
Total (n ¼ 881)
(72) (50e62) (21e26) (11.0e28.0) (22.5) (17.1) (17.1) (100e128) (62e74) (35e53)
681 53 23.8 14.0 201 122 174 110 69 40
(8.5e12.0) (38e88) (1.5e18.4) (2.8e3.5) (0.70e1.31) (0.006e0.008) (134e139) (1.27e2.02) (20e168)
11.0 58 2.4 3.1 0.80 0.006 138 1.45 46
P
(77) (48e57) (21.6e26.2) (10.0e21.0) (22.8) (13.8) (19.8) (100e121) (62e77) (34e48)
.199 <.001 .173 .002 >.999 .358 .537 .239 .099 .004
(9.0e12.0) (39e84) (1.3e6.8) (2.7e3.6) (0.60e1.00) (0.002e0.006) (134e140) (1.21e1.85) (19e110)
.002 .416 .088 .465 .001 <.001 .059 .098 .035
Data are presented as counts (percentages) or median (25the75th percentiles). Abbreviations: BNP, B-type natriuretic peptide; INR, international normalized ratio; MELD, Model for End-stage Liver Disease.
time of 56 20 months, 111 patients (12.6%) died, of whom 76 (8.6%) had died by 1 year after LT. Baseline characteristics and laboratory and outcome variables among the tertile groups of HR are summarized in Table 1. Patients with higher resting HR showed increases of Model for End-stage Liver Disease (MELD) score, total bilirubin, prothrombin time (all P < .001). The 1-year mortality rate was 3.9%, 8.2%, and 13.9% (P < .001) and overall mortality rate was 8.8%, 12.7%, and 16.4% (P ¼ .025) from lower to higher HR tertiles, respectively. Table 2 presents baseline characteristics and laboratory variables between the groups of patients who survived and patients who died. Patients who died were older (P < .001) and had higher MELD score (P ¼ .002), higher pulse pressure (P ¼ .004), lower hemoglobin level (P ¼ .031), and higher creatinine level (P ¼ .001). In the univariate Cox regression analysis, tertile 3 group showed a significantly higher overall mortality rate compared with tertile 1 group (hazard ratio 2.01; 95% confidence interval, 1.23e3.27; P ¼ .005) (Table 3). After adjusting for clinically significant variables in univariate analysis such as age, MELD score, hemoglobin, B-type natriuretic peptide, and pulse pressure, the multivariable Cox regression analysis showed that increased resting HR (tertile group 3) was significantly associated with higher risk for all-cause mortality (hazard ratio 1.83; 95% confidence interval, 1.10e3.07; P ¼ .021) compared with low resting HR group (tertile group 1) (Table 3). Kaplan-Meier analysis is shown on Fig 1 and revealed that the all-cause mortality rate was significantly higher for patients with increased HR (P ¼ .016, log-rank test).
DISCUSSION
In this retrospective study, the principle finding of our results is that increased preoperative HR is a prognostic risk factor for overall mortality in end-stage liver disease patients who underwent LT. Higher 1-year mortality rate
Table 3. Univariate and Multivariable Analysis of Risk Factors Associated With All-Cause Mortality After Liver Transplantation Univariate HR (95% CI)
Age Sex Body mass index Hypertension Diabetes b-blocker use BNP (log) Hemoglobin Platelet Creatinine MELD score Heart rate Tertile 1 Tertile 2 Tertile 3 Pulse pressure
1.07 1.35 0.97 1.27 0.98 0.82 1.56 0.87 1.00 1.33 1.04
Multivariable P
HR (95% CI)
P
(1.04e1.10) <.001 1.07 (1.04e1.10) <.001 (0.89e2.04) .159 (0.92e1.03) .328 (0.78e2.08) .344 (0.63e1.52) .923 (0.50e1.35) .441 (1.14e2.15) .006 (0.79e0.95) .002 (0.97e1.00) .921 (1.17e1.51) <.001 (1.02e1.05) <.001 1.03 (1.01e1.05) .006
1.43 (0.86e2.36) .169 1.46 (0.88e2.42) 2.01 (1.23e3.27) .005 1.83 (1.10e3.07) 1.03 (1.01e1.05) <.001 1.02 (1.00e1.03)
.147 .021 .060
Adjusted for age, MELD score, and laboratory variables including log-BNP, hemoglobin, and pulse pressure. Abbreviations: BNP, B-type natriuretic peptide; CI, confidence interval; HR, hazard ratio; MELD, Model for End-stage Liver Disease; tertile 1 group, heart rate 65 beats per minutes (bpm); tertile 2 group, 65 bpm < heart rate 80 bpm; tertile 3 group, heart rate >80 bpm.
PRETRANSPLANT RESTING HEART RATE AND MORTALITY
1095
Fig 1. Kaplan-Meier survival analysis of overall survival rate after liver transplantation in tertile groups of heart rate. Abbreviation: bpm, beats per minute.
(3.9% vs 8.2% vs 13.5%, from low to high HR, P < .001) and increased overall mortality rate during a mean followup time of 56 20 months (8.8% vs 13% vs 15.8%, from low to high HR, P ¼ .025) were observed. The relationship between HR and mortality persisted after adjusting for clinically significant factors such as MELD score, age, and B-type natriuretic peptide. Several hypothesized pathophysiological mechanisms had been proposed for the explanation of the relationship between increased HR and mortality. One of most acknowledged mechanism is myocardial oxygen supplyedemand imbalance. Tachycardia promotes higher oxygen consumption, which exposes the patient to the higher risk of perioperative myocardial injury, which greatly attributes to higher mortality [6,11]. Abbott et al have shown association between preoperative HR and myocardial injury, myocardial infarct, and overall mortality after noncardiac surgery [11]. Another proposed theory is that tachycardia could be a marker of elevated circulating levels of inflammatory markers or sympathetic autonomic dysfunction caused by underlying chronic disease states. Hyperdynamic circulation is a typical presentation in patients with liver cirrhosis, which is characterized by low overall systemic vascular resistance, increased cardiac output, and increased HR. It is caused by increased blood volume (increased preload), the presence of arteriovenous communications, and increased sympathetic nervous activity, which depends on severity of liver cirrhosis [8]. Henriksen et al have shown that increased circulating catecholamines is directly related to the severity of the disease in patients with cirrhosis, and Floras et al have shown a close association between circulating levels of catecholamine and increased burst frequency of skeletal muscle, which leads
to the conclusion that patients with liver cirrhosis exhibit enhanced activity of the sympathetic nervous system [6,8,12,13]. Therefore, a plausible mechanism by which tachycardia may be a marker of severity of a patient’s preoperative sympathetic cardiac neuropathy could be offered, which may affect the mortality as observed in our study. As far as we know, this is the first study to identify HR as a prognostic factor for mortality in end-stage liver disease patients after LT. Average preoperative HR in our study population was 73 bpm, ranging from 43 to 130 bpm. After categorizing the patients into groups of tertile proportionally, the median HR of highest tertile was 88 bpm, ranging from 81 to 130 bpm. Compared with the other study populations regarding the association between mortality and HR, our study populations tended to have higher resting HR, which may be caused by hyperactive sympathetic activation found in cirrhotic patients. This study has several limitations. Influence of other factors such as patient’s medications (eg, b-blockers) or emotional stress could have affected HR, which is well established. The effect of b-blockers on mortality has been actively studied but remains controversial. As for our study, the P value for univariate analysis for b-blocker was insignificant, so the factor was not included in multivariate analysis. Further studies are needed to clarify the effect of b-blockers on mortality in cirrhotic patients. We did not measure other sympathetic variables, such as blood levels of epinephrine and norepinephrine. In conclusion, pretransplant increased resting HR is associated with all-cause mortality in cirrhotic patients. This result allows us to identify end-stage liver disease patients at high risk of mortality following LT. Further research is
1096
needed to clarify the relationship between HR burden and sympathetic cardiac neuropathy in cirrhotic patients.
REFERENCES [1] Jensen MT, Suadicani P, Hein HO, Gyntelberg F. Elevated resting heart rate, physical fitness and all-cause mortality: a 16-year follow-up in the Copenhagen Male Study. Heart 2013;99:882e7. [2] Greenland P, Daviglus ML, Dyer AR, Liu K, Huang CF, Goldberger JJ, et al. Resting heart rate is a risk factor for cardiovascular and noncardiovascular mortality: the Chicago Heart Association Detection Project in Industry. Am J Epidemiol 1999;149:853e62. [3] Fox K, Ford I, Steg PG, Tendera M, Robertson M, Ferrari R, et al. Heart rate as a prognostic risk factor in patients with coronary artery disease and left-ventricular systolic dysfunction (BEAUTIFUL): a subgroup analysis of a randomised controlled trial. Lancet 2008;372:817e21. [4] Jensen MT, Marott JL, Allin KH, Nordestgaard BG, Jensen GB. Resting heart rate is associated with cardiovascular and all-cause mortality after adjusting for inflammatory markers: the Copenhagen City Heart Study. Eur J Prev Cardiol 2012;19:102e8. [5] Dyer AR, Persky V, Stamler J, Paul O, Shekelle RB, Berkson DM, et al. Heart rate as a prognostic factor for coronary heart disease and mortality: findings in three Chicago epidemiologic studies. Am J Epidemiol 1980;112:736e49.
KWON, JUN, JUNG ET AL [6] Landesberg G, Beattie WS, Mosseri M, Jaffe AS, Alpert JS. Perioperative myocardial infarction. Circulation 2009;119:2936e44. [7] Landesberg G, Mosseri M, Zahger D, Wolf Y, Perouansky M, Anner H, et al. Myocardial infarction after vascular surgery: the role of prolonged stress-induced, ST depression-type ischemia. J Am Coll Cardiol 2001;37:1839e45. [8] Trevisani F, Sica G, Mainqua P, Santese G, De Notariis S, Caraceni P, et al. Autonomic dysfunction and hyperdynamic circulation in cirrhosis with ascites. Hepatology 1999;30:1387e92. [9] Groszmann RJ. Hyperdynamic circulation of liver disease 40 years later: pathophysiology and clinical consequences. Hepatology 1994;20:1359e63. [10] Watt KD, Pedersen RA, Kremers WK, Heimbach JK, Charlton MR. Evolution of causes and risk factors for mortality post-liver transplant: results of the NIDDK long-term follow-up study. Am J Transplant 2010;10:1420e7. [11] Abbott TE, Ackland GL, Archbold RA, Wragg A, Kam E, Ahmad T, et al. Preoperative heart rate and myocardial injury after non-cardiac surgery: results of a predefined secondary analysis of the VISION study. Br J Anaesth 2016;117:172e81. [12] Henriksen JH, Ring-Larsen H, Christensen NJ. Sympathetic nervous activity in cirrhosis. A survey of plasma catecholamine studies. J Hepatol 1985;1:55e65. [13] Floras JS, Legault L, Morali GA, Hara K, Blendis LM. Increased sympathetic outflow in cirrhosis and ascites: direct evidence from intraneural recordings. Ann Intern Med 1991;114: 373e80.