Does Preoperative Beta-blocker Use Influence Intraoperative Hemodynamic Profile and Post-operative Course of Liver Transplantation? Z. Milana,*, C. Taylorb, D. Armstrongc, P. Daviesd, S. Robertse, B. Rupnikf, and A. Suddleg a Leeds Teaching Hospital, Leeds, United Kingdom; now at King’s College Hospital, London, United Kingdom; bSchool of Mathematics, University of Leeds, Leeds, United Kingdom; cLeeds Teaching Hospital, Leeds, United Kingdom; now at: Scarborough General Hospital, Scarborough, United Kingdom; dLeeds Teaching Hospital, Leeds, United Kingdom; now at: Gartnaval General Hospital, Glasgow, Scotland, United Kingdom; eLeeds Teaching Hospital, Leeds, United Kingdom; now at: University Hospital of South Manchester, Wythenshawe, United Kingdom; fMaribor University Hospital, Maribor, Slovenia; and gKing’s College Hospital, Denmark Hill, London, United Kingdom
ABSTRACT Objective. We compared the intraoperative hemodynamic profiles and outcomes of liver transplant (LT) patients receiving preoperative beta-blockers for portal hypertension with those of patients not receiving beta-blockers. Methods. Beat-to-beat hemodynamic data were recorded with the use of a lithium dilution cardiac output monitor, including heart rate (HR), cardiac output (CO), cardiac index (CI), systemic vascular resistance (SVR), SVR index, mean arterial pressure, stroke volume (SV), and SV index. Perioperative data included age, sex, etiology of cirrhosis, Model of End-Stage Liver Disease score, warm and cold ischemia times, and intraoperative blood and blood products replacement. Postoperative data included Acute Physiology and Chronic Health Evaluation II score, Intensive Care National Audit and Research Centre (ICNARC) score, length of intensive care unit stay, renal replacement treatment, and mortality. Results. After exclusion, 215 patients not receiving and 93 patients receiving betablockers for portal hypertension were included. There was no significant difference between groups in demographic data. HR, CO, and CI were significantly lower in the beta-blocker group (P ¼ .0001). Regarding outcomes, the ICNARC score was significantly higher in the beta-blocker group (P < .05). Conclusions. Preoperative administration of beta-blockers attenuates the hyperdynamic circulation in LT patients; this is clearly seen through a reduction in intraoperative CO, CI, and HR. This decreased hyperdynamic state was not shown to relate to outcomes in any of the variables except for a higher ICNARC score reported in the group treated with beta-blockers.
L
IVER TRANSPLANTATION (LT) is the treatment of choice for many patients with end-stage liver disease (ESLD). Despite developments in pre-, intra- and postoperative care, LT remains a high-risk procedure, owing to rapid fluid shift, intraoperative bleeding, reperfusion syndrome, rapid hemodynamic changes, and a hyperdynamic, hyporeactive circulation characteristic of ESLD [1]. Massive intraoperative bleeding and fluid replacement, combined with the use of inotropes, can contribute to negative outcomes of LT [2].
ª 2016 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
Transplantation Proceedings, 48, 111e115 (2016)
Nonselective beta-adrenergic blockers have been widely used in the management of portal hypertension caused by cirrhosis. Beta-blockers lower the portal pressure by reducing portal blood flow as a consequence of a decreased
*Address correspondence to Dr Zoka Milan, King’s College Hospital, Denmark Hill, London SE5 9RS. E-mail: zoka.milan@ kcl.ac.uk 0041-1345/16 http://dx.doi.org/10.1016/j.transproceed.2015.12.027
111
112
cardiac output (b1-receptor blockade) and arteriolar splanchnic vasoconstriction (b2-receptor blockade) [3]. The role of beta-blockers in ESLD in the pre-transplant period is therefore accepted [4,5]. However, the effect of this prophylactic measure on intraoperative hemodynamics is poorly understood. In the present study, we compared the intraoperative hemodynamic profiles and outcomes of liver transplant patients receiving preoperative beta-blockers for portal hypertension with those not receiving beta-blockers. PATIENTS AND METHODS Exclusion criteria for the study were: use of beta-blockers for treatment of hypertension or cardiac failure, patients who were .05 >.05 >.05 >.05 .33 .91 .92 .17 .32 .55 .09
Abbreviations: MELD, Model of End-stage Liver Disease; CIT, cold ischemia time; WIT, warm ischemia time; RBC; red blood cells.
PREOPERATIVE BETA-BLOCKER USE
113
Table 2. Intraoperative Data Derived With the Use of LIDCO Monitor b-Blocker (n ¼ 93)
Characteristic
HR (beats/min) CO (L/min) CI (L/min/m2) MAP (mm Hg) SV (mL/beat) SVI (mL/m2/beat) SVR (dyne-s/cm5) SVRI (dyne-s/cm5/m2)
83.1 9.5 5 68.5 117.1 61.5 608.7 1,093
15.9 3.0 1.3 8.9 37.3 16.2 296 308
No b-blocker (n ¼ 215)
P Value
.05 >.05 >.05
90 10.9 5.9 67.9 121.3 65.1 534.6 972
12.7 4.2 2.4 8.2 45.1 23.6 242.8 387
Abbreviations: LIDCO, lithium dilution cardiac output; HR, heart rate; CO, cardiac output; CI, cardiac index; MAP, mean arterial pressure; SV, stroke volume; SVI, stroke volume index; SVR, systemic vascular resistance; SVRI, systemic vascular resistance index.
significant difference (with the use of a 2-sample student t test) occurred for HR, CO, and CI. None of the other LIDCO parameters (SV, SVI, MAP, SVR, and SVRI) were found to be significantly different between 2 groups. Regarding the outcomes of LT (Table 3), the ICNARC score was significantly higher in the beta-blocker patient group. The differences in number of patients who required RRT and in mortality were not statistically significant. DISCUSSION
Intraoperative hemodynamic variables measured with the use of a calibrated LIDCO monitor demonstrated significant reductions in intraoperative HR (15.4%), CO (10%), and CI (10%) in patients treated preoperatively with betablockers. Perioperative hemodynamic data were recorded on a beat-to-beat basis rather than at certain time intervals during the LT (ie, dissection, reperfusion, and after reperfusion). Continuous monitoring was advantageous, in that all hemodynamic events were captured; however, important hemodynamic events that are short in duration, such as reperfusion, could have been lost as signals during long dissection or during the post-reperfusion phase. Siniscalchi et al’s study indicated that usage of betablockers resulted in a lower incidence of hyperdynamic circulation characteristic of ESLD patients during LT. They used a Swan-Ganz catheter for the intermittent measurement of hemodynamic parameters and used a SVR of .05 .044* >.05 >.05 >.05
Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; ICNARC, Intensive Care National Audit and Research Centre; ICU, intensive care unit; RRT, renal replacement treatment.
In our study, blood and blood product requirements were analyzed to determine whether the observed change in hemodynamic status had an effect on these parameters. With the decrease in HR and CO in the beta-blocker group, it was hypothesized that there would be lower fluid requirements within the beta-blocker group. However, requirements for blood, platelets, plasma, and cryoprecipitate were not lower in the group treated with beta-blockers. Our finding is different from the Sinicalchi et al study, which found that intraoperative hyperdynamic circulation correlates with hemodynamic alterations, requiring more blood products and vasopressor use [2]. The same study demonstrated worse outcome in patients with HC [2], which was not confirmed in our study. Although beta-blockers were discontinued briefly after LT, we compared the postoperative data (APACHE II and ICNARC scores, duration of ICU stay, RRT, and mortality) between the 2 groups of patients. Only the ICNARC score was significantly higher (P ¼ .044) in the beta-blocker group, whereas APACHE II score, RRT, and mortality were not different between the 2 groups. Scores for both ICNARC and APACHE II are based largely on the same sets of standard observations and blood results. Ten postoperative parameters were identical in both of these scoring systems, and the remaining 2 (serum urea and urine output in ICNARC and potassium and hematocrit in APACHE II) were not significantly different. We assessed APACHE II scores taken on admission into ICU and immediately after LT, and potassium and hematocrit levels are usually corrected intraoperatively, and serum urea and urine output can not be easily controlled. Differences between these scores are also due to different weights assigned by the scoring systems, because the maximal scores are 73 for APACHE II and 100 for ICNARC [6]. For our data, patients from the beta-blocker group had greater disease severity at the end of the procedure. This is in agreement with a recent meta-analysis on perioperative usage of betablockers in noncardiac surgery, which found an increase in all-cause mortality and stroke with the use of beta-blockers, although patients included in the meta-analysis were on selective beta-blockers for cardiovascular problems and underwent different surgical procedures [7]. In contrast, perioperative beta-blocker was significantly protective for postoperative cardiac events and increased the likelihood of event-free survival in LT patients [8]. However, Safadi et al examined patients that underwent LT from 2001 to 2005, considerably earlier than our study. We speculated that the incidence of cardiovascular events had decreased by the time we conducted our study, because preassessment and preoperative cardiovascular optimization has improved [9]. The MELD scoring system determines the severity of liver disease and is the basis for organ allocation in LT [10]. It is calculated on the basis of serum bilirubin, creatinine, and international normalized ratio. In the present study, the comparison between the MELD scores of the 2 groups was vital to indicate a similar extent of disease between groups,
114
thereby eliciting the true effect of interventions on outcomes. There was no statistically significant difference in ICU mortality between the 2 patient groups. The numbers of postoperative ICU deaths in each group were very low (3.2%) for this type of surgery. STUDY LIMITATIONS
Despite the many factors that can affect intraoperative hemodynamics in LT patients [1,2], our goal was to compare patients who were on beta-blockers with those who were not. We were aware that division of the groups in this manner may add inherent bias. Regarding the study size, this was a relatively small cohort of 308 patients. Although we used LIDCO monitoring for all LT patients, only 36.4% of the LT patients had retrievable LIDCO data. The LIDCO monitor automatically saves the data and keeps them for a couple of months before deletion. We lost data in several ways: 1) Some saved data were unusable; 2) some data were deleted when the monitor was updated (about which the research team was not informed); and 3) some data were deleted before they were able to be saved. Therefore, there were no biases in data selection. Because it was a retrospective study, we did not calculate the power of the study, and there is likely to be heterogeneity and bias in selecting patients without and with betablocker treatment for portal hypertension. Of the patients treated with beta-blockers, 31% had no documented esophageal varices, and 25% of patients not treated with beta-blockers had documented esophageal varices. Our clinicians’ decision to use nonselective betablockers was based on the belief that the risk of variceal bleeding correlates with several factors, not only the portal pressure, but also infections and the severity of liver dysfunction, suggesting the absence of a pure mechanical model [11]. Beta-blockers were used in decompensated patients with medium to large varices but not in patients with early or end-stage cirrhosis with refractory ascites. This explains why there were no differences in severity of liver disease between groups. In addition, LT programs have different intra- and postoperative care; therefore, it may not be possible to extrapolate the findings in our specific population to a mixed general population of patients scheduled to undergo LT. Patients received beta-blockers for prophylaxis of variceal bleeding without standardized intervention of drug type or dose. Although propranolol, a noncardioselective drug, would have been the primary intervention, all previous investigations used cardioselective agents [7]. Further trials should explore this more potent and evidence-based subclass of drugs in this unique surgical group, such as nadolol and carvedilol, which are gradually replacing propranolol for portal hypertension management in patients with ESLD [12]. In addition, the fact that this study was retrospective means that it is impossible to comment on preoperative beta-blocker compliance. However, the highly significant
MILAN, TAYLOR, ARMSTRONG ET AL
differences in physiology between the groups would have been unlikely unless there had been a satisfactory degree of compliance. It was also impossible to standardize group characteristics during allocation. We have already shown that there were no significant differences in the general make-up of the groups. However, we do not have data on preoperative cardiovascular diseases and cardiac outcomes in these patients to explore the incidence of cardiovascular events. Indeed, the reduction in the incidence of postoperative myocardial infarction (fatal/nonfatal) was a key finding in the Safadi et al study [8]. This represents a key outcome that we were unable to explore owing to lack of relevant information in the database. Most data were limited to the immediate ICU stay and did not encompass a full 30-day perioperative period Although we analyzed immediate perioperative parameters, a longer follow-up with a rigorous analysis of full clinical sequelae would have elicited more valuable information, for example, there is some evidence of limited persistent hyperdynamic circulation after transplantation compared with normal control subjects. However, return of physiologic parameters toward normalization is generally seen; for example, one cohort demonstrated a net decrease of CI by 35% [2]. It is clearly shown that the aforementioned confounding variables in this study would have been largely avoided with the implementation of a prospective cohort study. Indeed, a large randomized control trial (RCT) would be the method of choice owing to its superiority in the hierarchy of evidence. A large prospective RCT that uses a standard-dose cardioselective beta-blocker with long-term follow-up would be the optimal method of investigation. CONCLUSION
This study has demonstrated that the preoperative administration of beta-blockers to some degree attenuates the hyperdynamic circulation associated with ESLD, as clearly shown by a reduction in intraoperative CO, CI, and HR during LT. These hemodynamic changes were not shown to relate to outcome in any of the investigated variables, with no significant difference in any of the outcome parameters (ICU stay, RRT, postoperative mortality, and physiologic scoring methods) except for a higher ICNARC score in the group treated with beta-blockers. There is evidence to suggest that a cardioprotective effect may present postoperatively in the medium to long term. However, this study can not advise on treatment recommendations for cardiovascular prophylaxis, and further investigation via a prospective RCT with the use of a cardioselective beta-blocker class is required to further this work. ACKNOWLEDGMENTS The authors thank Ms Helen Aldersley, Leeds LT coordinator, for helping with data collection and Dr David Green from King’s College Hospital for useful suggestions before final submission of the manuscript.
PREOPERATIVE BETA-BLOCKER USE
REFERENCES [1] Milan Z, Taylor C, Duncan B, Kedilaya H, Sylvester D. Statistical modeling of hemodynamic changes during orthotopic liver transplantation: predictive value for outcome and effect of marginal donors. Transplant Proc 2011;43(5):1711e5. [2] Siniscalchi A, Aurini L, Spedicato S, Bernardi E, Zanoni A, Dante A, et al. Hyperdynamic circulation in cirrhosis: predictive factors and outcome following liver transplantation. Minerva Anestesiol 2013;79:15e23. [3] Vilas-Boas WW, Ribeiro-Oliveira Jr A, Ribeiro Rda C, Vieira RL, Almeida J, Nadu AP, et al. Effect of propranolol on the splanchnic and peripheral renin angiotensin system in cirrhotic patients. World J Gastroenterol 2008;14:6824e30. [4] Leithead JA, Rajoriya N, Tehami N, Hodson J, Gunson BK, Tripathi D, Ferguson JW. Nonselective b-blockers are associated with improved survival in patients with ascites listed for liver transplantation. Gut 2015;64:1111e9. [5] Mandorfer M, Bota S, Schwabl P, Bucsics T, Pfisterer N, Kruzik M, et al. Nonselective b-blockers are associated with improved survival in patients with ascites listed for liver transplantation. Gastroenterology 2014;146:1680e90. [6] Harrison DA, Parry GJ, Carpenter JR, Short A, Rowan K. A new risk prediction model for critical care: the Intensive Care
115 National Audit & Research Centre (ICNARC) model. Crit Care Med 2007;35:1091e8. [7] Blessberger H, Kammler J, Domanovits H, Schlager O, Wildner B, Azar D, et al. Perioperative beta-blockers for preventing surgery-related mortality and morbidity. Cochrane Database Syst Rev 2014;(9):CD004476. [8] Safadi A, Homsi M, Maskoun W, Lane KA, Singh I, Sawada SG, Mahenthiran J. Perioperative risk predictors of cardiac outcomes in patients undergoing liver transplantation surgery. Circulation 2009;120:1189e94. [9] Kong YG, Kang JW, Kim YK, Seo H, Lim TH, Hwang S, et al. Preoperative coronary calcium score is predictive of early post-operative cardiovascular complications in liver transplant recipients. Br J Anaesth 2015;114:437e43. [10] Klein KB, Stafinski TD, Menon D. Predicting survival after liver transplantation based on pre-transplant MELD score: a systematic review of the literature. PLoS One 2013;8:e80661. [11] Ferrarese A, Tsochtzis E, Burroughs AK, Senzolo M. Betablockers in cirrhosis: therapeutic window or an aspirin for all? J Hepatol 2014;61:449e50. [12] Lo GH, Chen WC, Wang HM, Yu HC. Randomized, controlled trial of carvedilol versus nadolol plus isosorbide mononitrate for the prevention of variceal rebleeding. J Gastroenterol Hepatol 2012;27:1681e7.
ABSTRACT Objective. We compared the intraoperative hemodynamic profiles and outcomes of liver transplant (LT) patients receiving preoperative beta-blockers for portal hypertension with those of patients not receiving beta-blockers. Methods. Beat-to-beat hemodynamic data were recorded with the use of a lithium dilution cardiac output monitor, including heart rate (HR), cardiac output (CO), cardiac index (CI), systemic vascular resistance (SVR), SVR index, mean arterial pressure, stroke volume (SV), and SV index. Perioperative data included age, sex, etiology of cirrhosis, Model of End-Stage Liver Disease score, warm and cold ischemia times, and intraoperative blood and blood products replacement. Postoperative data included Acute Physiology and Chronic Health Evaluation II score, Intensive Care National Audit and Research Centre (ICNARC) score, length of intensive care unit stay, renal replacement treatment, and mortality. Results. After exclusion, 215 patients not receiving and 93 patients receiving betablockers for portal hypertension were included. There was no significant difference between groups in demographic data. HR, CO, and CI were significantly lower in the beta-blocker group (P ¼ .0001). Regarding outcomes, the ICNARC score was significantly higher in the beta-blocker group (P < .05). Conclusions. Preoperative administration of beta-blockers attenuates the hyperdynamic circulation in LT patients; this is clearly seen through a reduction in intraoperative CO, CI, and HR. This decreased hyperdynamic state was not shown to relate to outcomes in any of the variables except for a higher ICNARC score reported in the group treated with beta-blockers.
L
IVER TRANSPLANTATION (LT) is the treatment of choice for many patients with end-stage liver disease (ESLD). Despite developments in pre-, intra- and postoperative care, LT remains a high-risk procedure, owing to rapid fluid shift, intraoperative bleeding, reperfusion syndrome, rapid hemodynamic changes, and a hyperdynamic, hyporeactive circulation characteristic of ESLD [1]. Massive intraoperative bleeding and fluid replacement, combined with the use of inotropes, can contribute to negative outcomes of LT [2].
ª 2016 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
Transplantation Proceedings, 48, 111e115 (2016)
Nonselective beta-adrenergic blockers have been widely used in the management of portal hypertension caused by cirrhosis. Beta-blockers lower the portal pressure by reducing portal blood flow as a consequence of a decreased
*Address correspondence to Dr Zoka Milan, King’s College Hospital, Denmark Hill, London SE5 9RS. E-mail: zoka.milan@ kcl.ac.uk 0041-1345/16 http://dx.doi.org/10.1016/j.transproceed.2015.12.027
111
112
cardiac output (b1-receptor blockade) and arteriolar splanchnic vasoconstriction (b2-receptor blockade) [3]. The role of beta-blockers in ESLD in the pre-transplant period is therefore accepted [4,5]. However, the effect of this prophylactic measure on intraoperative hemodynamics is poorly understood. In the present study, we compared the intraoperative hemodynamic profiles and outcomes of liver transplant patients receiving preoperative beta-blockers for portal hypertension with those not receiving beta-blockers. PATIENTS AND METHODS Exclusion criteria for the study were: use of beta-blockers for treatment of hypertension or cardiac failure, patients who were .05 >.05 >.05 >.05 .33 .91 .92 .17 .32 .55 .09
Abbreviations: MELD, Model of End-stage Liver Disease; CIT, cold ischemia time; WIT, warm ischemia time; RBC; red blood cells.
PREOPERATIVE BETA-BLOCKER USE
113
Table 2. Intraoperative Data Derived With the Use of LIDCO Monitor b-Blocker (n ¼ 93)
Characteristic
HR (beats/min) CO (L/min) CI (L/min/m2) MAP (mm Hg) SV (mL/beat) SVI (mL/m2/beat) SVR (dyne-s/cm5) SVRI (dyne-s/cm5/m2)
83.1 9.5 5 68.5 117.1 61.5 608.7 1,093
15.9 3.0 1.3 8.9 37.3 16.2 296 308
No b-blocker (n ¼ 215)
P Value
.05 >.05 >.05
90 10.9 5.9 67.9 121.3 65.1 534.6 972
12.7 4.2 2.4 8.2 45.1 23.6 242.8 387
Abbreviations: LIDCO, lithium dilution cardiac output; HR, heart rate; CO, cardiac output; CI, cardiac index; MAP, mean arterial pressure; SV, stroke volume; SVI, stroke volume index; SVR, systemic vascular resistance; SVRI, systemic vascular resistance index.
significant difference (with the use of a 2-sample student t test) occurred for HR, CO, and CI. None of the other LIDCO parameters (SV, SVI, MAP, SVR, and SVRI) were found to be significantly different between 2 groups. Regarding the outcomes of LT (Table 3), the ICNARC score was significantly higher in the beta-blocker patient group. The differences in number of patients who required RRT and in mortality were not statistically significant. DISCUSSION
Intraoperative hemodynamic variables measured with the use of a calibrated LIDCO monitor demonstrated significant reductions in intraoperative HR (15.4%), CO (10%), and CI (10%) in patients treated preoperatively with betablockers. Perioperative hemodynamic data were recorded on a beat-to-beat basis rather than at certain time intervals during the LT (ie, dissection, reperfusion, and after reperfusion). Continuous monitoring was advantageous, in that all hemodynamic events were captured; however, important hemodynamic events that are short in duration, such as reperfusion, could have been lost as signals during long dissection or during the post-reperfusion phase. Siniscalchi et al’s study indicated that usage of betablockers resulted in a lower incidence of hyperdynamic circulation characteristic of ESLD patients during LT. They used a Swan-Ganz catheter for the intermittent measurement of hemodynamic parameters and used a SVR of .05 .044* >.05 >.05 >.05
Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; ICNARC, Intensive Care National Audit and Research Centre; ICU, intensive care unit; RRT, renal replacement treatment.
In our study, blood and blood product requirements were analyzed to determine whether the observed change in hemodynamic status had an effect on these parameters. With the decrease in HR and CO in the beta-blocker group, it was hypothesized that there would be lower fluid requirements within the beta-blocker group. However, requirements for blood, platelets, plasma, and cryoprecipitate were not lower in the group treated with beta-blockers. Our finding is different from the Sinicalchi et al study, which found that intraoperative hyperdynamic circulation correlates with hemodynamic alterations, requiring more blood products and vasopressor use [2]. The same study demonstrated worse outcome in patients with HC [2], which was not confirmed in our study. Although beta-blockers were discontinued briefly after LT, we compared the postoperative data (APACHE II and ICNARC scores, duration of ICU stay, RRT, and mortality) between the 2 groups of patients. Only the ICNARC score was significantly higher (P ¼ .044) in the beta-blocker group, whereas APACHE II score, RRT, and mortality were not different between the 2 groups. Scores for both ICNARC and APACHE II are based largely on the same sets of standard observations and blood results. Ten postoperative parameters were identical in both of these scoring systems, and the remaining 2 (serum urea and urine output in ICNARC and potassium and hematocrit in APACHE II) were not significantly different. We assessed APACHE II scores taken on admission into ICU and immediately after LT, and potassium and hematocrit levels are usually corrected intraoperatively, and serum urea and urine output can not be easily controlled. Differences between these scores are also due to different weights assigned by the scoring systems, because the maximal scores are 73 for APACHE II and 100 for ICNARC [6]. For our data, patients from the beta-blocker group had greater disease severity at the end of the procedure. This is in agreement with a recent meta-analysis on perioperative usage of betablockers in noncardiac surgery, which found an increase in all-cause mortality and stroke with the use of beta-blockers, although patients included in the meta-analysis were on selective beta-blockers for cardiovascular problems and underwent different surgical procedures [7]. In contrast, perioperative beta-blocker was significantly protective for postoperative cardiac events and increased the likelihood of event-free survival in LT patients [8]. However, Safadi et al examined patients that underwent LT from 2001 to 2005, considerably earlier than our study. We speculated that the incidence of cardiovascular events had decreased by the time we conducted our study, because preassessment and preoperative cardiovascular optimization has improved [9]. The MELD scoring system determines the severity of liver disease and is the basis for organ allocation in LT [10]. It is calculated on the basis of serum bilirubin, creatinine, and international normalized ratio. In the present study, the comparison between the MELD scores of the 2 groups was vital to indicate a similar extent of disease between groups,
114
thereby eliciting the true effect of interventions on outcomes. There was no statistically significant difference in ICU mortality between the 2 patient groups. The numbers of postoperative ICU deaths in each group were very low (3.2%) for this type of surgery. STUDY LIMITATIONS
Despite the many factors that can affect intraoperative hemodynamics in LT patients [1,2], our goal was to compare patients who were on beta-blockers with those who were not. We were aware that division of the groups in this manner may add inherent bias. Regarding the study size, this was a relatively small cohort of 308 patients. Although we used LIDCO monitoring for all LT patients, only 36.4% of the LT patients had retrievable LIDCO data. The LIDCO monitor automatically saves the data and keeps them for a couple of months before deletion. We lost data in several ways: 1) Some saved data were unusable; 2) some data were deleted when the monitor was updated (about which the research team was not informed); and 3) some data were deleted before they were able to be saved. Therefore, there were no biases in data selection. Because it was a retrospective study, we did not calculate the power of the study, and there is likely to be heterogeneity and bias in selecting patients without and with betablocker treatment for portal hypertension. Of the patients treated with beta-blockers, 31% had no documented esophageal varices, and 25% of patients not treated with beta-blockers had documented esophageal varices. Our clinicians’ decision to use nonselective betablockers was based on the belief that the risk of variceal bleeding correlates with several factors, not only the portal pressure, but also infections and the severity of liver dysfunction, suggesting the absence of a pure mechanical model [11]. Beta-blockers were used in decompensated patients with medium to large varices but not in patients with early or end-stage cirrhosis with refractory ascites. This explains why there were no differences in severity of liver disease between groups. In addition, LT programs have different intra- and postoperative care; therefore, it may not be possible to extrapolate the findings in our specific population to a mixed general population of patients scheduled to undergo LT. Patients received beta-blockers for prophylaxis of variceal bleeding without standardized intervention of drug type or dose. Although propranolol, a noncardioselective drug, would have been the primary intervention, all previous investigations used cardioselective agents [7]. Further trials should explore this more potent and evidence-based subclass of drugs in this unique surgical group, such as nadolol and carvedilol, which are gradually replacing propranolol for portal hypertension management in patients with ESLD [12]. In addition, the fact that this study was retrospective means that it is impossible to comment on preoperative beta-blocker compliance. However, the highly significant
MILAN, TAYLOR, ARMSTRONG ET AL
differences in physiology between the groups would have been unlikely unless there had been a satisfactory degree of compliance. It was also impossible to standardize group characteristics during allocation. We have already shown that there were no significant differences in the general make-up of the groups. However, we do not have data on preoperative cardiovascular diseases and cardiac outcomes in these patients to explore the incidence of cardiovascular events. Indeed, the reduction in the incidence of postoperative myocardial infarction (fatal/nonfatal) was a key finding in the Safadi et al study [8]. This represents a key outcome that we were unable to explore owing to lack of relevant information in the database. Most data were limited to the immediate ICU stay and did not encompass a full 30-day perioperative period Although we analyzed immediate perioperative parameters, a longer follow-up with a rigorous analysis of full clinical sequelae would have elicited more valuable information, for example, there is some evidence of limited persistent hyperdynamic circulation after transplantation compared with normal control subjects. However, return of physiologic parameters toward normalization is generally seen; for example, one cohort demonstrated a net decrease of CI by 35% [2]. It is clearly shown that the aforementioned confounding variables in this study would have been largely avoided with the implementation of a prospective cohort study. Indeed, a large randomized control trial (RCT) would be the method of choice owing to its superiority in the hierarchy of evidence. A large prospective RCT that uses a standard-dose cardioselective beta-blocker with long-term follow-up would be the optimal method of investigation. CONCLUSION
This study has demonstrated that the preoperative administration of beta-blockers to some degree attenuates the hyperdynamic circulation associated with ESLD, as clearly shown by a reduction in intraoperative CO, CI, and HR during LT. These hemodynamic changes were not shown to relate to outcome in any of the investigated variables, with no significant difference in any of the outcome parameters (ICU stay, RRT, postoperative mortality, and physiologic scoring methods) except for a higher ICNARC score in the group treated with beta-blockers. There is evidence to suggest that a cardioprotective effect may present postoperatively in the medium to long term. However, this study can not advise on treatment recommendations for cardiovascular prophylaxis, and further investigation via a prospective RCT with the use of a cardioselective beta-blocker class is required to further this work. ACKNOWLEDGMENTS The authors thank Ms Helen Aldersley, Leeds LT coordinator, for helping with data collection and Dr David Green from King’s College Hospital for useful suggestions before final submission of the manuscript.
PREOPERATIVE BETA-BLOCKER USE
REFERENCES [1] Milan Z, Taylor C, Duncan B, Kedilaya H, Sylvester D. Statistical modeling of hemodynamic changes during orthotopic liver transplantation: predictive value for outcome and effect of marginal donors. Transplant Proc 2011;43(5):1711e5. [2] Siniscalchi A, Aurini L, Spedicato S, Bernardi E, Zanoni A, Dante A, et al. Hyperdynamic circulation in cirrhosis: predictive factors and outcome following liver transplantation. Minerva Anestesiol 2013;79:15e23. [3] Vilas-Boas WW, Ribeiro-Oliveira Jr A, Ribeiro Rda C, Vieira RL, Almeida J, Nadu AP, et al. Effect of propranolol on the splanchnic and peripheral renin angiotensin system in cirrhotic patients. World J Gastroenterol 2008;14:6824e30. [4] Leithead JA, Rajoriya N, Tehami N, Hodson J, Gunson BK, Tripathi D, Ferguson JW. Nonselective b-blockers are associated with improved survival in patients with ascites listed for liver transplantation. Gut 2015;64:1111e9. [5] Mandorfer M, Bota S, Schwabl P, Bucsics T, Pfisterer N, Kruzik M, et al. Nonselective b-blockers are associated with improved survival in patients with ascites listed for liver transplantation. Gastroenterology 2014;146:1680e90. [6] Harrison DA, Parry GJ, Carpenter JR, Short A, Rowan K. A new risk prediction model for critical care: the Intensive Care
115 National Audit & Research Centre (ICNARC) model. Crit Care Med 2007;35:1091e8. [7] Blessberger H, Kammler J, Domanovits H, Schlager O, Wildner B, Azar D, et al. Perioperative beta-blockers for preventing surgery-related mortality and morbidity. Cochrane Database Syst Rev 2014;(9):CD004476. [8] Safadi A, Homsi M, Maskoun W, Lane KA, Singh I, Sawada SG, Mahenthiran J. Perioperative risk predictors of cardiac outcomes in patients undergoing liver transplantation surgery. Circulation 2009;120:1189e94. [9] Kong YG, Kang JW, Kim YK, Seo H, Lim TH, Hwang S, et al. Preoperative coronary calcium score is predictive of early post-operative cardiovascular complications in liver transplant recipients. Br J Anaesth 2015;114:437e43. [10] Klein KB, Stafinski TD, Menon D. Predicting survival after liver transplantation based on pre-transplant MELD score: a systematic review of the literature. PLoS One 2013;8:e80661. [11] Ferrarese A, Tsochtzis E, Burroughs AK, Senzolo M. Betablockers in cirrhosis: therapeutic window or an aspirin for all? J Hepatol 2014;61:449e50. [12] Lo GH, Chen WC, Wang HM, Yu HC. Randomized, controlled trial of carvedilol versus nadolol plus isosorbide mononitrate for the prevention of variceal rebleeding. J Gastroenterol Hepatol 2012;27:1681e7.
