LIVER TRANSPLANTATION 14:235-239, 2008 ORIGINAL ARTICLE The Effect of Liver Transplantation on Autonomic Dysfunction in Patients with End-Stage Liver Disease Elizabeth J. Carey, Manjushree Gautam, 3 Timothy Ingall, 2 and David D. Douglas Divisions of 1 Transplantation Medicine and 2 Neurology, Mayo Clinic Arizona, Phoenix, AZ; and 3 Division of Gastroenterology and Hepatology, University of Iowa, Iowa City, IA. Autonomic dysfunction is a recognized complication of end-stage liver disease (ESLD), but there is little information on how liver transplantation (LT) affects this problem. We sought to prospectively evaluate autonomic function in patients with ESLD before and after LT. Autonomic reflex screen (ARS) was performed on 30 patients with ESLD prior to transplantation. A 10-point composite autonomic score (CAS) was calculated from these data. ARS was repeated after LT, and these scores were compared with the pre-lt ARS. Thirty patients (25 male, 5 female) with cirrhosis that were listed for LT were enrolled in the study and underwent ARS prior to LT. The average age was 55.4 9.1 years. Indications for LT included hepatitis C virus (14), cryptogenic cirrhosis (5), alcoholic cirrhosis (4), and other (7). The mean native Model for End-Stage Liver Disease (MELD) score at ARS was 17.0 5.0. Prior to LT, 86.7% of patients had evidence of autonomic dysfunction. Mean CAS was 2.7 2.2. Sudomotor function was disturbed in 66%, parasympathetic function was disturbed in 57%, and adrenergic function was disturbed in 37%. There was no relationship between pre-lt CAS and age, gender, diabetes, etiology of liver disease, or MELD score. Twenty-one patients (17 male, 4 female) had repeat ARS a mean of 9 6.2 months after LT. The mean native MELD score at the time of ARS testing was 18.1 4.3. Mean pre-lt CAS in this group was 3.0 2.4. Pretransplant CAS was not related to age, gender, diabetes, or MELD score. Autonomic dysfunction improved after LT (CAS pre-lt, 3.0, versus CAS post-lt, 1.9, P 0.02). There was no relationship between post-lt CAS and age, gender, diabetes, etiology of liver disease, immunosuppression, or type of transplant. In conclusion, autonomic dysfunction is common in patients with ESLD, with over 86% having abnormal testing. Sixty-three percent of patients with cirrhosis with autonomic dysfunction show improvement after LT. Liver Transpl 14:235-239, 2008. 2008 AASLD. Received June 29, 2007; accepted August 29, 2007. Autonomic neuropathy is recognized in up to 68%-80% of patients with cirrhosis. 1-3 It occurs with comparable frequency in all etiologies of liver disease 1,2,4,5 and worsens with the degree of hepatic dysfunction. 2 Autonomic dysfunction has been associated with hemodynamic instability during surgery 6 and is considered an independent risk factor for mortality in patients with cirrhosis. 4 There is little information regarding the effect of liver transplantation (LT) on autonomic neuropathy in patients with cirrhosis. With the corrected QT interval on the electrocardiograph used as a marker, improvement in autonomic function after LT has been demonstrated. 7 Improvement in sympathetic and parasympathetic function after LT has also been reported. 1 The aim of this study was to complete formal autonomic testing on patients with end-stage liver disease and to compare this with repeat formal testing after LT. PATIENTS AND METHODS Patients Thirty patients with end-stage liver disease listed for LT at a single institution were enrolled in the study. Pa- Abbreviations: ALD, alcoholic liver disease; ARS, autonomic reflex screen; CAS, composite autonomic score; ESLD, end-stage liver disease; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HH, hereditary hemochromatosis; LT, liver transplantation; MELD, model for end-stage liver disease; NASH, nonalcoholic steatohepatitis; PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; QSART, quantitative sudomotor axon reflex test; SD, standard deviation. Supported by the Edson Foundation. Address reprint requests to Elizabeth J. Carey, M.D., Assistant Professor of Medicine, Division of Transplantation Medicine, Mayo Clinic Arizona, 5779 East Mayo Boulevard, Phoenix, AZ 85054. Telephone: 480-342-1094; FAX: 480-342-2324; E-mail: carey.elizabeth@mayo.edu DOI 10.1002/lt.21350 Published online in Wiley InterScience (www.interscience.wiley.com). 2008 American Association for the Study of Liver Diseases.
236 CAREY ET AL. tients were offered enrollment regardless of the etiology of liver disease, age, gender, or status on the waitlist. Patients with a history of neurologic disease were excluded. Patients with diabetes were included, and their charts were reviewed for evidence of diabetes-related complications. All patients were abstinent from alcohol for at least 6 months, and none had a history or laboratory evidence of other causes of neuropathy such as family history, toxin exposure, vitamin B12 deficiency, thiamine deficiency, or human immunodeficiency virus. None of the hepatitis C patients had signs or symptoms of cryoglobulinemia. Model for End-Stage Liver Disease (MELD) scores were calculated from laboratory values on the day of autonomic reflex screen (ARS) testing. For the purposes of this study, only native MELD scores were considered. Autonomic Testing A formal ARS was performed on all enrolled patients prior to and after LT. ARS tests sudomotor, cardiovagal, and adrenergic nervous system function and includes the following: Quantitative sudomotor axon reflex test (QSART) at 4 sites. Heart rate response to Valsalva maneuver and deep breathing. Blood pressure and heart rate responses to head-up tilt. QSART measures quantitative sweat response at 4 sites (medial forearm, proximal leg, distal leg, and proximal foot) after 5 minutes of exposure to a stimulus of a 2-mA constant current. This process tests the integrity of the postganglionic sympathetic sudomotor axon. Cardiovagal function is determined with the heart rate response to Valsalva and deep breathing. This is performed after 20 minutes of lying supine (QSART testing is usually performed during this interval). Electrodes placed in the interscapular areas measure the heart rate, which is calculated by the R-R interval. The subject is taught to breathe maximally at a rate of 6 breaths per minute. Eight inspiratory/expiratory cycles are recorded, and the process is repeated after a 5-minute rest. The 5 largest consecutive responses are averaged, and the heart rate range is computed. To perform a Valsalva maneuver, the subject is asked to maintain a column of mercury at 40 mm Hg for 15 seconds via a bugle with an air leak. The subject performs 3 Valsalva maneuvers, with a 2-minute rest between each one. The Valsalva ratio is derived from the maximum heart rate generated during the Valsalva maneuver divided by the lowest heart rate occurring within 30 seconds of the peak heart rate. Adrenergic function is tested by the blood pressure and heart rate response to head-up tilt. Two baseline blood pressure recordings are taken with the patient in the supine position on the tilt table. The patient is then tilted to 70 degrees, and the blood pressure is rechecked after 1 and 5 minutes in this position. Adrenergic function is determined by the beat-to-beat variation in blood pressure caused by the Valsalva and tilt-up maneuvers. All ARS testing was performed in the Neurophysiology Laboratory at Mayo Clinic Arizona by trained and experienced neurophysiology technologists. All ARS testing was done in a standard and protocolized fashion with QSART testing performed first, followed by the tests of cardiovagal function and finally by the tests of adrenergic function. Precise details of autonomic testing have been published elsewhere. 8 All medications that might affect the study were stopped 48 hours prior to testing. Interpretation of ARS was performed by experienced neurologists who were unaware of the patients transplant status. A 10-point composite autonomic score (CAS) was calculated from the results of the individual subscores [maximal subscores: sudomotor (3), cardiovagal (3), and adrenergic (4)]. This scoring system takes into account normal differences in autonomic function as a result of age and gender. 9 Timing of Autonomic Testing Subjects underwent initial ARS at the time of enrollment into the study. The post-lt ARS was scheduled at the time of the 4-month posttransplant evaluation. If the post-lt ARS was unable to be done at the 4-month evaluation, it was performed as soon as logistically possible. No post-lt ARS testing was done prior to the 4-month post-lt follow-up. Statistics Statistical Software for Professionals version 8.2 (Stata, StataCorp LP, College Station, TX) was used for all data analysis. Values are expressed as mean and standard deviation. A paired Student t test was used to detect differences in continuous variables between 2 groups. Univariate and multivariate analyses were done with linear regression to assess the association between the outcome and the dependent variables. Written informed consent was obtained from each subject prior to enrollment. This study was approved by the Mayo Foundation Institutional Review Board. RESULTS Entire Cohort Thirty patients (25 men, 5 women) with cirrhosis and listed for LT underwent ARS prior to LT. The average age was 55.4 9.1 years. Indications for LT included hepatitis C virus (14), cryptogenic cirrhosis (5), alcoholic cirrhosis (4), and other (7; Table 1). The mean native MELD score at the time of ARS was 17.0 5.0. Three patients had diabetes, 2 without any end-organ damage and 1 with mild coronary heart disease. Prior to LT, 86.7% of patients had evidence of autonomic dysfunction. Mean CAS was 2.7 2.2. Sudomotor function was disturbed in 66%, parasympathetic function was disturbed in 57%, and adrenergic func-
AUTONOMIC DYSFUNCTION BEFORE AND AFTER LT 237 TABLE 1. Patient Demographics in 30 Patients Undergoing Autonomic Testing Prior to Liver Transplantation n 30 Age, mean SD 55.4 9.1 years Gender 25 male/5 female Diabetes 3/30 (10%) Etiology of liver disease HCV 14 (including 4 with HCV/HCC) Cryptogenic 5 ALD 4 NASH 2 PSC 1 HBV 1 HCV/ALD 1 HH/PBC 1 HH/ALD 1 MELD at time of ARS 17.0 5.0 testing, mean SD Time between pre-ars and 25 30 days transplantation, mean SD Composite autonomic 3.0 2.4 score, mean SD Abbreviations: ALD, alcoholic liver disease; ARS, autonomic reflex screen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HH, hereditary hemochromatosis; MELD, model for end-stage liver disease; NASH, nonalcoholic steatohepatitis; PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; SD, standard deviation. tion was disturbed in 37% of patients. Autonomic dysfunction was mild (CAS 3) in the majority of patients (16/30, 53.3%), moderate (CAS 4-6) in 9 of 30 (30%), and severe (CAS 7) in 1 of 30 (3.3%). Four patients (13.3%) did not have evidence of autonomic dysfunction at baseline. In univariate analysis, the pre-lt CAS was not associated with age (P 0.88), gender (P 0.56), presence of diabetes (P 0.06), etiology of liver disease (P 0.8), or native MELD score (P 0.1). Pre-LT and Post-LT Comparisons Of the 30 patients enrolled in the study, 21 patients (17 men, 4 women) had paired ARS testing before and after LT (Table 2). Nine patients did not have post-lt ARS testing for the following reasons: 4 patients were lost to follow-up, 4 died after transplantation but before the 4-month post-lt visit [complications of surgery (1), recurrent hepatitis C virus with fibrosing cholestatic hepatitis (1), recurrent hepatocellular carcinoma (1), and disseminated coccidioidomycosis (1)], and 1 died of complications of cirrhosis prior to LT. Pre-LT CAS scores were similar between those who were alive 4 months after transplantation and those who died prior to this time point (P 0.9). In the group of 21 patients with paired ARS testing, TABLE 2. Patient Demographics in 21 Patients Who Had Autonomic Testing Before and After Liver Transplantation n 21 Age, mean SD 54.8 9.5 Gender 17 male/4 female Diabetes 3 (14.3%) Etiology of liver disease HCV 10 HCV/HCC 1 Cryptogenic 3 ALD 2 Other 5 MELD at time of ARS testing, 18.1 4.3 mean SD Time between transplantation and 7 (4 17) months post-ars, median (range) Abbreviations: ALD, alcoholic liver disease; ARS, autonomic reflex screen; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; MELD, model for end-stage liver disease; SD, standard deviation. the mean pre-lt CAS was 3.0 2.4. The mean native MELD score at the time of ARS testing was 18.1 4.3. Pretransplant CAS was not associated with age (P 0.6), gender (P 0.85), presence of diabetes (P 0.1), or MELD score (P 0.9). Alcoholic cirrhosis was significantly associated with pre-lt CAS in univariate analysis (P 0.02), but this association was not significant in multivariate analysis. Post-LT ARS testing was performed a median of 7 (4-17) months after LT. Nine patients had deceaseddonor LT, and 12 underwent living-donor LT. There was no difference in the MELD score in patients undergoing deceased-donor LT versus living-donor LT (P 0.36). Mean post-lt CAS was 1.9 1.5. Post-LT CAS was not associated with age (P 0.3), gender (P 0.6), presence of diabetes (P 0.6), etiology of liver disease (P 0.4), or type of immunosuppression (P 0.2). Immunosuppression regimens consisted of tacrolimus plus mycophenolate mofetil (13), tacrolimus monotherapy (6), cyclosporine plus mycophenolate mofetil (1), and sirolimus plus mycophenolate mofetil (1). Autonomic dysfunction, as measured by CAS, improved after LT (CAS pre-lt, 3.0, versus CAS post-lt, 1.9, P 0.02). The adrenergic subscore improved after LT (P 0.04; Table 3). Cardiovagal improvement did not reach statistical significance (P 0.06). Improvement of autonomic function was not associated with age (P 0.9), gender (P 0.9), presence of diabetes (P 0.2), etiology of liver disease (P 0.2), type of liver transplant (P 0.05), pre-lt MELD (P 0.1), or type of immunosuppression (P 0.4). Sixty-three percent (12/19) of patients with autonomic dysfunction prior to transplant had an improvement in CAS post-transplant.
238 CAREY ET AL. TABLE 3. Pre-LT and Post-LT Autonomic Testing Scores Pre-LT Post-LT P Sudomotor, mean SD 1.45 1.2 1.2 1.2 0.44 Cardiovagal, mean SD 0.86 0.8 0.48 0.7 0.06 Adrenergic, mean SD 0.71 1.0 0.24 0.5 0.04 Composite autonomic score, mean SD 3.0 2.4 1.9 1.6 0.02 Abbreviations: LT, liver transplantation; SD, standard deviation. DISCUSSION Our study confirms that autonomic dysfunction is common in patients with cirrhosis and that it improves after LT. Formal autonomic testing shows improvement in overall autonomic function, with the greatest improvement in adrenergic function. A trend toward improvement in parasympathetic/cardiovagal function after LT was also seen. The pathogenesis of autonomic dysfunction in cirrhosis remains poorly understood. Autonomic dysfunction has consistently been shown to occur in all etiologies of liver disease, and this suggests that nervous system damage occurs as a result of hepatic dysfunction or portal hypertension. Although alcoholic liver disease was associated with higher pre-lt CAS score in univariate analysis, this relationship did not persist in multivariate analysis. A generalized source is suspected, as adrenergic, cardiovagal, and peripheral systems are affected. Although the majority of autonomic dysfunction in patients with liver disease occurs in cirrhosis, it has been described in a series of patients with noncirrhotic portal hypertension as well. 10 The prevalence of autonomic dysfunction in patients with cirrhosis is high. Using sensitive autonomic testing, we found that over 85% of patients listed for LT have some degree of autonomic dysfunction. Most have mild changes, but up to 30% may have evidence of moderate autonomic dysfunction. Autonomic dysfunction in cirrhosis has been shown to be an independent risk factor of mortality. Patients with Child A cirrhosis and normal autonomic studies have a 6% 4-year mortality, whereas Child A patients with abnormal autonomic studies have a 30% 4-year mortality. 11 Similarly, patients awaiting transplant demonstrate higher mortality if they have autonomic dysfunction. 4 The cause of this increased mortality is not known. The autonomic dysfunction of cirrhosis is associated with greater parasympathetic dysfunction than sympathetic dysfunction; this finding was confirmed by our study. 3,12-14 This is in contrast to patients with autonomic dysfunction as a result of diabetes, when both systems are affected equally. In the current study, sudomotor dysfunction was found more commonly than either parasympathetic or sympathetic dysfunction. Few studies have included sudomotor function when investigating the autonomic dysfunction of cirrhosis. Oliver et al., 5 using the silicone imprint technique to quantify the number of functioning sweat glands, found that sudomotor dysfunction occurred in 63% of patients with cirrhosis and occurred more often than parasympathetic or adrenergic dysfunction. Hendrickse et al. 15 reported a 39% frequency of sudomotor dysfunction, using the acetylcholine sweatspot test. Impaired sudomotor function may be the initial indicator of autonomic neuropathy in patients with cirrhosis. 15 The course of autonomic dysfunction after LT is receiving increasing attention. The QT interval on electrocardiography, a surrogate marker of autonomic dysfunction, is prolonged in patients with cirrhosis. This abnormality resolves in the majority of patients after LT, although a small percentage of patients remain abnormal after transplant. 7,12,16,17 There are only a few published reports of more rigorous autonomic testing after LT. Mohamed et al. 12 tested parasympathetic and sympathetic function in 53 patients before and after LT. Parasympathetic function was measured by heart rate variation during deep breathing, with Valsalva maneuver and with standing, in a manner quite similar to our study. Sympathetic function was assessed by blood pressure response to standing and handgrip. The majority of patients had improvement in the indices of parasympathetic and sympathetic function 3 months after LT, although a minority of patients showed no improvement. Perez-Pena et al. 1 found that parasympathetic dysfunction was more common than sympathetic dysfunction (60% versus 21%) and that the overall prevalence of autonomic dysfunction decreased 6 months after transplant. We found that 57% of patients had evidence of parasympathetic dysfunction prior to transplantation and 37% had adrenergic dysfunction. After transplantation, the prevalence decreased to 38% and 19%, respectively. The observation that autonomic dysfunction can reverse within a few months of transplantation suggests a metabolic rather than structural cause. A small number of patients do not have improvement of autonomic dysfunction after transplantation and may even have worsened function. This finding has been reported elsewhere. 1,7 The reasons that some patients fail to improve are poorly understood. In our study, only 3 patients had worsening of autonomic function after LT. In all 3 cases, there was mild (1-2 point) worsening of sudomotor function. All had normal postoperative recovery, and none suffered complications of transplantation or recurrent disease. None of the patients in this series had worsened cardiovagal or adrenergic dysfunction after LT.
AUTONOMIC DYSFUNCTION BEFORE AND AFTER LT 239 There are 2 major limitations to this study. The first is that patients were not evaluated for symptoms of autonomic dysfunction. Prior investigators 14 have not found a relationship between neurologic symptoms and the presence of autonomic dysfunction in patients with cirrhosis, but the current study cannot contribute to the understanding of this issue. A second limitation is related to the follow-up ARS, which was done at a relatively short interval after LT. It is possible that the degree of improvement documented in this study was limited by the short follow-up interval and that further testing done a year or more after LT would show more significant improvement. In this series, improvement was demonstrated within 6-7 months of transplantation in the majority of patients who had autonomic dysfunction prior to LT. We did not find an association between improvement of autonomic dysfunction and the timing of post-lt ARS, although this observation is limited by small patient numbers and autonomic testing that was performed only once after LT. A minor limitation of this study was the inclusion of diabetic patients because it is possible that diabetes may have accounted for autonomic dysfunction rather than liver disease. Only 3 patients with diabetes were included, 2 without any end-organ damage from diabetes and 1 with mild coronary heart disease. In the first patient, autonomic dysfunction was minimal (CAS 1) and involved sudomotor function. The second patient had mild-to-moderate parasympathetic and sympathetic dysfunction prior to transplant. After transplantation, the sympathetic dysfunction resolved, although parasympathetic dysfunction did not change. In the third patient, there was severe autonomic dysfunction (CAS 10), which improved considerably after transplantation (CAS 3) with resolution of the parasympathetic and sympathetic abnormalities. Such improvement would not be expected if these patients autonomic dysfunction was due to diabetes. In conclusion, we have shown that autonomic dysfunction involving sudomotor, cardiovagal, and adrenergic systems is common in patients with cirrhosis. This complication occurs independently of age, gender, or the etiology of liver disease and improves significantly after LT. REFERENCES 1. Perez-Pena J, Rincon D, Banares R, Olmedilla L, Garutti I, Grigorov I, et al. Autonomic neuropathy in end-stage cirrhotic patients and evolution after liver transplantation. Transplant Proc 2003;35:1834-1835. 2. Bajaj BK, Agarwal MP, Krishna Ram B. Autonomic neuropathy in patients with hepatic cirrhosis. Postgrad Med 2003;79:408-411. 3. 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;30:1387-1392. 4. Fleckenstein JF, Frank SM, Thuluvath PJ. Presence of autonomic neuropathy is a poor prognostic indicator in patients with advanced liver disease. Hepatology 1996;23: 472-475. 5. Oliver MI, Miralles R, Rubies-Prat J, Navarro X, Espadaler JM, Sola R, et al. Autonomic dysfunction in patients with non-alcoholic chronic liver disease. J Hepatol 1997;26: 1242-1248. 6. Perez-Pena J, Rincon D, Banares R, Olmedilla L, Garutti D, Arnal D, et al. Autonomic neuropathy is associated with hemodynamic instability during human liver transplant. Transplant Proc 2003;35:1866-1868. 7. Carey EJ, Douglas DD. Effects of orthotopic liver transplantation on the corrected QT interval in patients with end-stage liver disease. Dig Dis Sci 2005;50:320-323. 8. Low PA. Testing the autonomic nervous system. Semin Neurol 2003;23:407-421. 9. Low PA. Composite autonomic scoring scale for laboratory quantification of generalized autonomic failure. Mayo Clinic Proc 1993;68:748-752. 10. Rangari M, Sinha S, Kapoor D, Mohan JC, Sarin SK. Prevalence of autonomic dysfunction in cirrhotic and noncirrhotic portal hypertension. Am J Gastroenterol 2002; 97:707-713. 11. Hendrickse MT, Thuluvath PJ, Triger DR. Natural history of autonomic neuropathy in chronic liver disease. Lancet 1992;339:1462-1464. 12. Mohamed R, Forsey PR, Davies MK, Neuberger JM. Effect of liver transplantation on QT interval prolongation and autonomic dysfunction in end-stage liver disease. Hepatology 1996;23:1128-1134. 13. Thuluvath PJ, Triger DR. Autonomic neuropathy and chronic liver disease. Q J Med 1989;72:737-747. 14. Chaudhry V, Corse AM, O Brian R, Cornblath DR, Klein AS, Thuluvath PJ. Autonomic and peripheral (sensorimotor) neuropathy in chronic liver disease: a clinical and electrophysiologic study. Hepatology 1999;29:1698-1703. 15. Hendrickse MT, Triger DR. Peripheral and cardiovascular autonomic impairment in chronic liver disease: prevalence and relation to hepatic function. J Hepatol 1992;26: 177-183. 16. Bal JS, Thuluvath PJ. Prolongation of the QTc interval: relationship with etiology and severity of liver disease, mortality, and liver transplantation. Liver Int 2003,23: 243-248. 17. Garcia Gonzalez M, Hernandez-Madrid A, Lopez-Sanroman A, Candela A, Nuno J, Barcena R. Reversal of QT interval electrocardiographic alterations in patients with cirrhosis undergoing liver transplantation. Transplant Proc 1999;31:2366-2367.