Chronic severe hepatitis (CSH) can cause

Original Article / Liver The MELD scoring system for predicting prognosis in patients with severe hepatitis after plasma exchange treatment Jian-Wu Yu, Gui-Qiang Wang, Yong-Hua Zhao, Li-Jie Sun, Shu-Qin Wang and Shu-Chen Li Harbin, China BACKGROUND: Hepatic failure caused by severe hepatitis is a clinical syndrome where the major liver functions, particularly detoxification, synthetic functions, and metabolic regulation are impaired to different degrees, and may result in major life-threatening complications such as hepatic encephalopathy, ascites, jaundice, cholestasis, bleeding and hepatorenal syndrome. Plasma exchange (PE) has been found useful in treating patients with fulminant hepatic failure by removing hepatic toxins and replacement of clotting factors, so PE treatment has temporary supportive effects on liver failure caused by severe viral hepatitis. In this study, our aim was to predict the prognosis of patients with severe hepatitis after PE treatment using the end-stage liver disease (MELD) scoring system. METHODS: Two hundred and twenty patients were randomly divided into PE and control groups, and the was calculated for each patient according to the original formula. The efficacy of PE was assessed by or improvement in biochemical parameters and. RESULTS: The levels of total bilirubin and international normalised ratio (INR) in patients whose s were between 30 and 39 were lower than those before PE treatment, as those in patients whose s were 40 Author Affiliations: Department of Infectious Diseases, Second Affiliated Hospital, Harbin Medical University, Harbin 150086, China (Yu JW, Zhao YH, Sun LJ, Wang SQ and Li SC); and Department of Infectious Diseases, First Hospital of Peking University, Beijing 100034, China (Wang GQ) Corresponding Author: Shu-Chen Li, MD, Department of Infectious Diseases, Second Affiliated Hospital, Harbin Medical University, Harbin 150086, China (Tel: 86-451-86605614; Fax: 86-451-86605614; Email: shuchenli@yahoo.com.cn) 2007, Hepatobiliary Pancreat Dis Int. All rights reserved. or higher. The of patients in the PE group with s from 30 to 39 was 50.0%, while it was 83.3% in the control group (P<0.01). The of patients with s higher than 40 was 90.0% in the PE group and 98.0% in the control group (P>0.05). CONCLUSIONS: PE treatment can decrease the serum total bilirubin level and INR and of patients with severe hepatitis and improve liver function. Compared with the control group, PE can significantly decrease the of patients with s from 30 to 39, but has no effect in patients with s of 40 or higher. (Hepatobiliary Pancreat Dis Int 2007; 6: 492-496) KEY WORDS: severe hepatitis; model for end-stage liver disease; plasma exchange; Introduction Chronic severe hepatitis (CSH) can cause irreversible hepatic failure resulting from the severe impairment of liver function, and is associated with a rate of over 70% if liver transplantation is not available. [1] Liver failure is the clinical syndrome that results from severe impairment of liver function. This syndrome encompasses the development of encephalopathy, coagulopathy, jaundice, and hepatorenal syndrome. [2] Plasma exchange (PE) is efficacious in removing toxic substances which accumulate in patients with fulminant hepatic failure (evidenced by a decrease in plasma ammonia, bilirubin and bile acid levels), halting the progression of liver injury (as shown by decreased plasma alanine aminotransferase and lactate dehydrogenase levels), retaining useful materials (evidenced by higher hepatocyte growth factor retention), improving liver function (as shown 492 Hepatobiliary Pancreat Dis Int,Vol 6,No 5 October 15,2007 www.hbpdint.com

The MELD scoring system for predicting prognosis in patients with severe hepatitis after PE treatment by a better indocyanine green clearance value), promoting liver regeneration, and prolonging survival of patients. [3] PE is one of the reliable therapies for patients with CSH, but the reported efficacy is very different. Forman et al [4] reported that the model for end-stage liver disease (MELD) score is related to the prognosis of patients with severe hepatitis. In this study, we used the MELD scoring system to predict the 3-month prognosis of patients with CSH after PE treatment. Methods Patients Two hundred and twenty hospitalized patients with CSH were recruited from the Department of Infectious Diseases, Second Affiliated Hospital, Harbin Medical University, China, between January 2003 and December 2005. They were randomly divided into two groups (PE and control groups). The 110 patients in the PE group included 90 males and 20 females, with an average age of 44.0±10.3 years, while the 110 patients in the control group included 88 males and 22 females, with an average age of 45.9±10.9 years. The diagnosis of CSH patients was according to the criteria of the National Infectious Diseases Meeting in Xi'an 2000. [5] According to these criteria, there are three types of severe viral hepatitis in China: acute, subacute, and chronic-severe. Acute-severe hepatitis is characterized by malignant jaundice (total bilirubin >171 μmol/l), hepatic encephalopathy (above phase II), prolonged prothrombin time (prothrombin activity <40%), extreme fatigue, loss of appetite, nausea, abdominal distention or ascites and bleeding tendency within two weeks after the clinical manifestations. Subacute-severe hepatitis is characterized by symptoms of acute-severe hepatitis from two weeks to six months. Chronicsevere hepatitis is clinically similar to subacutesevere hepatitis on the basis of known history as a HBV or HCV carrier, chronic hepatitis or cirrhosis, signs of chronic liver disease (such as liver palmer or spider angioma) or with results from imaging or biopsy indicating the existence of chronic hepatitis. Malignant jaundice induced by obstructive jaundice and hemolytic jaundice and prolonged prothrombin time induced by blood system disease were excluded. Drug hepatitis, Wilson disease, alcoholic liver disease, and autoimmune hepatitis were also excluded. Of the 220 patients with CSH, 198 were hepatitis B virus positive, 22 were non-a-g hepatitis virus positive. Before treatment, 23 patients had encephalopathy, 21 had spontaneous bacterial peritonitis and 6 had hepatorenal syndrome. Therapeutic methods The 110 patients in the control group were given standard medical treatment. This included absolute bed rest; supplement of energy and vitamins; intravenous drop infusion of hepatocyte growthpromoting factors, transmetil, prostaglandin E, albumin, plasma, and prothrombin complex; maintenance of water, electrolyte and acid-base equilibrium; and prevention and treatment of complications. The patients with severe hepatitis B (HBV DNA 1 10 4 copies/ml) were treated with lamivudine, 100 mg daily. The 110 patients in the PE group were given standard medical treatment combined with PE therapy. All patients offered written informed consent before treatment. The PE treatment room was thoroughly sterilized by UV light. Before PE treatment, blood access for extracorporeal circulation was established with a double-lumen catheter inserted into the patient's jugular and femoral veins, and heparin was used to block the catheter after every session. PE was performed using the WLXGX-8888 WeiLi Plasmapheresis-Artificial Liver Support System, a matched plasma separator (WLXGX-8888) and channels (Beijing WeiLi Co., Ltd.). Routinely before therapy, 5 mg dexamethasone (i.v.) and 25 mg diprazine (i.m.) were injected to prevent an allergic response. During each session of PE treatment, which lasted for 4-6 hours, the total volume of exchanged plasma was 3000-4000 ml. The flow rate of blood was adjusted to 60-130 ml/min, and the flow rate of filtration was adjusted to 15-30 ml/min. The ratio of filtration pump to blood pump was 0.25-0.30. During the session, the prothrombin time was tested constantly to allow the dose of heparin to be adjusted. Blood pressure, pulses and ECG were recorded continuously. Transmembrane, arterial and venous pressures were closely watched, and dexamethasone and calcium gluconate were used routinely during sessions. Five-ten mg heparin was injected routinely before therapy, and the total dose of heparin was 5-25 mg. Protamine was not used during sessions. PE therapy was carried out twice every week. For patients at the critical stage, PE therapy was carried out daily or every other day until the patient's condition was stable. [6] Parameters The level of serum creatinine, the international normalised ratio (INR) for prothrombin time, and the Hepatobiliary Pancreat Dis Int,Vol 6,No 5 October 15,2007 www.hbpdint.com 493

Hepatobiliary & Pancreatic Diseases International Table 1. Comparison of clinical and biochemical characteristics and s of patients in the PE and control groups s 30-39 Gender (M/F) Age (yrs) n INR Creatinine (μmol/l) Total bilirubin (μmol/ L) ALT (U/L) AST (U/L) Albumin (g/l) MELD score Days of hospital stay PE 50/10 44.2±3.1 60 3.4±0.3 121.3±30.2 507.7±65.6 356.4±54.7 401.7±60.2 29.9±3.6 37.1±3.5 37.7±6.9 Control 47/13 44.9±3.7 60 3.5±0.4 114.8±33.1 499.3±57.8 348.9±60.7 409.4±65.2 29.1±3.8 36.7±3.7 38.4±7.2 s 40 PE 40/10 44.3±3.2 50 4.4±0.7 202.6±39.4 651.6±65.8 563.7±80.5 604.8±97.2 26.5±3.0 44.9±4.2 30.2±5.4 Control 41/9 44.1±3.6 50 4.5±0.6 208.6±43.2 643.2±64.3 597.5±86.5 597.8±94.7 25.9±3.3 45.7±4.7 30.2±6.0 Compared with the control group, P>0.05. level of serum total bilirubin of each CSH patient were recorded. The was calculated according to the original formula proposed by the Mayo Clinic group: 3.8 log e [bilirubin (mg/dl)]+11.2 log e (INR)+ 9.6 log e [creatinine (mg/dl)]+6.4 (etiology: 0 if cholestatic or alcoholic, 1 otherwise). [7] Clinical and biochemical characteristics Blood samples were collected before and 24 hours after each treatment to check liver function, renal function, INR and routine blood tests. The results of the clinical data and s before and after PE treatment were compared. Analysis of The 220 patients with CSH were followed up for at least three months. The outcome (recovery, bridging to liver transplantation, or death) of each patient was recorded. Statistical analysis The clinical and biochemical characteristics and s of the patients are expressed as mean± SD. Analysis of used the Chi-square test, the comparison of the clinical data and s before and after PE treatment used Student's t test, and for all analyses a P value less than 0.05 was considered statistically significant. Results Clinical and biochemical parameters and MELD scores of patients with and without PE treatment The s of all the patients were over 30. The patients were divided into two groups according to their scores: 30-39, and 40. There was no significant difference between the PE and control groups in clinical and biochemical characteristics and MELD scores (Table 1). Table 2. Clinical data and of patients pre- and post-pe treatment Patients in PE group Total bilirubin (μmol/l) INR Creatinine (μmol/ L) MELD score Pre-PE treatment 549.3±61.7 3.8±0.5 162.4±35.1 41.2±3.8 Post-PE treatment 422.8±53.1 * 3.0±0.3 * 153.7±42.0 # 37.6±3.7 * s 30-39 Pre-PE treatment 507.7±65.6 3.4±0.3 121.3±30.2 37.1±3.5 Post-PE treatment 380.4±40.6 * 2.4±0.3 * 109.4±22.7 # 30.6±3.7 * s 40 Pre-PE treatment 651.6±65.8 4.4±0.7 202.6±39.4 44.9±4.2 Post-PE treatment 593.7±60.8 * 3.7±0.4 * 142.8±25.1 # 39.4±3.6 * *: Compared with pre-treatment, P<0.05; #: compared with pretreatment, P>0.05. Table 3. Relationship between and 3-month Overall Mortality of patients with 30-39 Mortality of patients with 40 PE 68.2 (75/110) * 50.0 (30/60) ** 90.0 (45/50) # Control 90.0 (99/110) 83.3 (50/60) 98.0 (49/50) *: Compared with the control group, χ 2 =15.832, P<0.01; **: compared with the control group, χ 2 =15.00, P<0.01; #: compared with the control group, χ 2 =2.837, P>0.05. Clinical and biochemical characteristics and MELD scores after PE treatment The levels of total bilirubin and INR in patients whose s were between 30 and 39 were significantly lower than those before PE treatment, as were the levels in patients with s 40 (Table 2). The levels of serum creatinine were low before PE treatment, but there was no significant difference afterwards, which may be due to PE treatment with little effect on renal function in CSH patients. There was no significant difference in the 494 Hepatobiliary Pancreat Dis Int,Vol 6,No 5 October 15,2007 www.hbpdint.com

The MELD scoring system for predicting prognosis in patients with severe hepatitis after PE treatment levels of white blood cells, red blood cells, hemoglobin and platelets before and after therapy. Renal electrolytes showed no significant changes pre- and post-treatment. Mortality in patients with and without PE treatment During the 3-month follow-up, 174 patients died. The causes of death were all related to liver disease: 82 patients died of hepatic encephalopathy, 38 of hepatorenal syndrome, 42 of variceal bleeding; and 11 of sepsis due to liver dysfunction. Six patients bridged to liver transplantation, of whom one died during follow-up because of liver failure. The of patients in the PE group with s from 30 to 39 was 50.0%, while it was significantly higher (83.3%) in the control group (χ 2 =15.000, P<0.01). The of patients with s 40 was 90.0% in the PE group and 98.0% in the control group, showing no significant difference (χ 2 =2.837, P>0.05) (Table 3). PE treatment significantly decreased the of patients with s from 30 to 39, but had no effect on patients with s 40. Discussion In patients with CSH, the massive accumulation of metabolic toxins such as middle or micromolecule substitutes (blood ammonia, direct bilirubin, indirect bilirubin and aromatic amino acids), and macromolecule substitutes (benzopyrrole, mercaptan and endotoxin) in the circulation delays the recovery of hepatic function and causes many kinds of severe complications, which progress to hepatic failure. [8] PE treatment being the most common non-bioartificial liver has been widely applied in clinical treatment. PE separates and discards the plasma to remove the toxic substances (especially those binding with proteins) and compensates with an identical volume of normal freshfrozen plasma to supplement essential substances such as coagulation factors, albumin and immunoglobin so as to ameliorate the microenvironment of the liver and accelerate liver regeneration and functional [9, 10] recovery. Our study showed that PE treatment decreased the serum total bilirubin level and INR and in patients with severe hepatitis and improving liver function, which further suggested that PE can temporarily support liver function and gain precious time for hepatocytes to regenerate and thus decrease the of patients with CSH. At present, prediction of the prognosis in patients with CSH is largely limited to small-sample and single factor studies, with few objective parameters. The reports about clinical and biochemical characteristics and of patients with CSH after PE treatment [11, 12] are conflicting. The MELD was formulated to predict 3-month for cirrhotic patients undergoing transjugular intrahepatic portosystemic shunt. [13] In the United States, the MELD scoring system is mainly used in patients with end-stage cirrhosis awaiting liver transplantation. [14] At present, the MELD has been validated by the same authors on a broad series of patients with liver diseases of various etiology and severity. [15] The takes into consideration objective parameters (serum creatinine, INR for prothrombin time and bilirubin levels) and is computed with statistically derived coefficients on a continuous scale with no upper or lower limits, thus avoiding many drawbacks of the Child-Pugh score. [16] The short-term survival rates of patients with endstage liver disease decrease gradually as the increases. [7] This MELD scoring system is not only objective, but also consistent with the characteristics of CSH (for example, hyperbilirubinemia, coagulopathy, and renal function failure), so it may be used to predict the prognosis of CSH patients. [17] We studied the 3-month prognosis of patients with severe hepatitis after PE treatment using the MELD scoring system. The 3-month of patients with scores of 30-39 given standard medical treatment was 83%, while for those given PE, it was 50.0%, suggesting that PE significantly increased the shortterm survival rate of these patients, so PE treatment and other artificial liver support system therapy should be started as early as possible. The 3-month of the patients with s 40 given standard medical treatment was 98%, whereas for those given PE, it was 90.0%, suggesting that PE did not significantly increase the short-term survival rate in these patients. Liver transplantation may be the only effective therapeutic modality to save these patients' lives. While awaiting liver transplantation, PE treatment's temporary support of liver function is expected to be effective to reduce metabolic disturbance and coagulopathy, improve liver function before operation, and increase the success rate of the operation and survival rate of patients. In conclusion, PE treatment decreases the serum total bilirubin level as well as INR and s of patients with fulminant hepatitis and improves liver function. Compared with controls, PE can significantly decrease the of patients with s from 30 to 39, but it is not effective in patients with s 40. Hepatobiliary Pancreat Dis Int,Vol 6,No 5 October 15,2007 www.hbpdint.com 495

Hepatobiliary & Pancreatic Diseases International Funding: This study was supported by a grant from the Science and Technology Research Foundation of the Department of Education, Heilongjiang Province, China (No. 11511233). Ethical approval: Not needed. Contributors: YJW and LSC proposed the study and YJW wrote the first draft. SLJ analyzed the data. All authors contributed to the design and interpretation of the study and to further drafts. LSC is the guarantor. Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1 Marrero J, Martinez FJ, Hyzy R. Advances in critical care hepatology. Am J Respir Crit Care Med 2003;168:1421-1426. 2 Polson J, Lee WM; American Association for the Study of Liver Disease. AASLD position paper: the management of acute liver failure. Hepatology 2005;41:1179-1197. 3 Ho DW, Fan ST, To J, Woo YH, Zhang Z, Lau C, et al. Selective plasma filtration for treatment of fulminant hepatic failure induced by D-galactosamine in a pig model. Gut 2002;50:869-876. 4 Forman LM, Lucey MR. Predicting the prognosis of chronic liver disease: an evolution from child to MELD. Mayo Endstage Liver Disease. Hepatology 2001;33:473-475. 5 Chinese Society of Infectious Diseases and Parasitology and Chinese Society of Hepatology of Chinese Medical Association. The programme of prevention and cure for viral hepatitis. Zhonghua Chuan Ran Bing Za Zhi 2001;19: 56-62. 6 Artificial Liver Chinese Association of Infectious and Parasitic Diseases. Operating guide for artificial liver support system. Zhonghua Gan Zang Bing Za Zhi 2002;10: 329-332. 7 Kamath PS, Wiesner RH, Malinchoc M, Kremers W, Therneau TM, Kosberg CL, et al. A model to predict survival in patients with end-stage liver disease. Hepatology 2001;33:464-470. 8 Mullin EJ, Metcalfe MS, Maddern GJ. Artificial Liver Support: potential to retard regeneration? Arch Surg 2004;139:670-677. 9 Nakae H, Yonekawa T, Narita K, Endo S. Are proinflammatory cytokine concentrations reduced by plasma exchange in patients with severe acute hepatic failure? Res Commun Mol Pathol Pharmacol 2001;109: 65-72. 10 Nakamura T, Ushiyama C, Suzuki S, Shimada N, Ebihara I, Suzaki M, et al. Effect of plasma exchange on serum tissue inhibitor of metalloproteinase 1 and cytokine concentrations in patients with fulminant hepatitis. Blood Purif 2000;18:50-54. 11 Shang J, Jia BL, Zhang HP, Chen P, Jin X. Effects of artificial liver support system on chronic sever hepatitis patients. Zhonghua Gan Zang Bing Za Zhi 2003;11:506. 12 Li L, Yang Q, Huang J, Xu X, Chen Y, Chen Y, et al. Treatment of hepatic failure with artificial liver support system. Chin Med J (Engl) 2001;114:941-945. 13 Fejfar T, Safka V, Hulek P, Vanasek T, Krajina A, Jirkovsky V. in prediction of early in patients suffering refractory ascites treated by TIPS. Vnitr Lek 2006; 52:771-776. 14 Freeman RB Jr, Wiesner RH, Harper A, McDiarmid SV, Lake J, Edwards E, et al. The new liver allocation system: moving toward evidence-based transplantation policy. Liver Transpl 2002;8:851-858. 15 Dunn W, Jamil LH, Brown LS, Wiesner RH, Kim WR, Menon KV, et al. MELD accurately predicts in patients with alcoholic hepatitis. Hepatology 2005;41:353-358. 16 Huo TI, Lin HC, Wu JC, Lee FY, Hou MC, Lee PC, et al. Proposal of a modified Child-Turcotte-Pugh scoring system and comparison with the model for end-stage liver disease for outcome prediction in patients with cirrhosis. Liver Transpl 2006;12:65-71. 17 Yu JW, Wang GQ, Li SC. Prediction of the prognosis in patients with acute-on-chronic hepatitis using the MELD scoring system. J Gastroenterol Hepatol 2006;21:1519-1524. Received March 12, 2007 Accepted after revision June 25, 2007 496 Hepatobiliary Pancreat Dis Int,Vol 6,No 5 October 15,2007 www.hbpdint.com