- Email: [email protected]
189 MILD HEPATIC ENCEPHALOPATHY (HE) ASSESSED BY THE REPEATABLE BATTERY FOR THE ASSESSMENT OF NEUROPSYCHOLOGICAL STATUS (RBANS) IS HIGHLY PREVALENT IN AMBULATORY PATIENTS WITH CIRRHOSIS
POSTERS 187 A PHASE I/II, OPEN-LABEL, DOSE-ESCALATION TRIAL USING THE ONCE-DAILY ORAL CHELATOR DEFERASIROX TO TREAT IRON OVERLOAD IN HFE-RELATED HEREDITARY HEMOCHROMATOSIS: FINAL RESULTS P. Phatak1 , P. Brissot2 , M. Wurster3 , P.C. Adams4 , H.L. Bonkovsky5 , J. Gross6 , P. Malfertheiner7 , G.D. McLaren8,9 , C. Niederau10 , A. Piperno11 , L.W. Powell12 , M. Russo13 , U. Stoelzel14 , W. Stremmel15 , L. Griffel16 , N. Lynch16 , Y. Zhang16 , A. Pietrangelo17 . 1 Rochester General Hospital, Rochester, NY, USA; 2 Rennes University Hospital, Rennes, France; 3 Ohio State University Medical Center, Columbus, OH, USA; 4 University Hospital, London, ON, Canada; 5 Cannon Research Center at Carolinas Health Care System, Charlotte, NC, 6 Mayo Clinic, Rochester, MN, USA; 7 Medizinische Fakultat der Otto-von-Guericke-Universit¨ at Magdeburg, Magdeburg, Germany; 8 University of California, Irvine, 9 VA Medical Center, Long Beach, CA, USA; 10 St Josef Hospital, University of Essen, Oberhausen, Germany; 11 Department of Clinical Medicine and Prevention, University of Milano-Bicocca, Monza, Italy; 12 Royal Brisbane and Women’s Hospital and the University of Queensland, Brisbane, QLD, Australia; 13 Carolinas Medical Center, Charlotte, NC, USA; 14 Klinikum Chemnitz, Chemnitz, 15 University Hospital, Heidelberg, Germany; 16 Novartis Pharmaceuticals Corp., East Hanover, NJ, USA; 17 University of Modena and Reggio Emilia, Modena, Italy E-mail: [email protected] Background and Aims: Hereditary hemochromatosis (HH) is characterized by increased intestinal iron absorption. Consequences of iron overload include organ dysfunction, cirrhosis and hepatocellular carcinoma. Although phlebotomy is the standard of care, it is contraindicated in patients with severe heart disease or anemia, venous access may be difficult and compliance can be variable. The once-daily, oral iron chelator, deferasirox (Exjade® ) may provide an alternative treatment. Results of the 6-month core trial were previously reported; 6-month extension data are reported here. Methods: Patients with HFE C282Y homozygous HH with serum ferritin (SF) 300–2000 ng/mL, transferrin saturation ≥45% and no known history of cirrhosis were enrolled in this dose-escalation study (deferasirox 5, 10 and 15 mg/kg/day). This multicenter study comprised a core and extension phase (both 24 weeks). The primary endpoint was the incidence and severity of adverse events (AEs). Secondary endpoints included change in SF. Results: 49 patients were enrolled (33 men, 16 women; mean age 50.6 years; mean 3.1 years since HH diagnosis) and received deferasirox 5 (n = 11), 10 (n = 15) or 15 mg/kg/day (n = 23). 37 (75.5%) patients completed the core trial. Of 26 patients who entered the extension, 23 (88.5%) completed 48 weeks. The most common reasons for discontinuation were AEs, mainly in the 10 and 15 mg/kg/day cohorts. Overall, AEs were dose dependent, including diarrhea, headache and nausea (n = 18, n = 10 and n = 8 in the core and n = 1, n = 1, n = 0 in the extension, respectively). Throughout 48 weeks in the 15 mg/kg/day cohort, six patients experienced ALT >3 x baseline and > upper limit of normal (ULN), and eight patients had serum creatinine ≥33% above baseline and >ULN on two consecutive occasions. SF declined in all cohorts; median reduction was 63.5%, 74.8% and 74.1% in the 5, 10 and 15 mg/kg/day cohorts, respectively; in all cohorts, median SF was <250 ng/mL after receiving deferasirox for 48 weeks. Conclusions: Results suggest that deferasirox doses of 5, 10 and 15 mg/kg/day can reduce iron burden in HH patients. Based on the safety and efficacy results, deferasirox 10 mg/kg/day appears to be the most appropriate doses for further study in this patient population.
188 FACTORS PREDICTING IN-HOSPITAL MORTALITY IN CIRRHOTIC PATIENTS PRESENTING WITH SEPSIS A. Rajnakova1 , L.G. Lim1 , E. Tan2 , Y.Y. Dan1,3 , Y.M. Lee1 , V. Lai1 , S.G. Lim1,3 . 1 National University Health System, 2 National University High School, 3 Yong Yoo Lin School of Medicine, Singapore, Singapore E-mail: [email protected] Patients with cirrhosis are at risk of sepsis and increased mortality. We examined factors associated with increased inhospital mortality in cirrhotic patients admitted for sepsis. Methods: All cirrhotic patients admitted from 2004–2007 for sepsis, were identified from the hospital electronic database. Patients with liver cirrhosis and sepsis, defined as at least one sources of infection, and fever, or altered total white cell count, or raised CRP, were included. Liver disease was assessed by the Child–Pugh and the MELD scores. Results: Of total 324 admissions 116 were excluded as they did not fulfill criteria, leaving a total of 208 admissions in 156 patients. One and two sites of infection were identified in 182 and 26 admissions respectively. The infection sites were pneumonia (n = 66), empyema (n = 6), urinary tract infection (n = 63), cutaneous (n = 42), intraabdominal sepsis (n = 56) and endocarditis (n = 1). A positive culture was present in 63%: MRSA (13%), Escherichia coli (11%), ESBL-producing Klebsiella pneumonia (10%), Klebsiella pneumonia (10%), and ESBL-producing Escherichia coli (9%). The inhospital mortality rate was 25%. The mean CP score was 9.0±3.0 and the mean MELD score was 16.5±7.3. In predicting in-hospital death, CP score has an area under the receiver operating characteristic curve (AUROC) of 0.933, with the optimum cut-off at ≥10, while for MELD score was 0.757, with the optimum cut-off ≥17. Four factors were significantly associated with in-hospital mortality on multivariate analysis: more than one site of infection (p = 0.003, OR 27.8, 95% CI = 3.2–242.7), pneumonia (p = 0.000, OR 13.2, 95% CI = 3.8–45.7), Child’s C status (p = 0.000, OR 29.4, 95% CI = 5.9–147.2), and MELD score≥17 (p = 0.024, OR 4.6, 95% CI = 1.2–17.3). As the number of factors increased, the in-hospital mortality rate increased: 0% with no factor, 6.9% with 1 factor, 23.7% with 2 factors, 87.5% with 3 factors, and 100% with 4 factors. By Kaplan Meier analysis, the survival of those with less than 3 factors was significantly better than those with 3 or more factors (p = 0.000). Conclusion: Septic cirrhotic patients with pneumonia, more than one site of infection, Child’s C cirrhosis and high MELD score had a high mortality risk, especially those with at least 3 of these 4 factors. 189 MILD HEPATIC ENCEPHALOPATHY (HE) ASSESSED BY THE REPEATABLE BATTERY FOR THE ASSESSMENT OF NEUROPSYCHOLOGICAL STATUS (RBANS) IS HIGHLY PREVALENT IN AMBULATORY PATIENTS WITH CIRRHOSIS C. Randolph1 , J. Bajaj2 , M.Y. Sheikh3 , R. Vemuru4 , G. Morelli5 , L.A. Balart6 , M. Chojkier7 , M.S. Harris8 , J.D. Bornstein8 , K. Mullen9 . 1 Loyola University Med. Center, Chicago, IL, 2 McGuire DVAMC, Richmond, VA, 3 University of California at San Francisco (UCSF), Fresno Medical Educational Program, Fresno, CA, 4 Permian Research Foundation, Odessa, TX, 5 University of Florida, Gainesville, FL, 6 Tulane University, New Orleans, LA, 7 Veterans Medical Center San Diego, 8 Ocera Therapeutics, San Diego, CA, 9 Case Western Reserve University, Cleveland, OH, USA E-mail: [email protected] Mild HE impairs quality of life and increases the risk of job loss and motor vehicle accidents. The prevalence of mild HE in cirrhotics ranges from 30% to 80%. The ISHEN recommends the RBANS as the tool of choice for detecting mild HE in the United States since it has well-established population norms and correlates with impaired daily functioning and job loss. We are using RBANS to qualify subjects for the ASTUTE trial, a phase 2B study assessing the potential
Journal of Hepatology 2010 vol. 52 | S59–S182
S81
POSTERS for AST-120 (spherical carbon adsorbent) to treat patients with mild HE, and data is now available on the prevalence of neurocognitive dysfunction by RBANS for the first 206 screened subjects. Methods: Trained RBANS raters used the RBANS to determine eligibility for participation in the trial. Subjects with cirrhosis between 18–70 years old and with MELD scores ≤25 were eligible for inclusion provided they did not have a TIPS or surgical shunt, had not had an episode of overt HE in the previous 3 months or taken lactulose, rifaximin or neomycin in the previous 7 days. Subjects were eligible for randomization if they scored below the 10th percentile on the RBANS total scale score, after adjusting for age and education. Results: Of 206 subjects screened to date, 61% were male, the mean age was 55.4 years and hepatitis C was the most common underlying liver disease. 82% of all screened subjects were high school graduates and 55% had attended college. Of the 206, 115 (56%) subjects were eligible for randomization based on a qualifying RBANS score. Qualifying rate was similar regardless of education level – 54% of college attendees, 57% of high school grads and 59% of high school dropouts qualified. RBANS total score could not be predicted based on the subject’s platelet count, bilirubin, MELD score or known history of esophageal varices. Conclusions: Using the RBANS, we found a neurocognitive impairment rate of 56% in a population of well-compensated cirrhotics, suggesting that mild HE is highly prevalent in these patients. Mild HE was not predicted by age, education, MELD score or indicators of portal hypertension. 190 PREVALENCE OF DIASTOLIC DYSFUNCTION AND OTHER SUPPORTIVE CRITERIA OF CIRRHOTIC CARDIOMYOPATHY H. Ribeiro1 , J. Presa2 , R. Margato1 , S. Carvalho1 , C. Ferreira1 , P. Mateus1 , A. Ferreira1 , J.I. Moreira1 . 1 Cardiology, 2 Internal Medicine, CHTMAD, Vila Real, Portugal E-mail: [email protected] Introduction: Cirrhotic cardiomyopathy (CC) is the term used to describe a constellation of features indicative of abnormal heart structure and function in patients with liver cirrhosis (LC). These include systolic and diastolic dysfunction, electrophysiological changes, and macroscopic and microscopic structural changes. Purpose: Our aim was to assess the prevalence of diastolic dysfunction and other supportive criteria of CC, in a cohort of patients with LC. Methods: 37 patients with LC (75.7% male; mean age 62.3±9.3 years old; 64.9% with alcoholic cirrhosis and 35.1% with cirrhosis of nonalcoholic etiology; MELD score >15 in 18.9% and MELD <15 in 18.1%) were enrolled, after excluding those with known cardiovascular risk factors (hypertension, diabetes, dislipidemia, tabagism or obesity), heart or pulmonary diseases. We collect data from patients with LC who as part of their routine care realized blood tests (including natriuretic peptide – BNP and troponin I levels), an electrocardiogram and an echocardiogram. Left ventricular diastolic function was studied by Pulse Wave and Pulsed Doppler Tissue imaging (DTI) and the diagnosis of diastolic dysfunction (DD) was made as defined by the recent consensus statement on the diagnosis of heart failure with normal left ventricular ejection fraction by the Heart Failure and Echocardiography Associations of the European Society of Cardiology. Results: Brain natriuretic peptide (BNP > 100 pg/ml) was elevated in 48.4% and troponin I (TnI >0.1 pg/ml) levels was increased in 5.4% LC patients. Prolonged Q-Tc interval (Q-Tc >440 mseg) was observed in 18.9%. Left atrium enlargement (LA volume >20 cm2) was registered in 83.8% and increased left ventricle end diastolic diameter (LVEDd >53 mm) and left ventricle end systolic diameter (LVESd >32 mm) was present in 21.6% and 27.0% respectively. S82
Increased left ventricular myocardial mass indice (LVMMI >95 g/m2 in Women and >115g/m2 in men) was identified in 67.6%. None of the patients had depressed left ventricular ejection fraction (LVEF <50%). DD was detected in 24.3% of the studied patients. Conclusions: This study showed that DD and other supportive criteria of cirrhotic cardiomyopathy are not rare in LC patients. Left cavities enlargement, ventricular hypertrophy, elevated BNP, prolonged Q-Tc, DD and conserved LVEF are the main abnormalities suggestive of CC found in patients with LC. 191 CLINICAL EFFICACY OF TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT CREATED WITH COVERED STENTS WITH DIFFERENT DIAMETERS: RESULTS OF A RANDOMIZED CONTROLLED TRIAL O. Riggio1 , L. Ridola1 , S. Angeloni1 , F. Cerini1 , F. Fanelli2 , F.M. Salvatori2 , M. Merli1 . 1 Clinical Medicine, 2 Radiology, Sapienza University of Rome, Rome, Italy E-mail: [email protected] Introduction and Aim: The incidence of post-TIPS hepatic encephalopathy (HE) could be reduced by using stents with a small diameter, since post-TIPS HE showed to be related to the amount of blood shunted. Unfortunately, no pharmacological treatment proved to be able to prevent the incidence of post-TIPS HE. Aim of this study was to compare the incidence of HE and the clinical efficacy of TIPS created with 8- or 10-mm PTFE-covered stents. Patients and Methods: Consecutive cirrhotics submitted to TIPS for variceal bleeding or refractory ascites were randomized to receive a 8-mm or 10-mm covered stent. As recommended by our Ethical Committee, the trial was stopped after inclusion of 45 patients. Results: 22 patients were allocated to TIPS implantation with a 8 mm covered stent and 23 patients with a 10 mm covered stent. The two groups were comparable for age, sex, etiology, and psychometric performance. After TIPS placement, the portosystemic pressure gradient significantly decreased in both groups (from 21.3±4.9 to 8.9±2.7 mmHg in the 8-mm stent group and from 22.1±7.1 to 6.9±4.1 mmHg in the 10-mm stent group). The portosystemic pressure after TIPS tended to be lower in the 10mm than in the 8-mm stent group. The probability of remaining free of HE was similar in both groups (log-rank test, p = 0.48) (Figure 1). After TIPS, venous ammonia levels increased significantly only in the 10-mm stent group; psychometric performance was unmodified in both groups. The probability of remaining free of recurrence/persistence of complications due to portal hypertension was significantly higher in the 10-mm than in the 8-mm stent group: 82.9% versus 41.9% at one year; log-rank test, p = 0.002 (Figure 2). Cumulative survival rate was similar in both groups.
Figure 1.
Journal of Hepatology 2010 vol. 52 | S59–S182
188 FACTORS PREDICTING IN-HOSPITAL MORTALITY IN CIRRHOTIC PATIENTS PRESENTING WITH SEPSIS A. Rajnakova1 , L.G. Lim1 , E. Tan2 , Y.Y. Dan1,3 , Y.M. Lee1 , V. Lai1 , S.G. Lim1,3 . 1 National University Health System, 2 National University High School, 3 Yong Yoo Lin School of Medicine, Singapore, Singapore E-mail: [email protected] Patients with cirrhosis are at risk of sepsis and increased mortality. We examined factors associated with increased inhospital mortality in cirrhotic patients admitted for sepsis. Methods: All cirrhotic patients admitted from 2004–2007 for sepsis, were identified from the hospital electronic database. Patients with liver cirrhosis and sepsis, defined as at least one sources of infection, and fever, or altered total white cell count, or raised CRP, were included. Liver disease was assessed by the Child–Pugh and the MELD scores. Results: Of total 324 admissions 116 were excluded as they did not fulfill criteria, leaving a total of 208 admissions in 156 patients. One and two sites of infection were identified in 182 and 26 admissions respectively. The infection sites were pneumonia (n = 66), empyema (n = 6), urinary tract infection (n = 63), cutaneous (n = 42), intraabdominal sepsis (n = 56) and endocarditis (n = 1). A positive culture was present in 63%: MRSA (13%), Escherichia coli (11%), ESBL-producing Klebsiella pneumonia (10%), Klebsiella pneumonia (10%), and ESBL-producing Escherichia coli (9%). The inhospital mortality rate was 25%. The mean CP score was 9.0±3.0 and the mean MELD score was 16.5±7.3. In predicting in-hospital death, CP score has an area under the receiver operating characteristic curve (AUROC) of 0.933, with the optimum cut-off at ≥10, while for MELD score was 0.757, with the optimum cut-off ≥17. Four factors were significantly associated with in-hospital mortality on multivariate analysis: more than one site of infection (p = 0.003, OR 27.8, 95% CI = 3.2–242.7), pneumonia (p = 0.000, OR 13.2, 95% CI = 3.8–45.7), Child’s C status (p = 0.000, OR 29.4, 95% CI = 5.9–147.2), and MELD score≥17 (p = 0.024, OR 4.6, 95% CI = 1.2–17.3). As the number of factors increased, the in-hospital mortality rate increased: 0% with no factor, 6.9% with 1 factor, 23.7% with 2 factors, 87.5% with 3 factors, and 100% with 4 factors. By Kaplan Meier analysis, the survival of those with less than 3 factors was significantly better than those with 3 or more factors (p = 0.000). Conclusion: Septic cirrhotic patients with pneumonia, more than one site of infection, Child’s C cirrhosis and high MELD score had a high mortality risk, especially those with at least 3 of these 4 factors. 189 MILD HEPATIC ENCEPHALOPATHY (HE) ASSESSED BY THE REPEATABLE BATTERY FOR THE ASSESSMENT OF NEUROPSYCHOLOGICAL STATUS (RBANS) IS HIGHLY PREVALENT IN AMBULATORY PATIENTS WITH CIRRHOSIS C. Randolph1 , J. Bajaj2 , M.Y. Sheikh3 , R. Vemuru4 , G. Morelli5 , L.A. Balart6 , M. Chojkier7 , M.S. Harris8 , J.D. Bornstein8 , K. Mullen9 . 1 Loyola University Med. Center, Chicago, IL, 2 McGuire DVAMC, Richmond, VA, 3 University of California at San Francisco (UCSF), Fresno Medical Educational Program, Fresno, CA, 4 Permian Research Foundation, Odessa, TX, 5 University of Florida, Gainesville, FL, 6 Tulane University, New Orleans, LA, 7 Veterans Medical Center San Diego, 8 Ocera Therapeutics, San Diego, CA, 9 Case Western Reserve University, Cleveland, OH, USA E-mail: [email protected] Mild HE impairs quality of life and increases the risk of job loss and motor vehicle accidents. The prevalence of mild HE in cirrhotics ranges from 30% to 80%. The ISHEN recommends the RBANS as the tool of choice for detecting mild HE in the United States since it has well-established population norms and correlates with impaired daily functioning and job loss. We are using RBANS to qualify subjects for the ASTUTE trial, a phase 2B study assessing the potential
Journal of Hepatology 2010 vol. 52 | S59–S182
S81
POSTERS for AST-120 (spherical carbon adsorbent) to treat patients with mild HE, and data is now available on the prevalence of neurocognitive dysfunction by RBANS for the first 206 screened subjects. Methods: Trained RBANS raters used the RBANS to determine eligibility for participation in the trial. Subjects with cirrhosis between 18–70 years old and with MELD scores ≤25 were eligible for inclusion provided they did not have a TIPS or surgical shunt, had not had an episode of overt HE in the previous 3 months or taken lactulose, rifaximin or neomycin in the previous 7 days. Subjects were eligible for randomization if they scored below the 10th percentile on the RBANS total scale score, after adjusting for age and education. Results: Of 206 subjects screened to date, 61% were male, the mean age was 55.4 years and hepatitis C was the most common underlying liver disease. 82% of all screened subjects were high school graduates and 55% had attended college. Of the 206, 115 (56%) subjects were eligible for randomization based on a qualifying RBANS score. Qualifying rate was similar regardless of education level – 54% of college attendees, 57% of high school grads and 59% of high school dropouts qualified. RBANS total score could not be predicted based on the subject’s platelet count, bilirubin, MELD score or known history of esophageal varices. Conclusions: Using the RBANS, we found a neurocognitive impairment rate of 56% in a population of well-compensated cirrhotics, suggesting that mild HE is highly prevalent in these patients. Mild HE was not predicted by age, education, MELD score or indicators of portal hypertension. 190 PREVALENCE OF DIASTOLIC DYSFUNCTION AND OTHER SUPPORTIVE CRITERIA OF CIRRHOTIC CARDIOMYOPATHY H. Ribeiro1 , J. Presa2 , R. Margato1 , S. Carvalho1 , C. Ferreira1 , P. Mateus1 , A. Ferreira1 , J.I. Moreira1 . 1 Cardiology, 2 Internal Medicine, CHTMAD, Vila Real, Portugal E-mail: [email protected] Introduction: Cirrhotic cardiomyopathy (CC) is the term used to describe a constellation of features indicative of abnormal heart structure and function in patients with liver cirrhosis (LC). These include systolic and diastolic dysfunction, electrophysiological changes, and macroscopic and microscopic structural changes. Purpose: Our aim was to assess the prevalence of diastolic dysfunction and other supportive criteria of CC, in a cohort of patients with LC. Methods: 37 patients with LC (75.7% male; mean age 62.3±9.3 years old; 64.9% with alcoholic cirrhosis and 35.1% with cirrhosis of nonalcoholic etiology; MELD score >15 in 18.9% and MELD <15 in 18.1%) were enrolled, after excluding those with known cardiovascular risk factors (hypertension, diabetes, dislipidemia, tabagism or obesity), heart or pulmonary diseases. We collect data from patients with LC who as part of their routine care realized blood tests (including natriuretic peptide – BNP and troponin I levels), an electrocardiogram and an echocardiogram. Left ventricular diastolic function was studied by Pulse Wave and Pulsed Doppler Tissue imaging (DTI) and the diagnosis of diastolic dysfunction (DD) was made as defined by the recent consensus statement on the diagnosis of heart failure with normal left ventricular ejection fraction by the Heart Failure and Echocardiography Associations of the European Society of Cardiology. Results: Brain natriuretic peptide (BNP > 100 pg/ml) was elevated in 48.4% and troponin I (TnI >0.1 pg/ml) levels was increased in 5.4% LC patients. Prolonged Q-Tc interval (Q-Tc >440 mseg) was observed in 18.9%. Left atrium enlargement (LA volume >20 cm2) was registered in 83.8% and increased left ventricle end diastolic diameter (LVEDd >53 mm) and left ventricle end systolic diameter (LVESd >32 mm) was present in 21.6% and 27.0% respectively. S82
Increased left ventricular myocardial mass indice (LVMMI >95 g/m2 in Women and >115g/m2 in men) was identified in 67.6%. None of the patients had depressed left ventricular ejection fraction (LVEF <50%). DD was detected in 24.3% of the studied patients. Conclusions: This study showed that DD and other supportive criteria of cirrhotic cardiomyopathy are not rare in LC patients. Left cavities enlargement, ventricular hypertrophy, elevated BNP, prolonged Q-Tc, DD and conserved LVEF are the main abnormalities suggestive of CC found in patients with LC. 191 CLINICAL EFFICACY OF TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNT CREATED WITH COVERED STENTS WITH DIFFERENT DIAMETERS: RESULTS OF A RANDOMIZED CONTROLLED TRIAL O. Riggio1 , L. Ridola1 , S. Angeloni1 , F. Cerini1 , F. Fanelli2 , F.M. Salvatori2 , M. Merli1 . 1 Clinical Medicine, 2 Radiology, Sapienza University of Rome, Rome, Italy E-mail: [email protected] Introduction and Aim: The incidence of post-TIPS hepatic encephalopathy (HE) could be reduced by using stents with a small diameter, since post-TIPS HE showed to be related to the amount of blood shunted. Unfortunately, no pharmacological treatment proved to be able to prevent the incidence of post-TIPS HE. Aim of this study was to compare the incidence of HE and the clinical efficacy of TIPS created with 8- or 10-mm PTFE-covered stents. Patients and Methods: Consecutive cirrhotics submitted to TIPS for variceal bleeding or refractory ascites were randomized to receive a 8-mm or 10-mm covered stent. As recommended by our Ethical Committee, the trial was stopped after inclusion of 45 patients. Results: 22 patients were allocated to TIPS implantation with a 8 mm covered stent and 23 patients with a 10 mm covered stent. The two groups were comparable for age, sex, etiology, and psychometric performance. After TIPS placement, the portosystemic pressure gradient significantly decreased in both groups (from 21.3±4.9 to 8.9±2.7 mmHg in the 8-mm stent group and from 22.1±7.1 to 6.9±4.1 mmHg in the 10-mm stent group). The portosystemic pressure after TIPS tended to be lower in the 10mm than in the 8-mm stent group. The probability of remaining free of HE was similar in both groups (log-rank test, p = 0.48) (Figure 1). After TIPS, venous ammonia levels increased significantly only in the 10-mm stent group; psychometric performance was unmodified in both groups. The probability of remaining free of recurrence/persistence of complications due to portal hypertension was significantly higher in the 10-mm than in the 8-mm stent group: 82.9% versus 41.9% at one year; log-rank test, p = 0.002 (Figure 2). Cumulative survival rate was similar in both groups.
Figure 1.
Journal of Hepatology 2010 vol. 52 | S59–S182