- Email: [email protected]
709 FACTORS INFLUENCING THE ACCURACY OF LIVER STIFFNESS MEASUREMENTS BY TRANSIENT ELASTOGRAPHY (TE) AND BY ACOUSTIC RADIATION FORCE IMPULSE (ARFI) ELASTOGRAPHY FOR LIVER FIBROSIS EVALUATION
POSTERS were performed with a FibroScan device (M probe – Echosens, France); a median value was calculated and the results were expressed in kilopascals (kPa). Failure of TE measurements was defined if no valid measurement was obtained after at least 10 shots; and unreliable if: fewer than 10 valid shots obtained; success rate (SR) <60% and/or interquartile range interval (IQR) ≥30%. We analyzed the factors associated with failed and unreliable measurements. Results: Failed and unreliable LSMs were observed in 29.2% (2404/8218) cases. In univariant analysis, the following risk factors were associated with failed and unreliable measurements: age>50 years (OR 2.04; 95% CI 1.84–2.26, p < 0.0001); female gender (OR 1.32; 95% CI 1.20–1.45, p = 0.0001); BMI>27.7kg/m2 (OR 2.89, 95% CI 2.62–3.19, p < 0.0001); weight>77kg (OR 2.17; 95% CI 1.97–2.40, p < 0.0001); and height <162cm (OR 1.26; 95% CI 1.14–1.40, p < 0.0001). In multivariate analysis all factors mentioned above were independently associated with the risk of failed and unreliable measurements. If all negative predictive factors were present (woman>50 years, BMI>27.7kg/m2 , heavier than 77 kg and shorter than 162 cm) the rate of failed and unreliable measurements was 58.5% (185/316). In obese patients (BMI>30 kg/m2 ), the rate of failed and unreliable measurements reached almost 50% (969/1956 – 49.5%). If all positive factors were present (men <50 years, BMI ≤27.7 kg/m2 , lighter than 77 kg and taller than 162cm) the rate of failed and unreliable measurements was 11.6% (79/676). Conclusion: Failed and unreliable LSMs were observed in 29.1% of the patients. Female gender, older age, higher BMI, higher weight and lower height were significantly associated with failed and unreliable LSMs. This study underlines the need to use the XL probe in overweight patients to increase the rate of valid measurements in this type of patients. 709 FACTORS INFLUENCING THE ACCURACY OF LIVER STIFFNESS MEASUREMENTS BY TRANSIENT ELASTOGRAPHY (TE) AND BY ACOUSTIC RADIATION FORCE IMPULSE (ARFI) ELASTOGRAPHY FOR LIVER FIBROSIS EVALUATION S. Bota1 , I. Sporea1 , M. Peck-Radosavljevic2 , R. Sirli1 , H. Saito3 , H. Ebinuma3 , M. Lupsor4 , R. Badea4 , M. Friedrich-Rust5 , C. Sarrazin5 , F. Piscaglia6 , S. Marinelli6 , P. Salzl2 , A. Popescu1 , M. Danila1 . 1 Gastroenterology and Hepatology, ‘Victor Babes’ University of Medicine and Pharmacy, Timisoara, Romania; 2 Internal Medicine III, Div. of Gastroenterology and Hepatology, Medical University, Vienna, Austria; 3 Internal Medicine, School of Medicine, Keio University, Tokyo, Japan; 4 IIIrd Medical Clinic, University of Medicine and Pharmacy, Cluj-Napoca, Romania; 5 Div. Internal Medicine, Dept. Clinical Medicine, Department of Internal Medicine 1, J.W. Goethe University, Frankfurt/Main, Germany; 6 Div. Internal Medicine, Dept. Clinical Medicine, University and General Hospital S. Orsola-Malpighi, Bologna, Italy E-mail: [email protected] Aim: To identify the factors associated with discordance between liver stiffness values assessed by TE and ARFI elastography and histological fibrosis. Material and Methods: Our retrospective international study included 512patients from 6 centres, with various etiologies of chronic hepatopathies. We performed liver biopsy (LB), ARFI and TE measurements. A discordance of at least two stages of fibrosis (METAVIR) between TE and ARFI results and LB was defined as significant. The performance of TE was assessed using the liver stiffness cut-off values (kPa) (Tsochatzis-J Hepatol. 2011): F1–6; F2–7.2; F3–9.6; F4–14.5, and for ARFI the cut-offs (m/s) (FriedrichRust-J Viral Hepat. 2012): F1–1.19; F2–1.34; F3–1.55; F4–1.80. S288
Table 1. Transient Elastography Parameter
Non-discordant patients (73.2%)
Discordant patients (26.8%)
p
Age (years) Gender:female BMI (kg/m2 ) European patients Asian patients ALT (×ULN) Fibrosis: F0–2 Length of LB specimen (cm)
53 (19–77) n = 150 (47.4%) 25.6±3.8 n = 280 (80.7%) n = 94 (19.3%) 1.3 (0.2–22) n = 182 (52.4%) 3 (0.8–6)
52 (21–78) n = 69 (54.4%) 25.2±3.8 n = 90 (70.8%) n = 37 (29.2%) 1.3 (0.3–26) n = 94 (74.1%) 2.5 (0.7–5)
0.97 0.21 0.47 0.02 0.02 0.41 <0.0001 0.16
ULN, upper limit of normal.
Table 2. ARFI Elastography
Age (years) Gender: female BMI (kg/m2 ) European patients Asian patients ALT (xULN) Fibrosis: F0–2 Length of LB specimen (cm)
Non-discordant patients (74.4%)
Discordant patients (25.6%)
p
53 (19–77) n = 152 (43.1%) 25.2±3.7 n = 256 (72.5%) n = 97 (27.5%) 1.3 (0.2–22) n = 190 (53.8%) 2.5 (0.8–6)
53 (19–78) n = 67 (55.4%) 26.1±3.9 n = 87 (71.9%) n = 34 (28.1%) 1.3 (0.3–26) n = 86 (71.1%) 2.5 (0.7–6)
0.77 0.02 0.08 0.99 0.99 0.96 0.001 0.20
Results: In 92.5% of patients reliable measurements were obtained both by elastographic methods. Discordance with biopsy of at least 2 stages of fibrosis was observed in a similar number of cases in TE and ARFI: 26.8% vs. 25.6%, p = 0.78. The factors associated with discordances are presented in tables. Conclusions: Mild stages of fibrosis were associated with discordance in both TE and ARFI elastography. For TE the Asian race and for ARFI female gender were also associated with discordances. 710 TIMP-1 IN PATIENTS WITH CIRRHOSIS: RELATION TO LIVER DYSFUNCTION, PORTAL HYPERTENSION, AND HAEMODYNAMIC CHANGES 3 T.M. Busk1 , F. Bendtsen2 , V. Jensen3 , H.J. Nielsen4 , N. Brunner ¨ , 1 1 S. Møller . Centre of Functional Imaging and Research, Department of Clinical Physiology and Nuclear Medicine, 2 Gastro Unit, Medical Division, Copenhagen University Hospital Hvidovre, Hvidovre, 3 Section of Pathobiology, Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, 4 Department of Surgical Gastroenterology, Copenhagen University Hospital Hvidovre, Hvidovre, Denmark E-mail: [email protected] Background: Patients with cirrhosis and portal hypertension often develop complications relating to haemodynamic changes. The diagnosis and consequences hereof are often based on invasive procedures, such as liver biopsy and measurements of the hepatic venous pressure gradient. Thus, there is obviously a need to identify non-invasive markers of the severity of liver cirrhosis. Matrix metalloproteinases and their specific inhibitors, tissue inhibitors of metalloproteinases (TIMPs) play a pivotal role in hepatic fibrogenesis and fibrolysis. Aim: To investigate plasma TIMP-1 in liver cirrhosis patients undergoing catheterization and relate the findings to the degree of liver dysfunction, portal hypertension and haemodynamic changes. Methods: We included 96 patients with verified cirrhosis (Child A/B/C: 31/33/32) and 15 matched controls without liver disease. All individuals underwent a liver vein catheterisation with haemodynamic assessment. TIMP-1 was determined in arterial and hepatic venous plasma using the in-house MAC-15 TIMP-1 ELISA, which measures total (complexed and uncomplexed TIMP-1) with high analytical performance.
Journal of Hepatology 2013 vol. 58 | S229–S407
Table 1. Transient Elastography Parameter
Non-discordant patients (73.2%)
Discordant patients (26.8%)
p
Age (years) Gender:female BMI (kg/m2 ) European patients Asian patients ALT (×ULN) Fibrosis: F0–2 Length of LB specimen (cm)
53 (19–77) n = 150 (47.4%) 25.6±3.8 n = 280 (80.7%) n = 94 (19.3%) 1.3 (0.2–22) n = 182 (52.4%) 3 (0.8–6)
52 (21–78) n = 69 (54.4%) 25.2±3.8 n = 90 (70.8%) n = 37 (29.2%) 1.3 (0.3–26) n = 94 (74.1%) 2.5 (0.7–5)
0.97 0.21 0.47 0.02 0.02 0.41 <0.0001 0.16
ULN, upper limit of normal.
Table 2. ARFI Elastography
Age (years) Gender: female BMI (kg/m2 ) European patients Asian patients ALT (xULN) Fibrosis: F0–2 Length of LB specimen (cm)
Non-discordant patients (74.4%)
Discordant patients (25.6%)
p
53 (19–77) n = 152 (43.1%) 25.2±3.7 n = 256 (72.5%) n = 97 (27.5%) 1.3 (0.2–22) n = 190 (53.8%) 2.5 (0.8–6)
53 (19–78) n = 67 (55.4%) 26.1±3.9 n = 87 (71.9%) n = 34 (28.1%) 1.3 (0.3–26) n = 86 (71.1%) 2.5 (0.7–6)
0.77 0.02 0.08 0.99 0.99 0.96 0.001 0.20
Results: In 92.5% of patients reliable measurements were obtained both by elastographic methods. Discordance with biopsy of at least 2 stages of fibrosis was observed in a similar number of cases in TE and ARFI: 26.8% vs. 25.6%, p = 0.78. The factors associated with discordances are presented in tables. Conclusions: Mild stages of fibrosis were associated with discordance in both TE and ARFI elastography. For TE the Asian race and for ARFI female gender were also associated with discordances. 710 TIMP-1 IN PATIENTS WITH CIRRHOSIS: RELATION TO LIVER DYSFUNCTION, PORTAL HYPERTENSION, AND HAEMODYNAMIC CHANGES 3 T.M. Busk1 , F. Bendtsen2 , V. Jensen3 , H.J. Nielsen4 , N. Brunner ¨ , 1 1 S. Møller . Centre of Functional Imaging and Research, Department of Clinical Physiology and Nuclear Medicine, 2 Gastro Unit, Medical Division, Copenhagen University Hospital Hvidovre, Hvidovre, 3 Section of Pathobiology, Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, 4 Department of Surgical Gastroenterology, Copenhagen University Hospital Hvidovre, Hvidovre, Denmark E-mail: [email protected] Background: Patients with cirrhosis and portal hypertension often develop complications relating to haemodynamic changes. The diagnosis and consequences hereof are often based on invasive procedures, such as liver biopsy and measurements of the hepatic venous pressure gradient. Thus, there is obviously a need to identify non-invasive markers of the severity of liver cirrhosis. Matrix metalloproteinases and their specific inhibitors, tissue inhibitors of metalloproteinases (TIMPs) play a pivotal role in hepatic fibrogenesis and fibrolysis. Aim: To investigate plasma TIMP-1 in liver cirrhosis patients undergoing catheterization and relate the findings to the degree of liver dysfunction, portal hypertension and haemodynamic changes. Methods: We included 96 patients with verified cirrhosis (Child A/B/C: 31/33/32) and 15 matched controls without liver disease. All individuals underwent a liver vein catheterisation with haemodynamic assessment. TIMP-1 was determined in arterial and hepatic venous plasma using the in-house MAC-15 TIMP-1 ELISA, which measures total (complexed and uncomplexed TIMP-1) with high analytical performance.
Journal of Hepatology 2013 vol. 58 | S229–S407