Ultrasound in Med. & Biol., Vol. -, No. -, pp. 1–8, 2015 Copyright Ó 2015 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$ - see front matter
http://dx.doi.org/10.1016/j.ultrasmedbio.2015.02.005
d
Original Contribution USING ULTRASONIC TRANSIENT ELASTOMETRY (FIBROSCAN) TO PREDICT ESOPHAGEAL VARICES IN PATIENTS WITH VIRAL LIVER CIRRHOSIS ZHONGWEI HU,* YUYUAN LI,y CHUO LI,* CHUNMING HUANG,* ZHITAO OU,* JIAWEI GUO,* HONGBIN LUO,* and XIAOPING TANGz * Department of Internal Medicine, Guangzhou No. 8 People’s Hospital, Guangzhou, China; y Department of Gastroenterology Hepatology, Guangzhou First People’s Hospital, Guangzhou Medical University, Guangzhou, China; and z Institute for Infectious Diseases, Guangzhou No. 8 People’s Hospital, Guangzhou, China (Received 17 August 2014; revised 10 February 2015; in final form 13 February 2015)
Abstract—The correlation between liver stiffness (LS), measured by ultrasonic transient elastometry (FibroScan), and the presence and severity of esophageal varices (EV) in patients with viral cirrhosis of the liver has not been well documented to date. The study described here investigated the value of using FibroScan to predict EV. Patients with cirrhosis (200 patients: 167 cases caused by hepatitis B virus and 33 cases caused by hepatitis C virus) underwent both upper gastrointestinal endoscopy and FibroScan. Demographic, clinical, biochemical and endoscopic data and FibroScan-obtained LS parameters were collected. The mean LS value in patients with EV (33.2 kPa) was significantly higher than the mean LS value in patients without EV (18.6 kPa) (p , 0.05). The mean LS value in patients with grade 2 and 3 EV (38.3 kPa) was significantly higher than that in patients with grade 1 EV (24.8 kPa) (p , 0.05). Overall, FibroScan was 86.4% sensitive and 72.2% specific in predicting the presence of EV, with an area under the receiver operating characteristic curve (AUROC) of 0.84. The sensitivity and specificity for the patients with grade 2 or 3 EV were 84% and 73% (AUROC 5 0.86). When FibroScan was combined with platelet count, the overall sensitivity and specificity of prediction increased to 84% and 80% (AUROC 5 0.88), respectively, and 84% and 75% (AUROC 5 0.89), respectively, in patients with grade 2 and 3 EV. FibroScan alone or combined with platelet count might predict the presence and severity of EV in patients with hepatitis B or C-related viral cirrhosis. (E-mail: [email protected]) Ó 2015 World Federation for Ultrasound in Medicine & Biology. Key Words: Esophageal varices, Liver cirrhosis, FibroScan, Transient elastography.
(HCV) present with cirrhosis at their first clinical examination. As many as 20%–30% of patients who do not present with cirrhosis at that initial visit will develop cirrhosis within at least a decade (Benvegn u et al. 2004; Di Marco et al. 1999; Ikeda et al. 1998; Niederau et al. 1998). Major complications of cirrhosis include liver failure, portal hypertension (PHT), hepatorenal syndrome and esophageal varices (EV), all of which are poor prognostic indicators (Gines et al. 2004). Up to 90% of patients with cirrhosis develop EV, which may bleed. The incidence of bleeding EV is around 5% in patients with small EV (SEV) and up to 15% in patients with large EV (LEV) (Jensen 2002). Mortality per bleeding episode is around 10%–20% (Carbonell et al. 2004). Therefore, screening for EV in patients with cirrhosis is strongly recommended across guidelines and consensus statements (de Franchis 2005). Accordingly, it is generally recommended that patients with cirrhosis undergo active surveillance for
INTRODUCTION Progressive hepatic fibrosis caused by the accumulation of collagen in the extracellular matrix is characteristic of almost all chronic liver diseases (Bataller and Brenner 2005). Liver cirrhosis is the final phase of the hepatic fibrosis process, combining extensive fibrosis and regenerative nodules. In the East, chronic infection with hepatotropic viruses is a predominant problem (Margolis et al. 1991; Wang et al. 2014). Approximately 10%–20% of patients infected with chronic hepatitis B virus (HBV) or hepatitis C virus
Address correspondence to: Xiaoping Tang, Institute for Infectious Diseases, Guangzhou No.8 People’s Hospital, No. 627 Dongfengdong Road Yuexiu District, Guangzhou 510060, China. E-mail: [email protected] Conflicts of Interest: The authors declare no conflicts of interest, real or perceived, financial or non-financial. 1
2
Ultrasound in Medicine and Biology
early detection of EV via upper gastrointestinal (GI) endoscopy; however, endoscopic examination can be considered an unnecessary burden for both patients and doctors. Approximately half of patients with cirrhosis will not develop EV in the 10 y after the initial cirrhosis diagnosis (Addley et al. 2012). A common consequence of virus-related chronic liver disease is PHT, which results in EV. The hepatic venous pressure gradient (HVPG) is considered the gold standard for assessing PHT in patients with cirrhosis; however, assessments using this parameter are invasive and not routinely performed (Thabut et al. 2011). More non-invasive methods have been proposed as alternatives to endoscopy for EV screening. Measurement of liver stiffness (LS) with ultrasonic transient elastometry has been recognized as a rapid, non-invasive technique in which the measurement correlates well with the underlying stage of fibrosis. A particularly strong correlation has been observed between LS measured by FibroScan (Echosens, Paris, France), the first tool introduced to assess fibrosis through LS (Echosens, Paris, France), and HVPG. Thus, FibroScan has potential to be used for the non-invasive evaluation of EV (Castera et al. 2012). Although the value of FibroScan in assessment of the degree of LS in patients with chronic HCV infection has been well documented, few studies have been conducted on the use of FibroScan for patients with chronic HBV infection (Castera et al. 2005; Cui et al. 2013; Dolmazashvili et al. 2008; Marcellin et al. 2009; Ziol et al. 2005). Moreover, few studies have focused on the relationship between LS and the presence and severity of EV, especially in patients with HBV infection. This is concerning, as the most common cause of cirrhosis with high mortality in China is HBV infection (Calvaruso et al. 2013; Castera et al. 2009; Kazemi et al. 2006). In addition, the results of different studies still lack sensitivity and specificity (Calvaruso et al. 2013). The aim of this study was to investigate the relationship between degree of LS and EV and to assess the use of FibroScan to predict the presence and severity of EV.
METHODS Participants Between July 2007 and October 2012, 200 consecutive inpatients from Guangzhou No. 8 People’s Hospital, China, were initially included. Patients were older than 18 y, included both sexes, had chronic HBV- or HCVrelated liver cirrhosis, had undergone upper GI endoscopy, and agreed to a FibroScan examination. Cirrhosis of the liver was diagnosed using the criteria from the Chinese Association of Liver Diseases practice guidelines, which, in turn, are adapted from the American
Volume -, Number -, 2015
Association for the Study of Liver Diseases guidelines (Runyon 2013). The diagnosis was based mainly on abdominal ultrasonic findings combined with medical history, clinical symptoms and laboratory results. If the diagnosis was unclear from these parameters, liver biopsy was considered a secondary diagnostic measure. In this study, 22 patients underwent liver biopsy. Serologic detection of hepatitis B surface antigens and hepatitis C antibodies was used to diagnose HBV or HCV infection. Patients were excluded from this study if they were infected with both HBV and HCV, reported heavy alcohol intake (defined as .40 g of alcohol/d for $5 y), or had comorbid chronic liver diseases, aspartate aminotransferase (AST)/alanine aminotransferase (ALT) levels greater than three times the normal upper limit, hepatocellular carcinoma, acute liver failure or decompensated congestive heart failure. The clinical parameters collected from each patient were: age, sex, history of ascites, bleeding varices, spontaneous bacterial peritonitis, hepatic encephalopathy, organic kidney failure, or hepatorenal syndrome. Collected biological parameters included a complete blood count and levels of AST, ALT, alkaline phosphatases, total bilirubin, albumin, gamma globulins, prothrombin time (PT), creatinine, urea and a-fetoprotein. The research protocol was approved by the ethics committee of Guangzhou No. 8 People’s Hospital. Written informed consent was obtained from each participant. The study was registered in the Chinese Clinical Trial Registry. Endoscopy All patients underwent upper GI endoscopy (Fujinon EG-590 WR, Japan). Varices were graded according to their size and appearance by an endoscopic physician experienced in the following grading scale (Chinese Medical Association 2003): grade 0 5 no varices; grade 1 5 small, straight EV; grade 2 5 enlarged, tortuous EV occupying less than one-third of the lumen; and grade 3 5 large, coil-shaped EV occupying more than onethird of the lumen. Liver stiffness measurement FibroScan was used to measure LS during the same week the patients underwent endoscopic examinations. An experienced doctor blinded to the patients’ clinical data and endoscopic results performed all measurements. The FibroScan device was equipped with a probe that included a vibrating component. The vibrator transmitted vibrations to the tissues, inducing an elastic shear wave that propagated through the tissue. The propagation was followed by pulse-echo ultrasound acquisitions. Their velocity, directly related to tissue stiffness, was then measured (Friedrich-Rust et al. 2010; Sandrin et al.
Prediction of esophageal varices with FibroScan d Z. HU et al.
2003). The measurement depth was between 25 and 65 mm, and the results were expressed in kilopascals (kPa). Faster shear wave speeds were correlated with higher LS scores and indicated a stiffer liver. As suggested by the manufacturer, 10 successful acquisitions were performed for each patient. To determine how reliable the transient elastometry results were, the following criteria were used: very reliable 5 interquartile range/median ratio (IQR/ M) # 0.10; reliable 5 0.10 , IQR/M # 0.30 or IQR/ M . 0.30 with LS median , 7.1 kPa; and poorly reliable 5 IQR/M . 0.30 with LS median $ 7.1 kPa (Boursier et al. 2013). The poorly reliable results were excluded from the final analysis. Statistical analysis Data were analyzed using SPSS Version 10.0 (SPSS, Chicago, IL, USA). Continuous data were expressed as means 6 standard deviations (SD) and examined with a one-way analysis of variance. The Kruskal–Wallis nonparametric analysis of variance and Bonferroni corrections were used to compare LS among the different EV stages. Mann–Whitney U-tests were performed to compare LS among HBV and HCV subgroups. Categorical variables were expressed as percentages and analyzed with the c2 test. Multiple backward regression was used to assess the association between EV grade (dependent variable) and independent variables, namely, demographic (age, sex) and clinical (chronic HBVor HCV, kidney failure, hepatorenal syndrome, history and biochemical criteria) characteristics. All independent variables were entered into the prediction model and removed in a backward stepwise procedure if their pvalues were .0.05. Thus, a predictive model containing all significant independent variables and dependent variables (EV grades) was established. The receiver operating characteristic (ROC) curve, a plot of sensitivity versus 1 2 specificity, was created for all possible cutoff
3
values. The optimal cutoff value was chosen when the sum of sensitivity and specificity was maximized. The Hanley and McNeil test in the SPSS software was performed for ROC curve analysis. Moreover, the Delong test was employed to compare ROC curves using MedCalc software (http://www.medcalc.org). A p-value , 0.05 was considered to indicate statistical significance. RESULTS Among 210 patients who initially met the inclusion criteria, 200 patients with complete data were included in this study’s final analysis. The 10 patients who were excluded had ‘‘poorly reliable’’ transient elastometry results. The mean age of patients without EV was 44.8 6 10.1 y, including 65 (32.5%) men and 25 (12.5%) women. The mean age of patients with EV was 45.39 6 10.4 y, including 76 (38%) men and 34 (17%) women. In patients without EV, HBV caused 75 (37.5%) cirrhosis cases, and HCV caused 15 (7.5%) cirrhosis cases. In patients with EV, HBV caused 92 (46%) cirrhosis cases, and HCV caused 18 (9%) cirrhosis cases. No significant differences in age, sex and HBV/ HCV incidence were observed in the three graded groups (grade 0, 1 and 2/3 EV) (p . 0.05) (Table 1). There were significant differences in history of ascites, hepatic encephalopathy, GI bleeding and mosaic pattern of the gastric mucosa between the three groups (p , 0.05). For the biochemical variables (Table 2), there were significant differences in albumin, platelet count, bilirubin, prothrombin time and urea between the three groups (p , 0.05). Of the 200 patients in the study, 90 (45%) did not have EV (V0), 41 (20.5%) had grade 1 EV (V1) and 69 (34.5%) had grade 2 or 3 EV (V2/3). The overall mean LS value in patients with EV (33.2 kPa, range: 10.6– 77.0 kPa) was significantly higher than that in patients without EV (18.6 kPa, range: 8.6–38.8 kPa) (V0–V1:
Table 1. Demographic, clinical and endoscopic data Patients with esophageal varices Variable
All patients (n 5 200)
Grade 0 (n 5 90)
Grade 1 (n 5 41)
Grades 2 and 3 (n 5 69)
p-Value
Age (y) Males Chronic hepatitis B Kidney failure Hepatorenal syndrome History of ascites History of leg edema Hepatic encephalopathy Gastrointestinal bleeding Spontaneous bacterial peritonitis Right pleural effusion Mosaic pattern in gastric mucosa
45.1 6 10.2 141 (71%) 167 (84%) 9 (5%) 1 (1%) 61 (31%) 43 (22%) 12 (6%) 21 (11%) 3 (%) 3 (2%) 122 (61%)
44.8 6 10.1 65 (72%) 75 (83%) 4 (4%) 0 22 (24%) 19 (21%) 2 (2%) 0 0 0 27 (30%)
45.3 6 10.4 28 (68%) 35 (85%) 2 (5%) 0 19 (46%) 13 (32%) 6 (15%) 9 (22%) 0 1 (2%) 29 (71%)
45.4 6 10.5 48 (70%) 57 (83%) 3 (4%) 1 (1%) 20 (29%) 11 (16%) 4 (6%) 12 (17%) 3 (%) 2 (3%) 66 (96%)
0.89 0.88 0.92 0.99 0.99 0.04 0.14 0.02 0.01 0.18 0.22 0.01
4
Ultrasound in Medicine and Biology
Volume -, Number -, 2015
Table 2. Biochemical characteristics Patients with EV Variable
All patients (n 5 200)
Patients without EV (grade 0) (n 5 90)
Grade 1 (n 5 41)
Grade 2/3 (n 5 69)
p-Value
Total bilirubin (mmol/L) Albumin (g/L) Gamma globulins (g/L) Alanine aminotransferase (IU) Aspartate aminotransferase (IU) Alkaline phosphatases (IU/L) Prothrombin time (%) Leukocytes (3 103/mm3) Polynuclear neutrophils (3 103/mm3) Hemoglobin (g/dL) Platelet count (103/mm3) Creatinine (mmol/L) Urea (mmol/L) Natremia (mmol/L) a-Fetoprotein (ng/mL)
39.8 6 26.7 33.5 6 7.2 16.9 6 7.8 63.8 6 24.3 70.2 6 37.6 142.3 6 87.6 60.9 6 16.9 5.1 6 2.3 3.0 6 2.1 12.3 6 2.3 124.1 6 70.9 79.0 6 64.5 5.95 6 5.35 137.0 6 3.44 9.4 6 9.0
32.6 6 23.2 35.6 6 7.4 15.5 6 7.6 61.0 6 21.1 72.0 6 52.7 151.4 6 96.2 68.4 6 18.5 5.3 6 2.5 3.2 6 2.4 12.5 6 2.3 167.8 6 75.8 77.1 6 66.8 5.29 6 5.13 138.5 6 3.69 9.2 6 9.0
40.3 6 19.1 34.2 6 6.8 17.9 6 7.1 63.4 6 19.5 56.5 6 13.6 140.8 6 83.3 57.7 6 14.4 5.1 6 2.3 3.0 6 2.2 12.6 6 2.1 107.3 6 40.7 82.3 6 63.9 6.34 6 5.66 135.6 6 3.11 9.4 6 8.8
49.1 6 31.8 30.3 6 6.1 18.2 6 8.5 67.9 6 30.0 76.0 6 26.2 131.4 6 79.1 53.0 6 16.1 4.9 6 2.0 2.8 6 1.5 11.8 6 2.4 77.3 6 36.1 79.5 6 61.9 6.56 6 5.45 135.8 6 3.32 9.6 6 9.2
0.04 0.04 0.55 0.71 0.03 0.10 0.03 0.82 0.36 0.49 0.01 0.68 0.04 0.93 0.84
EV 5 esophageal varices.
p , 0.001, and V0–V2/3: p , 0.001, respectively). There was a significant difference between the in LS values between groups V1 (24.8 kPa, 10.6–40.3 kPa) and V2/3 (38.3 kPa, 13.9–77.0 kPa) (p , 0.001). The mean LS values in V0, V1 and V2/3 patients with HBV-related cirrhosis were 18.8 (8.6–38.8), 25.2 (11.5–40.3) and 39.0 (15.9–77.0) kPa. The LS values in V0, V1 and V2/ 3 patients with HCV-related cirrhosis were 17.8 (10.7– 32.0), 22.2 (10.6–27.8) and 34.9 (13.9–66.4) kPa, respectively. No significant difference was detected between the HBV and HCV subgroups (p-values were 0.618, 0.306 and 0.218, respectively). Distribution of liver stiffness (FibroScan) for each grade (0–3) of EV was showed in Figure 2. With endoscopic findings as the ‘‘gold standard’’ for EV diagnosis, a ROC curve for the presence of EV as predicted by LS values was generated. The cutoff values are listed in Table 3. When the cutoff value was set at 20.3 kPa, the area under the ROC curve (AUROC) was 0.843, and FibroScan’s sensitivity and specificity for predicting the presence of EV were 86.4% and 72.2%, respectively (Fig. 1a, Table 3). The sensitivity and specificity for patients with grade 2 and 3 EV were 84.1% and 72.5%, respectively, with an AUROC of 0.855 (Fig. 1b, Table 3). The same AUROC results were obtained with the Hanley and McNeil test and Delong test. A multiple regression analysis for the predictors of EV grade (dependent variable) was conducted using demographic variables (age, sex); presence of HBV or HCV; kidney failure; hepatorenal syndrome; medical history (ascites, leg edema, hepatic encephalopathy, GI bleeding, spontaneous bacterial peritonitis, right pleural effusion, mosaic pattern in gastric mucosa); biochemical criteria (total bilirubin, albumin, gamma globulins, ALT, AST, alkaline phosphatase, prothrombin time, leuko-
cytes, polynuclear neutrophils, hemoglobin, platelet count, creatinine, urea, natremia, a-fetoprotein) and FibroScan value. Only the FibroScan value and platelet count had a confirmed predictive value for the presence of EV. The correlation coefficients for FibroScan value and platelet count were 0.048 (95% confidence interval: 0.039 to 0.057) and –0.005 (95% confidence interval: –0.007 to 0.004), respectively, and the constant was 0.581. ROC curves and data for predicting the presence of EV using platelet count and FibroScan scores are provided in Figure 1c and Table 3. Overall, combining FibroScan values with platelet count yielded a better AUROC (0.879) for diagnosis of EV. For diagnosis of EV grade $2, the AUROC increased to 0.886 (Fig. 1d, Table 3). The Hanley and McNeil test and Delong test yielded the same AUROC results. DISCUSSION Patients with cirrhosis are at risk for morbidity and mortality from GI bleeding through EV. Screening for EV, rather than treating patients after a crisis, ultimately conserves resources and time. Finding non-invasive methods to screen for EV directly caused by PHT is a recent research focus, however, and still poses some clinical challenges. About 40% of patients with Child– Pugh A and 60% of patients with ascites develop varices as a result of PHT, with an expected incidence of new varices of about 5% per year (Addley et al. 2012). Because PHT occurs as a complication of liver cirrhosis, the degree of PHT can be correlated with the severity of cirrhosis (Thabut et al. 2011); therefore, PHT might be a potential tool for evaluating the relationship between EV and cirrhosis. Unfortunately, the current gold standard for assessing the presence and severity of PHT,
Prediction of esophageal varices with FibroScan d Z. HU et al.
5
Table 3. Diagnostic performances of the FibroScan with respect to the presence of EV EV grade
Sensitivity (%)
FibroScan alone 2/3 84.1 1–3 86.4 FibroScan 1 platelet counts 2/3 84.1 1–3 83.6
Specificity (%)
Positive predictive value (%)
Negative predictive value (%)
AUROC (%)
Cutoff value (kPa)
72.5 72.2
71.7 79.2
90.8 81.3
85.5 (0.80–0.91) 84.3 (0.79–0.90)
25.55 20.25
74.8 80.0
63.7 83.6
89.9 80.0
88.6 (0.84–0.93) 87.9 (0.83–0.93)
1.24 1.02
AUROC 5 area under receiver operating characteristic curve; EV 5 esophageal varices.
HVPG measurement, is invasive and performed only in expert centers. Although a variety of blood tests for EV screening have been proposed, including albumin level and prothrombin time (Kim et al. 2012a; Yeh et al.
2003), these biochemical markers are more strongly associated with the severity of cirrhosis and not with the presence or absence of EV at a clinically useful level.
Fig. 1. ROC curves for diagnosis of EV: (a) Diagnosis of EV using FibroScan. (b) Diagnosis of grade $2 EV using FibroScan. (c) Diagnosis of EV using the FibroScan and platelet count. (d) Diagnosis of grade $2 EV using the FibroScan and platelet count. ROC 5 receiver operating characteristic; EV 5 esophageal varices.
6
Ultrasound in Medicine and Biology
Volume -, Number -, 2015
Fig. 2. Distribution of liver stiffness (FibroScan) for each grade (0–3) of EV. (a) Scatter plot of liver stiffness values (FibroScan) for each EV grade (0–3). (b) Error bar represents 95% confidence interval of mean liver stiffness value (FibroScan). (c) Box plot illustrates distribution of liver stiffness values (FibroScan) by EV grade (0–3). The top and bottom of the boxes are the first and third quartiles. The length of the box represents the interquartile range, within which are located 50% of the values. The lines through the middle of the boxes represent median values. Unfilled circles are data points that were either 1.5 3 IQR or more above the third quartile or 1.5 3 IQR or more below the first quartile. CI 5 confidence interval; EV 5 esophageal varices; IQR 5 interquartile range.
Studies using different ultrasonic methods to estimate liver fibrosis, including the use of transient elastometry (e.g., FibroScan), real-time tissue elastography and shear wave elastography (SWE) using acoustic radiation force impulse (ARFI), have been published. In patients with obesity-related liver disease, ARFI has been successfully used to distinguish fibrosis stages from low to high (90% sensitivity and specificity) (Palmeri et al. 2011). Another study reported that LS measurements obtained with ARFI had a similar correlation strength as fibrosis stage values obtained by transient elastometry methods in patients with chronic HBV and HCV (Sporea et al. 2012). On the other hand, a different study reported that FibroScan was a much better predictor of liver cirrhosis than the AST-to-platelet ratio index (APRI) and suggested real-time tissue elastography as a promising method for determining the stage of hepatic fibrosis (Tatsumi et al. 2008). Among transient elastometry methods, FibroScan was the first to be introduced and remains the most common tool in clinical practice (Cui et al. 2013). Fibroscan has been considered an accurate, reliable and reproducible method for assessing liver fibrosis through LS measurements (Fransen van de Putte et al. 2011; Jang et al. 2012; Kim et al. 2012a, 2012b). Some studies have reported that FibroScan is a good diagnostic tool for detecting PHT, a common complication of liver fibrosis (Bureau et al. 2008; Robic et al. 2011; Thabut et al. 2007). As PHT worsens, varices can form and worsen as well. LS values correlate with HVPG and are an excellent predictor of PHT in patients with compensated cirrhosis (Lemoine et al. 2008; Vizzutti et al. 2007). Thus, FibroScan values may be correlated with the degree of EV. To date, two studies have reported significant correlation
between FibroScan-measured LS values and the presence of LEVs (Nguyen-Khac et al. 2010; Thabut et al. 2006). In these studies, however, the etiology of cirrhosis was predominantly alcohol abuse. Liver cirrhosis caused by viral infection (viral hepatitis, especially hepatitis C, a major endemic disease) is a major cause of death in China (He et al. 2005). In the present study, patients with chronic HBV and HCV infections were emphasized; to our knowledge, this was a novel approach. In addition, the relationship between SEVs detected by endoscopy and liver cirrhosis evaluated by FibroScan remains obscure. It is, however, worth noting that although LS has been considered an excellent non-invasive parameter for assessing stages of liver cirrhosis, the presence of hepatic congestion is known to drastically increase LS independent of fibrosis stage (Millonig et al. 2009, 2010). For this reason, patients with decompensated congestive heart failure were excluded from this study. In this study, cutoff values for diagnosis of EV were different from values in previous studies. Other studies have reported cutoff values of 19, 18 and 48 kPa for diagnosing LEV (Bureau et al. 2008; Kazemi et al. 2006; Nguyen-Khac et al. 2010; Vizzutti et al. 2007). In our study, the cutoff value was 25.55 kPa for LEV and 20.25 kPa for any EV. The differences might be due to the varied study populations across trials, which included patients with chronic liver diseases of various etiologies. In the previous studies, patients were primarily or exclusively those with alcoholic cirrhosis or a chronic HCV infection (Bureau et al. 2008; Kazemi et al. 2006; Nguyen-Khac et al. 2010; Vizzutti et al. 2007). Other preliminary studies have suggested that some non-invasive biochemical markers for liver fibrosis could also predict the presence of EV. One study found that a
Prediction of esophageal varices with FibroScan d Z. HU et al.
platelet count .150,000/mm3 was associated with a negative predictive value of 99% for EV (Sanyal et al. 2006). Another study reported that the platelet count/ spleen diameter ratio (cutoff value 5 909) had a diagnostic accuracy of 86% for EV (Giannini et al. 2006). Platelet count, AST-to-ALT ratio (AAR), Lok index, AST-to-platelet ratio index (APRI), Forns index, Fib-4 and FibroIndex were analyzed in a multicenter study. The results indicated that the Lok index (cutoff value 5 1.5) could be used to exclude LEV with a 96% negative predictive value. A combination of the Lok index (cutoff 5 0.9) and Forns index (cutoff 5 8.5) predicted the presence of EV of any grade with an AUROC of 0.82 (Sebastiani et al. 2010). In contrast, our study found that FibroScan-obtained LS values alone were valuable, but not reliable enough to be used as a screening method for detecting the presence of EV in patients with liver cirrhosis. Combining FibroScan values and platelet counts might improve prediction of the presence of EV. In addition, the Hanley and McNeil test and Delong test yielded the same AUROC results, strengthening the reliability of the findings. The present study had some limitations. First, there may have been a lack of reliable and reproducible color ultrasound results. Abdominal color ultrasonography scans that were completed by several examiners were excluded because of a lack of normalcy and objectivity. The second limitation was the small number of patients with HCV, possibly limiting generalizability for this population. CONCLUSIONS In this prospective study, we found a useful FibroScan cutoff value for the non-invasive diagnosis of EV in patients with cirrhosis. In particular, when a subanalysis of patients with cirrhosis and severe EV (grade 2/3) was conducted, the predictive usefulness increased. FibroScan values or a combination of FibroScan values and platelet counts might be an acceptable reference marker for the prediction of severe EV in patients with cirrhosis when endoscopic evaluation is not indicated because of poor compliance and high cost. Acknowledgments—This study was funded by Guangdong Province Science and Technology Plan Project No. 2010 B031600008 and Natural Science Foundation of China Grant 81200903.
REFERENCES Addley J, Tham TC, Cash WJ. Use of portal pressure studies in the management of variceal haemorrhage. World J Gastrointest Endosc 2012;4:281–289. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115:209–218. Benvegn u L, Gios M, Boccato S, Alberti A. Natural history of compensated viral cirrhosis: A prospective study on the incidence and hierarchy of major complications. Gut 2004;53:744–749. Boursier J, Zarski JP, de Ledinghen V, Rousselet MC, Sturm N, Lebail B, Fouchard-Hubert I, Gallois Y, Oberti F, Bertrais S,
7
Cales P. Multicentric Group from ANRS/HC/EP23 FIBROSTAR Studies. Determination of reliability criteria for liver stiffness evaluation by transient elastography. Hepatology 2013;57: 1182–1191. Bureau C, Metivier S, Peron JM, Selves J, Robic MA, Gourraud PA, Rouquet O, Dupuis E, Alric L, Vinel JP. Transient elastography accurately predicts presence of significant portal hypertension in patients with chronic liver disease. Aliment Pharmacol Ther 2008;27: 1261–1268. Calvaruso V, Bronte F, Conte E, Simone F, Craxı A, Di Marco V. Modified spleen stiffness measurement by transient elastography is associated with presence of large oesophageal varices in patients with compensated hepatitis C virus cirrhosis. J Viral Hepat 2013;20: 867–874. Carbonell N, Pauwels A, Serfaty L, Fourdan O, Levy VG, Poupon R. Improved survival after variceal bleeding in patients with cirrhosis over the past two decades. Hepatology 2004;40:652–659. Castera L, Le Bail B, Roudot-Thoraval F, Bernard PH, Foucher J, Merrouche W, Couzigou P, de Ledinghen V. Early detection in routine clinical practice of cirrhosis and oesophageal varices in chronic hepatitis C: Comparison of transient elastography (FibroScan) with standard laboratory tests and non-invasive scores. J Hepatol 2009;50:59–68. Castera L, Pinzani M, Bosch J. Non-invasive evaluation of portal hypertension using transient elastography. J Hepatol 2012;56:696–703. Castera L, Vergniol J, Foucher J, Le Bail B, Chanteloup E, Haaser M, Darriet M, Couzigou P, De Ledinghen V. Prospective comparison of transient elastography, FibroTest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005;128:343–350. Chinese Medical Association. Trial scheme of diagnosing and treating gastroesophageal varices under endoscopy. Chin J Dig Endosc 2003;2004(21):149–151. Cui XW, Friedrich-Rust M, De Molo C, Ignee A, Schreiber-Dietrich D, Dietrich CF. Liver elastography, comments on EFSUMB elastography guidelines 2013. World J Gastroenterol 2013;19:6329–6347. De Franchis R. Evolving consensus in portal hypertension: Report of the Baveno IV consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 2005;43:167–176. Di Marco V, Lo Iacono O, Camma C, Vaccaro A, Giunta M, Martorana G, Fuschi P, Almasio PL, Craxı A. The long-term course of chronic hepatitis B. Hepatology 1999;30:257–264. Dolmazashvili E, Zhamutashvili M, Svanidze M, Nizharadze N, Abutidze A. FibroScan and FibroTest/FibroMax to assess liver fibrosis/cirrhosis in patients with chronic HBV and HCV infection in Georgia. Georgian Med News 2008;165:83–87. Fransen van de Putte D, Blom R, van Soest H, Mundt M, Verveer C, Arends J, de Knegt RE, Mauser-Bunschoten E, van Erpecum K. Impact of FibroScan on management of chronic viral hepatitis in clinical practice. Ann Hepatol 2011;10:469–476. Friedrich-Rust M, Rosenberg W, Parkes J, Herrmann E, Zeuzem S, Sarrazin C. Comparison of ELF, FibroTest and FibroScan for the non-invasive assessment of liver fibrosis. BMC Gastroenterol 2010;10:103. Giannini EG, Zaman A, Kreil A, Floreani A, Dulbecco P, Testa E, Sohaey R, Verhey P, Peck-Radosavljevic M, Mansi C, Savarino V, Testa R. Platelet count/spleen diameter ratio for the non-invasive diagnosis of esophageal varices: Results of a multicenter, prospective, validation study. Am J Gastroenterol 2006;101:2511–2519. Gines P, Cardenas A, Arroyo V, Rodes J. Management of cirrhosis and ascites. N Engl J Med 2004;350:1646–1654. He J, Gu D, Wu X, Reynolds K, Duan X, Yao C, Wang J, Chen CS, Chen J, Wildman RP, Klag MJ, Whelton PK. Major causes of death among men and women in China. N Engl J Med 2005;353: 1124–1134. Ikeda K, Saitoh S, Suzuki Y, Kobayashi M, Tsubota A, Koida I, Arase Y, Fukuda M, Chayama K, Murashima N, Kumada H. Disease progression and hepatocellular carcinogenesis in patients with chronic viral hepatitis: A prospective observation of 2215 patients. J Hepatol 1998;28:930–938. Jang HW, Kim SU, Park JY, Ahn SH, Han KH, Chon CY, Park YN, Choi EH, Kim DY. How many valid measurements are necessary
8
Ultrasound in Medicine and Biology
to assess liver fibrosis using FibroScan in patients with chronic viral hepatitis? An analysis of patients with at least 10 valid measurements. Yonsei Med J 2012;53:337–345. Jensen DM. Endoscopic screening for varices in cirrhosis: Findings, implications, and outcomes. Gastroenterology 2002;122:1620–1630. Kazemi F, Kettaneh A, N’kontchou G, Pinto E, Ganne-Carrie N, Trinchet JC, Beaugrand M. Liver stiffness measurement selects patients with cirrhosis at risk of bearing large oesophageal varices. J Hepatol 2006;45:230–235. Kim BK, Kim HS, Park JY, Ahn SH, Chon CY, Park YN, Han KH, Kim SU. Prospective validation of ELF test in comparison with FibroScan and FibroTest to predict liver fibrosis in Asian patients with chronic hepatitis B. PLoS One 2012a;7:e41964. Kim SU, Lee JH, Kim DY, Ahn SH, Jung KS, Choi EH, Park YN, Han KH, Chon CY, Park JY. Prediction of liver-related events using FibroScan in chronic hepatitis B patients showing advanced liver fibrosis. PLoS One 2012b;7:e36676. Lemoine M, Katsahian S, Ziol M, Nahon P, Ganne-Carrie N, Kazemi F, Grando-Lemaire V, Trinchet JC, Beaugrand M. Liver stiffness measurement as a predictive tool of clinically significant portal hypertension in patients with compensated hepatitis C virus or alcohol-related cirrhosis. Aliment Pharmacol Ther 2008;28: 1102–1110. Marcellin P, Ziol M, Bedossa P, Douvin C, Poupon R, de Ledinghen V, Beaugrand M. Non-invasive assessment of liver fibrosis by stiffness measurement in patients with chronic hepatitis B. Liver Int 2009;29: 242–247. Margolis HS, Alter MJ, Hadler SC. Hepatitis B: Evolving epidemiology and implications for control. Semin Liver Dis 1991;11:84–92. Millonig G, Friedrich S, Adolf S, Fonouni H, Golriz M, Mehrabi A, Buchler M. Liver stiffness is directly influenced by central venous pressure. Ultrasound Med Biol 2010;52:206–210. Millonig G, Friedrich S, Adolf S, Poschl G, Stiefel P, Seitz HK, Muelle S. Increased Venous pressure increases liver stiffness (FibroScan) independent of liver fibrosis. J Hepatol 2009;50:204. Nguyen-Khac E, Saint-Leger P, Tramier B, Coevoet H, Capron D, Dupas JL. Noninvasive diagnosis of large esophageal varices by FibroScan: Strong influence of the cirrhosis etiology. Alcohol Clin Exp Res 2010;34:1146–1153. Niederau C, Lange S, Heintges T, Erhardt A, Buschkamp M, H€urter D, Nawrocki M, Kruska L, Hensel F, Petry W, H€aussinger D. Prognosis of chronic hepatitis C: Results of a large, prospective cohort study. Hepatology 1998;28:1687–1695. Palmeri ML, Wang MH, Rouze NC, Abdelmalek MF, Guy CD, Moser B, Diehl AM, Nightingale KR. Noninvasive evaluation of hepatic fibrosis using acoustic radiation force-based shear stiffness in patients with nonalcoholic fatty liver disease. J Hepatol 2011;55: 666–672. Robic MA, Procopet B, Metivier S, Peron JM, Selves J, Vinel JP, Bureau C. Liver stiffness accurately predicts portal hypertension related complications in patients with chronic liver disease: A prospective study. J Hepatol 2011;55:1017–1024. Runyon BA. Introduction to the revised American Association for the Study of Liver Diseases Practice Guideline management of adult pa-
Volume -, Number -, 2015 tients with ascites due to cirrhosis 2012. Hepatology 2013;57: 1651–1653. Sandrin L, Fourquet B, Hasquenoph JM, Yon S, Fournier C, Mal F, Christidis C, Ziol M, Poulet B, Kazemi F, Beaugrand M, Palau R. Transient elastography: A new non-invasive method for assessment of hepatic fibrosis. Ultrasound Med Biol 2003;29: 1705–1713. Sanyal AJ, Fontana RJ, Di Bisceglie AM, Everhart JE, Doherty MC, Everson GT, Donovan JA, Malet PF, Mehta S, Sheikh MY, Reid AE, Ghany MG, Gretch DR. The prevalence and risk factors associated with esophageal varices in patients with hepatitis C and advanced fibrosis. Gastrointest Endosc 2006;64:855–864. Sebastiani G, Tempesta D, Fattovich G, Castera L, Halfon P, Bourliere M, Noventa F, Angeli P, Saggioro A, Alberti A. Prediction of oesophageal varices in hepatic cirrhosis by simple serum noninvasive markers: Results of a multicenter, large-scale study. J Hepatol 2010;53:630–638. Sporea I, Sirli R, Bota S, Popescu A, Sendroiu M, Jurchis A. Comparative study concerning the value of acoustic radiation force impulse elastography (ARFI) in comparison with transient elastography (TE) for the assessment of liver fibrosis in patients with chronic hepatitis B and C. Ultrasound Med Biol 2012;38:1310–1316. Tatsumi C, Kudo M, Ueshima K, Kitai S, Takahashi S, Inoue T, Minami Y, Chung H, Maekawa K, Fujimoto K, Akiko T, Takeshi M. Noninvasive evaluation of hepatic fibrosis using serum fibrotic markers, transient elastography (FibroScan) and real-time tissue elastography. Intervirology 2008;51(Suppl 1):27–33. Thabut D, Imbert-Bismut F, Cazals-Hatem D, Messous D, Muntenau M, Valla DC, Moreau R, Poynard T, Lebrec D. Relationship between the FibroTest and portal hypertension in patients with liver disease. Aliment Pharmacol Ther 2007;26:359–368. Thabut D, Moreau R, Lebrec D. Noninvasive assessment of portal hypertension in patients with cirrhosis. Hepatology 2011;53:683–694. Thabut D, Trabut JB, Massard J, Rudler M, Muntenau M, Messous D, Poynard T. Non-invasive diagnosis of large oesophageal varices with FibroTest in patients with cirrhosis: A preliminary retrospective study. Liver Int 2006;26:271–278. Vizzutti F, Arena U, Romanelli RG, Rega L, Foschi M, Colagrande S, Petrarca A, Moscarella S, Belli G, Zignego AL, Marra F, Laffi G, Pinzani M. Liver stiffness measurement predicts severe portal hypertension in patients with HCV-related cirrhosis. Hepatology 2007;45: 1290–1297. Wang L, Xing J, Chen F, Yan R, Ge L, Qin Q, Wang L, Ding Z, Guo W, Wang N. Spatial analysis on hepatitis C virus infection in mainland china: from 2005 to 2011. PLoS One 2014;9:e110861. Yeh JL, Peng YC, Tung CF, Yang DY, Hu WH, Chow WK, Yeh HZ, Chen GH. Clinical significance and prediction factors of gastric varices in patients with hepatocellular carcinoma. Hepatogastroenterology 2003;50:1603–1608. Ziol M, Handra-Luca A, Kettaneh A, Christidis C, Mal F, Kazemi F, de Ledinghen V, Marcellin P, Dhumeaux D, Trinchet JC, Beaugrand M. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology 2005;41:48–54.
http://dx.doi.org/10.1016/j.ultrasmedbio.2015.02.005
d
Original Contribution USING ULTRASONIC TRANSIENT ELASTOMETRY (FIBROSCAN) TO PREDICT ESOPHAGEAL VARICES IN PATIENTS WITH VIRAL LIVER CIRRHOSIS ZHONGWEI HU,* YUYUAN LI,y CHUO LI,* CHUNMING HUANG,* ZHITAO OU,* JIAWEI GUO,* HONGBIN LUO,* and XIAOPING TANGz * Department of Internal Medicine, Guangzhou No. 8 People’s Hospital, Guangzhou, China; y Department of Gastroenterology Hepatology, Guangzhou First People’s Hospital, Guangzhou Medical University, Guangzhou, China; and z Institute for Infectious Diseases, Guangzhou No. 8 People’s Hospital, Guangzhou, China (Received 17 August 2014; revised 10 February 2015; in final form 13 February 2015)
Abstract—The correlation between liver stiffness (LS), measured by ultrasonic transient elastometry (FibroScan), and the presence and severity of esophageal varices (EV) in patients with viral cirrhosis of the liver has not been well documented to date. The study described here investigated the value of using FibroScan to predict EV. Patients with cirrhosis (200 patients: 167 cases caused by hepatitis B virus and 33 cases caused by hepatitis C virus) underwent both upper gastrointestinal endoscopy and FibroScan. Demographic, clinical, biochemical and endoscopic data and FibroScan-obtained LS parameters were collected. The mean LS value in patients with EV (33.2 kPa) was significantly higher than the mean LS value in patients without EV (18.6 kPa) (p , 0.05). The mean LS value in patients with grade 2 and 3 EV (38.3 kPa) was significantly higher than that in patients with grade 1 EV (24.8 kPa) (p , 0.05). Overall, FibroScan was 86.4% sensitive and 72.2% specific in predicting the presence of EV, with an area under the receiver operating characteristic curve (AUROC) of 0.84. The sensitivity and specificity for the patients with grade 2 or 3 EV were 84% and 73% (AUROC 5 0.86). When FibroScan was combined with platelet count, the overall sensitivity and specificity of prediction increased to 84% and 80% (AUROC 5 0.88), respectively, and 84% and 75% (AUROC 5 0.89), respectively, in patients with grade 2 and 3 EV. FibroScan alone or combined with platelet count might predict the presence and severity of EV in patients with hepatitis B or C-related viral cirrhosis. (E-mail: [email protected]) Ó 2015 World Federation for Ultrasound in Medicine & Biology. Key Words: Esophageal varices, Liver cirrhosis, FibroScan, Transient elastography.
(HCV) present with cirrhosis at their first clinical examination. As many as 20%–30% of patients who do not present with cirrhosis at that initial visit will develop cirrhosis within at least a decade (Benvegn u et al. 2004; Di Marco et al. 1999; Ikeda et al. 1998; Niederau et al. 1998). Major complications of cirrhosis include liver failure, portal hypertension (PHT), hepatorenal syndrome and esophageal varices (EV), all of which are poor prognostic indicators (Gines et al. 2004). Up to 90% of patients with cirrhosis develop EV, which may bleed. The incidence of bleeding EV is around 5% in patients with small EV (SEV) and up to 15% in patients with large EV (LEV) (Jensen 2002). Mortality per bleeding episode is around 10%–20% (Carbonell et al. 2004). Therefore, screening for EV in patients with cirrhosis is strongly recommended across guidelines and consensus statements (de Franchis 2005). Accordingly, it is generally recommended that patients with cirrhosis undergo active surveillance for
INTRODUCTION Progressive hepatic fibrosis caused by the accumulation of collagen in the extracellular matrix is characteristic of almost all chronic liver diseases (Bataller and Brenner 2005). Liver cirrhosis is the final phase of the hepatic fibrosis process, combining extensive fibrosis and regenerative nodules. In the East, chronic infection with hepatotropic viruses is a predominant problem (Margolis et al. 1991; Wang et al. 2014). Approximately 10%–20% of patients infected with chronic hepatitis B virus (HBV) or hepatitis C virus
Address correspondence to: Xiaoping Tang, Institute for Infectious Diseases, Guangzhou No.8 People’s Hospital, No. 627 Dongfengdong Road Yuexiu District, Guangzhou 510060, China. E-mail: [email protected] Conflicts of Interest: The authors declare no conflicts of interest, real or perceived, financial or non-financial. 1
2
Ultrasound in Medicine and Biology
early detection of EV via upper gastrointestinal (GI) endoscopy; however, endoscopic examination can be considered an unnecessary burden for both patients and doctors. Approximately half of patients with cirrhosis will not develop EV in the 10 y after the initial cirrhosis diagnosis (Addley et al. 2012). A common consequence of virus-related chronic liver disease is PHT, which results in EV. The hepatic venous pressure gradient (HVPG) is considered the gold standard for assessing PHT in patients with cirrhosis; however, assessments using this parameter are invasive and not routinely performed (Thabut et al. 2011). More non-invasive methods have been proposed as alternatives to endoscopy for EV screening. Measurement of liver stiffness (LS) with ultrasonic transient elastometry has been recognized as a rapid, non-invasive technique in which the measurement correlates well with the underlying stage of fibrosis. A particularly strong correlation has been observed between LS measured by FibroScan (Echosens, Paris, France), the first tool introduced to assess fibrosis through LS (Echosens, Paris, France), and HVPG. Thus, FibroScan has potential to be used for the non-invasive evaluation of EV (Castera et al. 2012). Although the value of FibroScan in assessment of the degree of LS in patients with chronic HCV infection has been well documented, few studies have been conducted on the use of FibroScan for patients with chronic HBV infection (Castera et al. 2005; Cui et al. 2013; Dolmazashvili et al. 2008; Marcellin et al. 2009; Ziol et al. 2005). Moreover, few studies have focused on the relationship between LS and the presence and severity of EV, especially in patients with HBV infection. This is concerning, as the most common cause of cirrhosis with high mortality in China is HBV infection (Calvaruso et al. 2013; Castera et al. 2009; Kazemi et al. 2006). In addition, the results of different studies still lack sensitivity and specificity (Calvaruso et al. 2013). The aim of this study was to investigate the relationship between degree of LS and EV and to assess the use of FibroScan to predict the presence and severity of EV.
METHODS Participants Between July 2007 and October 2012, 200 consecutive inpatients from Guangzhou No. 8 People’s Hospital, China, were initially included. Patients were older than 18 y, included both sexes, had chronic HBV- or HCVrelated liver cirrhosis, had undergone upper GI endoscopy, and agreed to a FibroScan examination. Cirrhosis of the liver was diagnosed using the criteria from the Chinese Association of Liver Diseases practice guidelines, which, in turn, are adapted from the American
Volume -, Number -, 2015
Association for the Study of Liver Diseases guidelines (Runyon 2013). The diagnosis was based mainly on abdominal ultrasonic findings combined with medical history, clinical symptoms and laboratory results. If the diagnosis was unclear from these parameters, liver biopsy was considered a secondary diagnostic measure. In this study, 22 patients underwent liver biopsy. Serologic detection of hepatitis B surface antigens and hepatitis C antibodies was used to diagnose HBV or HCV infection. Patients were excluded from this study if they were infected with both HBV and HCV, reported heavy alcohol intake (defined as .40 g of alcohol/d for $5 y), or had comorbid chronic liver diseases, aspartate aminotransferase (AST)/alanine aminotransferase (ALT) levels greater than three times the normal upper limit, hepatocellular carcinoma, acute liver failure or decompensated congestive heart failure. The clinical parameters collected from each patient were: age, sex, history of ascites, bleeding varices, spontaneous bacterial peritonitis, hepatic encephalopathy, organic kidney failure, or hepatorenal syndrome. Collected biological parameters included a complete blood count and levels of AST, ALT, alkaline phosphatases, total bilirubin, albumin, gamma globulins, prothrombin time (PT), creatinine, urea and a-fetoprotein. The research protocol was approved by the ethics committee of Guangzhou No. 8 People’s Hospital. Written informed consent was obtained from each participant. The study was registered in the Chinese Clinical Trial Registry. Endoscopy All patients underwent upper GI endoscopy (Fujinon EG-590 WR, Japan). Varices were graded according to their size and appearance by an endoscopic physician experienced in the following grading scale (Chinese Medical Association 2003): grade 0 5 no varices; grade 1 5 small, straight EV; grade 2 5 enlarged, tortuous EV occupying less than one-third of the lumen; and grade 3 5 large, coil-shaped EV occupying more than onethird of the lumen. Liver stiffness measurement FibroScan was used to measure LS during the same week the patients underwent endoscopic examinations. An experienced doctor blinded to the patients’ clinical data and endoscopic results performed all measurements. The FibroScan device was equipped with a probe that included a vibrating component. The vibrator transmitted vibrations to the tissues, inducing an elastic shear wave that propagated through the tissue. The propagation was followed by pulse-echo ultrasound acquisitions. Their velocity, directly related to tissue stiffness, was then measured (Friedrich-Rust et al. 2010; Sandrin et al.
Prediction of esophageal varices with FibroScan d Z. HU et al.
2003). The measurement depth was between 25 and 65 mm, and the results were expressed in kilopascals (kPa). Faster shear wave speeds were correlated with higher LS scores and indicated a stiffer liver. As suggested by the manufacturer, 10 successful acquisitions were performed for each patient. To determine how reliable the transient elastometry results were, the following criteria were used: very reliable 5 interquartile range/median ratio (IQR/ M) # 0.10; reliable 5 0.10 , IQR/M # 0.30 or IQR/ M . 0.30 with LS median , 7.1 kPa; and poorly reliable 5 IQR/M . 0.30 with LS median $ 7.1 kPa (Boursier et al. 2013). The poorly reliable results were excluded from the final analysis. Statistical analysis Data were analyzed using SPSS Version 10.0 (SPSS, Chicago, IL, USA). Continuous data were expressed as means 6 standard deviations (SD) and examined with a one-way analysis of variance. The Kruskal–Wallis nonparametric analysis of variance and Bonferroni corrections were used to compare LS among the different EV stages. Mann–Whitney U-tests were performed to compare LS among HBV and HCV subgroups. Categorical variables were expressed as percentages and analyzed with the c2 test. Multiple backward regression was used to assess the association between EV grade (dependent variable) and independent variables, namely, demographic (age, sex) and clinical (chronic HBVor HCV, kidney failure, hepatorenal syndrome, history and biochemical criteria) characteristics. All independent variables were entered into the prediction model and removed in a backward stepwise procedure if their pvalues were .0.05. Thus, a predictive model containing all significant independent variables and dependent variables (EV grades) was established. The receiver operating characteristic (ROC) curve, a plot of sensitivity versus 1 2 specificity, was created for all possible cutoff
3
values. The optimal cutoff value was chosen when the sum of sensitivity and specificity was maximized. The Hanley and McNeil test in the SPSS software was performed for ROC curve analysis. Moreover, the Delong test was employed to compare ROC curves using MedCalc software (http://www.medcalc.org). A p-value , 0.05 was considered to indicate statistical significance. RESULTS Among 210 patients who initially met the inclusion criteria, 200 patients with complete data were included in this study’s final analysis. The 10 patients who were excluded had ‘‘poorly reliable’’ transient elastometry results. The mean age of patients without EV was 44.8 6 10.1 y, including 65 (32.5%) men and 25 (12.5%) women. The mean age of patients with EV was 45.39 6 10.4 y, including 76 (38%) men and 34 (17%) women. In patients without EV, HBV caused 75 (37.5%) cirrhosis cases, and HCV caused 15 (7.5%) cirrhosis cases. In patients with EV, HBV caused 92 (46%) cirrhosis cases, and HCV caused 18 (9%) cirrhosis cases. No significant differences in age, sex and HBV/ HCV incidence were observed in the three graded groups (grade 0, 1 and 2/3 EV) (p . 0.05) (Table 1). There were significant differences in history of ascites, hepatic encephalopathy, GI bleeding and mosaic pattern of the gastric mucosa between the three groups (p , 0.05). For the biochemical variables (Table 2), there were significant differences in albumin, platelet count, bilirubin, prothrombin time and urea between the three groups (p , 0.05). Of the 200 patients in the study, 90 (45%) did not have EV (V0), 41 (20.5%) had grade 1 EV (V1) and 69 (34.5%) had grade 2 or 3 EV (V2/3). The overall mean LS value in patients with EV (33.2 kPa, range: 10.6– 77.0 kPa) was significantly higher than that in patients without EV (18.6 kPa, range: 8.6–38.8 kPa) (V0–V1:
Table 1. Demographic, clinical and endoscopic data Patients with esophageal varices Variable
All patients (n 5 200)
Grade 0 (n 5 90)
Grade 1 (n 5 41)
Grades 2 and 3 (n 5 69)
p-Value
Age (y) Males Chronic hepatitis B Kidney failure Hepatorenal syndrome History of ascites History of leg edema Hepatic encephalopathy Gastrointestinal bleeding Spontaneous bacterial peritonitis Right pleural effusion Mosaic pattern in gastric mucosa
45.1 6 10.2 141 (71%) 167 (84%) 9 (5%) 1 (1%) 61 (31%) 43 (22%) 12 (6%) 21 (11%) 3 (%) 3 (2%) 122 (61%)
44.8 6 10.1 65 (72%) 75 (83%) 4 (4%) 0 22 (24%) 19 (21%) 2 (2%) 0 0 0 27 (30%)
45.3 6 10.4 28 (68%) 35 (85%) 2 (5%) 0 19 (46%) 13 (32%) 6 (15%) 9 (22%) 0 1 (2%) 29 (71%)
45.4 6 10.5 48 (70%) 57 (83%) 3 (4%) 1 (1%) 20 (29%) 11 (16%) 4 (6%) 12 (17%) 3 (%) 2 (3%) 66 (96%)
0.89 0.88 0.92 0.99 0.99 0.04 0.14 0.02 0.01 0.18 0.22 0.01
4
Ultrasound in Medicine and Biology
Volume -, Number -, 2015
Table 2. Biochemical characteristics Patients with EV Variable
All patients (n 5 200)
Patients without EV (grade 0) (n 5 90)
Grade 1 (n 5 41)
Grade 2/3 (n 5 69)
p-Value
Total bilirubin (mmol/L) Albumin (g/L) Gamma globulins (g/L) Alanine aminotransferase (IU) Aspartate aminotransferase (IU) Alkaline phosphatases (IU/L) Prothrombin time (%) Leukocytes (3 103/mm3) Polynuclear neutrophils (3 103/mm3) Hemoglobin (g/dL) Platelet count (103/mm3) Creatinine (mmol/L) Urea (mmol/L) Natremia (mmol/L) a-Fetoprotein (ng/mL)
39.8 6 26.7 33.5 6 7.2 16.9 6 7.8 63.8 6 24.3 70.2 6 37.6 142.3 6 87.6 60.9 6 16.9 5.1 6 2.3 3.0 6 2.1 12.3 6 2.3 124.1 6 70.9 79.0 6 64.5 5.95 6 5.35 137.0 6 3.44 9.4 6 9.0
32.6 6 23.2 35.6 6 7.4 15.5 6 7.6 61.0 6 21.1 72.0 6 52.7 151.4 6 96.2 68.4 6 18.5 5.3 6 2.5 3.2 6 2.4 12.5 6 2.3 167.8 6 75.8 77.1 6 66.8 5.29 6 5.13 138.5 6 3.69 9.2 6 9.0
40.3 6 19.1 34.2 6 6.8 17.9 6 7.1 63.4 6 19.5 56.5 6 13.6 140.8 6 83.3 57.7 6 14.4 5.1 6 2.3 3.0 6 2.2 12.6 6 2.1 107.3 6 40.7 82.3 6 63.9 6.34 6 5.66 135.6 6 3.11 9.4 6 8.8
49.1 6 31.8 30.3 6 6.1 18.2 6 8.5 67.9 6 30.0 76.0 6 26.2 131.4 6 79.1 53.0 6 16.1 4.9 6 2.0 2.8 6 1.5 11.8 6 2.4 77.3 6 36.1 79.5 6 61.9 6.56 6 5.45 135.8 6 3.32 9.6 6 9.2
0.04 0.04 0.55 0.71 0.03 0.10 0.03 0.82 0.36 0.49 0.01 0.68 0.04 0.93 0.84
EV 5 esophageal varices.
p , 0.001, and V0–V2/3: p , 0.001, respectively). There was a significant difference between the in LS values between groups V1 (24.8 kPa, 10.6–40.3 kPa) and V2/3 (38.3 kPa, 13.9–77.0 kPa) (p , 0.001). The mean LS values in V0, V1 and V2/3 patients with HBV-related cirrhosis were 18.8 (8.6–38.8), 25.2 (11.5–40.3) and 39.0 (15.9–77.0) kPa. The LS values in V0, V1 and V2/ 3 patients with HCV-related cirrhosis were 17.8 (10.7– 32.0), 22.2 (10.6–27.8) and 34.9 (13.9–66.4) kPa, respectively. No significant difference was detected between the HBV and HCV subgroups (p-values were 0.618, 0.306 and 0.218, respectively). Distribution of liver stiffness (FibroScan) for each grade (0–3) of EV was showed in Figure 2. With endoscopic findings as the ‘‘gold standard’’ for EV diagnosis, a ROC curve for the presence of EV as predicted by LS values was generated. The cutoff values are listed in Table 3. When the cutoff value was set at 20.3 kPa, the area under the ROC curve (AUROC) was 0.843, and FibroScan’s sensitivity and specificity for predicting the presence of EV were 86.4% and 72.2%, respectively (Fig. 1a, Table 3). The sensitivity and specificity for patients with grade 2 and 3 EV were 84.1% and 72.5%, respectively, with an AUROC of 0.855 (Fig. 1b, Table 3). The same AUROC results were obtained with the Hanley and McNeil test and Delong test. A multiple regression analysis for the predictors of EV grade (dependent variable) was conducted using demographic variables (age, sex); presence of HBV or HCV; kidney failure; hepatorenal syndrome; medical history (ascites, leg edema, hepatic encephalopathy, GI bleeding, spontaneous bacterial peritonitis, right pleural effusion, mosaic pattern in gastric mucosa); biochemical criteria (total bilirubin, albumin, gamma globulins, ALT, AST, alkaline phosphatase, prothrombin time, leuko-
cytes, polynuclear neutrophils, hemoglobin, platelet count, creatinine, urea, natremia, a-fetoprotein) and FibroScan value. Only the FibroScan value and platelet count had a confirmed predictive value for the presence of EV. The correlation coefficients for FibroScan value and platelet count were 0.048 (95% confidence interval: 0.039 to 0.057) and –0.005 (95% confidence interval: –0.007 to 0.004), respectively, and the constant was 0.581. ROC curves and data for predicting the presence of EV using platelet count and FibroScan scores are provided in Figure 1c and Table 3. Overall, combining FibroScan values with platelet count yielded a better AUROC (0.879) for diagnosis of EV. For diagnosis of EV grade $2, the AUROC increased to 0.886 (Fig. 1d, Table 3). The Hanley and McNeil test and Delong test yielded the same AUROC results. DISCUSSION Patients with cirrhosis are at risk for morbidity and mortality from GI bleeding through EV. Screening for EV, rather than treating patients after a crisis, ultimately conserves resources and time. Finding non-invasive methods to screen for EV directly caused by PHT is a recent research focus, however, and still poses some clinical challenges. About 40% of patients with Child– Pugh A and 60% of patients with ascites develop varices as a result of PHT, with an expected incidence of new varices of about 5% per year (Addley et al. 2012). Because PHT occurs as a complication of liver cirrhosis, the degree of PHT can be correlated with the severity of cirrhosis (Thabut et al. 2011); therefore, PHT might be a potential tool for evaluating the relationship between EV and cirrhosis. Unfortunately, the current gold standard for assessing the presence and severity of PHT,
Prediction of esophageal varices with FibroScan d Z. HU et al.
5
Table 3. Diagnostic performances of the FibroScan with respect to the presence of EV EV grade
Sensitivity (%)
FibroScan alone 2/3 84.1 1–3 86.4 FibroScan 1 platelet counts 2/3 84.1 1–3 83.6
Specificity (%)
Positive predictive value (%)
Negative predictive value (%)
AUROC (%)
Cutoff value (kPa)
72.5 72.2
71.7 79.2
90.8 81.3
85.5 (0.80–0.91) 84.3 (0.79–0.90)
25.55 20.25
74.8 80.0
63.7 83.6
89.9 80.0
88.6 (0.84–0.93) 87.9 (0.83–0.93)
1.24 1.02
AUROC 5 area under receiver operating characteristic curve; EV 5 esophageal varices.
HVPG measurement, is invasive and performed only in expert centers. Although a variety of blood tests for EV screening have been proposed, including albumin level and prothrombin time (Kim et al. 2012a; Yeh et al.
2003), these biochemical markers are more strongly associated with the severity of cirrhosis and not with the presence or absence of EV at a clinically useful level.
Fig. 1. ROC curves for diagnosis of EV: (a) Diagnosis of EV using FibroScan. (b) Diagnosis of grade $2 EV using FibroScan. (c) Diagnosis of EV using the FibroScan and platelet count. (d) Diagnosis of grade $2 EV using the FibroScan and platelet count. ROC 5 receiver operating characteristic; EV 5 esophageal varices.
6
Ultrasound in Medicine and Biology
Volume -, Number -, 2015
Fig. 2. Distribution of liver stiffness (FibroScan) for each grade (0–3) of EV. (a) Scatter plot of liver stiffness values (FibroScan) for each EV grade (0–3). (b) Error bar represents 95% confidence interval of mean liver stiffness value (FibroScan). (c) Box plot illustrates distribution of liver stiffness values (FibroScan) by EV grade (0–3). The top and bottom of the boxes are the first and third quartiles. The length of the box represents the interquartile range, within which are located 50% of the values. The lines through the middle of the boxes represent median values. Unfilled circles are data points that were either 1.5 3 IQR or more above the third quartile or 1.5 3 IQR or more below the first quartile. CI 5 confidence interval; EV 5 esophageal varices; IQR 5 interquartile range.
Studies using different ultrasonic methods to estimate liver fibrosis, including the use of transient elastometry (e.g., FibroScan), real-time tissue elastography and shear wave elastography (SWE) using acoustic radiation force impulse (ARFI), have been published. In patients with obesity-related liver disease, ARFI has been successfully used to distinguish fibrosis stages from low to high (90% sensitivity and specificity) (Palmeri et al. 2011). Another study reported that LS measurements obtained with ARFI had a similar correlation strength as fibrosis stage values obtained by transient elastometry methods in patients with chronic HBV and HCV (Sporea et al. 2012). On the other hand, a different study reported that FibroScan was a much better predictor of liver cirrhosis than the AST-to-platelet ratio index (APRI) and suggested real-time tissue elastography as a promising method for determining the stage of hepatic fibrosis (Tatsumi et al. 2008). Among transient elastometry methods, FibroScan was the first to be introduced and remains the most common tool in clinical practice (Cui et al. 2013). Fibroscan has been considered an accurate, reliable and reproducible method for assessing liver fibrosis through LS measurements (Fransen van de Putte et al. 2011; Jang et al. 2012; Kim et al. 2012a, 2012b). Some studies have reported that FibroScan is a good diagnostic tool for detecting PHT, a common complication of liver fibrosis (Bureau et al. 2008; Robic et al. 2011; Thabut et al. 2007). As PHT worsens, varices can form and worsen as well. LS values correlate with HVPG and are an excellent predictor of PHT in patients with compensated cirrhosis (Lemoine et al. 2008; Vizzutti et al. 2007). Thus, FibroScan values may be correlated with the degree of EV. To date, two studies have reported significant correlation
between FibroScan-measured LS values and the presence of LEVs (Nguyen-Khac et al. 2010; Thabut et al. 2006). In these studies, however, the etiology of cirrhosis was predominantly alcohol abuse. Liver cirrhosis caused by viral infection (viral hepatitis, especially hepatitis C, a major endemic disease) is a major cause of death in China (He et al. 2005). In the present study, patients with chronic HBV and HCV infections were emphasized; to our knowledge, this was a novel approach. In addition, the relationship between SEVs detected by endoscopy and liver cirrhosis evaluated by FibroScan remains obscure. It is, however, worth noting that although LS has been considered an excellent non-invasive parameter for assessing stages of liver cirrhosis, the presence of hepatic congestion is known to drastically increase LS independent of fibrosis stage (Millonig et al. 2009, 2010). For this reason, patients with decompensated congestive heart failure were excluded from this study. In this study, cutoff values for diagnosis of EV were different from values in previous studies. Other studies have reported cutoff values of 19, 18 and 48 kPa for diagnosing LEV (Bureau et al. 2008; Kazemi et al. 2006; Nguyen-Khac et al. 2010; Vizzutti et al. 2007). In our study, the cutoff value was 25.55 kPa for LEV and 20.25 kPa for any EV. The differences might be due to the varied study populations across trials, which included patients with chronic liver diseases of various etiologies. In the previous studies, patients were primarily or exclusively those with alcoholic cirrhosis or a chronic HCV infection (Bureau et al. 2008; Kazemi et al. 2006; Nguyen-Khac et al. 2010; Vizzutti et al. 2007). Other preliminary studies have suggested that some non-invasive biochemical markers for liver fibrosis could also predict the presence of EV. One study found that a
Prediction of esophageal varices with FibroScan d Z. HU et al.
platelet count .150,000/mm3 was associated with a negative predictive value of 99% for EV (Sanyal et al. 2006). Another study reported that the platelet count/ spleen diameter ratio (cutoff value 5 909) had a diagnostic accuracy of 86% for EV (Giannini et al. 2006). Platelet count, AST-to-ALT ratio (AAR), Lok index, AST-to-platelet ratio index (APRI), Forns index, Fib-4 and FibroIndex were analyzed in a multicenter study. The results indicated that the Lok index (cutoff value 5 1.5) could be used to exclude LEV with a 96% negative predictive value. A combination of the Lok index (cutoff 5 0.9) and Forns index (cutoff 5 8.5) predicted the presence of EV of any grade with an AUROC of 0.82 (Sebastiani et al. 2010). In contrast, our study found that FibroScan-obtained LS values alone were valuable, but not reliable enough to be used as a screening method for detecting the presence of EV in patients with liver cirrhosis. Combining FibroScan values and platelet counts might improve prediction of the presence of EV. In addition, the Hanley and McNeil test and Delong test yielded the same AUROC results, strengthening the reliability of the findings. The present study had some limitations. First, there may have been a lack of reliable and reproducible color ultrasound results. Abdominal color ultrasonography scans that were completed by several examiners were excluded because of a lack of normalcy and objectivity. The second limitation was the small number of patients with HCV, possibly limiting generalizability for this population. CONCLUSIONS In this prospective study, we found a useful FibroScan cutoff value for the non-invasive diagnosis of EV in patients with cirrhosis. In particular, when a subanalysis of patients with cirrhosis and severe EV (grade 2/3) was conducted, the predictive usefulness increased. FibroScan values or a combination of FibroScan values and platelet counts might be an acceptable reference marker for the prediction of severe EV in patients with cirrhosis when endoscopic evaluation is not indicated because of poor compliance and high cost. Acknowledgments—This study was funded by Guangdong Province Science and Technology Plan Project No. 2010 B031600008 and Natural Science Foundation of China Grant 81200903.
REFERENCES Addley J, Tham TC, Cash WJ. Use of portal pressure studies in the management of variceal haemorrhage. World J Gastrointest Endosc 2012;4:281–289. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115:209–218. Benvegn u L, Gios M, Boccato S, Alberti A. Natural history of compensated viral cirrhosis: A prospective study on the incidence and hierarchy of major complications. Gut 2004;53:744–749. Boursier J, Zarski JP, de Ledinghen V, Rousselet MC, Sturm N, Lebail B, Fouchard-Hubert I, Gallois Y, Oberti F, Bertrais S,
7
Cales P. Multicentric Group from ANRS/HC/EP23 FIBROSTAR Studies. Determination of reliability criteria for liver stiffness evaluation by transient elastography. Hepatology 2013;57: 1182–1191. Bureau C, Metivier S, Peron JM, Selves J, Robic MA, Gourraud PA, Rouquet O, Dupuis E, Alric L, Vinel JP. Transient elastography accurately predicts presence of significant portal hypertension in patients with chronic liver disease. Aliment Pharmacol Ther 2008;27: 1261–1268. Calvaruso V, Bronte F, Conte E, Simone F, Craxı A, Di Marco V. Modified spleen stiffness measurement by transient elastography is associated with presence of large oesophageal varices in patients with compensated hepatitis C virus cirrhosis. J Viral Hepat 2013;20: 867–874. Carbonell N, Pauwels A, Serfaty L, Fourdan O, Levy VG, Poupon R. Improved survival after variceal bleeding in patients with cirrhosis over the past two decades. Hepatology 2004;40:652–659. Castera L, Le Bail B, Roudot-Thoraval F, Bernard PH, Foucher J, Merrouche W, Couzigou P, de Ledinghen V. Early detection in routine clinical practice of cirrhosis and oesophageal varices in chronic hepatitis C: Comparison of transient elastography (FibroScan) with standard laboratory tests and non-invasive scores. J Hepatol 2009;50:59–68. Castera L, Pinzani M, Bosch J. Non-invasive evaluation of portal hypertension using transient elastography. J Hepatol 2012;56:696–703. Castera L, Vergniol J, Foucher J, Le Bail B, Chanteloup E, Haaser M, Darriet M, Couzigou P, De Ledinghen V. Prospective comparison of transient elastography, FibroTest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005;128:343–350. Chinese Medical Association. Trial scheme of diagnosing and treating gastroesophageal varices under endoscopy. Chin J Dig Endosc 2003;2004(21):149–151. Cui XW, Friedrich-Rust M, De Molo C, Ignee A, Schreiber-Dietrich D, Dietrich CF. Liver elastography, comments on EFSUMB elastography guidelines 2013. World J Gastroenterol 2013;19:6329–6347. De Franchis R. Evolving consensus in portal hypertension: Report of the Baveno IV consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 2005;43:167–176. Di Marco V, Lo Iacono O, Camma C, Vaccaro A, Giunta M, Martorana G, Fuschi P, Almasio PL, Craxı A. The long-term course of chronic hepatitis B. Hepatology 1999;30:257–264. Dolmazashvili E, Zhamutashvili M, Svanidze M, Nizharadze N, Abutidze A. FibroScan and FibroTest/FibroMax to assess liver fibrosis/cirrhosis in patients with chronic HBV and HCV infection in Georgia. Georgian Med News 2008;165:83–87. Fransen van de Putte D, Blom R, van Soest H, Mundt M, Verveer C, Arends J, de Knegt RE, Mauser-Bunschoten E, van Erpecum K. Impact of FibroScan on management of chronic viral hepatitis in clinical practice. Ann Hepatol 2011;10:469–476. Friedrich-Rust M, Rosenberg W, Parkes J, Herrmann E, Zeuzem S, Sarrazin C. Comparison of ELF, FibroTest and FibroScan for the non-invasive assessment of liver fibrosis. BMC Gastroenterol 2010;10:103. Giannini EG, Zaman A, Kreil A, Floreani A, Dulbecco P, Testa E, Sohaey R, Verhey P, Peck-Radosavljevic M, Mansi C, Savarino V, Testa R. Platelet count/spleen diameter ratio for the non-invasive diagnosis of esophageal varices: Results of a multicenter, prospective, validation study. Am J Gastroenterol 2006;101:2511–2519. Gines P, Cardenas A, Arroyo V, Rodes J. Management of cirrhosis and ascites. N Engl J Med 2004;350:1646–1654. He J, Gu D, Wu X, Reynolds K, Duan X, Yao C, Wang J, Chen CS, Chen J, Wildman RP, Klag MJ, Whelton PK. Major causes of death among men and women in China. N Engl J Med 2005;353: 1124–1134. Ikeda K, Saitoh S, Suzuki Y, Kobayashi M, Tsubota A, Koida I, Arase Y, Fukuda M, Chayama K, Murashima N, Kumada H. Disease progression and hepatocellular carcinogenesis in patients with chronic viral hepatitis: A prospective observation of 2215 patients. J Hepatol 1998;28:930–938. Jang HW, Kim SU, Park JY, Ahn SH, Han KH, Chon CY, Park YN, Choi EH, Kim DY. How many valid measurements are necessary
8
Ultrasound in Medicine and Biology
to assess liver fibrosis using FibroScan in patients with chronic viral hepatitis? An analysis of patients with at least 10 valid measurements. Yonsei Med J 2012;53:337–345. Jensen DM. Endoscopic screening for varices in cirrhosis: Findings, implications, and outcomes. Gastroenterology 2002;122:1620–1630. Kazemi F, Kettaneh A, N’kontchou G, Pinto E, Ganne-Carrie N, Trinchet JC, Beaugrand M. Liver stiffness measurement selects patients with cirrhosis at risk of bearing large oesophageal varices. J Hepatol 2006;45:230–235. Kim BK, Kim HS, Park JY, Ahn SH, Chon CY, Park YN, Han KH, Kim SU. Prospective validation of ELF test in comparison with FibroScan and FibroTest to predict liver fibrosis in Asian patients with chronic hepatitis B. PLoS One 2012a;7:e41964. Kim SU, Lee JH, Kim DY, Ahn SH, Jung KS, Choi EH, Park YN, Han KH, Chon CY, Park JY. Prediction of liver-related events using FibroScan in chronic hepatitis B patients showing advanced liver fibrosis. PLoS One 2012b;7:e36676. Lemoine M, Katsahian S, Ziol M, Nahon P, Ganne-Carrie N, Kazemi F, Grando-Lemaire V, Trinchet JC, Beaugrand M. Liver stiffness measurement as a predictive tool of clinically significant portal hypertension in patients with compensated hepatitis C virus or alcohol-related cirrhosis. Aliment Pharmacol Ther 2008;28: 1102–1110. Marcellin P, Ziol M, Bedossa P, Douvin C, Poupon R, de Ledinghen V, Beaugrand M. Non-invasive assessment of liver fibrosis by stiffness measurement in patients with chronic hepatitis B. Liver Int 2009;29: 242–247. Margolis HS, Alter MJ, Hadler SC. Hepatitis B: Evolving epidemiology and implications for control. Semin Liver Dis 1991;11:84–92. Millonig G, Friedrich S, Adolf S, Fonouni H, Golriz M, Mehrabi A, Buchler M. Liver stiffness is directly influenced by central venous pressure. Ultrasound Med Biol 2010;52:206–210. Millonig G, Friedrich S, Adolf S, Poschl G, Stiefel P, Seitz HK, Muelle S. Increased Venous pressure increases liver stiffness (FibroScan) independent of liver fibrosis. J Hepatol 2009;50:204. Nguyen-Khac E, Saint-Leger P, Tramier B, Coevoet H, Capron D, Dupas JL. Noninvasive diagnosis of large esophageal varices by FibroScan: Strong influence of the cirrhosis etiology. Alcohol Clin Exp Res 2010;34:1146–1153. Niederau C, Lange S, Heintges T, Erhardt A, Buschkamp M, H€urter D, Nawrocki M, Kruska L, Hensel F, Petry W, H€aussinger D. Prognosis of chronic hepatitis C: Results of a large, prospective cohort study. Hepatology 1998;28:1687–1695. Palmeri ML, Wang MH, Rouze NC, Abdelmalek MF, Guy CD, Moser B, Diehl AM, Nightingale KR. Noninvasive evaluation of hepatic fibrosis using acoustic radiation force-based shear stiffness in patients with nonalcoholic fatty liver disease. J Hepatol 2011;55: 666–672. Robic MA, Procopet B, Metivier S, Peron JM, Selves J, Vinel JP, Bureau C. Liver stiffness accurately predicts portal hypertension related complications in patients with chronic liver disease: A prospective study. J Hepatol 2011;55:1017–1024. Runyon BA. Introduction to the revised American Association for the Study of Liver Diseases Practice Guideline management of adult pa-
Volume -, Number -, 2015 tients with ascites due to cirrhosis 2012. Hepatology 2013;57: 1651–1653. Sandrin L, Fourquet B, Hasquenoph JM, Yon S, Fournier C, Mal F, Christidis C, Ziol M, Poulet B, Kazemi F, Beaugrand M, Palau R. Transient elastography: A new non-invasive method for assessment of hepatic fibrosis. Ultrasound Med Biol 2003;29: 1705–1713. Sanyal AJ, Fontana RJ, Di Bisceglie AM, Everhart JE, Doherty MC, Everson GT, Donovan JA, Malet PF, Mehta S, Sheikh MY, Reid AE, Ghany MG, Gretch DR. The prevalence and risk factors associated with esophageal varices in patients with hepatitis C and advanced fibrosis. Gastrointest Endosc 2006;64:855–864. Sebastiani G, Tempesta D, Fattovich G, Castera L, Halfon P, Bourliere M, Noventa F, Angeli P, Saggioro A, Alberti A. Prediction of oesophageal varices in hepatic cirrhosis by simple serum noninvasive markers: Results of a multicenter, large-scale study. J Hepatol 2010;53:630–638. Sporea I, Sirli R, Bota S, Popescu A, Sendroiu M, Jurchis A. Comparative study concerning the value of acoustic radiation force impulse elastography (ARFI) in comparison with transient elastography (TE) for the assessment of liver fibrosis in patients with chronic hepatitis B and C. Ultrasound Med Biol 2012;38:1310–1316. Tatsumi C, Kudo M, Ueshima K, Kitai S, Takahashi S, Inoue T, Minami Y, Chung H, Maekawa K, Fujimoto K, Akiko T, Takeshi M. Noninvasive evaluation of hepatic fibrosis using serum fibrotic markers, transient elastography (FibroScan) and real-time tissue elastography. Intervirology 2008;51(Suppl 1):27–33. Thabut D, Imbert-Bismut F, Cazals-Hatem D, Messous D, Muntenau M, Valla DC, Moreau R, Poynard T, Lebrec D. Relationship between the FibroTest and portal hypertension in patients with liver disease. Aliment Pharmacol Ther 2007;26:359–368. Thabut D, Moreau R, Lebrec D. Noninvasive assessment of portal hypertension in patients with cirrhosis. Hepatology 2011;53:683–694. Thabut D, Trabut JB, Massard J, Rudler M, Muntenau M, Messous D, Poynard T. Non-invasive diagnosis of large oesophageal varices with FibroTest in patients with cirrhosis: A preliminary retrospective study. Liver Int 2006;26:271–278. Vizzutti F, Arena U, Romanelli RG, Rega L, Foschi M, Colagrande S, Petrarca A, Moscarella S, Belli G, Zignego AL, Marra F, Laffi G, Pinzani M. Liver stiffness measurement predicts severe portal hypertension in patients with HCV-related cirrhosis. Hepatology 2007;45: 1290–1297. Wang L, Xing J, Chen F, Yan R, Ge L, Qin Q, Wang L, Ding Z, Guo W, Wang N. Spatial analysis on hepatitis C virus infection in mainland china: from 2005 to 2011. PLoS One 2014;9:e110861. Yeh JL, Peng YC, Tung CF, Yang DY, Hu WH, Chow WK, Yeh HZ, Chen GH. Clinical significance and prediction factors of gastric varices in patients with hepatocellular carcinoma. Hepatogastroenterology 2003;50:1603–1608. Ziol M, Handra-Luca A, Kettaneh A, Christidis C, Mal F, Kazemi F, de Ledinghen V, Marcellin P, Dhumeaux D, Trinchet JC, Beaugrand M. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology 2005;41:48–54.
