MEDICINE

REVIEW ARTICLE

Non-Invasive Investigation of Liver Cirrhosis

Julia Lrke, Andreas Erhardt, Christoph Vogt, Dieter Hussinger

SUMMARY
Introduction: The investigation of cirrhosis of the liver requires regular screening aimed at the detection or prevention of complications. This article reviews the usefulness of transient elastography in making a diagnosis of liver cirrhosis, against a background of other methods. Transient elastography is a novel, easily applicable and painless method to assess cirrhosis by measuring liver stiffness. Methods: All publications found in a PubMed search using the key words elastography and liver up to December 2006 were considered. Results: Depending on the chosen cut-off and the etiology of liver disease, sensitivities of 77 to 100 % and specificities of 82 to 97 % were reported for the diagnosis of liver cirrhosis by transient elastography. Thus higher sensitivities were found for transient elastography compared to other non-invasive fibrosis detection methods despite similar specificities. Discussion: In the context with clinical signs, ultrasound and laboratory parameters, transient elastography might be particularly helpful for the non-invasive diagnosis of liver cirrhosis making histology dispensable in many cases. For differentiation of earlier stages of liver fibrosis, liver biopsy still remains superior to transient elastography. Dtsch Arztebl 2007; 104(24): A 17527.
Key words: liver cirrhosis, non-invasive diagnostic, serum marker of fibrosis, liver biopsy

round 4 to 5 million people suffer from chronic liver disease in Germany (1), which can lead to hepatic cirrhosis and in turn to complications such as ascites, bleeding oesophageal varices, hepatic encephalopathy, hepatorenal syndrome, hepatopulmonary syndrome and hepatocellular carcinoma (HCC). Patients with hepatic cirrhosis have a significantly reduced life expectancy relative to non-cirrhotic patients (2). Hepatic cirrhosis is considered a precursor to hepatocellular carcinoma. The high (1 to 6% per year) incidence of hepatocellular carcinomas secondary to hepatic cirrhosis contribute significantly to the high mortality seen in these patients, hence many expert bodies recommend regular screening for hepatocellular carcinoma in patients with cirrhosis of the liver (3). The limitations of available diagnostic methods for hepatic cirrhosis (clinical signs, imaging techniques and liver biopsy) are presently driving interest in a range of non-invasive tests, which are currently under evaluation. The focus of this review is research relating to transient elastography. We searched all articles appearing on Pubmed up to December 2006 using the search terms "elastography" and "liver." No meta-analyses on transient elastography are available to date.

Liver biopsy and ultrasound

Liver biopsy is a common element of diagnostic workup in hepatic cirrhosis, alongside clinical examination and abdominal ultrasound, and is the accepted diagnostic gold standard. Its sensitivity and specificity can be further increased by the combination of liver histology with macroscopic liver examination via laparoscopy (4). It should be noted that there is no unitary system of histological classification used to evaluate the degree of fibrosis, and individual systems are often validated only for individual disease entities (table 1). A fundamental difficulty with histological classification arises whenever a continuous process such as hepatic fibrosis is translated into an ordinally scaled system of points. This inevitably leads to incongruities in the histological evaluation of the stage of fibrosis. The variance in evaluation can be as high as 20% even among experienced pathologists (5). This is compounded by the fact that fibrotic change in liver tissue does not always occur homogeneously, so that a degree of error is introduced
Klinik fr Gastroenterologie, Hepatologie und Infektiologie, Heinrich-Heine-Universitt Dsseldorf: Dr. med. Lrke, PD Dr. med. Erhardt, Prof. Dr. med. Hussinger; St. Josef Krankenhaus, Moers: Dr. med. Vogt

Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de

MEDICINE

even at the level of the biopsy specimen. Hence in 33% of cases, differences of up to one stage of fibrosis are found within the same patient between biopsy specimens taken from the right and left lobes of the liver, respectively (6). A further disadvantage of liver biopsy is that it is an invasive and possibly painful intervention, with a serious complication rate of 1 to 3 % (5). Ultrasound allows accurate diagnosis of hepatic cirrhosis in 82 to 88 % of cases (7). However, its value depends heavily on practitioner experience and the quality of the instrument. Ultrasonographic diagnosis of hepatic cirrhosis relies on indirect parameters such as evidence of portal hypertension.

Serum markers and scores for fibrosis

There are numerous laboratory measures which individually or in combination mark the extent of fibrosis (box). However, most of these markers bear no causal relation with fibrogenesis. In comparing the various non-invasive methods, it should be remembered that they are often derived from differing patient populations, often only as part of pilot studies, without independent validation, and often using differing diagnostic threshold values (table 2). In addition, these non-invasive markers and scores are measured against liver biopsy, itself a variable and error-prone gold standard. Of all the individual measures, the platelet count is the most convenient rough guide to the evaluation of the stage of fibrosis. In advanced disease of the liver parenchyma, or cirrhosis, the platelet count often falls. This finding correlates moderately with the histologically determined degree of fibrosis, with a correlation coefficient of 0.46 to 0.50 (8). In addition to the platelet count, procollagen III peptide has found a place in routine investigation (box). Another measure of serum fibrosis is the APRI ("aspartate aminotransferase to platelet ratio index") score, which is calculated from the GOT level and the platelet count (box). The APRI score has thus far only been evaluated in patients with viral hepatitis. It is not suitable for the evaluation of the degree of fibrosis in other liver parenchymal diseases such as ethyltoxicity with high GOT levels. A threshold level for significant fibrosis has been

TABLE 1
Histological classification systems for evaluating the stage of fibrosis in chronic viral hepatitis (21)

Stage of fibrosis 0 1

Desmet & Scheuer No increase in fibrosis Portal fibrosis, no septa Incomplete or complete portoportal septa, architecture preserved Fibrosis with septum formation and architectural disruption, but no evidence of complete cirrhotic change Probable or definite cirrhosis

Knodell No fibrosis Portal fibrosis

Ishak No fibrosis Fibrosis of isolated portal areas with or without short septa Increased fibrosis in most portal areas with or without short septa Portal fibrosis with portoportal septa

Metavir* No fibrosis Portal fibrosis

n. d.

Portal fibrosis with scattered septa Numerous septa without cirrhosis

Portoportal or portocentral septa

Cirrhosis

Portal fibrosis with marked porportoportal or portocentral septa Marked septum formation (portoportal or portocentral) with some nodule formation (incomplete cirrhosis) Probable or definite cirrhosis

Cirrhosis

n. d.

n. d.

n. d.

n. d.

n. d.

n. d.
n. d. = not defined; * only validated in chronic hepatitis C

Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de

MEDICINE

BOX

Laboratory tests for degree of hepatic fibrosis

Individual tests: b Platelet count b Procollagen-III-peptide b Type 4 collagen b Hyaluronic acid b Thrombopoetin b TGF1 b Laminin Combined serological markers of fibrosis: b Pohl score (13) AST/ALT and platelet count Diagnostic range: Advanced fibrosis (Metavir F3/F4): AST/ALT > 1 and platelets < 150 000/L, Exclusion of advanced fibrosis (F3/F4): AST/ALT < 1 and platelets > 150 000/L b APRI index (9, 10) Relation of measured GOT [U/L] to reference value [U/L] divided by platelets [109/L] x 100 Diagnostic range: Advanced fibrosis (Ishak 3): score > 1.5 Exclusion of advanced fibrosis (Ishak < 3): score ( 0.5 b Foras index (12) 7.8113.131xln (platelets [109/1]) + 0.781xln (gamma GT) + 3.467xln (age) 0.014 x cholesterol [mg/dL] Diagnostic range: Advanced fibrosis (F2, F3, F4 according to Scheuer score): score > 6.9 Exclusion of advanced fibrosis: score < 4.21 b FibroTest (10) F = 4.467 x log [alpha 2 makroglobulin (g/L)] 1.357 x log [Haptoglobin (g/L)] + 1.017 x log[gamma GT (IU/L)] + 0.281 x [age (years)] + 1.737 x log [bilirubin (mol/L)] 1.184 [apolipoprotein A1 (g/L)] + 0.301 x sex (0 for female, 1 for male) 5.540 Diagnostic range: Advanced fibrosis (Metavir F2, F3, F4): score 0.60 to 1.00 Exclusion of advanced fibrosis: score 0 to 0.10
AST, aspartate aminotransferase; ALT, alanine aminotransferase

defined (Ishak score F3 to F6) as a score of > 1.5 (9), whereas a score of < 0.5 is said to rule out significant fibrosis. In hepatic cirrhosis (Ishak score F5 to F6) the threshold for significant disease is > 2.0, whereas a score of < 1.0 is thought to exclude cirrhosis. The area under the receiver operator curve (AUROC: graphic representation of sensitivity and specificity) for the APRI index in identifying significant fibrosis or cirrhosis is in the order of 0.80 or 0.89, respectively (9, 10). The larger the area under the curve, the better the diagnostic test, where an area of 1.0 equates to a perfect test with a sensitivity and specificity of 100%. An additional combined marker for serum fibrosis is the FibroTest, which is based on a mathematical algorithm made up of various serological values (box). Scores of 0 to 0.10 are held to exclude significant fibrosis (Metavir F2, F3 or F4), whereas values between 0.60 and 1.00 show at least F2 level fibrosis. These threshold levels yield a negative predictive Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de

MEDICINE

value of 100% and a positive predictive value of > 90% for the diagnosis of advanced fibrosis. The AUROC values for the detection of significant fibrosis were in the region of 0.83 and 0.87 respectively (11). A more recent study reports AUROC values of 0.90 and 0.87 respectively for significant hepatic fibrosis (Metavir F3) or cirrhosis (Metavir F4) (10). These findings are roughly comparable with APRI score data. The Forns index is calculated using the platelet count, the gamma GT, the age of the patient and the cholesterol level (box). A result of under 4.21 effectively rules out significant fibrosis (Metavir F2, F3 or F4), with a high degree of probability (negative predictive value 96%). The positive predictive value and hence the probability of fibrosis in the presence of a value over 6.9 was 79%, significantly lower (12) than for elastography. The Pohl score, another combined serum marker of fibrosis, is derived from the GOT to GPT ratio, in combination with the platelet count (box). A GOT:GPT ratio > 1 and a platelet count below 150 000/L yielded a positive predictive value for advanced fibrosis of 93.1% according to Pohl et al.'s data (Metavir F3/F4). The sensitivity, specificity and negative predictive values were 41.2 %, 99.1 % and 85.0 % respectively. The Pohl score was validated in patients with chronic HCV infection, and is not transferable to patients with alcohol abuse (13).
TABLE 2
Comparison of the diagnostic value of non-invasive methods for evaluating hepatic fibrosis

Non-invasive tests for liver fibrosis Platelets (8) APRI score (8, 22) FibroTest (22) Forns index (12) Pohl Score (13) Transient elastography (10, 1419, 2325)

AUROC values for diagnosing Sensitivity for cirrhosis cirrhosis diagnosis 0.89 0.900.94 0.87 No data No data 0.910.97 77 % 3857 % 50 % 30 % 1841 % 77100 %

Specificity for cirrhosis diagnosis 88 % 8793 % 93 % 95 % 9899 % 8297 %

Transient elastography
Transient elastography is a new imaging modality for detecting hepatic fibrosis. The measuring instrument comprises a computer driven control unit and a measuring probe containing a low frequency vibration emitter together with a 5 Mhz ultrasound probe. Transient elastography is carried out with the patient supine, with his/her right arm behind their head. The measuring probe is positioned at the level of the xiphoid in the right midaxillary line, at 90 degrees to the body (diagram 1). The vibrating probe emits a mechanical signal on the body surface, which generates a low frequency elastic wave of 50 Hz, which is transmitted into liver tissue at a speed of around 1 m/s. Using the high frequency ultrasound wave of 1 500 m/s the transmission velocity of the elastic wave in the liver tissue can be measured (diagram 2). This allows liver stiffness to be measured, which correlates with the degree of fibrosis, a higher degree of fibrosis implying greater stiffness. Stiffness is measured in kPa, with maximal possible values of up to 75 kPa. At least 6 valid measurements should be taken. A single measurement takes no more than 15 seconds. A final reading is derived from the median of the individual readings, in order to exclude outlying measurements. Transient elastography can be performed on an outpatient basis, and is both non-invasive and painless. It is risk free, comparable to ultrasound examination, and can be repeated freely. It has not been validated in children and adolescents under 18. The price of a measuring instrument is currently in the region of 80 000 euros, with minimal maintenance costs. Some early large studies have generated threshold values for the diagnosis of hepatic cirrhosis of between 12.5 and 14.6 kPa with sensitivities of 77 to 87 % and specificities of 91 to 97 % (10, 14, 15). Our own results in 147 patients yield a sensitivity of 90% and a specificity of 82% using a diagnostic threshold value of 13 kPa (table 3). The resulting AUROC curve for patients with advanced fibrosis (Desmet und Scheuer stage F3) was Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de

MEDICINE

DIAGRAM 1

Schematic representation of the positioning of the probe in relation to the area of liver under investigation

DIAGRAM 2

Principles of transient elastography. The spreading of a mechanical wave is measured ultrasonographically.

0.88, and for patients with cirrhosis, 0.92 (95% confidence interval 0.87 to 0.98). As optimal threshold values appear to vary between studies, it seems wise to establish individual threshold values for each newly acquired piece of equipment. Until adequate patient numbers have been investigated to generate these values, it seems advisable to use the lowest threshold values quoted in the literature of 12.5 kPa. Transient elastography is not only helpful in diagnosing hepatic cirrhosis with viral hepatitis. In 380 patients with chronic alcohol abuse, positive predictive values for cirrhosis of 97% were achieved, using a threshold value of 13 kPa (16). A study of 150 patients with primary biliary cirrhosis reports AUROC values of 0.93 (17). In HIV/HCV coinfected patients similarly high AUROC values of 0.97 were reported for a diagnosis of hepatic cirrhosis (18) (table 3). Elastography appears unreliable in differentiating between milder levels of fibrosis, however. AUROC values of between 0.72 and 0.81 have been reported (18). In our own studies, it was impossible to distinguish reliably between F0 and F1, between F1 and F2, and between F2 and F3 fibrosis (Desmet and Scheuer classification) (19). Elastography is also limited in patients with a high body mass index (BMI), primarily due to impeded penetration of the mechanical impulse into tissue. One large study reported invalid results in 4.5% of 2 114 patients (20). Both the univariate and multivariate analyses showed a significant correlation between invalid measurement and a BMI of over 28 kg/m2. Other impediments to transient elastography include narrow intercostal spaces, ascites, space occupying lesions in the liver, acute hepatitis and scarring following liver resection, which can restrict the progress of the elastic wave. The effects of liver fat content, vascular changes or variations in portal pressure on liver stiffness have not been systematically evaluated. Improvements in the sensitivity and specificity of non-invasive techniques for evaluating fibrosis could probably be achieved via a combination of individual methods. In patients with hepatic cirrhosis, the fibrosis test alone generated 80% agreement with the histological findings, for elastography alone this figure was 90%, but with a combination of the two, an Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de

MEDICINE

TABLE 3
Efficacy of transient elastography in the diagnosis of hepatic cirrhosis review of literature referring to liver biopsy

Author Ziol et al., (15) Foucher et al., (14) Castera et al., (10) Melin et al., (16) Pares et al., (17) de Ledinghen et al., (18) Corpechot et al., (23) Erhardt et al., (19)

Number of patients 327 354 183 380 150 72 101 147

Disease HCV Differing etiologies HCV Alcohol abuse PBC HIV/HCV coinfection PBC, PSC Differing etiologies

Threshold value for F = 4 14.6 kPa 17.6 kPa 12.5 kPa 13 kPa 14.45 kPa 11.8 kPa 17,3 kPa 13 kPa

Results Sensitivity 0.86 Specificity 0.96 Sensitivity 77 % Specificity 97 % Sensitivity 87 % Specificity 91 % PPV: 97 % AUROC for F 3: 0.93 AUROC for F = 4: 0.97 AUROC for F = 4: 0.96 Sensitivity 90 % Specificity 82 % PPV: 71 % NPV: 95 % AUROC: 0.92 Sensitivity 100 % Specificity 87 % PPV: 50 % NPV: 100 % AUROC: 0.98 Sensitivity 89 % Specificity 96 % PPV: 80 % NPV: 98 %

Carrion et al., (24)

124

HCV

12.5 kPa

Gomez-Dominguez et al., (25)

103

Differing etiologies

16 kPa

PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; PPV, positive predictive value; NPV, negative predictive value

agreement of 94% was achieved (10). In summary, transient elastography, used in combination with other non-invasive tests and with refinement of mathematical methods, can therefore diagnose cirrhosis of the liver reliably, rendering liver biopsy unnecessary. The marginal value of transient elastography over and above history, examination findings, laboratory results and other imaging, where these are available, has not been studied and remains uncertain.
Conflict of Interest Statement The authors declare no conflict of interests in the terms of the International Committee of Medical Journal Editors. Manuscript submitted on 19 July 2006, final version accepted on 18 January 2007. Translated from the original German by Dr Sandra Goldbeck-Wood.

REFERENCES
1. Bundesgesundheitsbericht fr Deutschland (1998): Chronische Leberkrankheit und -zirrhose. Kap. 5.22. http://www.gbe-bund.de, 1998. 2. Niederau C, Lange S, Heintges T et al.: Prognosis of chronic hepatitis C: results of a large, prospective cohort study. Hepatology 1998; 28: 168795. 3. Bruix J, Sherman M: Management of hepatocellular carcinoma. Hepatology 2005; 42: 120836. 4. Nord HJ: Biopsy diagnosis of cirrhosis: blind percutaneous versus guided direct vision techniques a review. Gastrointest Endosc 1982; 28: 1024. 5. Afdhal NH: Diagnosing fibrosis in hepatitis C: is the pendulum swinging from biopsy to blood tests? Hepatology 2003; 37: 9724. 6. Regev A, Berho M, Jeffers LJ et al.: Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002; 97: 26148. 7. Aube C, Oberti F, Korali N et al.: Ultrasonographic diagnosis of hepatic fibrosis or cirrhosis. J Hepatol 1999; 30: 4728. 8. Lackner C, Struber G, Liegl B et al.: Comparison and validation of simple noninvasive tests for prediction of fibrosis in chronic hepatitis C. Hepatology 2005; 41: 137682.

Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de

MEDICINE

9. Wai CT, Greenson JK, Fontana RJ et al.: A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003; 38: 51826. 10. Castera L, Vergniol J, Foucher J et al.: Prospective comparison of transient elastography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005; 128: 34350. 11. Imbert-Bismut F, Ratziu V, Pieroni L, Charlotte F, Benhamou Y,Poynard T: Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001; 357: 106975. 12. Forns X, Ampurdanes S, Llovet JM et al.: Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 2002; 36: 98692. 13. Pohl A, Behling C, Oliver D, Kilani M, Monson P, Hassanein T: Serum aminotransferase levels and platelet counts as predictors of degree of fibrosis in chronic hepatitis C virus infection. Am J Gastroenterol 2001; 96: 31426. 14. Foucher J, Chanteloup E, Vergniol J et al.: Diagnosis of cirrhosis by transient elastography (FibroScan): a prospective study. Gut 2006; 55: 4038. 15. Ziol M, Handra-Luca A, Kettaneh A et al.: Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology 2005; 41: 4854. 16. Melin P, Dacon A, Gauchet A et al.: Cirrhosis screening in alcoholism consultation using Fibroscan. Hepatology 2006; 42: 753. 17. Pares A, Llorenc C, Lazaro E, Garcia-Criado MA, Navasa M, Gilabert R: Transient elastography: a new and useful non-invasive method for assessing liver damage progression in primary biliary cirrhosis. Hepatology 2005; 42: 679. 18. de Ledinghen V, Douvin C, Kettaneh A et al.: Diagnosis of hepatic fibrosis and cirrhosis by transient elastography in HIV/hepatitis C virus-coinfected patients. J Acquir Immune Defic Syndr 2006; 41: 1759. 19. Erhardt A, Lrke J, Vogt C et al.: Transient elastography for diagnosing liver cirrhosis. Dtsch Med Wochenschr 2006; 131: 27659. 20. Foucher J, Castera L, Bernard PH et al.: Prevalence and factors associated with failure of liver stiffness measurement using FibroScan in a prospective study of 2114 examinations. Eur J Gastroenterol Hepatol 2006; 18: 4112. 21. Hubscher SG: Histological grading and staging in chronic hepatitis: clinical applications and problems. J Hepatol 1998; 29: 101522. 22. Sebastiani G, Vario A, Guido M et al.: Stepwise combination algorithms of non-invasive markers to diagnose significant fibrosis in chronic hepatitis C. J Hepatol 2006; 44: 68693. 23. Corpechot C, El Naggar A, Poujol-Robert A et al.: Assessment of biliary fibrosis by transient elastography in patients with PBC and PSC. Hepatology 2006; 43: 111824. 24. Carrion JA, Navasa M, Bosch J, Bruguera M, Gilabert R, Forns X: Transient elastography for diagnosis of advanced fibrosis and portal hypertension in patients with hepatitis C recurrence after liver transplantation. Liver Transpl 2006; 12: 17918. 25. Gomez-Dominguez E, Mendoza J, Rubio S, Moreno-Monteagudo JA, Garcia-Buey L, Moreno-Otero R: Transient elastography: a valid alternative to biopsy in patients with chronic liver disease. Aliment Pharmacol Ther 2006; 24: 5138.
Corresponding author Prof. Dr. med. Dieter Hussinger Heinrich-Heine-Universitt Dsseldorf Klinik fr Gastroenterologie, Hepatologie und Infektiologie Moorenstr. 5 40225 Dsseldorf, Germany E-Mail: werth@med.uni-duesseldorf.de

Dtsch Arztebl 2007; 104(24): A 17527 www.aerzteblatt.de