Transient elastography: The bigger we are, the harder to scan
Version of Record online: 25 JAN 2011
© 2011 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd
Journal of Gastroenterology and Hepatology
Volume 26, Issue 2, pages 217–219, February 2011
How to Cite
Hodge, A. and Kemp, W. (2011), Transient elastography: The bigger we are, the harder to scan. Journal of Gastroenterology and Hepatology, 26: 217–219. doi: 10.1111/j.1440-1746.2010.06589.x
- Issue online: 25 JAN 2011
- Version of Record online: 25 JAN 2011
- Accepted manuscript online: 17 NOV 2010 03:20AM EST
- Accepted for publication 5 November 2010.
Establishing the presence of hepatic fibrosis or cirrhosis is of paramount importance in the management of individuals with chronic liver disease, as it can be useful in guiding treatment as well as predicting liver related complications and mortality. Although liver biopsy has an established role in the assessment of many liver diseases, recognition of its limitations have resulted in the emergence of non-invasive strategies to assess liver fibrosis. Transient elastography (TE) using FibroScan (Echsens, Paris, France) is a non-invasive method of determining liver fibrosis based on the measurement of liver stiffness.1 The technique uses an ultrasound transducer probe to determine the speed of a shear wave emitted from a vibrator. The velocity of the shear wave emitted from the vibrator is proportional to tissue stiffness. TE is rapid to carry out (5–10 min), painless, and because it assesses the liver stiffness from a volume of liver tissue 1 × 4 cm (100 times the size of a core liver biopsy), it is more representative of the hepatic parenchyma. Despite these relative advantages, it is important to recognize and understand the potential limitations of this technology.
To obtain a reading, the ultrasound transducer is placed perpendicular to the skin in an intercostal space in the mid-axillary line with the patient lying in the supine position. Currently, three probe types are available that differ in the depth at which they assess the velocity of the shear wave (S probe, 15–50 mm; M probe 25–65 mm; XL probe 35–75 mm). The liver stiffness is reported in kiloPascals (kPa) and can range from 2.5 to 75 kPa. The validity of a FibroScan assessment is dependent on obtaining a minimum of 10 readings, having a success rate for reading acquisitions of at least 60% and an interquartile range (IQR) to median liver stiffness ratio of less than 30%.
The evidence supporting the use of TE in clinical practice is strongest for the prediction of significant fibrosis and cirrhosis in the chronic hepatitis C population.2 Potential future applications of this technology might extend to a role in the assessment of portal hypertension and to stratify the risk of complications of chronic liver disease, such as varices, decompensation and development of hepatocellular carcinoma.
However, as the global experience with this technology increases, it has become apparent that TE is an ineffective tool for the assessment of hepatic fibrosis among certain patient subgroups. A prospective study of 2114 FibroScan examinations identified failure to measure liver stiffness in 96 cases (4.5%).3 Body mass index (BMI) greater than 28 was identified as the only variable associated with such technical failure of liver stiffness determination. Four years later, the same group reported their experience in a prospective review of 13 369 examinations.4 Almost 1 in 5 scans was uninterpretable and, more specifically, 3.1% were measurement failures (no valid readings obtained); an additional 15.8% of results were unreliable (< 10 valid readings, an (IQR/liver stiffness measurement [LSM] greater than 30%, or a success rate less than 60%). These unreliable or failed assessments were associated with increasing age (> 52 years-of-age), increasing BMI (> 30 kg/m2), presence of type 2 diabetes and operator inexperience (completed less than 500 FibroScan determinations). Similar results were echoed in a Romanian study of 3459 cases reported by Sporea et al.,5 which had a 5.3% failure rate and had 16% unreliable reading results. Furthermore, studies from France6 and China7 had similar failure rates of ∼5%. In this edition of the Journal, Wong et al.8 reviewed the factors limiting FibroScanmeasurements in 3205 Chinese patients. They found both unreliable and failure of LSM rates of 11.6% and 2.7%, respectively. This failure rate of LSM is slightly lower than observed in other FibroScan studies, which might be reflected in the different ethnic populations observed. These studies all implicated obesity as the primary cause for unreliable or failed LSM.
Obesity has consistently been shown to be associated with diminished success of LSM readings. With BMI greater than 28 kg/m2, the odds ratio (OR) for LSM failure is as high as 10.3 The adipose tissue associated with obesity can increase the distance between the FibroScan probe and liver, which increases the likelihood of failure. Although the majority of TE studies use BMI as a marker for obesity, waist circumference (WC) has been shown to more accurately reflect central obesity. This would suggest WC is a more accurate predictor of LSM difficulties. Castéra et al.4 in their French cohort found BMI > 30 kg/m2 had an OR of 7.5 (95% CI 5.6–10.2, P = 0.0001) for LSM failure. In a subgroup analysis of 2835 patients with metabolic syndrome, they found WC was the most important determinant of LSM failure with an OR of 25 (95% CI 7.8–79.3 P = 0.0001). Wong et al.8 also noted central obesity (WC > 80 cm in woman and > 90 cm in men) is an independent predictor for LSM failure (OR 5.8, 95% CI 2.9–11.5). However, BMI ≥ 28 kg/m2 was determined to be the primary predictor of TE difficulty with a 29% failure rate (OR 10.1 95% CI 6.4–14.2, P < 0.0001). These differences are likely accounted for by the differing ethnicities, comorbidities and subsequent different body fat distributions of the patient cohorts between the two studies.
What is not addressed by these studies is whether the number of failed or unreliable readings can be reduced by the use of the XL probe. de Ledinghen et al.9 showed that the number of successful readings in patients with a BMI ≥ 30 kg/m2 could be increased by almost 60% using the XL probe as compared with the M probe. This is supported by our own observations,10 whereby valid LSM could be achieved in 94% of patients by integrating the use of the M and the XL probe in a clinical setting. This has significant implications for the use FibroScan in a Western population, given that the frequency of obesity in countries such as Australia exceeds 20% among adults.
The paper by Wong et al. also identified a potential limitation of FibroScan in patients with low BMI (< 17 kg/m2). This subgroup had higher rates of unreliable or failed LSM compared to those with normal BMI. This finding is likely associated with narrow intercostal spacing in smaller patients that impedes proper probe placement. Narrow intercostal spaces are a known limitation of FibroScan. In clinical practice, various patient maneuvers can be used to widen the intercostal space and allow unobstructed readings.
Several other factors have been shown to limit the performance of TE in the assessment of hepatic fibrosis. Ascites prevents the propagation of shear waves, thereby preventing the acquisition of a liver stiffness. Furthermore, liver stiffness increases during the alanine aminotransferase (ALT) flares of chronic viral hepatitis and during liver injury associated with acute viral, drug related or autoimmune causes.11 An appreciation of the impact of hepatic necro-inflammation on liver stiffness might be critical in the accurate interpretation of TE. Several groups have shown that the performance of FibroScan varies according to ALT levels.12 In addition to these factors, elevated LSM independent of hepatic fibrosis is seen in conditions including cholestasis13 and congestive cardiac failure.14
Despite the aforementioned limitations, TE is gaining popularity throughout the world as a tool for predicting or ruling out cirrhosis, particularly in patients with chronic hepatitis C. It is also gaining acceptance in other chronic liver diseases, and much attention of late has been turned towards staging fibrosis in patients with non-alcoholic fatty liver disease. Obesity is common in this patient group, and is becoming an increasingly prevalent problem in many of our patients with other liver diseases, including hepatitis C. Because obesity accounts for the majority of unreliable or failed LSM, future studies will undoubtedly need to use the XL probe to avoid excluding this important patient subgroup.
In summary, FibroScan has consistently been shown to be superior to other non-invasive assessment techniques in the prediction of advanced fibrosis/cirrhosis.6,15 Transient elastography is quick, reproducible and non-invasive, and thus is likely to be increasingly used as a clinical tool in the assessment of hepatic fibrosis. As our collective experience with FibroScan grows, its role in clinical practice will become further clarified.
- 7Prospective evaluation of transient elastography for the diagnosis of hepatic fibrosis in Asians: comparison with liver biopsy and aspartate transaminase platelet ratio index. Aliment. Pharmacol. Ther. 2008; 28: 51–61., , et al.
- 10Comparison of the M and XL probe in transient elastography using fibroscan. J. Gastroenterol. Hepatol. 2010; 25: A29., , .