Elsevier

Clinical Radiology

Volume 64, Issue 12, December 2009, Pages 1181-1187
Clinical Radiology

Original paper
Reliability of the bright liver echo pattern in diagnosing steatosis in patients with cryptogenic and HCV-related hypertransaminasaemia

https://doi.org/10.1016/j.crad.2009.06.013Get rights and content

Aim

To evaluate the reliability of the bright liver (BL) echo pattern on ultrasound to detect histological steatosis in chronic cryptogenic hypertransaminasaemia (CCH) and hepatitis C virus (HCV)-related forms of hypertransaminasaemia.

Materials and methods

One hundred and fifty patients, 54 with CCH and 96 with HCV hypertransaminasaemia (76 genotype 1/2 and 20 genotype 3), were enrolled. Histological steatosis was measured as the percentage of hepatocytes involved. The reliability of the BL sign was estimated using the sensitivity, specificity, positive and negative predictive values.

Results

Histological steatosis was present in 102/150 patients (68%) divided into 59/96 (62%) in the HCV group and 43/54 (79.6%) in the CCH group (χ2 = 4.4; p = 0.035). In a multivariate analysis, the variable associated with the BL echo pattern was steatosis percentage (p = 0.0018). Steatosis percentage was higher in CCH group than in the HCV genotype 1/2 and 3 groups (p = 0.02). The sensitivity of the BL echo pattern was 88% in the CCH group [confidence interval (CI) 95% 74–95] versus 61% (CI 95% 44–73) in the HCV genotype 1/2 group. The CI indicates that ultrasound can provide evidence for steatosis in a statistically significant way in the CCH versus HCV genotype 1/2 patients. In the genotype 3 group, the sensitivity was high (90%), but the limited number of cases limited the statistical significance due to the high CI.

Conclusion

In CCH the BL echo pattern has excellent reliability in diagnosing steatosis, better than in HCV hypertransaminasaemia because of the higher prevalence and extent of steatosis.

Introduction

In our geographical area the most frequent causes of chronic hypertransaminasaemia are the hepatitis viruses [C (HCV) and B (HBV)] and alcohol.1 Non-alcoholic and non-virus-correlated hypertransaminasaemias have a prevalence of between 3 and 5%.1, 2 Among these, some have serum markers of autoimmunity (autoimmune liver disease, coeliac disease) or metabolic disorders (Wilson' s disease, haemochromatosis) which help to define their aetiology, while others have no certain aetiological indicators and are, therefore, termed chronic cryptogenic hypertransaminasaemias (CCH). In western countries, most are accompanied by a histological picture of liver steatosis,2 which is referred to as non-alcoholic fatty liver disease (NAFLD) when the aetiological factors are unknown. The frequency of this form is on the increase because of the improvements in economic and living conditions (greater availability of food, more sedentary life style). In some of the more severe cases it can evolve towards end-stage liver disease and, in view of the decrease in the diffusion of HBV and HCV hepatitis, it is likely to become an increasingly important cause of chronic liver disease in the future. Steatosis is also found in a certain quota of chronic hepatitis C patients where its presence depends on viral genotype in some cases (genotype 3), whereas in others, it is independent of genotype and, in fact, conditions the response to antiviral treatment to some extent.3

The real prevalence of steatosis in CCH is not well defined. Ultrasound studies report a prevalence of 24% in the general population and 63% in obese patients, whereas biopsy studies report a prevalence ranging between 66 and 90%.1, 2, 4, 5

The diagnostic reference standard for steatosis is undoubtedly liver biopsy. However, in conditions such as CCH, it should probably only be performed in selected cases,6 and in HCV-related forms its relationship to the evolution of the disease cannot be detected with repeated biopsies; therefore, a non-invasive diagnostic technique would be worthwhile.

For this reason, other diagnostic tools have been sought to diagnose steatosis and among these the most reliable is magnetic resonance imaging (MRI), but its high cost limits its feasibility. Ultrasound is, therefore, the most frequently used technique.7, 8

The bright liver (BL) ultrasound echo pattern is considered to be diagnostic of liver steatosis,5, 6, 7, 8 but its reliability is controversial. In fact, BL has a highly variable reported performance both in terms of sensitivity (40–92%) and specificity (75–100%).9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 Moreover, it has a variable interobserver agreement,13, 14, 17, 20 so that some authors have also concluded that BL is not an expression of steatosis.19, 20

The present study evaluated1 the prevalence of liver steatosis in a patient population with CCH and in HCV-correlated liver disease2; the factors determining the presence of BL3; the diagnostic reliability of the BL echo pattern in detecting histological steatosis in a population of patients with CCH or with HCV-related liver disease.

Section snippets

Patients and methods

In the present prospective study, 232 patients with persistent hypertransaminasaemia for at least 6 months were consecutively selected between January 2004 and December 2007. Only patients with CCH and patients positive for HCV were selected. CCH was diagnosed after being demonstrated monthly for at least 6 months in patients negative for HCV, HBV viral markers, liver-specific auto-antibodies, or anti-transglutaminase (anti-tTG) and anti-endomysium (EmA), with normal serum levels of

Results

Table 1 shows the anthropometrical, serological, clinical, and histological data of the two study groups. In the HCV group, the age and HDL-C were significantly higher (p = 0.037 and p = 0.0001, respectively), whereas the serum levels of GGT (p = 0.038), cholesterol (p = 0.0043), and triglycerides (p = 0.0001) were significantly lower than in the CCH group. Hypertension (p = 0.0001), diabetes mellitus, or fasting glucose serum levels above 110 mg/dl were significantly more frequent in the CCH group (p = 

Discussion

CCH and HCV-associated chronic liver diseases are frequently associated with liver steatosis.1, 2, 4, 5 The most appropriate method for non-invasive diagnosis remains unresolved. Liver biopsy, which is the reference standard, is not always practical, especially in the cryptogenic forms, as they are usually benign, and their diagnosis, generally NAFLD, does not carry any implications for treatment.6 Ultrasound is the most frequently used diagnostic tool because it is a simple, non-invasive, and

Acknowledgements

The authors thank Dr Gaetano Leto for his statistical advice and Carole Greenall for revising the English text. This work was supported by grant of MIUR 60% 2006 to M.S.

References (27)

  • S. Bellantani et al.

    Prevalence of chronic liver disease in the general population of northern Italy: the Dionysos Study

    Hepatology

    (1994 Dec)
  • A. Lonardo et al.

    Hepatitis C and steatosis: a reappraisal

    J Viral Hepat

    (2006 Feb)
  • C. Bersain et al.

    Pathological and virological findings in patients with persistent hypertransaminasaemia of unknown aetiology

    Gut

    (2000)
  • Cited by (18)

    • Ultrasonographic fatty liver indicator detects mild steatosis and correlates with metabolic/histological parameters in various liver diseases

      2017, Metabolism: Clinical and Experimental
      Citation Excerpt :

      In our study, we found five false positive patients with HCV and other five with different liver diseases, almost all with a low level of hepatic inflammation and fibrosis on histology. Nevertheless, in our study a subgroup analysis in patients with HCV (Table S2 and Fig. S3) showed a performance of US-FLI in detecting steatosis that was better than that reported in other ultrasound studies of HCV patients [32–37]. Our study showed for the first time a semi-quantitative ultrasonographic score cut-off (US-FLI ≥ 5) for the detection of severe (> 66%) liver steatosis.

    • Body mass index and liver stiffness affect accuracy of ultrasonography in detecting steatosis in patients with chronic hepatitis c virus genotype 1 infection

      2014, Clinical Gastroenterology and Hepatology
      Citation Excerpt :

      In our study on G1 CHC patients, US had an NPV of 70.3% for the diagnosis of steatosis of 5% or greater on liver biopsy. This value is in line with that reported in the earlier-mentioned small series of CHC patients,12–14 and it reflects the limited diagnostic potential of US for steatosis detection. Furthermore, considering moderate and severe steatosis, the accuracy of US was even lower, as mirrored by an unacceptably low PPV (50.0% for steatosis ≥20%, and 31.7% for steatosis ≥30%), which is also an expression of the low prevalence of steatosis in patients with G1 CHC, despite a good accuracy to rule out steatosis of 30% or greater (NPV, 93.8%).

    • Prospective evaluation of hepatic steatosis in HIV-infected patients with or without hepatitis C virus co-infection

      2012, International Journal of Infectious Diseases
      Citation Excerpt :

      Ideally HS would be assessed by liver biopsy, but because of its invasiveness and risks this may be considered unethical for asymptomatic HIV mono-infected patients. In addition, ultrasonography has demonstrated a higher sensitivity in the detection of HS in patients without known liver disease and obesity,26,40 as was the case for our HIV mono-infected patients with low BMI. However, the sensitivity of US in the diagnosis of HS is lower when infection with HCV genotype 1–2 occurs.

    View all citing articles on Scopus
    View full text