Silvia Colombo

Head-to-head comparison of transient elastography (TE), real-time tissue elastography (RTE), and acoustic radiation force impulse (ARFI) imaging in the diagnosis of liver fibrosis

BackgroundReal-time tissue elastography (RTE), acoustic radiation force impulse (ARFI) imaging, and transient elastography (TE) are new technologies that are used for liver stiffness evaluation. The aim of this study was to compare these methods in the same population and to determine their diagnostic accuracy in the prediction of liver fibrosis.MethodsForty-five consecutive, previously biopsied, patients with chronic liver disease and 27 normal subjects underwent TE, RTE, and ARFI on the right liver lobe. Correlation coefficients between measurements, Metavir fibrosis stage, and histological necro-inflammatory activity (adjusted for fibrosis stage) were evaluated via Spearman’s rank order correlation coefficients. Areas under the receiver operating characteristic curve (AUROCs) were calculated to predict each fibrosis stage.ResultsFailure or inconsistent results occurred in 12.5% of the attempts at TE, but in none of the attempts at RTE and ARFI. The three methods showed high correlation with fibrosis and poor correlation with necro-inflammatory activity. TE and ARFI exhibited high diagnostic accuracy (AUROCs ≥0.9) in diagnosing cirrhosis (F4 Metavir). All three methods presented fair (AUROCs >0.7) to good (AUROCs >0.8) diagnostic accuracy in diagnosing fibrosis (F1–4 Metavir) and significant fibrosis (F2–4 Metavir), with TE showing the best performance (AUROCs were 0.878 for fibrosis and 0.897 for significant fibrosis).ConclusionsTE and ARFI provide high diagnostic accuracy in the diagnosis of cirrhosis. When feasible, TE may perform better than RTE and ARFI in predicting fibrosis and significant fibrosis, but larger studies are needed.

Normal liver stiffness and its determinants in healthy blood donors

BACKGROUND Several studies in healthy populations have investigated normal liver stiffness on transient elastography, but none has excluded subjects with fatty liver. AIMS To define normal liver stiffness and its determinants in 923 healthy voluntary blood donors with and without fatty liver. METHODS Seven hundred and forty six subjects were analyzed with transient elastography according to the absence (602) or presence of fatty liver (144) at ultrasonography. The cut-off for significant fibrosis was a liver stiffness of 7.9kPa. RESULTS Normal subjects had significantly lower liver stiffness (median 4.4kPa) than fatty liver subjects (median 5.3, p<0.001). In normal livers male gender was significantly associated with increased liver stiffness at multiple linear regression analysis. Nine (1.4%) blood donors with normal liver and 9 with fatty liver (6.2%) had >7.9kPa. Subjects with verified liver stiffness >7.9kPa, were further investigated with liver biopsy or non-invasive fibrosis markers: only 1 patient with fatty liver had >F1 fibrosis. CONCLUSIONS Liver stiffness in normal liver is lower than in fatty liver, and gender is the only influencing variable. Transient elastography has a very low false positive rate for significant fibrosis and may have a role in screening populations at risk for liver disease.

The mysterious case of spontaneous disappearance of hepatocellular carcinoma

We describe the case of a 77-year-old woman with chronic hepatitis C and well compensated cirrhosis in whom a single encapsulated 5.5 cm hepatocellular carcinoma was found in the right liver lobe. The patient was symptomatic with left upper quadrant pain and had elevated alfa-fetoprotein levels (3133 ng/ml). While she was waiting for liver resection and 2 months after the initial diagnosis the pain improved and alfa-fetoprotein levels normalized. A computerized tomography scan showed reduction in size of the lesion to 2.5 cm, with no central arterial enhancement, but with the demonstration of a peripheral rim enhancing in all dynamic phases. Follow up computerized tomography and magnetic resonance imaging examinations showed further reduction in size of the lesion to 1.3 cm with persistence of the enhancing rim 20 months after the initial diagnosis. The spontaneous and durable regression of the HCC and the persistent peripheral enhancing rim could be explained by a strong and persistent activation of the immune system directed against the neoplastic cells.

Immunoglobulin M anti-hepatitis D virus in monitoring chronic hepatitis delta.

To the Editor: In his interesting review on hepatitis delta, Wedemeyer (1) reports on the recent Hep-Net International hepatitis D international trial (HIDIT-1) (2). In this study, a 48-week course of peginterferon a-2a had significant anti-viral activity against hepatitis D virus (HDV) and was effective in clearing HDV RNA in 28% of patients. Sustained response was defined as persistent ALT normalization and negative HDV RNA 24 weeks after the end of treatment. HDV RNA, however, may not be the best indicator to monitor the outcome of chronic hepatitis delta, because correlation with disease activity is poor (3). In fact, only 6 out of 14 patients with negative viraemia at week 48 had decreased histological activity and ALT normalization was not paralleled by HDV RNA negativity at week 48 and 72. Moreover, 9 out of 17 sustained responders became HDV RNA negative during the follow-up, indicating that the effect of pegylated interferon in controlling viraemia may be delayed and unrelated to the improvement of disease activity. Immunoglobulin M anti-HDV has been proposed in the past as a surrogate marker for HDV activity (4) and we think it may still be useful in the present. We treated eight patients with chronic hepatitis delta (six with conventional interferon for 96 weeks and two with peginterferon a-2a for 48 weeks). All were IgM anti-HDV positive (DiaSorin, Saluggia, Italy), had increased ALT and demonstrable HDAg in pretreatment liver biopsy. Three patients treated with conventional interferon had a sustained response with durable (7–10 years) ALT normalization and absent or minimal activity on post-treatment liver biopsy. In all cases, IgM anti-HDV became persistently negative 3–6 months after starting treatment. IgM titre fell rapidly in responders, concomitant to ALT normalization, while no change was observed in non-responders. Two other patients with normal ALT and negative IgM anti-HDV were followed without treatment for a period of 3–6 years and did not show any re-activation of the disease. We think that IgM anti-HDV has been prematurely dismissed as a diagnostic tool in chronic hepatitis delta and merits further evaluation as an adjunct to HDV RNA monitoring.