Date Published: May 24, 2019
Publisher: Public Library of Science
Author(s): Sosthene Somda, Amandine Lebrun, Hadrien Tranchart, Karima Lamouri, Sophie Prevot, Micheline Njike-Nakseu, Martin Gaillard, Panagiotis Lainas, Axel Balian, Ibrahim Dagher, Gabriel Perlemuter, Sylvie Naveau, Cosmin Sebastian Voican, Pavel Strnad.
The controlled attenuation parameter (CAP) using FibroScan (Echosens, Paris, France) M or XL probe has been developed for liver steatosis assessment. However, CAP performs poorly in patients with high body mass index. The aim of our study was to assess whether CAP is overestimated using the standard XL probe in patients with morbid obesity, and in the case of an overestimation, to reprocess the data at a greater depth to obtain the appropriate CAP (CAPa).
We conducted an observational prospective cohort study on a total of 249 severely obese patients admitted to our institution to undergo sleeve gastrectomy. Patients had a liver biopsy performed during the surgery and a CAP measurement during the 15 days preceding biopsy. Patient files were reprocessed retrospectively by an algorithm, blinded to the patients’ clinical data. The algorithm automatically assessed the probe-to-capsula distance (PCD) by analysing the echogenicity of ultrasound signals on the time-motion mode. In the case of a distance >35 mm, the algorithm automatically selected a deeper measurement for CAP (CAPa). When PCD was less than 35 mm, the measured CAP was considered as appropriated (CAPa) and no further reprocessing was performed.
CAP recording was not performed at a sufficient depth in 130 patients. In these patients, the CAPa obtained at the adapted depth was significantly lower than CAP (298±3.9 versus 340±4.2 dB/m; p< 0.0001) measured at the standard depth (35 to 75 mm). Multiple linear regression analysis revealed that both body mass index and hepatic steatosis were independently correlated with CAP values. After reprocessing the CAP in patients with PCD > 35 mm, steatosis stage was the only parameter independently correlated with CAP values. For the diagnosis of steatosis (S≥1), moderate to severe steatosis (S≥2) and severe steatosis (S = 3), the AUROC curves of CAPa (measured CAP in patients with PCD<35 mm and reprocessed CAP in those with PCD>35 mm) were 0.86, 0.83 and 0.79, respectively. The Obuchowski measure for the diagnosis of steatosis was 0.90±0.013.
CAP was overestimated in a half of morbidly obese patients using an XL probe, but CAP can be performed correctly in these patients after adapting the measurement depth.
Non-alcoholic fatty liver disease (NAFLD) is already a major public health issue with an estimated worldwide prevalence between 25% in general population and more than 80% in populations with obesity and metabolic syndrome . Its burden is expected to rise in the context of ongoing increase of obesity prevalence. NAFLD comprises histological lesions ranging from pure steatosis (NAFL) to steatosis plus necroinflammation (NASH, non-alcoholic steatohepatitis) with or without fibrosis . Steatosis is generally associated with a benign outcome and no mortality increase compared to general population. In contrast, evolution to NASH stage increases the risk of progression to advanced liver disease (cirrhosis and/or hepatocellular carcinoma) and mortality . The presence of steatosis is a mandatory finding to support the diagnosis of NAFLD. An accurate diagnosis of steatosis is therefore important for early identification of NAFLD in risk populations. Liver biopsy is considered to be the gold standard for the evaluation of steatosis. Nevertheless, liver biopsy is an invasive test with risk of complications and low acceptability, unsuitable for screening. Furthermore, steatosis severity may change within just a few weeks of therapeutic intervention, and liver biopsy cannot be suitable for patient follow-up. Therefore, a number of imaging technics have been developed and provide potential alternatives for the diagnosis of steatosis.
The PCD was greater than 35 mm in more than a half of our cohort of patients with morbid obesity. In these patients, the CAP measured at an adjusted depth was lower than that measured at the standard depth of XL probe. Our results suggest that CAP measurements are probably overestimated in morbidly obese patients despite the use of the XL probe. The poor performance and high failure rate of CAP measurement in patients with high BMI was already proved by studies using M probe [8, 15]. Furthermore, CAP measured by M probe showed a low diagnostic value in patients with PCD greater than 25 mm . In accordance with these data, our study finds a poor CAP performance in differentiating steatosis using XL probe in patients with PCD greater than 35 mm. The PCD can be very large in severely obese patients because of the thick layer of subcutaneous adipose tissue. The subcutaneous fat is involved in the measurement using the XL probe in patients with a PCD greater than 35 mm, strengthening the degree of attenuation and overestimating CAP. In a previous study, we showed that presence of nonhepatic tissue in the volume explored by the XL probe also attenuates the transmission of shear waves into the liver and the ultrasonic signals used to measure their speed of propagation, leading to an overestimation of stiffness values .