Liver biopsy with histological visualization of hepatocellular fat vacuoles remains the reference method in order to determine the grade of steatosis in chronic liver diseases,
but it is an invasive procedure,
which can study only a small portion of the liver (i.e.
1/50,000 of the total volume) [4,
31].
Discomfort and bleeding are well known procedure–related complications.
In addition to sampling errors,
routine histological examination is semi–quantitative,
observer–dependent,
and grading is performed with broad severity brackets [38].
Therefore,
a non–invasive and objective assessment on a continuous scale may be preferable to biopsy in both clinical practice and research.
Different non–invasive imaging methods,
including US,
CT and MRI,
have been employed to provide an estimate of liver steatosis.
It causes reduced liver attenuation at CT,
resulting in low hepatic density compared to spleen during pre–contrast and portal venous phase imaging [5].
Despite the development of quantitative methods of image analysis to assess the severity of hepatic steatosis with CT [5],
the clinical implementation of this imaging modality is hampered by exposure to ionizing radiation,
which limits its application for repeated measurements in monitoring disease progression [9,
15].
Using B–mode US imaging,
an indirect estimate of hepatic steatosis is obtained by comparing the echogenicity of the liver parenchyma with that of the cortex of the right kidney.
This comparison may be performed in either semi–quantitative (i.e.
normal liver echotexture,
minimal,
mild,
moderate,
severe hyperechogenicity [5]) or quantitative modality (i.e.
hepatorenal index [39]).
Hepatorenal index calculation has been presented as an effective tool for differentiating patients with steatosis from those without steatosis [39],
showing a strong correlation with the histological FF (r=0.71,
p<0.0001).
However,
it has to be kept in mind that a high echogenicity of the liver parenchyma in not synonymous of steatosis.
In fact,
this appearance of the liver at B–mode US may also be related to the presence of parenchymal fibrosis and liver iron overload,
leading to overestimation of the true steatosis grade or misdiagnosis.
MRI–based techniques have been widely employed to determine the presence and grade the severity of hepatic steatosis,
and MRS is regarded as the most accurate non–invasive method for assessing this condition [11–14].
In fact,
FF calculated from spectroscopy–determined proton densities has shown a strong direct correlation with the intracellular triglyceride content [14,
15].
However,
this expensive and time–consuming technique is not widely available and mainly limited to research settings.
Advanced multiecho and multi–interference MRI techniques allow measurement of PDFFs that is corrected for confounding factors,
including B0 inhomogeneity,
T1 bias,
T2* decay and multi–frequency signal interference effects caused by protons in fat [10–13,
18–20].
The most recent studies are giving encouraging results on clinical grounds,
demonstrating a strong correlation between MRI PDFF and hepatic steatosis grade determined by histological validation and proposing MRI PDFF as a valid noninvasive biomarker for assessing liver fat content [22,
24].
In our work,
we performed the quantification of MRI PDFF by means of a comprehensive model derived from that proposed by Bydder et al.
[20],
incorporating correction for T1– and T2* relaxation effects,
B0 inhomogeneity and spectral complexity of fat.
This method of analysis has never been employed in a homogeneous cohort of patients with chronic C hepatitis.
The prevalence of steatosis in chronic C hepatitis is about 40%,
which represents an approximately 2–fold increase compared to the prevalence of steatosis in chronic B hepatitis (i.e.
20%) [1,
40].
According to the literature,
we found in our cohort of patients a prevalence of steatosis of 40.26%.
In HCV–related steatosis,
the percentage of fat–containing hepatocytes is usually mild to moderate (i.e.
10–20%) [35],
as it was observed in our study,
with a median histological FF in patients with relevant steatosis (S≥1) of 15%.
In addition,
we observed a lack of patients with grade 3 steatosis (i.e.
>66% of fat–containing hepatocytes).
The severity of steatosis seems to correlate with the level of HCV replication (i.e.
HCV RNA levels in serum) [3],
and it significantly reduces or disappears when patients are successfully treated with antivirals [41].
Interestingly,
and according to our results,
as the liver disease progresses to cirrhosis (i.e.
F4 METAVIR stage of fibrosis),
there is a trend of reduction of parenchymal steatosis [42],
a phenomenon already observed in NAFLD [43].
Some longitudinal studies underscored the role of steatosis in fibrosis progression.
In a recent study on paired liver biopsies performed in 135 untreated patients with chronic C hepatitis [44],
steatosis was the only independent factor predictive of fibrosis progression.
The progression of fibrosis was significantly related to the percentage of hepatocytes with steatosis [44].
Given the clinical importance of steatosis detection and grading in chronic viral C hepatitis,
we aimed to assess the clinical value of MRI PDFF as a non–invasive biomarker of fatty liver,
finding a significant,
strong correlation of the MRI PDFF with the histological FF.
According to the results of Tang et al.
[10],
we noticed that MRI PDFF values are lower than histological figures,
and MRI PDFF cutoff values to distinguish between different steatosis grades are not comparable with the histological ones.
This is not surprising,
since histologic examination assesses the percentage of fat–containing cells in the biopsy specimens and does not measure the volumetric fat content in a wide portion of liver parenchyma.
With MRI PDFF,
the proportion of mobile protons contained within fat molecules of three–dimensional liver voxels is quantified [8,
10,
12].
Therefore,
MRI PDFF and histological FF assess different aspects of steatosis.
Our study has some limitations.
As mentioned above,
the lack of patients with a grade 3 steatosis may be considered an intrinsic limitation when examining a cohort of patients affected by chronic viral C hepatitis.
Therefore,
we were not able to assess the diagnostic performance of PDFF for discriminating between S0–S2 vs.
S3 patients.
In addition,
we did not find cases of clinically significant MRI–detectable iron overload (i.e.
MRI T2* values <6.3 ms [30]),
and the presence of hemosiderin deposits was appreciable in only few cases.
This may be due to the low number of cirrhotic patients in our cohort; in fact,
it is known that histologically detectable iron is more frequently associated with advanced parenchymal fibrosis and cirrhosis [45].
Therefore,
we were not able to reliably assess the influence of hepatic iron accumulation on the MRI PDFF measurements.
Nevertheless,
we decided to introduce T2* decay as a confounding covariate in the partial correlation model,
finding that its influence on the correlation between MRI PDFF and histological FF was not significant.
A point of strength of our study is that we kept a reasonably low time–interval between MRI,
liver biopsy and TE (<10 days),
thus avoiding any meaningful change in the hepatic fat content during the biopsy–MRI imaging interim.
In addition,
we performed a double check of the influence of parenchymal fibrosis on MRI PDFF measurements,
introducing TE values of liver stiffness in the partial correlation model,
and both TE values and METAVIR stage of fibrosis in the multiple linear regression analysis.
MRI PDFF is a promising technique for the non–invasive assessment of liver steatosis in patients with chronic viral C hepatitis.
In particular,
MRI PDFF has shown a strong correlation with the histological FF,
and this correlation seems to be influenced by neither the stage of parenchymal fibrosis nor the necroinflammatory activity.
In addition,
MRI PDFF allows for discrimination between different histological grades of steatosis with good diagnostic accuracy.
Further studies on larger cohort of patients involving adequate control groups are needed to get a complete clinical validation of this technique in patients with chronic viral C hepatitis.