Dynamic contrast-enhanced MRI is now widely used as a reference test and has a role in the management of HCC according to the previously mentioned AASLD and EASL practice guidelines (2-4).
DW-MRI is a newly described imaging method and has a potential use in oncologic imaging.
In our study,
in addition to other studies,
we evaluated the use of DW-MRI in the differential diagnosis between DN and HCC nodules in patients with chronic liver disease and showed that high signal intensity on DW-MRI images,
low ADC and ADC ratios,
regardless of serum AFP level and tumor size,
were related to HCC and could be used for discrimination.
Early detection of HCCs and their differentiation from DN in patients with chronic liver disease is crucial for curative treatment procedures (2,
4,).
Differentiating nodules in patients with chronic liver disease by imaging methods has gained importance and dynamic contrast enhanced imaging by computed tomography,
MRI and ultrasonography (US) are evaluated and hyperenhancement on arterial phase and washout of portal venous or late phases are now used for HCC diagnosis without histopathological confirmation.
Among these,
MRI is the generally preferred imaging method because of the use of radiofrequency waves,
low risk of allergic reactions and giving more data by multiplanar imaging.
But the long time for image acquisition,
which can be problem for elderly or claustrophobic patients,
sensitivity to motion of the patient,
intestinal peristaltism or cardiac activity that can produce artifacts on images,
and also the high cost of MRI-contrast material and adverse effects of contrast agents such as nephrogenic systemic fibrosis can limit the utilization of contrast enhanced MRI in liver masses (9).
Recently DW-MRI,
a fast and unenhanced MRI sequence,
was evaluated and it could be shown that HCC nodules were easily detectable and visible by suppressing the signals from the liver parenchyma and vascular structures,
and could be used for differentiation from other benign liver masses (3,
10).
DN,
which is assumed to be a precursor lesion for HCC in patients with chronic liver disease,
were hypo or isointense on DW-MRI in our study.
Xu et al found that most of the DNs were hypo or isointense while most of the HCCs were hyperintense on DW-MRI,
similar to our results (7).
The difference between histological maturation grade and the amount of cells in DNs and HCCs alter the SI on DW-MRI.
Contrast to the surrounding liver parenchyma and microstructure of the HCCs is different because of the increased density of cells,
thickened cellular walls and arterial vascular supply.
Thus,
movement of water molecules is progressively restricted and results in a high SI on DW-MRI while normal appearing liver parenchyma is hypointense (7,
8,
11).
The sensitivity of hyperintensity on DW-MRI varied between 67.5% and 95.2% according to recent studies and it was shown that HCCs were generally hyperintense relative to the surrounding liver parenchyma on DW-MRI (3,
7,
8,
10).
In agreement with previously published reports we obtained a value of 89%.
Although they were significant,
the wide range of sensitivity values might be due to different b factors and sequences of DW-MRIs.
Tumor type,
maturation grade and necrosis could alter the diffusivity of water molecules and this should be kept in mind with oncologic imaging by DW-MRI (12).
Although we did not evaluate and compare the SI properties of different histopathological grades of HCC on DW-MRI,
the histopathological grade of HCCs might affect the SI on DWI and may explain the wide range of sensitivity values.
The SI on DWI is affected by the T2 shine-through effect while the ADC value is not,
therefore the exact value of restriction in diffusion can be assessed by the ADC measurements of the lesions,
which could be used for differentiating nodules in chronic liver disease (7,11).
ADC and ADC ratio values of the HCC nodules were significantly lower than the DN in our study but lower than previously published reports (3,
7).
ADC is a measure of signal loss related to the value of a b factor,
thus this discrepancy may be related to the high value of b factor we used compared to other studies (13).
We found an ADC value of ≤ 1.10 and an ADC ratio of ≤ 0.86 as the best cut off points to differentiate HCCs from DN with high sensitivity,
specifity,
PPV,
NPV values.
Serum AFP level is a screening test for HCC with ultrasonography (US),
with sensitivity values between 41-65% (14).
In our study,
we found a negative and insignificant correlation between HCC size and serum AFP levels and HCC ADC and ADC ratios.
Using serum AFP level for screening HCC has some limitations because of false positive and negative situations.
During pregnancy and for embryonic tumors,
AFP can be positive.
Also some HCCs are AFP negative (15) and could explain the insignificant correlation.
This study had some limitations.
Firstly,
inter and intra observer agreement for SI on DW-MRI and measurement of ADC and ADC ratio were not investigated,
which could show that our results were reproducible.
Secondly,
the histopathological confirmation of diagnosis of HCC and dysplastic liver nodules was made for some of the cases.
But AASLD and EASL practice guidelines for MRI criteria as a reference test for HCC screening and histopathological confirmation can be challenging in clinical practice,
even at a reference hospital.
We could also make more homogeneous groups of HCCs,
as small or large,
and grading of HCCs,
to interpret the effect of necrosis and maturation of tumors on ADC values.
In conclusion,
our results showed that DW-MRI can be used for differentiating HCC and DN in chronic liver disease on the basis of SI pattern and ADC values.
Due to the lack of contrast medium administration and faster imaging,
DW-MRI should be added to dynamic contrast enhanced MRI in patients with chronic liver disease