European Radiology: To compare the predictive power of CT volumetric method and hepatic dynamic CT for liver fibrosis in patients with liver cancer

It is well known that liver resection in the presence of cirrhosis often leads to serious postoperative complications
.
Liver biopsy is the standard method for evaluating liver fibrosis (LF), but has drawbacks such as sampling errors, invasiveness, and complications.

CT volume analysis (CTV) and cellular volumetric integral number (ECV) are highly sensitive and specific tools
for diagnosing LF.
In CTV analysis, the ratio of total liver volume (TLV) to body surface area (BSA), spleen volume (SV) to BSA (SV/BSA), TLV to SV (TLV/SV), right liver to SV (RV/SV), and Couinaud I-III to IV-VIII volume (liver volume ratio: LSVR) all help assess LF grade
.
ECV is an index calculated using hematocrit (Hct), non-contrast CT, and equilibrium CT and has important clinical value for estimating the extent of LF.

ECV requires CT values to be measured by placing regions of interest (ROI), which are manually set by the observer (manual-ECV).

Rigid registration of ECV (rigid-ECV) refers to the alignment between objects with three-dimensional shapes, and only by rotation and translation
.
Nonrigid-registered ECV (nonrigid-ECV) is a method of
aligning objects by deforming them through biomimetic transformations.
At this stage, the use of non-rigidly registered ECVs has attracted great clinical attention
in liver stiffness measurement.

A study published in the journal European Radiology compared the ability of CTV to predict LF using ECV measured using dynamic CT of the liver, and explored the method of manual placement of ROI and the prediction of LF by rigid and non-rigid ECV, providing imaging support
for rapid, non-invasive, and accurate clinical assessment of LF.

This study retrospectively analyzed 90 consecutive patients
receiving CTV and ECV.
The calculation method for manually placing ECVs (manual-ECV), rigid registered ECVs (rigid-ECV), and non-rigid registered ECVs (non-rigid-ECV) in the region of interest is: ECV (%) = (1-hemoglobin) × (ΔHUliver/ΔHUaorta), where ΔHU = flat sweep minus equilibrium phase (240s).

。 Manual-ECV was compared
with CTV in estimating LF.
The ratio of total liver volume to body surface area (TLV/BSA), spleen volume to BSA (SV/BSA), TLV to SV (TLV/SV), right liver volume to SV (RV/SV), and liver segment volume ratio (LSVR)
were measured.
ROC analysis
was performed on ECV and CTV.

After excluding 10 patients, 78 of the 80 patients (97.
5%) had a Child-Pugh score of 5 and 2 (2.
5%) had a Child-Pugh score of 6
.
There are no significant differences
in AUC for ECV between manual-ECV, rigid-ECV, and non-rigid-ECV.
TLV/BSA, SV/BSA, TLV/SV, and RV/SV are more correlated with LF ratings than manual ECV
.
In F0-1 vs F2-4 and F0-2 vs F3-4, the AUC of SV/BSA is significantly higher than that of manual ECV
.
The AUC (0.
76-0.
83) of SV/BSA is higher than that of manual ECVs of all LF grades (0.
61-0.
75), although manual ECVs distinguish between F0-3 and F4
with a higher AUC (0.
75).

CT image
of a 67-year-old woman with liver cancer.
A Axis image
of TLV.
After the raw data is manually sent to the workstation, TLV automatically implements the measurement
.
B-segmented TLV axis bit image
.
The sickle ligament and central hepatic vein are used as markers to separate II, III, and IV
from the medial and lateral sides, respectively.
Axis image
of C SV.
In this workstation (ver5.
5), the SV is measured manually

This study showed that CTV was better diagnostic than ECV
in surgical patients without severe liver dysfunction.
Although ECV has demonstrated the ability to assess cirrhosis (>F4), SV/BSA is a better way to
assess LF.

Original source:

Kenichiro Tago,Jitsuro Tsukada,Naohiro Sudo,et al.
Comparison between CT volumetry and extracellular volume fraction using liver dynamic CT for the predictive ability of liver fibrosis in patients with hepatocellular carcinoma.
DOI:10.
1007/s00330-022-08852-x