Ultrasonography and Computer Tomography Evaluation of Hepatocellular Carcinoma with Cytohistopathological Correlation

The prevalence of hepatocellular carcinoma (HCC) in Bangladesh is 35% among all liver diseases. Sonographic examinations were performed for the evaluation of 38 cases of HCC; then, CT examinations were done of these cases subsequently. Expert opinion was taken in each case for both modalities. Age, sex, clinical features, location of the hepatic lesion, multiplicity, echo-character, CT density, and, contrast enhancement were evaluated in all cases. Histocytopathology reports were collected from the patients and were correlated with the ultrasonography and CT findings. Thirty five cases were detected as HCC on ultrasonography and 36 cases in CT scan. In ultrasonography, most of the lesions (82.9%) were found in right lobe, maximum lesions (45.7%) were hypoechoic and lesion showed mosaic pattern in 68.6% cases, lateral shadowing in 34.3% and posterior acoustic enhancement in 45.7% cases. Significant difference found between mosaic pattern and lateral shadowing (p<0.05). On CT scan, majority of lesions (50%) were hypodense, 91.7% lesions were contrast enhanced. Pattern of enhancement was mostly heterogeneous. Both of the modalities found sensitive but CT was found more sensitive, specific and accurate than ultrasonography in detecting HCC.


Introduction
Hepatocellular carcinoma (HCC) causing an estimated 1 million deaths in the world per year.Its incidence in developing country is high.It is the third most malignancy among all types of cancers.Epidemiological study shows that HCC is very common in Asia and Sub-Saharan Africa and uncommon in developed countries 1,2 .In developing countries, incidence rate is two to three-fold higher than in developed countries.Excess liver cancer incidence among men compare to women is universal 2 .It is also found that HCC has an increasing incidence during the last century 3,4 .In Bangladesh its prevalence is 35% among all liver diseases 5 .Study shows HBV contribution 47% in causation of HCC 5 and HCV accounts for 56% 6 .Cirrhosis of liver also is a major (18.75%) contributor to HCC in Bangladeshi population 5 .This is similar to other studies in Himalayan subcontinent [7][8][9][10] .
Significant advances in cross sectional imaging modalities like ultrasonography and CT now allow not only detection but often non-invasive charac-terization of focal and diffuse hepatic processes.Several studies were undertaken regarding the sonographic or CT evaluation of HCC.The sensitivity of real time sonography for detection of small HCCs has been established 11,12 .
Ultrasonography had the highest detection rate (91.2%) for HCC <3 cm in diameter compare to CT (63.2%).For HCC 3-5 cm in diameter, the difference in detection rate becomes smaller which is 92.9% for ultrasonography and 81.8% for CT 13 .
For HCC ≤2 cm in diameter delectability rate for ultrasonography is 52% and that of contrast CT is 56%.For HCC ≤3 cm in diameter delectability rate for ultrasonography is 85% and that of contrast CT is 100% 14 .
As surgical resection is the most effective treatment of HCC, it needs early detection.Prognosis of HCC is extremely poor because of the difficulty in early detection.But small HCC discovered by a mass screening program have a very good prognosis when successfully resected.Therefore periodical follow up of high risk patients by imaging techniques is essential for the early detection of HCCs.
Among the modern imaging modalities, ultrasonography and CT scan was used to detect and evaluate HCCs.This study also evaluate whether ultrasonography can effectively used, in detection and follow up of case of HCC, as screening modality.

Materials and Methods
This prospective study was carried out in the Department of Radiology and Imaging, BSMMU, Dhaka from July 2004 to June 2005 based on an ethically cleared approved protocol.Thirty eight patients having complaints suggestive of HCC from 24 years to 70 years of age of either sex with raised serum alpha-fetoprotein beyond 20 ng/ml were included in this study.All patients were informed about the nature, objectives and procedure of the study in understandable language.Written informed consent was taken with the assurance of keeping the personal information concealed.A pre-tested questionnaire was used for data collection.Patients were evaluated by detailed history, clinical examination with emphasis on hepatobiliary system.Sonographic examination was performed for the evaluation of HCC.Then CT scan was done of the same patient, subsequently.Expert opinion was taken in each case for both the modalities.
Ultrasonographic technique: Every patient had been fasting for 8-10 hours previous to sonographic examination.Ultracarbon tablets were given for minimizing abdominal free gas.Ultrasonography intervention was performed by Sonoline Adora machine (SEIMENS) with 3.5 MHz curvilinear transducer.Scanning was performed with the patient in a supine, right and/left anterior oblique and/ right lateral decubitus position.Patient was asked to hold his/her breath at different phase of respiration and sagittal, transverse and oblique scans were done through sub costal, and intercostal spaces, in a symmetrical fashion to screen the whole liver.Number, site, nature, echogenicity, lobar distribution of the lesion were recorded.
CT technique: CT scan was done after US detection of the HCC.Scan was done by third generation helical CT machine (Hitachi/W2000).CT obtained in cranio-caudal direction with the following parameters-165 effective mA, 120 KVp, 1:1.5 pitch, collimation-5 mm, effective section thickness 3-5 mm at 5-10 mm interval.IV injection was given by 18G needle in ante-cubital vein.Contrast was used Iopamiro 60 ml.CT was interpreted by the researchers' panel and reviewed by professors of the department.Number, nature, density (after and before contrast), and lobar distribution of the lesion was recorded.
Cytopathology technique: Ultrasonography guided aspiration from the lesion was done by 18G needle under full aseptic precaution.Then the aspirate spread in a glass slide and fixed in absolute alcohol.Slides then sent for laboratory examination.The Papaniculaus' stain is done and examined under microscope.
Age, sex, clinical features, location of the lesion, multiplicity, echo character/CT density and contrast enhancement were evaluated in all cases.Data were tabulated and analyzed by SPSS by using computer.Cyto-histopathology reports were correlated with the ultrasonography/CT findings.Sensitivity, specificity, accuracy, along with positive/negative predictive values of both modalities was calculated by standard statistical formulae.
Among 38 cases, 35 cases were detected by ultrasonography as HCC.Most of the patients got right lobe lesions (82.9%).Lesions in left lobe were in 11.4% and in both lobes in 5.7% cases (Table I).II).Ten lesions showed mixed density.After injection of contrast agent 33 (91.7%) lesions were enhanced, which was mostly heterogenous.Three (8.3%) lesions were unenhanced.

Discussion
The purpose of the study was the determine accuracy of ultrasonography and CT in detection and evaluation of HCC.As the patients of HCC have rapidly deterioration courses, early detection is essential for surgical treatment 15 .Early detection of HCC has improved with non invasive imaging modalities, such as ultrasonography, CT and magnetic resonance imaging.
Ultrasonography is one of the useful tool for detecting focal liver diseases and its advantages over other modalities in early detection and characterization of HCC.It is safe, cheap, easily available, can be done rapidly and repeatedly without any radiation hazards.Real time imaging is also possible.
CT is quite valuable in diagnosing and evaluating HCC.However, the presence of isodense tumor should always by kept in mind false positive examination may result from very small lesion, low contrast between tumor and surroundings liver tissue or due to technical failure.In this study there are two false negative in ultrasonography and one in CT.With the spiral CT diagnostic decision tree for the management of HCC has dramatically changed 16 .Ultrasonography, CT, magnetic resonance imaging techniques seem to have a strong potential to improve detection and characterization techniques seen to have a strong potential to improve detection and characterization of HCC 30 .CT has been highly useful in detecting liver tumors and determining their extent 17 .
In this study, out of 38 cases 35 were diagnosed as HCC by ultrasonography and 36 cases were diagnosed by CT.Right lobe of the liver alone involved in 82.9% cases in ultrasonography and 80.5% cases in CT.In ultrasonography 11.4% lesions detected in left lobe, which was 11.1% in CT.Ultrasonography detects 5.7% lesions in both lobes whereas with CT 8.3% cases are detected.
Saad et al 1996 found 59% right lobe lesion and 23% left lobe lesion.This study is not similar to our findings, probably because of variation in patient selection 18 .
In Ultrasonography and CT have been used as modality for screening and evaluation.In regular survillance program it is found that, 75% to 90% of HCCs occur as a single lesion less than 5cm.The sensitivity of ultrasonography for detecting small HCC ranges from 55%-85% and specificity is between 90%-94%.The sensitivity and specificity of CT in detecting HCC varies from size to size of the lesion.Sensitivity of CT is greater than 90% for tumors greater than 1 cm in size.In general, sensitivity of contrast enhanced CT ranges from 53% to 95%.CT scan specificity in detecting HCC is approximately 94% 25 .Sensitivity, specificity, accuracy for CT and ultrasonography found in our study were in alignment with other studies.
CT is found more sensitive and accurate.But ultrasonography is the modality which is easily available in most areas of the country.As a screening modality ultrasonography may be used but for proper characterization of the hepatic lesion CT is necessary.

Table I :
Location and numeral of HCC

Table II :
Distribution of HCC patients by pattern of lesions

Table III :
Comparison of ultrasonography and CT with histocytopathological findings