|Home | About | Journals | Submit | Contact Us | Français|
Hepatic lesions identified by computed tomography (CT) during arterial portography (CTAP) or CT hepatic arteriography (CTHA) in hepatocellular carcinoma (HCC) patients are sometimes too small to be diagnosed as HCC. We undertook this cohort study to assess whether these small lesions are actually HCC, and to clarify the effectiveness of these imaging examinations in a clinical setting.
We assessed the characteristics of 74 tiny lesions detected by CTAP and/or CTHA, but not by CT in 67 patients.
Seven out of 10 nodules were histologically confirmed as HCC and 18 out of 64 lesions increased in size and showed typical findings of HCC during the follow-up period. Multivariate analysis revealed that the size of the main tumor (>30 mm in diameter) was associated with the presence of tiny additional HCC lesions (P = 0.002).
These findings indicate that CTAP and CTHA are recommended for determining the stage of HCC, especially when the HCC nodule is larger than 30 mm in diameter.
Hepatocellular carcinoma (HCC) ranks the fifth most common cancer in the world . The number of patients with HCC is expected to increase in developed countries . Hepatitis C virus (HCV) infection is one of the major risk factors for HCC in the West and Japan [3–6] and surveillance to detect early HCC in patients with HCV infection is recommended to decrease cancer-related death .
Although patients receive radical treatment, such as resection, liver transplantation and percutaneous ablation, the long-term prognosis is still disappointing owing to a high rate of HCC recurrence. The rates of intrahepatic recurrence at 1, 3 and 5 years after radical treatment are reported to reach 19, 50 and 64%, respectively  and recurrent HCC is a major factor contributing to the poor prognosis. It is possible that some cases of intrahepatic recurrence are due to tiny nodules that cannot be detected by computed tomography (CT) at the time of initial diagnosis.
Recent advances in imaging techniques, such as CT, magnetic resonance (MR) imaging, ultrasonography (US), Doppler US, CT during arterial portography (CTAP) and CT hepatic arteriography (CTHA) have enabled the diagnosis of small HCC [9–12]. Among these imaging techniques, CTAP is one of the most sensitive techniques available for detecting hemodynamic change [13, 14], while its distinct disadvantages include invasiveness, high cost and a high rate of false-positive results [15, 16]. We have frequently observed that some of the hepatic lesions identified by CTAP and/or CTHA in HCC patients are too small to be diagnosed as additional HCC lesions.
Hitherto, the precise evaluation of small nodules detected by CTAP and/or CTHA has not been demonstrated in a clinical cohort study. We undertook this cohort study to assess whether the tiny lesions identified by CTAP and/or CTHA, but not by multi-detector row CT (MDCT), are actually HCC and to clarify the effectiveness of these imaging modalities in a clinical setting.
Between February 2001 and January 2004, 364 patients with HCC were admitted to the Department of Gastroenterology and Hepatology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences. HCC nodules were detected by two imaging modalities consisting of US and MDCT. The diagnosis of HCC was confirmed using US-guided fine-needle biopsy specimens or from typical findings observed during MDCT (hyperattenuation area at the hepatic arterial phase, and hypoattenuation at the delayed phase) [17, 18]. Of these 364 patients, 67 consecutive patients diagnosed with HCC for the first time were enrolled in this study, and these patients underwent CTAP and CTHA during admission, within 4 weeks following MDCT.
Informed consent was obtained from all patients for the use of their clinical data. The study protocol conformed to the ethical guidelines of the World Medical Association Declaration of Helsinki, and was approved by the ethical committees of the institute.
Triphasic spiral CT was performed using MDCT (Aquilion, TOSHIBA, Tokyo, Japan) at an outpatient clinic. The scanning parameters were 120 kVp, 150 mAs, 2-mm section collimation and an 11.0 mm/s table speed, during a single-breath-hold helical acquisition period of 20–25 s, depending on liver size. Images were obtained in a craniocaudal direction and were reconstructed every 5 mm, to provide contiguous sections. The bolus tracking method was used for scanning in each patient. With a power injector, 100 mL of nonionic iodinated contrast agent (iopamidol, Iopamiron 370; Shering, Berlin, Germany) was injected in an antecubital vein, at a flow rate of 4.0 mL/s. The hepatic arterial phase, portal venous phase and delayed phase spiral scans were automatically started 18, 45 and 180 s, respectively after exceeding the contrast enhancement threshold level in the lumen of the descending aorta.
For CTAP and CTHA, we used an interventional radiology system with spiral CT (Infinix Activ, TOSHIBA, Tokyo, Japan). Selective catheterization was performed with right femoral artery punctures, using the Seldinger technique. The 4-Fr DSA catheter was placed in the superior mesenteric artery for CTAP, and in the proper hepatic artery or replaced right hepatic artery arising from the superior mesenteric artery, depending on arterial variance, for CTHA. A total of 100 mL of nonionic contrast material (iopamidol, Iopamiron 300; Shering, Berlin, Germany) diluted with saline (1:1 ratio) was used for CTAP; 30 mL of the same material, at the same dilution ratio was used for CTHA. CT scanning was performed 28 s after initiating injection of the contrast material, at a rate of 3 mL/s for CTAP and 10 s after initiating injection for CTHA at a rate of 1.5 mL/s.
Multi-detector row CT imaging studies were prospectively evaluated independently by two radiologists and two hepatologists, and the number, size, location and vascularity of all nodules were recorded. CTAP and CTHA studies were evaluated by these hepatologists (angiographers) at 1 week after the MDCT evaluations. All lesions, including typical HCC lesions were identified, and the number, size, location and vascularity of each lesion were recorded. When disagreement existed between the readers, a consensus was eventually attained until the discrepancies were resolved. These lesions were too small, and it was difficult to classify under the categories, such as conclusive, probable and non-diagnosis for HCC. The radiologists and the hepatologists were unaware of patient clinical data.
Tiny lesions that were detected by CTAP and/or CTHA, but not by MDCT were then evaluated histologically or in a long-term follow-up. When the tiny lesions were detected by US, tumor aimed biopsies were performed percutaneously under US guidance using a 21-gauge needle at least twice to avoid sampling errors. Histological diagnosis was performed by two pathologists and two hepatologists according to the criteria outlined by an International Working Party.
When the tiny lesions were not detected by US, a follow-up study of more than 12 months duration was conducted involving blood tests, including α-fetoprotein (AFP) and des-gamma-carboxy prothrombin (DCP) each month at outpatient clinics, with US or MDCT performed every 1–3 months. During the follow-up period, when the tiny lesions were observed to increase in size and to show the typical findings of HCC (hyperattenuation at the hepatic arterial phase, and hypoattenuation at the delayed phase) by two dynamic imaging studies, such as MDCT, MR imaging or angiography, they were diagnosed as HCC.
The frequency and predictors of tiny additional HCC lesions detected by CTAP and/or CTHA but not MDCT were analyzed. The Kaplan–Meier method was used to determine the cumulative probability of the appearance of additional HCC during the follow-up period. All parameters with a P < 0.05 in the univariate and multivariate analyses, which were made by means of the Cox proportional hazards model were considered significant to predict the presence of additional HCC lesions. All statistical analyses were performed using the JMP statistical software, version 5.1 (SAS Institute, Inc., Cary, NC, USA).
The patients enrolled in this study consisted of 52 males and 15 females with median age of 65 years (Table 1). Hepatitis B surface antigen (HBsAg) and antibodies against HCV (anti-HCV) were positive for 12 and 49 patients, respectively. Positivity for both HBsAg and anti-HCV was found in one patient and negativity for both in five. Of the 67 patients, 34, 29 and 4 patients belonged to Child-Pugh classification A, B and C, respectively. By MDCT, 41 patients presented a single tumor nodule, while 12, 9 and 5 patients showed 2, 3 and more than 3 nodules, respectively. The median size of the main tumor was 22 mm in diameter and 10 patients showed portal vein tumor thrombus on MDCT. The AFP level ranged from 2 to 11,006 IU/mL, and the DCP ranged from 10 to 66,700 mAU/mL.
One hundred and ninety-two lesions were detected by CTAP, CTHA, or MDCT (Fig. 1). Of these lesions, 116 lesions were detected by both MDCT and CTAP/CTHA (median size 16 mm, range 4–49 mm), 2 lesions were detected by MDCT, but not by CTAP and CTHA (median size 7 mm, range 6–8 mm), and 74 lesions were detected by CTAP and/or CTHA, but not by MDCT (median size 7.7 mm, range 3–26 mm). We evaluated whether these 74 lesions detected by CTHA and/or CTAP alone in 31 patients were actually HCC.
Of these 74 lesions, 10 (median size 11 mm, range 6–16 mm) were identified by US and 7 were histologically confirmed to be HCC (Table 2). The median diameter of these additional HCC lesions was 11 mm (range 11–16 mm). All seven tumors were well-differentiated HCC, and the remaining three lesions were low or high-grade dysplastic nodules. Of these seven additional HCC, five tumors were hypovascular nodules and two were hypervascular nodules.
Of the above 74 lesions, 64 lesions (median size 7.6 mm, range 3–26 mm) were not identified by abdominal US. No nodules showed typical finding of HCC, such as corona enhancement at the late-phase of CTHA. The median follow-up period was 630 days (range 427–1,396 days). Of these lesions, 18 nodules increased in size and began to show typical radiological findings of HCC on MDCT, and were thus diagnosed as HCC (Fig. 2). Two nodules and 16 nodules, respectively, had been hypovascular and hypervascular nodules during CTHA, at the time of initial diagnosis (Table 3). The median size of these 18 additional HCC lesions was 7 mm in diameter (range 3–15 mm). We found that, on a per-nodule basis, 25 of 74 (34%) lesions in 12 patients were determined to be additional HCC lesions.
Positive predictive value, negative predictive value and accuracy of CTAP were 40% (25/63), 100% (11/11) and 49% (36/74), respectively; and those of CTHA were 31% (21/68), 33% (2/6) and 31% (23/74), respectively.
On a per-patient basis, nine pretreatment clinical parameters were analyzed to clarify the predictors of the presence of tiny additional HCC lesions identified by CTAP and/or CTHA, but not by MDCT (Tables 4 and and5).5). In terms of the size of the main HCC nodule during MDCT prior to treatment, 44% (7/16) of patients with an HCC nodule larger than 30 mm in diameter had tiny additional HCC lesions detected by CTAP and/or CTHA, whereas only 10% (5/51) of patients with a tumor smaller than or equal to this size had additional HCC lesions. Univariate and multivariate analyses revealed that the size of the main HCC nodule (>30 mm in diameter) was significantly associated with the presence of tiny additional HCC lesions (P = 0.007 and 0.002, respectively).
Hepatocellular carcinoma staging is important for the decision of treatment procedure, such as resection, liver transplantation, radio frequency ablation, transcatheter arterial chemoembolization and chemotherapy. Underestimation of HCC extension results in early recurrence, and imaging procedures to detect tiny HCC lesions prior to treatment are important. Several studies have reported that CTAP is considered to be the most sensitive technique for the detection of hepatic tumors [13, 14], however, this technique is costly, invasive and has a high false-positive rate [15, 16]. There are several studies indicating that the combined use of CTAP and CTHA for patients with cirrhosis would be helpful in accurate diagnosis of small nodules [19–22], and is beneficial in the preoperative evaluation of patients with HCC . Hitherto, no definitive cohort study using a large number of patients has been reported. The present study was not intended to compare the imaging techniques, but rather to clarify the effectiveness of these imaging techniques, by evaluating tiny lesions detected by CTAP and/or CTHA but not MDCT in a histological confirmation or a long-term follow-up.
In this study, CTAP and CTHA revealed 74 additional lesions, of which 34% (25/74) proved to be HCC in 18% of the patients (12/67), with a median follow-up period of 21 months. However, 46 of the 64 lesions followed up were not confirmed to be HCC. Some of these lesions may be non-tumorous arterioportal shunts which showed wedge-shaped hyperattenuating lesions at the hepatic arterial phase and their attenuation during the portal venous phase never decreased below that of the surrounding liver parenchyma . Furthermore, some of these lesions may be regenerative nodules in cirrhotic liver, which were visualized as enhancing nodules surrounded by lower attenuation tin septa during CTAP and non-enhancing nodules surrounded by enhancing fibrous septa during CTHA .
In cases where the main HCC nodule was larger than 30 mm in diameter, as observed during MDCT, approximately half of the patients had tiny additional HCC lesions. This finding indicates that approximately half of the patients with HCC nodule larger than 30 mm diameter can result in early recurrence following curative treatment by resection or tumor ablation. This data is consistent with a previous histopathological study which reported that larger size of HCC nodule is usually associated with intrahepatic metastatic nodules .
MR imaging is increasingly being used as a diagnostic tool in liver tumors. Imai et al.  suggested that combined dynamic MDCT and superparamagnetic iron oxide (SPIO)-enhanced MR imaging showed a diagnostic accuracy comparable to CTAP and CTHA. However, as they discussed, CTAP and CTHA were superior in detecting HCC lesions smaller than or equal to 15 mm in diameter. Recently, gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MR imaging has been used for the diagnosis of HCC. The detection rate of liver nodules was reported to be higher than that of CT [28, 29]. However, the most lesions they assessed were metastatic tumors and the rate of detection for small HCC lesions (<10 mm) was low. To our knowledge, no study was reported comparing the detective efficacy of HCC lesions between Gd-EOB-DTPA-enhanced MR imaging and CTAP/CTHA. Additional studies are needed to confirm the usefulness of MR imaging for the screening of small HCC lesions.
One limitation of this study is that few patients underwent histological examination. Ten lesions (14%) detected by US were examined for histological confirmation, but most tiny lesions (86%) received a periodical medical check-up using MDCT. A further limitation is that tiny lesions which increased in size very slowly, and did not show the typical findings of HCC during the follow-up period might have been regarded as non-HCC.
In summary, for the pretreatment evaluation of HCC, it is not necessary to perform CTAP and CTHA for all HCC patients. We conclude that for the detection of tiny HCC lesions prior to treatment, CTAP and CTHA are appropriate and superior imaging modalities for HCC staging, especially when the main HCC nodule detected by MDCT is larger than 30 mm in diameter in a clinical setting.