Intratumoral pseudoaneurysms in hepatocellular carcinoma: do they occur de novo without any prior intervention? A tertiary care center experience of 6 years

Abstract

Background

Intratumoural pseudoaneurysms (ITPA) within hepatocellular carcinomas (HCC) usually arise as postprocedural complications of endovascular therapies or surgical procedures. Their de novo presence in HCC without any prior interventions has not been adequately described in medical literature.

Purpose

To evaluate and quantify the presence of intratumoural pseudoaneurysms (ITPA) within hepatocellular carcinomas (HCC) prior to any intervention.

Material and Methods

Retrospective cross-sectional review of 519 patients with HCC at a tertiary care university hospital with the purpose to evaluate and quantify the presence of ITPA present prior to any therapy. Patients’ baseline data along with viral marker status, alpha fetoprotein (AFP) levels, imaging findings, and any prior treatment provided were recorded. Multi-detector computed tomography (MDCT) scans of selected patients were reviewed for presence of any ITPA and their incidence was calculated.

Results

ITPAs without any prior therapy were found in 5% (25/519) of patients with HCC. Seventeen of 25 (68%) patients had liver cirrhosis while eight of 25 (32%) patients were non-cirrhotic on imaging. Multiple ITPAs were seen in 44% (11/25) of patients. Eight percent (2/25) of patients had pseudoaneurysm-associated hemorrhage, 20% (5/25) had lung metastasis, 12% (3/25) had portal vein thrombosis, 8% (2/25) had hepatic vein thrombosis, and 16% (4/25) had peritumoral hematoma. The incidence of de novo ITPAs occurring in patients with HCC without any prior therapy or intervention was 0.24%.

Conclusion

These cases provide a unique insight into an additional feature of HCC and usefulness of recognizing the ITPAs on imaging studies. Although de novo ITPAs in HCC are uncommon, occurring with an incidence rate of 0.24%, their presence in hypervascular hepatic lesion may point towards the diagnosis of HCC. Additionally, they should be accounted for in management planning as they can lead to complications of rupture and hemorrhage.

Keywords

Adults CT liver neoplasm – vascular aneurysms

Introduction

Hepatocellular carcinoma (HCC) is the most common primary malignancy of liver, resulting in significant morbidity and mortality. The incidence of hepatocellular carcinoma is rising worldwide, but especially in South East Asia (1). Intratumoural pseudoaneurysms (ITPA) within a hepatocellular carcinoma (HCC) usually arise as a postprocedural complication of minimally invasive therapies like radiofrequency ablation (RFA), transarterial chemoemblisation (TACE), percutaneous ethanol injection (PEI), and selective internal radiation therapy (SIRT) (2 –8). ITPA may also occur after surgical procedures (9,10). There is a paucity of data in medical literature regarding the presence of ITPAs arising de novo within HCC prior to any intervention. A recent case report described an ITPA which arose de novo within a HCC (9). This case report provided us the impetus to search for more cases of ITPA in HCC patients. We conducted a retrospective review of our patients diagnosed with HCC with a purpose to evaluate and quantify the presence of ITPAs in HCC prior to any surgical and/or endovascular therapy.

Material and Methods

This is a retrospective cross sectional single-center study. Owing to the retrospective nature of the study, patients’ written informed consent was waived by the ethical review board of our institution. We searched our database to identify all patients with a final diagnosis of HCC between June 2007 and July 2013. For inclusion into the study the final diagnosis was based on either histopathology or a combination of raised alpha fetoproteins levels (400 IU/L or higher), classic imaging appearances (enhancement of lesion on arterial phase and washout of contrast on portal venous phase), and/or clinical grounds. Patients without adequate imaging or inadequate supporting evidence of the final diagnosis of HCC were excluded. The CT scans of the selected patients were reviewed for the presence of any ITPAs. Patients’ baseline data along with viral marker status and any prior intervention or surgical treatment were also recorded. We identified 519 patients with a final diagnosis of HCC who met our inclusion criteria. ITPA were found in 30 cases. This subset was then subjected to detailed analysis.

Imaging protocol and interpretation technique

All available imaging for the 519 patients with HCC were retrieved from the Picture Archiving and Communication System (PACS) of our institution. In all patients CT scan examinations were performed on a 64-slice multidetector computed tomography (MDCT) (Aquilion 64, Toshiba Medical Systems, Otawara, Japan) unit, using a standard biphasic protocol that included an arterial phase scan timed at 25 s from the start of contrast injection and a portovenous phase scan 40 s later. The volume data were reconstructed at 5 mm slice thickness and 2.5 mm reconstruction interval. All patients were given intravenous non-ionic iodinated contrast agent, iohexol (Omnipaque 300, Amersham Health, NJ, USA) at a dose of 3 mL/kg body weight up to a maximum dose of 200 mL in a typical 70 kg patient. Contrast injection was given through an automatic CT controlled injector with a flow rate of 4 mL/s into an antecubital vein. Imaging studies were reviewed independently by two certified radiologists with at least 5 years of experience of abdominal imaging. This was followed by a consensus read. On the CT scans any well-defined rounded, saccular, or globular focal area of enhancement within a hepatocellular carcinoma which had an equal enhancement to the hepatic artery on both arterial and portovenous phases was defined as a “pseudoaneurysm”. It was rated according to its appearance, number, and diameter. If the number exceeded 3, the case was labeled as multiple pseudoaneurysms. The size of HCC and ITPA was measured using electronic calipers using a standard measurement scale in mm from outer border to outer border of both HCCs and pseudoaneurysms. Relevant additional findings such as any extravasation of contrast, adjacent tumor hemorrhage/hematoma, or other findings were also recorded.

Data analysis

Data analysis was performed on the Statistical Package for the Social Sciences (SPSS) software windows version 14 (Chicago, IL, USA). Frequency and percentages were calculated for categorical variables and means and standard deviations were calculated for continuous variables. Incidence was calculated for presence of pseudoaneurysms in HCC without prior treatment.

Results

Our initial search identified a total of 30 patients with ITPAs occurring in HCC. Among these 70% (21/30) of patients were men (age range, 48–94 years; mean age, 65 years) and 30% (9/30) of patients were women (age range, 28–76 years, mean age, 61 years). Seventeen percent (5/30) of patients had history of prior interventions. Three had TACE and two had RFA. In 83% (25/30) of patients there was no history of prior interventions, and ITPAs in these patients were deemed to have developed de novo.

Further detailed analysis of these 25 patients was performed. In 28% (7/25) of these patients there was histopathological proof of HCC while in the other 72% (18/25) patients, diagnosis of HCC was made on a combination of imaging findings, (hypervascular lesion with enhancement on arterial phase and washout of contrast on portal venous phase), raised alpha-feto protein levels (range, 401–4901 IU/L; mean, 1547 IU/L) and clinical findings. The largest HCC measured 147 mm in maximal diameter (range, 37–147 mm; mean, 83 mm). Sixty-eight percent (17/25) of patients had signs of liver cirrhosis while the remaining 32% (8/25) of the patients had no imaging evidence of cirrhotic changes. Twelve of 25 patients (48%) were hepatitis C virus (HCV) positive, 20% (5/25) were hepatitis B virus (HBV) positive. Thirty-two percent (8/25) were negative for both HBV and HCV. All 25 patients had biphasic MDCT images available including arterial and portovenous phases. Forty percent (10/25) of patients had additional ultrasound examinations. In four of 25 patients (16%) conventional celiac angiography was also performed as part of workup and further treatment. Histopathological diagnosis was available in seven of 25 patients (28%) who had no typical features of HCC on MDCT imaging and subsequently underwent ultrasound-guided core biopsy of these lesions. Multiple ITPAs were seen in 11 of 25 patients (44%). Figs. 1 and 2 show examples of multiple ITPAs within HCC. In seven of 25 patients (28%) a single ITPA was found while 20% (5/25) had two ITPAs and 8% (2/25) patients had three ITPAs. Further details are given in Table 1. The largest ITPA measured 6 mm (Fig. 3) (range, 2–6 mm; mean, 2.7 mm). Complication of pseudoaneurysmal hemorrhage was found in two of 25 patients (8%). Twenty percent (5/25) patients had lung metastasis, 12% (3/25) had portal vein thrombosis, 8% (2/25) had hepatic vein thrombosis, and 16% (4/25) had peritumoral rupture and hematoma (Fig. 4). The annual incidence of de novo ITPA occurring in patients with HCC in our set of patients was 0.24%.

Fig. 1.

Multicentric HCC showing multiple intratumoral pseudoaneurysms in patient with liver cirrhosis. (a) Contrast-enhanced (CE) MDCT axial image in arterial phase show multicentric heterogeneously enhancing lesions in segments VII and VIII of liver. Note multiple globular areas of contrast pooling within the lesion representing intratumoral pseudoaneurysms. Black arrow pointing to the largest pseudoaneurysm. (b) CE MDCT axial image in portovenous phase show patchy washout of contrast from multicentric heterogeneously enhancing lesions as well as from the multiple intratumoral pseudoaneurysms (black arrow). Note the liver do not show signs of cirrhosis. (c) Digital subtraction angiography image showing catheter tip in hepatic artery proper. Note abnormal contrast blush in areas of lesions corresponding to MDCT images. There are focal areas of contrast pooling representing intratumoral pseudoaneurysms (black arrow). (d) Digital subtraction angiography magnified image already shown in (c), showing the pseudoaneurysms more clearly (black arrow).

Fig. 2.

HCC showing multiple intratumoral pseudoaneurysms in patient with liver cirrhosis. (a) CE MDCT axial image in arterial phase show heterogeneously enhancing lesion in segments V and VI of liver. Note multiple focal areas of contrast pooling within the lesion representing multiple intratumoral pseudoaneurysms. Black arrow pointing to the most conspicuous pseudoaneurysm. (b) CE MDCT axial image in portovenous phase show patchy washout of contrast from heterogeneously enhancing lesion as well as from the multiple intratumoral pseudoaneurysms (black arrow). Also note that the liver show irregular nodular margins suggesting underlying liver cirrhosis.

Fig. 3.

HCC showing intratumoral pseudoaneurysms in non-cirrhotic patient. (a) CE MDCT axial image in arterial phase show heterogeneously enhancing lesion in segment VII of liver. At least three areas of focal contrast pooling within the lesion representing multiple intratumoral pseudoaneurysms. Black arrow pointing to the largest pseudoaneurysm. (b) CE MDCT axial image in portovenous phase show patchy washout of contrast from heterogeneously enhancing lesion and the focal areas of contrast pooling are less dense representing multiple intratumoral pseudoaneurysms (black arrow). The liver shows regular and smooth margins without signs of underlying liver cirrhosis.

Fig. 4.

HCC showing solitary intratumoral pseudoaneurysm in patient with liver cirrhosis. (a) CE MDCT axial image in arterial phase show heterogeneously enhancing lesion in segments VI of liver. Note single areas of contrast pooling within the lesion representing intratumoral pseudoaneurysm. Black arrow pointing to the pseudoaneurysm present at the edge of the leison. There is perilesional as well as perihepatic hematoma due to aneurysmal rupture marked by star. (b) CE MDCT axial image in portovenous phase show patchy washout of contrast from heterogeneously enhancing lesion as well as from the intratumoral pseudoaneurysm (black arrow). There is perilesional as well as perihepatic hematoma due to aneurysmal rupture marked by star. Also note that the liver appears shrunken with nodular margins suggesting underlying cirrhosis. (c) CE MDCT coronal image in portovenous phase redemonstrating findings of same patient shown in axial images (a, b). (d) Ultrasound-guided biopsy of the lesion performed after resolution of hematoma which confirmed the diagnosis of HCC. Black arrow head pointing to echogenic linear appearing biopsy needle.

Table 1.

Showing details of cases with ITPA in HCC without prior therapy.

Patient no.	Gender	Age (years)	Viral status	Alpha F proteins (IU/L)	Cirrhosis	MDCT	US	Angio	H/P	ITPA No.	ITPA size (mm)	HCC size (mm)
1	M	48	HBV	405	Yes	Yes	No	No	No	1	2.4	51
2	M	79	HCV	476	Yes	Yes	No	No	No	1	2	37
3	M	94	NBNC	616	Yes	Yes	Yes	No	No	1	2.4	67
4	M	85	HCV	628	No	Yes	No	No	No	1	2.9	79
5	M	67	HCV	4910	No	Yes	No	No	Yes	1	3.1	58
6	F	87	NBNC	887	No	Yes	Yes	No	No	1	2	71
7	F	76	HCV	403	Yes	Yes	No	No	No	1	3	64
8	M	69	HBV	879	Yes	Yes	No	No	No	2	3.5*	54
9	F	65	NBNC	659	Yes	Yes	No	No	Yes	2	2.8*	68
10	F	71	HCV	696	Yes	Yes	Yes	No	No	2	1.9*	70
11	M	28	HBV	3312	Yes	Yes	No	No	No	2	3*	61
12	M	60	HCV	783	Yes	Yes	Yes	No	No	2	2.7*	69
13	M	68	HCV	809	Yes	Yes	No	No	Yes	3	2.3*	66
14	M	66	HBV	401	No	Yes	No	Yes	No	3	3*	72
15	M	55	HCV	1201	Yes	Yes	Yes	No	No	Multiple	2.7*	83
16	F	60	NBNC	1037	Yes	Yes	Yes	No	Yes	Multiple	2.3*	88
17	M	56	HCV	2311	Yes	Yes	No	No	No	Multiple	3*	93
18	M	60	NBNC	536	No	Yes	No	No	No	Multiple	5.1*	96
19	M	60	HCV	4009	No	Yes	Yes	No	No	Multiple	2*	134
20	M	58	HBV	3538	Yes	Yes	Yes	No	No	Multiple	2.5*	109
21	F	61	NBNC	2986	Yes	Yes	Yes	Yes	No	Multiple	3.1*	119
22	M	72	NBNC	532	No	Yes	No	No	Yes	Multiple	2.5*	123
23	M	60	HCV	1041	Yes	Yes	Yes	No	Yes	Multiple	4*	147
24	F	60	NBNC	1431	No	Yes	No	No	Yes	Multiple	3*	112
25	M	74	HCV	4201	Yes	Yes	No	No	No	Multiple	3.2*	107

Angio, angiography; HBV, hepatitis B virus; HCV, hepatitis C virus; H/P, histopathology; NBNC, non B non C viral status; US, ultrasound.

Average size of ITPA.

Discussion

Our study describes ITPAs within an HCC which develop without any prior interventions, as evidenced by the 25 cases in our series. Previously pseudoaneurysms within an HCC have been reported as manifestation of interventional procedures like percutaneous biopsy, RFA, TACE, surgical procedures, PEI, and SIRT (2 –8). We found a significant number (n = 25) of patients having such pseudoaneurysms. Our results further potentiate the fact that ITPAs can develop de novo without any prior therapy. Their development is thought to be directly related to vessel injury, vessel erosion, and tumor angiogenesis, rather than being an incidental finding (11,12). Abdominal aortic aneurysms and infrarenal aneurysms are examples of other pretherapy aneurysms found in patients diagnosed with HCC prior to any therapy (12,13). However, these aneurysms point towards isolated events rather than induction by tumor angiogenesis or vascular wall injury. Other tumors that have been associated with pseudoaneurysms include osteochondroma, neurofibromatosis, choriocarcinoma, leukemia, and lymphoma (11,14,15). A case of kaposiform hemangioendothelioma and renal cell carcinoma has been reported recently (16,17). The incidence of pseudoaneurysm caused by malignant tumors is uncertain, however, choriocarcinoma has been reported to account for approximately one-quarter of neoplastic aneurysms (14). Doppler imaging is useful in the diagnosis of ITPAs specially when the size is large but was not performed in our series (18). Large ITPAs can be easily detected on contrast-enhanced CT, whereas small lesions can be easily overlooked and requires angiography for further evaluation (18). In our series angiography was performed in some patients as a part of endovascular treatment. In others angiography was undertaken to treat a complication of an ITPA rupture or hemorrhage. The results of our study suggest that neoplastic pseudoaneurysms in HCC are not rare; their incidence in our series turned out to be 0.24%. The reason that neoplastic pseudoaneurysms have not been reported earlier may be due to the limitations of the available technology. With advent of modern technology especially MDCT, these findings can be seen in more patients with HCC. The recognition of ITPAs is important as they are associated with complication of aneurysmal hemorrhage or rupture which occurred in 8% of our patients. Presence of ITPAs without any prior intervention found in a hyper vascular liver tumor increase the possibility of the lesion being a HCC. It may point towards an aggressive nature and underlying high grade of HCC and may provide a new insight towards prospective future research. In a recent study on enhancement patterns of histopathological proven HCCs by Lee et al., ITPAs were found in 14 of 243 (6%) HCCs. Interesting to note was that 13 of 14 (93%) HCCs were poorly differentiated while the other one (1/14, 7%) was a moderately differentiated HCC. In their study the presence of ITPA was labeled as a highly specific sign for a poorly differentiated HCC (19). Presence of ITPAs may be a helpful finding to suggest diagnosis of HCC if the underlying liver is noncirrhotic and tumor shows typical imaging features of early arterial enhancement and washout on portovenous phase, as in our series about 32% of patients did not show features of cirrhosis on imaging but had HCC and ITPA.

There were many limitations in our study. Two radiologists read all imaging separately and in consensus but no inter-observer variability was calculated. The diagnostic criteria for HCC in all patients were not uniform with the majority of patients being diagnosed on classical imaging appearances and clinical grounds. The absence of histopathological correlate in most of our patient has also not allowed us to correlate the presence of ITPA and the histological grade of the tumor. This aspect may be the subject of a further study.

In conclusion, de novo ITPAs in HCC are not rare and occur with an incidence rate of 0.24%. These cases may provide a unique insight into additional feature of HCC and usefulness of radiographic imaging in recognizing the intratumoral pseudoaneurysms. Their presence in hypervascular hepatic lesions may help point towards the diagnosis of HCC as well as in planning of treatment as ITPAs can lead to complication of rupture and bleeding.

Footnotes

Conflict of interest

None declared.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Farazi

DePinho

. Hepatocellular carcinoma pathogenesis: from genes to environment. Nat Rev Cancer 2006; 6: 674–687.

Chuang

Chen

Tsai

. Hepatic infarction and hepatic artery pseudoaneurysm with peritoneal bleeding after radiofrequency ablation for hepatoma. Clin Gastroenterol Hepatol 2005; 3: A23–A23.

Kodama

Shimizu

Endo

. Complications of per cutaneous transhepatic portal vein embolization. J Vasc Interv Radiol 2002; 13: 1233–1237.

Tamai

Furuse

Maru

. Intrahepatic pseudoaneurysm: a complication following radio-frequency ablation therapy for hepatocellular carcinoma. Eur J Radiol 2002; 44: 40–43.

Richard

3rd Silberzweig

Mitty

. Hepatic arterial complications in liver transplant recipients treated with pretransplantation chemoembolization for hepatocellular carcinoma. Radiology 2000; 214: 775–779.

Ahn

. Pseudoaneurysm formation after catheter dissection of the common hepatic artery: report of two cases. Am J Gastroenterol 1997; 92: 696–696.

Evangelos

Vassiliki

Dimitrios

. Hepatic artery aneurysm complicating intra-arterial chemotherapy for hepatocellular carcinoma. Hepatogastroenterology 2003; 50: 830–831.

Pross

Ridwelski

Reiher

. Hepatic artery aneurysm associated with upper gastrointestinal bleeding after intrahepatic artery chemotherapy. Hepatogastroenterology 1999; 46: 2285–2288.

Chingkoe

Chang

Legiehn

. Hepatic artery pseudoaneurysms arising from within a hepatocellular carcinoma. Br J Radiol 2010; 83: e252–e254.

10.

Shanley

Shah

Messina

. Common splanchnic artery aneurysms: splenic, hepatic, and celiac. Ann Vasc Surg 1996; 10: 315–322.

11.

Smith

Munschauer

Diamond

. Ruptured internal carotid aneurysm resulting from neurofibromatosis: treatment with intraluminal stent graft. J Vasc Surg 2000; 32: 824–828.

12.

Takeuchi

Komori

Ohta

. A simultaneous resection of a concomitant abdominal aortic aneurysm and hepatocellular carcinoma: two cases. Int Surg 2000; 85: 152–157.

13.

Bastounis

Felekouras

Arvelakis

. Abdominal aortic aneurysm and hepatocellular carcinoma: a one-stage approach. HPB 2002; 4: 195–197.

14.

Yang

Yoon

Kim

. Interrenal pseudoaneurysms complicating renal choriocarcinoma metastases: treatment with coil embolization. Clin Imaging 2000; 24: 217–220.

15.

Kim

Kang

. Nontraumatic hepatic artery pseudoaneurysm associated with acute leukemia: a possible complication of pyogenic liver abscess. Abdom Imaging 2002; 27: 458–460.

16.

Garcia-Monaco

Giachetti

Peralta

. Kaposiform hemangioendothelioma with Kasabach-Merritt phenomenon: successful treatment with embolization and vincristine in two newborns. J Vasc Interv Radiol 2012; 23: 417–422.

17.

Tsutsui

Lee

Tomoda

. Intratumoral giant pseudoaneurysm: a rare radiographic finding of renal cell carcinoma. Int J Urol 2011; 18: 256–257.

18.

Marshall

Muiesan

Srinivasan

. Hepatic artery pseudoaneurysms following liver transplantation: incidence, presenting features and management. Clin Radiol 2001; 56: 579–587.

19.

Lee

Kim

. Enhancement patterns of hepatocellular carcinomas on multiphasic multidetector row CT: comparison with pathological differentiation. Br J Radiol 2012; 85: e573–e583.