Hepatic arterial damage after transarterial chemoembolization for the treatment of hepatocellular carcinoma: comparison of drug-eluting bead and conventional chemoembolization in a retrospective controlled study

Abstract

Background

Transarterial chemoembolization (TACE) of hepatocellular carcinoma (HCC) frequently causes feeding artery stenosis or occlusion that may interfere with repeated treatment.

Purpose

To investigate the incidence and predictors of hepatic arterial damage (HAD) after drug-eluting bead-TACE (DEB-TACE) in comparison with conventional TACE (Conv-TACE).

Material and Methods

We retrospectively analyzed 54 patients who underwent DEB-TACE for HCC as an initial treatment with follow-up angiography and 54 patients who underwent Conv-TACE using doxorubicin-lipiodol mixture and gelfoam particles for comparison. HAD was evaluated after a single session of TACE and graded as follows: grade I, no significant wall irregularity; grade II, overt stenosis; grade III, occlusion.

Results

The incidence of HAD was significantly higher in the DEB-TACE group than the Conv-TACE group when analyzed per branch (odds ratio [OR], 6.36; P < 0.001) and per patient (OR, 3.15; P = 0.005). For each HAD grade, the mean doxorubicin dose was greater in the DEB-TACE group than in the Conv-TACE group (P < 0.001, P = 0.053, and P = 0.01 for grades I, II, and III, respectively). In multivariate analysis, risk factors of HAD included mean doxorubicin dose and selective embolization in the Conv-TACE group (P = 0.03 and P < 0.001, respectively) and mean doxorubicin dose in the DEB-TACE group (P = 0.004).

Conclusion

The incidence and grade of HAD were higher after DEB-TACE compared to Conv-TACE with doxorubicin dose as a possible risk factor. HAD was independent of overall survival in both groups.

Keywords

Drug-eluting bead transarterial chemoembolization doxorubicin hepatic arterial damage

Introduction

Hepatocellular carcinoma (HCC) is the most common primary liver cancer (1) and frequently occurs in the setting of chronic liver disease and cirrhosis. Thus, treatment selection is determined by tumor extent, multiplicity, and hepatic function reserve (2). Transarterial chemoembolization (TACE), which induces tumor ischemia by disrupting the blood supply to the tumor, has been utilized successfully for the treatment of unresectable HCC (3 –5). To decrease collateral damage to tumor-free parenchyma and increase the effect of the treatment, a selective arterial approach is preferable to a non-selective approach (6). However, TACE may cause hepatic arterial damage (HAD), such as stenosis and occlusion. Such arterial complications may interfere with selective catheterization during repeated TACE sessions (7,8).

Recently, embolic drug-eluting beads (DEBs) have been developed as drug delivery agents for TACE. After delivery of the beads and loaded drugs, such as doxorubicin or epirubicin, to target lesions, the beads elute the drugs in a controlled manner over a prolonged period of time (9 –11). Because DEB-TACE reduces the amount of the drug entering systemic circulation compared with conventional TACE (Conv-TACE), fewer drug-related adverse events such as post-embolization syndrome develop and more of the chemotherapeutic agent can be used without increasing the risk of adverse events (12 –15).

Maeda et al. reported that the incidence and the degree of HAD after Conv-TACE for HCC are associated with impaired liver function and a higher dose of the chemotherapeutic agent (16). In DEB-TACE, a greater amount of doxorubicin and permanent embolic property of the DEBs could increase the risk of hepatic arterial complications. However, a limited number of reports on HAD after DEB-TACE are currently available (17). The purpose of this study was to investigate the incidence and predictors of HAD after DEB-TACE in comparison with Conv-TACE.

Material and Methods

Participants

Between March 2010 and February 2012, 171 consecutive patients were treated with DEB-TACE for HCC. Among them, 93 patients were newly diagnosed with HCC based on the diagnostic criteria used by the American Association for the Study of Liver Disease and were treated with DEB-TACE as an initial treatment (18). The remaining 78 patients were excluded because of previous treatments (DEB-TACE in 49 patients; Conv-TACE in 12; resection in 17) or presence of distant metastasis (n = 6). A total of 54 patients were available for evaluation of HAD on follow-up angiography of the second TACE session. During the same period, 54 participants who underwent Conv-TACE as an initial treatment for HCC and follow-up angiography were selected as the control group. Patients in the Conv-TACE group were matched for age, sex, and Child-Pugh class with those in the DEB-TACE group (Fig. 1). No patient received sorafenib before the second TACE session. Our retrospective analysis of data was approved by the Institutional Review Board and the hospital’s ethics committee for all patients and the need for informed consent was waived.

Fig. 1.

Flow chart of patient selection.

TACE was selected as an initial treatment based on a multidisciplinary HCC conference involving radiologists, hepatologists, and surgeons. Selection of the embolic material used for TACE was based on the preference of the referring physicians and the patients’ economic status, as DEB was not yet fully reimbursed under our medical insurance scheme, rather than according to the characteristics of HCC. Patient characteristics are shown in Table 1. The mean patient age, sex, Child-Pugh class, incidence of viral hepatitis, maximal tumor size, and UICC stage were not significantly different between two groups.

Table 1.

Baseline characteristics of patients and procedures.

	Conventional group	DEB group	P value
Age (years)*	61.6 ± 10.4 (31–89)	63.3 ± 10.4 (34–88)	0.39^†
Sex (male/female)	45/9	44/10	0.80^‡
CP class (A/B/C)	47/7/0	51/3/0	0.18^‡
Hepatitis (B/C/non-B, non-C)	38/8/8	37/6/11	0.68^‡
Maximal tumor size (cm)	4.09 ± 2.52	3.34 ± 2.17	0.11^†
UICC stage (I/II/III/IV)	23/23/8	22/19/13	0.45^§
F/U interval (days)**	170 ± 116.8	214 ± 188.2	0.15^†
Targeted segments (patients number)			0.06^‡
1 segment	28	31
2 segments	10	15
3–4 segments	7	7
5–7 segments	9	1
Total	131 segments	90 segments
Level of embolization (patients number)			<0.001^††
Subsegmental	28	33
Segmental	18	21
Lobar	8	0
Doxorubicin dose per patient (mg)**	42 ± 14	65 ± 38	<0.001^†,††
Doxorubicin dose per branch (mg)**	17 ± 13	39 ± 23	<0.001^†,††

Data are mean values ± standard deviations. Numbers in parentheses are ranges.

†

Student’s t-test.

‡

χ² test.

Fisher’s exact test.

Data are mean values ± standard deviations.

††

Significant difference.

CP class, Child-Pugh class; DEB, drug eluting bead; UICC, Union for International Cancer Control.

Protocol of DEB-TACE and Conv-TACE

Before TACE, angiography of the superior mesenteric artery was performed to check for portal vein patency and hepatopetal portal flow. Angiography of the celiac artery and common hepatic artery were then performed to map the arterial anatomy and identify arterial feeders of the HCC (contrast volume, 30 mL for celiac arteriogram/15–20 mL for common hepatic arteriogram; tube angulation, anterior–posterior view, and 30° right anterior oblique view; image frequency, 2 frames/s). The feeding artery was catheterized as selectively as possible with a microcatheter (2.2-French Progreat β, Terumo; or 2.2-French Stride, Asahi, Japan) and embolization was performed. If the feeding artery was not clear on plain digital subtraction angiography (DSA), C-arm rotational angiography and cone-beam CT (Axiom Artis; Siemens, Erlangen, Germany) were performed with placement of a 5-French catheter at the common hepatic artery. In the case of multiple tumors involving more than three segments in a unilateral lobe, the lobar artery was catheterized. For DEB-TACE, the amount of doxorubicin and DEB (DC Bead; Biocompatibles, Farnham, UK) was chosen based on the extent of the tumor burden. Typically, 2 or 4 mL of DEB with a diameter of 100–300 or 300–500 µm was loaded with doxorubicin (35 mg/mL of DEB) and non-ionic contrast medium (Visipaque; GE Healthcare, Princeton, NJ, USA). If the HCC was larger than 5 cm with rich vascularity, DEB with a diameter of 300–500 µm was used. The injection rate was very slow to avoid reflux of DEB to adjacent arteries. The end point of embolization was near stasis of blood flow during at least 10 s and clearing of residual tumor staining. For Conv-TACE, a mixture of 3–20 mL of iodized oil (Lipiodol; Laboratoire Guerbet, Aulnay-Sous-Bois, France) and doxorubicin (maximum 70 mg) was infused into the selected feeding arteries. Subsequent embolization was performed using absorbable gelatin sponge particles (Gelfoam; Upjohn, Kalamazoo, MI, USA) cut into 1 mm² sized pieces. Injection was continued until target arterial flow ceased and pressure was felt on infusion. Second angiography and TACE was performed when a residual or recurrent HCC was found on follow-up CT or MRI.

Evaluation of hepatic arterial damage

HAD was evaluated on follow-up angiography of the second TACE session. HAD was graded in seven subsegmental branches of the hepatic artery, but not in the caudate lobe branch. Hepatic arteries were evaluated in consensus by two radiologists who were blinded to the initial treatment method by comparing the initial and follow-up angiography data. HAD grades were in accordance with the criteria proposed by Maeda et al. as follows: grade I, no or slight wall irregularity; grade II, overt stenosis (Fig. 2); and grade III, occlusion (Fig. 3). Grades II and III were regarded as significant HAD (16). Additionally, “overflow damage” was defined as HAD proximal to a selected branch for chemoembolization. We took the follow-up angiogram into account for overflow damage when HAD was noted in non-embolized segments of the hepatic artery proximal to the target branches (Fig. 4).

Fig. 2.

Grade II (overt stenosis) hepatic artery damage. A 79-year-old woman with HCC underwent DEB-TACE of the A8 hepatic artery. The arteriogram performed at the initial session (a) and follow-up after 6 weeks (b) show development of overt stenosis, grade 2 hepatic arterial damage of the A8 after TACE.

Fig. 3.

Grade III (total occlusion) Hepatic arterial damage. A 55-year-old man with HCC underwent DEB-TACE for multiple HCCs. The arteriogram performed at the initial session (a) and follow-up after 8 weeks (b) show development of total occlusion, grade 3 hepatic arterial damage of the A8 after TACE.

Fig. 4.

A 57-year-old woman with HCC underwent DEB-TACE of the A2 hepatic artery; Overflow damage. (a) The initial hepatic arteriogram showed patent A2 artery that supplied HCC located in segment 2. (b) Infusion of DEB was done at distal portion of A2 artery after passing the microcatheter through the proximal segment of the A2 artery. (c) Follow-up hepatic arteriogram after 20 months shows overflow damage of entire A2 artery with grade III total occlusion (arrows).

Statistical analysis

To compare the Conv-TACE group and the DEB-TACE group, patient and procedural characteristics including sex, Child-Pugh class, risk factors for HCC, UICC cancer stage, number of targeted segments, level of embolization, and HAD grade per branch and per patient were analyzed by the Chi-squared test or Fisher’s exact test. Student’s t-test was used to analyze continuous variables including age and doxorubicin dose per branch and per patient for each group of HAD grades. To investigate predictors of significant HAD, univariate association between HAD and factors including age, sex, Child-Pugh class, follow-up period, maximal tumor size, level of embolization, and doxorubicin dose was assessed by logistic regression analysis. On the basis of univariate analysis, potentially significant parameters including Child-Pugh class, level of embolization, and dose of doxorubicin were tested for possible interrelationships with multivariate logistic regression. Kaplan–Meier curves of overall survival (OS) were compared by log-rank test according to TACE method and presence of significant HAD in each TACE group. Statistical analyses were performed using SPSS 20.0 software (SPSS Inc. Chicago, IL, USA). P values less than 0.05 were considered to indicate a statistically significant difference.

Results

TACE procedure

The procedural characteristics are shown in Table 1. There was no significant difference in the mean time interval from the first TACE session to follow-up angiography or in the number of targeted segments between the Conv-TACE group and the DEB-TACE group. The mean dose of doxorubicin per patient was significantly greater in the DEB-TACE group than in the Conv-TACE group (70 ± 43 mg versus 41 ± 13 mg, respectively, P < 0.001). The mean dose of doxorubicin per branch was also significantly greater in the DEB-TACE group than in the Conv-TACE group (39 ± 25 mg versus 18 ± 14 mg, respectively, P < 0.001). There was a significant difference in the level of embolization between the two groups (P < 0.001).

Incidence of hepatic artery damage after TACE

Among 378 subsegmental arteries of each group of patients, 131 subsegmental arteries in the Conv-TACE group and 90 subsegmental arteries in the DEB-TACE group were targeted for chemoembolization. Table 2 shows the incidence of HAD after Conv-TACE and DEB-TACE. Both grade II and grade III HAD developed more frequently in the DEB-TACE group. The incidence of significant HAD was significantly higher in the DEB-TACE group than the Conv-TACE group (per branch: odds ratio [OR], 6.36; 95% confidence interval [CI], 3.51–11.55, P < 0.001; per patient: OR, 3.15; 95% CI, 1.39–7.17, P = 0.005). Overflow damage was observed in 7/54 patients (13.0%) in the Conv-TACE group and 21/54 patients (38.9%) in the DEB- TACE group, with statistical difference (OR, 4.70; 95% CI, 1.83–12.1, P < 0.001).

Table 2.

Hepatic arterial damage after conventional TACE and DEB-TACE.

	Conventional group			DEB group			P value*
Grade	I	II	III	I	II	III
Per branch	93 (71.0%)	19 (14.5%)	19 (14.5%)	25 (27.8%)	22 (24.4%)	43 (47.8%)	<0.001^†
Per patient	27 (50.0%)	12 (22.2%)	15 (27.8%)	13 (24.1%)	10 (18.5%)	31 (57.4%)	0.005^†

Numbers in parentheses are percentages.

χ² test comparing groups without stenosis (HAD grade I) and significant stenosis (HAD grades II and III).

†

Significant difference.

DEB, drug eluting bead; TACE, transarterial chemoembolization.

Dose of doxorubicin

Table 3 shows the dose of doxorubicin per branch for each HAD grade of the Conv-TACE group and the DEB-TACE group. The mean doxorubicin dose of the Conv-TACE group was 13 ± 10 mg in grade I HAD, 26 ± 16 mg in grade II HAD, and 31 ± 13 mg in grade III HAD. The mean doxorubicin dose of the DEB-TACE group was 34 ± 15 mg in grade I HAD, 36 ± 17 mg in grade II HAD, and 44 ± 29 mg in grade III HAD. For all HAD grades, the mean doxorubicin dose was significantly greater in the DEB-TACE group than the Conv-TACE group (P < 0.001, 0.053, and 0.01, respectively).

Table 3.

Comparison of doxorubicin dose in each grade of hepatic arterial damage after conventional and DEB-TACE.

Grade	Conventional group (mg)	DEB group (mg)	Difference (mg)	95% CI	P value*
I	13 ± 10	34 ± 15	21	16 ∼ 26	<0.001^†
II	26 ± 16	36 ± 17	10	0 ∼ 21	0.053
III	30 ± 13	44 ± 29	14	3 ∼ 25	0.01^†

Student’s t-test.

†

Significant difference

CI, confidence interval; DEB, drug eluting bead; TACE, transarterial chemoembolization.

Risk factor analysis

Table 4 shows the results of logistic regression analysis in the Conv-TACE group. In both univariate and multivariate analyses, subsegmental artery selection and dose of doxorubicin were statistically significant predictors (P < 0.001 and <0.001 in univariate analysis; P < 0.002 and 0.030 in multivariate analysis, respectively).

Table 4.

Results of univariate and multivariate logistic regression analysis in the conventional TACE group.

	Univariate analysis		Multivariate analysis
Variable	OR (95% CI)	P value	Adjusted OR (95% CI)	P value
Age	0.98 (0.95–1.01)	0.13	–	–
Male sex	0.86 (0.29–2.57)	0.78	–	–
Child-Pugh Class B	1.67 (0.27–10.40)	0.58	1.08 (0.13–9.41)	0.94
Maximal tumor size	0.96 (0.83–1.10)	0.53	–	–
Subsegmental TACE*	14.5 (5.42–38.85)	<0.001	6.70 (2.05–21.88)	0.002
Dose of doxorubicin (10 mg)	2.4 (1.68–3.40)	<0.001	1.57 (1.05–2.34)	0.03

Each result is expressed as the increase in the odds of having hepatic arterial damage associated with the variable. An OR > 1.0 indicates an increased risk of hepatic arterial damage. Numbers in parentheses are ranges of 95% CI.

Versus selective and non-selective artery selection.

TACE, transarterial chemoembolization.

Table 5 shows the results of logistic regression analysis in the DEB-TACE group. No factors were significant in univariate analysis. In multivariate analysis, the dose of doxorubicin was a significant predictor of HAD (P = 0.05).

Table 5.

Results of univariate and multivariate logistic regression analysis in the DEB-TACE group.

	Univariate analysis		Multivariate analysis
Variable	OR (95% CI)	P value	Adjusted OR (95% CI)	P value
Age	1.00 (0.95–1.05)	0.95	–	–
Male sex	1.58 (0.47–5.31)	0.46	–	–
CP Class B	2.88 (0.60–13.89)	0.19	3.11 (0.63–15.23)	0.16
Maximal tumor size	0.95 (0.78–1.16)	0.63	–	–
Subsegmental TACE*	0.50 (0.15–1.65)	0.25	0.30 (0.08–1.11)	0.07
Dose of doxorubicin (10 mg)	1.21 (0.93–1.58)	0.15	1.34 (1.00–1.80)	0.05

Versus selective and non-selective artery selection.

DEB, drug eluting bead; TACE, transarterial chemoembolization.

Survival analysis

Median follow-up time among survivors was 36.8 months (interquartile range, 21.0–43.7 months). During follow-up, the dropout rate was 9.2% in the DEB-TACE group and 11.1% in the Conv-TACE group. The Kaplan–Meier curves for different TACE methods and the presence of significant HAD are shown in Fig. 5. The cumulative OS at 1 and 3 years was 96% and 57% in the Conv-TACE group compared with 94% and 62% in the DEB-TACE group. Among the Conv-TACE group, OS at 1 and 3 years was 96% and 44% in patients without significant HAD compared with 96% and 70% for patients with significant HAD. In the DEB-TACE group, OS at 1 and 3 years was 100% and 75% without significant HAD compared with 93% and 58% with significant HAD. However, log-rank tests revealed no significant differences in OS between the groups.

Fig. 5.

Kaplan–Meier plots for overall survival after TACE for (a) the DEB-TACE group and the conventional TACE group; (b) the patients without HAD and the patients with HAD in the conventional TACE group; (c) the patients without HAD and the patients with HAD in the DEB-TACE.

Discussion

The aim of DEB-TACE is greater prolonged intra-tumoral drug retention with low-level release of the drug into the systemic circulation. Several studies have confirmed the superior pharmacokinetic profile of DEB-TACE, which allows a higher safety dose for therapy (9 –11). In several studies, DEB-TACE offered higher rates of treatment response compared to Conv-TACE. The local recurrence or progression rate, however, could not be ignored (12,13,19 –21) and the possibility of repetitive TACE should be considered even after a successful initial TACE.

Most patients with HCC require repeated TACE sessions. Thus, patency of the targeted hepatic arteries is important for the delivery of chemotherapeutic and embolic agents. In addition, compromised hepatic arterial perfusion can induce ischemic liver or biliary injury, especially in patients with decreased liver function (22 –24). HAD is related to toxic arteritis induced by chemotherapeutics, and histologic findings of previous reports revealed intimal fibrosis and chronic inflammation of hepatic arteries. As arteritis was found beyond the tip of the catheter, it was assumed that the drug played a major role in producing the damage (7,25). In our study the patient characteristics, except for dose of doxorubicin and the use of bead or gelatin sponge, did not vary between the two groups. The mean dose of doxorubicin was higher in the DEB-TACE group than the Conv-TACE group because the recommended dose of doxorubicin is much higher in DEB-TACE. A hypersensitivity reaction to the gelatin sponge is known to cause arteritis after TACE (26). However, none of the patients had allergies to porcine collagen or significant eosinophilia in the Conv-TACE group.

A previous study reported that significant HAD after Conv-TACE occurred in 16% of the branches and 48% of the patients (16). In our study, the incidence of significant HAD after Conv-TACE was similar to the results of this report. In comparison, the incidence of significant HAD after DEB-TACE was more than twice that reported previously. The permanent embolic effect of DEB, which is in contrast to the transient embolism induced by the gelatin sponge, might be a cause of the higher proportion of stenosis or occlusion in the DEB-TACE group. However, we cannot conclude that the use of DEB itself induces HAD as the mean dose of doxorubicin was significantly greater for DEB-TACE than for Conv-TACE in each HAD grade. Furthermore, there were trends toward increasing HAD grade with increasing doxorubicin dose in both groups with statistical significance. Reduction of the loading dose of doxorubicin in DEB-TACE should be considered for patients who are expected to undergo repeated TACE as long as it does not affect the efficacy of treatment.

In our study, subsegmental selection was a significant risk factor of HAD after Conv-TACE with a high OR. Subsegmental selection of the smaller feeding artery can lead to direct mechanical damage of the arterial wall by the catheter and the guidewire, which might be related to HAD. However, in the DEB-TACE group subsegmental selection was not a risk factor for HAD, and even with the higher proportion of subsegmental selection in the DEB-TACE group the dose of doxorubicin was still an important risk factor in the multivariate analysis. HAD in subsegmental TACE might be attributed to other factors such as the chemical characteristics of lipiodol or gelfoam rather than direct mechanical injury. Although there was no statistical significance, survival analysis showed a better prognosis in the patients with HAD compared the patients without HAD in the conventional TACE group, which is contrary to the DEB-TACE group. We think 54 patients of each group were not enough and more cases and longer follow-up periods are needed to assess the impact of HAD on the survival of the patients with HCC.

If the procedure follows the TACE protocols, the proximal part of the target artery should not be exposed to doxorubicin and embolic agent during the procedure. However, overflow damage happened frequently, especially in DEB-TACE. The higher incidence of overflow damage in the DEB TACE group might be due to the pharmacokinetic characteristics of the DEBs. Sustained release and a higher dose of doxorubicin with DEB-TACE might make the proximal artery vulnerable to arteritis and damage. Moreover, the permanent embolic effect of DEB could induce prolonged stasis of blood flow and thrombosis of the proximal artery.

There are several limitations to the current study. First, HAD grades were analyzed in a retrospective manner as we could only access the still images saved in the PACS system. Therefore, the possibility of over- or undergrading cannot be ignored. Second, the procedures were decided according to the preference of the referring physician and patient economic status. Although we selected the 54 patients of the Conv-TACE group to match the various conditions, selection bias is still present and because patients with advanced stage disease had a tendency to bear the greater cost of DEB-TACE, the real results of survival analyses might be masked by this bias. Third, the therapeutic effect of DEB might be concealed by the higher incidence of HAD after TACE. Although there was no statistical significance, survival analysis showed a better prognosis in patients without significant HAD and in the DEB-TACE group.

In conclusion, the incidence and grade of HAD were higher after DEB-TACE than after Conv-TACE. The possible risk factors for HAD are doxorubicin dose in DEB-TACE, and doxorubicin dose and level of selection in Conv-TACE. Despite the importance of arterial patency for repeated TACE, HAD was independent of OS in both groups.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Jemal

Bray

Center

. Global cancer statistics. CA Cancer J Clin 2011; 61: 69–90.

Vauthey

Dixon

Abdalla

. Pretreatment assessment of hepatocellular carcinoma: expert consensus statement. HPB (Oxford) 2010; 12: 289–299.

Brown

Nevins

Getrajdman

. Particle embolization for hepatocellular carcinoma. J Vasc Interv Radiol 1998; 9: 822–828.

Covey

Maluccio

Schubert

. Particle embolization of recurrent hepatocellular carcinoma after hepatectomy. Cancer 2006; 106: 2181–2189.

Morse

Hanks

Suhocki

. Improved time to progression for transarterial chemoembolization compared with transarterial embolization for patients with unresectable hepatocellular carcinoma. Clin Colorectal Cancer 2012; 11: 185–190.

Nishimine

Uchida

Matsuo

. Segmental transarterial chemoembolization with Lipiodol mixed with anticancer drugs for nonresectable hepatocellular carcinoma: follow-up CT and therapeutic results. Cancer Chemother Pharmacol 1994; 33 Suppl: S60–68.

Belli

Magistretti

Puricelli

. Arteritis following intra-arterial chemotherapy for liver tumors. Eur Radiol 1997; 7: 323–326.

Sakamoto

Aso

Nagaoki

. Complications associated with transcatheter arterial embolization for hepatic tumors. Radiographics 1998; 18: 605–619.

Sottani

Poggi

Quaretti

. Serum pharmacokinetics in patients treated with transarterial chemoembolization (TACE) using two types of epirubicin-loaded microspheres. Anticancer Res 2012; 32: 1769–1774.

10.

Lee

Liapi

Cornell

. Doxorubicin-loaded QuadraSphere microspheres: plasma pharmacokinetics and intratumoral drug concentration in an animal model of liver cancer. Cardiovasc Intervent Radiol 2010; 33: 576–582.

11.

Jordan

Denys

De Baere

. Comparative study of chemoembolization loadable beads: in vitro drug release and physical properties of DC bead and hepasphere loaded with doxorubicin and irinotecan. J Vasc Interv Radiol 2010; 21: 1084–1090.

12.

Song

Chun

Song

. Comparative study between Doxorubicin-eluting beads and conventional transarterial chemoembolization for treatment of hepatocellular carcinoma. J Hepatol 2012; 57: 1244–1250.

13.

Lammer

Malagari

Vogl

. Prospective randomized study of doxorubicin-eluting-bead embolization in the treatment of hepatocellular carcinoma: results of the PRECISION V study. Cardiovasc Intervent Radiol 2010; 33: 41–52.

14.

Vogl

Lammer

Lencioni

. Liver, gastrointestinal, and cardiac toxicity in intermediate hepatocellular carcinoma treated with PRECISION TACE with drug-eluting beads: results from the PRECISION V randomized trial. Am J Roentgenol 2011; 197: W562–570.

15.

Sacco

Bargellini

Bertini

. Conventional versus doxorubicin-eluting bead transarterial chemoembolization for hepatocellular carcinoma. J Vasc Interv Radiol 2011; 22: 1545–1552.

16.

Maeda

Osuga

Mikami

. Angiographic evaluation of hepatic arterial damage after transarterial chemoembolization for hepatocellular carcinoma. Radiation Medicine 2008; 26: 206–212.

17.

Seki

Hori

Kobayashi

. Transcatheter arterial chemoembolization with epirubicin-loaded superabsorbent polymer microspheres for 135 hepatocellular carcinoma patients: single-center experience. Cardiovasc Intervent Radiol 2011; 34: 557–565.

18.

Bruix

Sherman

. Management of hepatocellular carcinoma: an update. Hepatology 2011; 53: 1020–1022.

19.

Burrel

Reig

Forner

. Survival of patients with hepatocellular carcinoma treated by transarterial chemoembolisation (TACE) using drug eluting beads. Implications for clinical practice and trial design. J Hepatol 2012; 56: 1330–1335.

20.

Syha

Ketelsen

Heller

. Hepatocellular carcinoma: initial tumour response after short-term and long-interval chemoembolization with drug-eluting beads using modified RECIST. Eur J Gastroenterol Hepatol 2012; 24: 1325–1332.

21.

Wiggermann

Sieron

Brosche

. Transarterial chemoembolization of Child-A hepatocellular carcinoma: drug-eluting bead TACE (DEB TACE) vs. TACE with cisplatin/lipiodol (cTACE). Med Sci Monit 2011; 17: CR189–195.

22.

Horrow

Blumenthal

Reich

. Sonographic diagnosis and outcome of hepatic artery thrombosis after orthotopic liver transplantation in adults. Am J Roentgenol 2007; 189: 346–351.

23.

Bekker

Ploem

de Jong

. Early hepatic artery thrombosis after liver transplantation: a systematic review of the incidence, outcome and risk factors. Am J Transplant 2009; 9: 746–757.

24.

Plessier

Rautou

Valla

. Management of hepatic vascular diseases. J Hepatol 2012; 56(Suppl 1): S25–38.

25.

Roth

Wallner

Safi

. Arterial perfusion abnormalities of the liver after hepatic arterial infusion chemotherapy and their correlation with changes in the metastases: evaluation with CT and angiography. Am J Roentgenol 1989; 153: 751–754.

26.

Ishikura

Sotozaki

Adachi

. Granulomatous arteritis with massive eosinophilic leukocyte infiltration and transient peripheral eosinophilia subsequent to transarterial embolization therapy with a gelatin sponge. Acta Pathol Jpn 1991; 41: 618–622.