This article was presented in part at the 2011 annual meeting of the American Association for the Study of Liver Diseases in San Francisco, CA.
Morbidity and mortality following transarterial liver chemoembolization in patients with hepatocellular carcinoma and synthetic hepatic dysfunction
Article first published online: 12 DEC 2012
Copyright © 2012 American Association for the Study of Liver Diseases
Volume 19, Issue 2, pages 164–173, February 2013
How to Cite
Garwood, E. R., Fidelman, N., Hoch, S. E., Kerlan, R. K. and Yao, F. Y. (2013), Morbidity and mortality following transarterial liver chemoembolization in patients with hepatocellular carcinoma and synthetic hepatic dysfunction. Liver Transpl, 19: 164–173. doi: 10.1002/lt.23552
This study was financially supported by a seed grant from the University of California San Francisco Department of Radiology and Biomedical Imaging.
- Issue published online: 12 FEB 2013
- Article first published online: 12 DEC 2012
- Accepted manuscript online: 24 SEP 2012 12:20PM EST
- Manuscript Accepted: 29 AUG 2012
- Manuscript Received: 14 MAY 2012
The purpose of this study was to determine the rate and risk factors for the development of irreversible hepatotoxicity after transarterial chemoembolization (TACE) in patients with hepatocellular carcinoma (HCC) and synthetic hepatic dysfunction. Two hundred fifty-one consecutive patients with HCC and hepatic dysfunction who underwent 443 TACE procedures from 2005 to 2010 were retrospectively reviewed. The included patients met one of the following criteria: a pre-TACE bilirubin level ≥ 2 mg/dL, an international normalized ratio (INR) > 1.5, a creatinine level > 1.2 mg/dL, a platelet count ≤ 60,000/mL, a Model for End-Stage Liver Disease (MELD) score > 15, Child-Turcotte-Pugh class B or C, ascites, or portal vein thrombosis. Hepatotoxicity was defined as new or worsening ascites, encephalopathy, or grade 3 or 4 toxicity (bilirubin, aspartate aminotransferase, alanine aminotransferase, creatinine, or INR) according to the National Cancer Institute Common Terminology Criteria for Adverse Events. The rate and risk factors for death or urgent liver transplantation within 6 weeks of TACE and irreversible hepatotoxicity were determined with a generalized estimating equation analysis. Reversible hepatotoxicity developed after 90 procedures (20%) in 78 patients (31%). Irreversible hepatotoxicity developed after 41 procedures (9%) in 37 patients (15%). Six patients (2%) underwent urgent liver transplantation, and 11 (4%) died within 6 weeks of TACE. Patients at increased risk for procedure-related mortality or urgent liver transplantation within 6 weeks of TACE had a baseline serum bilirubin level ≥ 4.0 mg/dL (P = 0.01), an elevated INR (P < 0.001), hypoalbuminemia with an albumin level < 2.0 g/L (P = 0.01), a serum creatinine level > 2.0 mg/dL (P = 0.02), large ascites (P = 0.002), encephalopathy (P = 0.005), or a MELD score ≥ 20 (P < 0.001). In conclusion, TACE can be performed safely in patients with baseline hepatic dysfunction. However, a poor hepatic reserve increases the risk of irreversible hepatotoxicity, which may lead to death or the need for urgent liver transplantation. Liver Transpl 19:164–173, 2013. © 2012 AASLD.
Barcelona Clinic Liver Cancer
hepatitis C virus
international normalized ratio
Model for End-Stage Liver Disease
percutaneous ethanol injection
transjugular intrahepatic portosystemic shunt
United Network for Organ Sharing
Transarterial chemoembolization (TACE) is a standard treatment option for patients with unresectable, intermediate-stage hepatocellular carcinoma (HCC) without significant hepatic synthetic dysfunction. TACE produces an objective tumor response rate of 35% to 40% and confers a significant survival advantage in comparison with the best supportive care. Improved quality of life and reduced pain after the initiation of chemoembolization have been reported for patients undergoing sequential TACE. Current indications for TACE include the following: a first-line treatment for unresectable, intermediate-stage HCC; a bridging measure for managing disease progression in patients awaiting liver transplantation; and a method for down-staging or qualifying patients for liver transplantation or liver resection through the reduction of the disease burden.[6, 7] The patient population with intermediate HCC remains heterogeneous with varying degrees of hepatic dysfunction; as such, identifying patients who will benefit most from TACE or for whom the risks of irreversible hepatic dysfunction outweigh the potential benefits of TACE is crucial.[1, 8] Guidelines currently recommend against performing TACE in the setting of Child-Turcotte-Pugh (CTP) class C disease, main portal vein thrombosis, or a serum bilirubin level > 3.0 mg/dL unless segmental injection is possible.
Complication rates following TACE are high (25%-45%), with the majority being reversible elevations of hepatic transaminases and serum bilirubin without an impact on the long-term prognosis. However, for a small proportion of patients, irreversible hepatic decompensation may occur.[8, 10, 11] TACE-related irreversible hepatic decompensation is described as a significant and acute elevation of liver function tests or as evidence of new ascites, new encephalopathy, or the worsening of either. These serious complications, which are frequently accompanied by a deterioration in clinical status, may lead to the need for urgent liver transplantation in eligible patients or death. The reported rates of 30-day mortality after TACE range from 0% to 10%.
According to clinical experience gained early in the history of the TACE procedure, a number of risk factors for the development of serious complications following TACE have been identified, and they are predominantly based on laboratory values reflecting hepatic synthetic function or concomitant renal insufficiency. Relative contraindications to TACE focus on the exclusion of patients with hepatic synthetic dysfunction or concomitant renal insufficiency, which is defined as one or more laboratory criteria [elevated serum bilirubin, aspartate aminotransferase, alanine aminotransferase, prothrombin time/international normalized ratio (INR), or creatinine levels or low platelet counts or albumin levels], radiographic criteria (the presence of ascites, hepatofugal flow in the portal vein, or portal vein thrombosis), or clinical indicators of functional status (hepatic encephalopathy or a portosystemic shunt). Most physicians consider TACE to be an evolving procedure with a loosely defined target population; as Brown et al. stated in 2004, “the absolute upper limits of treatment remain undefined.” There are few contemporary studies establishing rates of morbidity and mortality among patients with high-risk features. Defining the level of risk associated with pursuing TACE in these populations would serve to inform the clinician's risk/benefit analysis and the patient's choice to undergo a procedure that may offer increased survival, improved symptom control, and improved quality of life to patients with severe disease for which few other treatment options exist.
TACE is commonly used for maintaining patients on a waiting list for liver transplantation and for down-staging the disease burden to the Milan criteria. Many patients listed for liver transplantation at our institution have hepatic dysfunction and have undergone TACE safely while they are waiting for an organ. This study was designed to determine the rate and risk factors for the development of irreversible severe liver toxicity after TACE in a contemporary cohort of patients with unresectable HCC and significant synthetic hepatic dysfunction for the purpose of informing the clinical decision-making process in the evaluation of patients for TACE eligibility. Rates of mortality and liver transplantation in this population were also examined.
PATIENTS AND METHODS
This single-center, retrospective study was approved by the committee on human research of the institutional review board at the University of California, San Francisco. The requirement for informed consent was waived. The medical records of adult patients with HCC confined to the liver who were treated with TACE between January 2005 and January 2010 were retrospectively reviewed. Subjects were eligible for inclusion if they had one of the following risk factors present immediately before they underwent TACE: a total serum bilirubin level ≥ 2.0 mg/dL, an INR > 1.5, a serum creatinine level > 1.2 mg/dL, a platelet count ≤ 60,000/mL, the presence of ascites on cross-sectional imaging, the disruption of normal portal vein flow by a transjugular intrahepatic portosystemic shunt (TIPS) or a surgical portosystemic shunt, evidence of hepatofugal portal vein blood flow on Doppler ultrasound imaging, or radiographic confirmation of a portal vein branch or main portal vein bland or tumor thrombus. All patients underwent abdominal contrast-enhanced computed tomography or magnetic resonance imaging before their first TACE procedure and every 3 months thereafter. The study population consisted of 251 patients [201 men and 50 women with a mean age of 59.2 years (range = 38.5-84.4 years)] who underwent 443 TACE procedures.
TACE was performed after patients provided written, informed consent. All TACE procedures were performed by fellowship-trained and experienced interventional radiologists. A 5-Fr catheter was inserted into the common femoral artery, and an angiographic survey of the celiac and superior mesenteric arteries was performed. Digital subtraction angiography was also performed after selective catheterization of the proper, right, and/or left hepatic arteries. Depending on the lesion distribution, subsegmental or segmental hepatic artery branches were selected with a coaxially placed microcatheter (Renegade HI-FLO, Boston Scientific, Natick, MA). The TACE regimen consisted of a combination of doxorubicin hydrochloride (25 mg), cisplatin (50 mg), and mitomycin C (10 mg), which were dissolved in 10 mL of a contrast agent (Omnipaque 350, Amersham Health, Princeton NJ) and mixed in a 50-50 emulsion with ethiodized oil (Ethiodol, Laboratoires Guerbet, Roissy, France). After the administration of the emulsion, the targeted hepatic artery branch was embolized to stasis with a gelatin sponge slurry (Gelfoam, Pharmacia Upjohn, Kalamazoo, MI). The dose of doxorubicin was reduced to 12.5 mg for patients with a serum bilirubin level > 3 mg/dL; cisplatin was withheld for patients with a serum creatinine level > 1.2 mg/dL; and patients with a white blood cell count < 3000/mL or a platelet count ≤ 60,000/mL did not receive mitomycin C.
Study Outcomes and Data Analysis
The primary study outcome was death or severe hepatotoxicity requiring liver transplantation within 6 weeks of a TACE procedure. The secondary outcome was irreversible severe hepatotoxicity. Severe hepatotoxicity was defined as the occurrence of one of the following laboratory, clinical, or radiographic criteria during the immediate follow-up period after TACE: new or worsening ascites, new or worsening encephalopathy, or grade 3 or 4 toxicity (total serum bilirubin, aspartate aminotransferase, alanine aminotransferase, creatinine, or INR) according to the National Cancer Institute Common Terminology Criteria for Adverse Events (version 4, May 2009).
Survival between individual patient subgroups was compared with the Kaplan-Meier method and the log-rank test. Relationships between individual risk factors and the study outcomes were evaluated via logistical regression with univariate and multivariate generalized estimating equation models. Generalized estimating equation models are a general method for incorporating within-subject correlations into a generalized linear model that accounts for the statistical dependence among multiple lesions in the same individual. Multivariate models were constructed through the combination of the Model for End-Stage Liver Disease (MELD) score, the CTP score, the presence of Barcelona Clinic Liver Cancer (BCLC) stage D disease, or hepatitis C virus (HCV) as the etiology of cirrhosis with individual variables for which the compound variables did not already account. Summary statistics (means, medians, ranges, and standard deviations) were used to describe the characteristics of the study population. The dates of death, liver transplantation, and last follow-up were extracted from medical records or the Social Security Death Index. Statistical analyses were performed with SAS 9.2 (SAS Institute, Inc., Cary, NC). A P value < 0.05 was considered statistically significant.
Two hundred fifty-one patients with HCC met the inclusion criteria and underwent 443 TACE procedures during the study period. The demographics and characteristics of the patient population are displayed in Table 1, and detailed inclusion criteria are listed in Table 2. Sixteen percent of the high-risk TACE procedures were performed in patients who had undergone at least 1 invasive liver-directed procedure for the treatment of HCC [radiofrequency ablation (RFA), percutaneous alcohol injection, or liver resection] before they were included in the study, and 55% were performed in patients who had undergone at least 1 prior TACE procedure. At the time of TACE, the mean total bilirubin level was 2.0 mg/dL, with 19% of the study population having a bilirubin level of 3.0 mg/dL or higher. The mean calculated MELD score in the study population was 13. At the time of TACE, 40.2% of the patients were classified as CTP class B, and 11.7% were CTP class C. In all, 55% of the procedures were performed in patients classified as United Network for Organ Sharing (UNOS) stage 2 or lower. The subjects included in this study had on average 3 high-risk characteristics (as defined in the Patients and Methods section; range = 1-8). Chemoembolization most commonly was confined to 1 segment of the liver, and in 28% of the procedures, more than 1 liver segment was embolized.
|Demographics by subject|
|Male [n (%)]||201 (80)|
|Female [n (%)]||50 (20)|
|Age (years)a||59.2 ± 9|
|HCC etiologies by procedure [n (%)]|
|Hepatitis B virus||68 (15.3)|
|Viral and alcohol||38 (8.6)|
|Laboratory values by procedurea|
|Total bilirubin (mg/dL)||2.0 ± 1.1|
|INR||1.30 ± 0.2|
|Creatinine (mg/dL)||1.08 ± 0.8|
|Platelet count (/mL)||93,000 ± 60,000|
|Alpha-fetoprotein (μg/L)||1210 ± 6046|
|Clinical characteristics by procedurec|
|BCLC stage [n (%)]|
|UNOS stage [n (%)]|
|MELD scorea||13 ± 4|
|CTP class [n (%)]|
|CTP scorea||8 ± 2|
|Ascites present [n (%)]||190 (42.9)|
|Encephalopathy present [n (%)]||49 (11.1)|
|Prior therapy by procedure [n (%)]|
|No prior interventions||176 (39.7)|
|Prior TACE||194 (43.8)|
|RFA ± TACE||58 (13.1)|
|Segmental or subsegmental resection or resections||11 (2.5)|
|Inclusion Criteria||n (%)|
|Total bilirubin ≥ 2.0 mg/dL||168 (37.9)|
|INR > 1.5||86 (19.4)|
|Creatinine > 1.2 mg/dL||66 (14.9)|
|Platelet count ≤ 60,000/mL||130 (29.3)|
|Ascites present||190 (42.9)|
|TIPS present||16 (3.6)|
|Portal vein thrombus present||66 (14.9)|
|Hepatofugal flow present||4 (0.9)|
|MELD score ≥ 15||86 (19.4)|
|CTP class B or C||230 (51.9)|
Reversible severe hepatotoxicity developed after 90 TACE procedures (20%) in 78 patients (31%). Abnormalities were resolved within 30 days after 63 procedures, in 1 to 3 months after 21 procedures, and in 3 to 6 months after 6 procedures. Irreversible severe hepatotoxicity developed after 41 procedures (9%) in 37 patients (15%). Six patients (2%) received urgent liver transplants, 5 (2%) died within 30 days of TACE, and 11 (4%) died within 6 weeks of TACE. All of these patients also experienced irreversible severe hepatotoxicity.
Predictors of Urgent Transplantation or Death Within 6 Weeks of TACE
The univariate predictors for procedure-related mortality or a need for liver transplantation within 6 weeks of TACE were a baseline total serum bilirubin level ≥ 4.0 mg/dL (P = 0.01), an elevated INR (P < 0.001), hypoalbuminemia with an albumin level < 2.0 g/L (P = 0.01), a serum creatinine level > 2.0 (P = 0.02), large ascites (P = 0.002), encephalopathy (P = 0.005), BCLC stage D disease (P < 0.001), the presence of more than 1 high-risk factor (P = 0.004), CTP class C cirrhosis (P = 0.007), and a MELD score ≥ 20 (P < 0.001; Table 3). Notably, an elevated baseline bilirubin level did not reach statistical significance as a predictor of mortality or a need for urgent transplantation until the serum bilirubin level was ≥4.0 mg/dL.
|Predictor||OR||95% CI||P Value|
|Mild to moderate||1.3||0.4–4.0||0.64|
|0, A, B, or C||Reference||Reference||Reference|
|Number of high-risk factors||1.4||1.1–1.8||0.004|
The 3 most robust predictors of TACE-related death or a need for urgent liver transplantation were an increasing INR as a continuous variable [odds ratio (OR) = 53, confidence interval (CI) = 8.2-343, P < 0.001], a MELD score ≥ 20 (OR = 21, CI = 3.9-116, P < 0.001), and a serum albumin level < 2.0 g/L (OR = 20, CI = 1.8-218, P = 0.01). The following characteristics were not statistically significant predictors of procedure-related mortality or a need for urgent liver transplantation in the 6-week period following TACE: a creatinine level of 1.2 to 2.0 mg/dL, thrombocytopenia, mild to moderate ascites, CTP class B cirrhosis, the etiology of liver disease, prior liver-directed therapy [defined as prior hepatic resection, RFA, or percutaneous ethanol injection (PEI); P = 0.26], at least 1 prior TACE procedure (P = 0.24), the number of prior TACE procedures (P = 0.38), the baseline estimated creatinine clearance (P = 0.16), and the presence of a portal vein flow abnormality (eg, portal vein thrombosis, hepatofugal flow in the portal vein, or a patent TIPS; P = 0.70).
A multivariate analysis showed that thrombocytopenia became a significant independent predictor of TACE-related death or urgent transplantation when it was adjusted for the MELD or CTP score (Table 4). The serum creatinine level was an independent predictor when it was adjusted for the CTP score or the presence of BCLC stage D disease. Encephalopathy was an independent predictor when the MELD score was also in the model, whereas INR continued to be an independent predictor when it was adjusted for BCLC stage D disease. Because of the relatively small numbers of patients with liver disease etiologies other than HCV, the analysis was focused on the group of patients with HCV-associated cirrhosis (with or without concomitant alcohol use). HCV cirrhosis was not found to be an independent predictor of death or urgent liver transplantation after high-risk TACE when we controlled for other possible risk factors.
|Predictor||OR||95% CI||P Value|
|MELD score +|
|Decreasing platelet count||1.01||1.003–1.02||0.01|
|CTP score +|
|Decreasing platelet count||1.008||1.001–1.02||0.04|
|BCLC stage D +|
Predictors of Irreversible Hepatotoxicity Within 6 Weeks of TACE
Similarly, a univariate analysis demonstrated that irreversible hepatotoxicity was associated with a baseline serum bilirubin level ≥ 2 mg/dL (P < 0.001), an INR > 1.5 (P < 0.001), a serum creatinine level > 2.0 mg/dL (P < 0.001), thrombocytopenia with a platelet count < 50,000/mL (P = 0.03), large ascites (P = 0.005), hepatic encephalopathy (P = 0.003), BCLC stage D disease (P = 0.03), CTP class B or C (P = 0.04 and P < 0.001), a MELD score > 15 (P < 0.001), an increasing number of prior high-risk TACE procedures (P = 0.03), a multifactorial cirrhosis etiology (viral hepatitis and alcohol; P = 0.03), and an increasing number of risk factors (P < 0.001; Table 5). Any degree of bilirubin elevation ≥ 2.0 mg/dL was associated with an increased risk of irreversible hepatotoxicity. Thrombocytopenia did not reach statistical significance as a risk factor until the platelet counts were <50,000 mL. There was a slightly reduced risk of irreversible hepatotoxicity with a higher initial estimated creatinine clearance (P = 0.049).
|Predictor||OR||95% CI||P Value|
|Estimated creatinine clearance (mL/minute)||0.98||0.97–1.0||0.049|
|Platelet count (109/L)|
|Mild to moderate||1.8||0.9–3.6||0.11|
|Cirrhotic etiology: mixed (hepatitis and alcohol)||5.1||1.2–21||0.03|
|Number of prior high-risk TACE procedures||1.26||1.02–1.56||0.03|
|Total number of high-risk factors||1.6||1.3–1.9||<0.001|
The 3 most robust predictors of irreversible hepatotoxicity according to the univariate analysis were an initial serum bilirubin level ≥ 4.0 mg/dL (OR = 73, CI = 8.9-601, P < 0.001), an increasing INR as a continuous variable (OR = 64, CI = 16.8-247, P < 0.001), and CTP class C (OR = 59, CI = 8.0-444, P < 0.001). Among the potential risk factors investigated, no statistically significant associations were found between irreversible hepatotoxicity and the number of hepatic segments targeted during a single TACE procedure; the number of prior liver-directed treatments, including resection, PEI, or RFA (P = 0.33), or any prior liver-directed treatments (P = 0.06); the number of prior TACE procedures (P = 0.11); or portal vein flow abnormalities such as portal vein thrombosis, hepatofugal flow in the portal vein, or a patent TIPS (P = 0.22).
A multivariate analysis showed that hypoalbuminemia continued to be a significant independent predictor of TACE-related irreversible hepatotoxicity when it was adjusted for the MELD score, whereas an elevated serum creatinine level was an independent predictor of irreversible hepatotoxicity when it was adjusted for the CTP score (Table 6). Both thrombocytopenia and an elevated serum creatinine level were significant predictors of irreversible severe hepatotoxicity in a model containing BCLC stage D disease. HCV cirrhosis was found to be independently predictive of irreversible hepatotoxicity when it was combined in a model with a serum creatinine level > 2.0 mg/dL, when there was clinical evidence of encephalopathy, and when the MELD score was ≥20.
|Predictor||OR||95% CI||P Value|
|MELD score + albumin||0.23||0.1–0.5||<0.001|
|CTP score + creatinine||1.5||1.3–1.7||<0.001|
|BCLC stage D +|
|Decreasing platelet count||1.02||1.004–1.03||0.01|
|HCV cirrhosis +|
|Creatinine > 2.0 mg/dL||2.5||1.02–5.9||0.045|
|MELD score ≥ 20||2.6||1.1–6.1||0.03|
The median follow-up was 12.5 months (range = 1-60 months). Overall, 79 deaths occurred among 251 patients (31%), and the median survival time was 10.3 months from the date of the first high-risk TACE procedure. The 1-, 2-, and 3-year survival rates for the entire study population were 81%, 72%, and 69%, respectively. Seventy-one of the 171 patients (42%) who did not undergo liver transplantation died during the follow-up period. Their median survival time was 8 months, whereas their 1-, 2-, and 3-year survival rates were 73%, 62%, and 59%, respectively. Six patients received urgent liver transplantation within 6 weeks of TACE, and 74 patients received nonurgent liver transplantation during the follow-up period. Seven of the 80 patients who underwent transplantation died, and the overall mortality rate was 8.8%. The median survival of the transplant recipients was 2.1 years. Their 1-, 2-, and 3-year survival rates were 100%, 94%, and 91% respectively, which were significantly better than the rates for those patients who did not undergo liver transplantation (P < 0.001; Fig. 1). Similarly, there were significantly fewer deaths in the group of patients with cirrhosis due to HCV who underwent transplantation (P < 0.001). Five of the 58 patients (8.6%) with cirrhosis due to HCV who underwent transplantation died, whereas 50 of the 108 patients (47%) with HCV cirrhosis who did not undergo transplantation died (Fig. 2).
This study included 16 patients with a pre-TACE MELD score ≥20. Nine of these patients underwent liver transplantation, and all were still living at the conclusion of the data collection. Seven patients did not undergo transplantation, and 4 died during the follow-up period. A log-rank test for equality of survivor functions was significant with P = 0.003 in this small subset (Fig. 3).
The presence of a portal vein thrombus in the main portal vein or portal vein branches (diagnosed by ultrasound or computed tomography imaging) was a risk factor also associated with worse survival in this study (Fig. 4). Forty-four percent of the patients with portal vein thrombosis at the time of high-risk TACE died during the follow-up period, whereas only 29% of the patients without portal vein thrombosis did (P = 0.006).
The effect of previous TACE procedures or other liver-directed therapies for HCC (including RFA, PEI, and hepatic resection) on survival was also examined (Fig. 5). The survival of 5 groups of patients was compared: (1) no prior TACE procedures or other liver-directed therapy (n = 176), (2) a single prior TACE procedure (n = 37), (3) multiple prior TACE procedures (n = 12), (4) prior TACE procedures and other liver-directed therapies (n = 6), and (5) prior surgical resection or ablative therapy other than TACE (n = 20). A log-rank test showed that the group of patients with a history of multiple prior TACE procedures had worse survival after high-risk TACE in comparison with the group of patients who had no previous liver-directed therapy (P = 0.007). The survival rates of the first, second, fourth, and fifth groups were similar.
Current guidelines recommend TACE for patients with intermediate-stage HCC (BCLC stage B), multinodular tumors without cancer-related symptoms, vascular invasion, or extrahepatic spread.[9, 12] These selection criteria may define the best candidates for TACE in terms of survival benefits and rates of treatment-related adverse events. The risk of complications may be greater in patients with more extensive disease or hepatic decompensation. However, the degree of increased risk and the point at which this risk outweighs the survival benefits of TACE remain undefined. TACE is “discouraged in patients with decompensated liver disease, advanced liver dysfunction, macroscopic invasion or extrahepatic spread,” and with respect to TACE, “those with liver decompensation or more advanced liver failure should be excluded since the ischaemic insult can lead to severe adverse events.”
Changes in the TACE procedure and particularly the development of the selective and superselective chemoembolization techniques, which limit collateral damage to the liver parenchyma and provide more direct access to the tumor vascular supply, have improved hepatotoxicity rates over time. In patients with preserved hepatic function (CTP class A or B and bilirubin level < 2.0 mg/dL), the rate of selective or superselective TACE-related irreversible hepatic decompensation is approximately 3%, whereas a rate of 20% has been observed in a series of nonselective TACE procedures.
Absolute and relative contraindications to TACE suggested by a literature review have included portal vein occlusion, the alteration of the normal directional portal vein flow, decompensated or severe hepatic disease, recent variceal bleeding, encephalopathy, refractory ascites, and the extrahepatic spread of disease. In the process of exploring the extent of hepatic dysfunction that is acceptable before TACE, many relative contraindications to TACE have been challenged. A portal vein thrombus, which is present in 8% to 26% of HCC patients, was initially considered an absolute contraindication. Several studies have now demonstrated that with alterations in technique, TACE can be safely and efficaciously performed in the setting of main portal vein thrombosis,[13, 14] particularly when there is established collateral circulation and there is good hepatic reserve. In concordance with these findings, our study did not establish a statistically significant association between irreversible hepatotoxicity and portal vein flow abnormalities such as portal vein thrombosis, hepatofugal flow in the portal vein, or a patent TIPS.
Similarly, the conventional exclusion of patients with concomitant extrahepatic disease is currently under investigation. One cohort study of 251 patients with extrahepatic metastases at the time of TACE found a survival advantage in comparison with conservative management, and this was increased further by the addition of sorafenib.
As more patients with high-risk features are considered for TACE, it is important to establish a firm understanding of the risk factors that are associated with hepatic decompensation following TACE. Although complication rates have been reported consistently for patients undergoing TACE in the setting of published prospective studies, retrospective studies, and randomized controlled trials, few contemporary studies have reported complication rates among high-risk patients. Among high-risk patients, selective or superselective TACE has been associated with procedure-related morbidity rates ranging from 2% to 10.8% and 30-day mortality rates ranging from 1% to 7.7%. In these studies, the definitions of high-risk features were based on a number of laboratory, radiographic, or clinical parameters, which varied widely. Kothary et al. in 2007 reported complication outcomes for 52 patients undergoing 65 TACE procedures. In this study, high-risk subjects demonstrated at least one of the following features: a bilirubin level > 2 mg/dL, an albumin level < 3.5 mg/dL, a CTP score > 9, multifocal disease, a TIPS, or a biliary obstruction. The procedure-related morbidity rate was 10.8%, and the 30-day mortality rate was 7.7%. Kiely et al. reported in 2006 on a cohort of 141 low- and high-risk patients undergoing 355 TACE procedures with high-risk features (a bilirubin level > 2.0 mg/dL, portal vein thrombosis or invasion, or intrahepatic biliary dilatation involving at least 1 lobe of the liver). A comparison of 36 subjects with low-risk features and subjects considered to be at high risk demonstrated no statistically significant differences in overall per-procedure complications (3.2% versus 8.2%) or 30-day mortality (1.4% versus 5.5%). In a study of 96 high-risk patients, Yoon et al. observed a 1% rate of procedure-related mortality and a 2% rate of irreversible complications after TACE among patients with diffuse, infiltrative, or multifocal tumors. A recent retrospective cohort study by Russell et al. examining periprocedural complications of transarterial embolization/TACE for HCV-related HCC in 141 procedures in 76 patients found no significant differences in complication rates between Child-Pugh class A patients and Child-Pugh class B patients.
Current research focusing on the reduction of complications in high-risk patients is underway. The use of chemotherapeutic delivery vehicles that allow controlled drug release (eg, doxorubicin-eluting microspheres) may allow prolonged chemotherapeutic exposure with lower systemic drug concentrations and thus lower the hepatotoxicity risk. The PRECISION V (Prospective randomized study of doxorubicin-eluting-bead embolization in the treatment of hepatocellular carcinoma) randomized controlled trial compared conventional TACE using doxorubicin and ethiodized oil to TACE using doxorubicin-eluting beads (DEBs), and it was found that the maximum postchemoembolization alanine aminotransferase increase was 50% less in the DEB group. Adverse hepatobiliary effects occurred in 16.1% of the patients in the DEB group versus 25% of the patients in the conventional TACE group.
The limitations of this study include its retrospective cohort design. Additionally, coexisting medical conditions such as diabetes are emerging as risk factors for complications following TACE. In this study, we did not collect information about concomitant diagnoses. Long-term follow-up data in our data set for evaluating the impact of elevated risk factors on the tumoral response to TACE and survival in this patient population would be valuable. Our hypothesis was that baseline hepatic synthetic dysfunction should not preclude patients from receiving TACE, and our findings support the notion that TACE can be performed safely in patients with a number of high-risk clinical features. However, prospective randomized trials for establishing whether there is a survival benefit from TACE among subjects with significant hepatic dysfunction would help to guide care for these patients because current studies demonstrating benefits from TACE have been performed in populations with preserved hepatic function.
This study describes morbidity and mortality outcomes for the largest reported cohort of HCC subjects undergoing high-risk TACE. Our data indicate that TACE can be performed safely in patients with baseline hepatic dysfunction. However, the complication of irreversible hepatotoxicity may lead to death or a need for urgent liver transplantation. The 30-day mortality rate in our high-risk cohort was 2%, and the rate of urgent liver transplantation was 2%; interestingly, this was analogous to the 2.4% rate of 30-day mortality reported in a recent meta-analysis of TACE clinical trials, which included 2858 subjects generally considered to be low-risk for TACE. Poor hepatic reserve does increase the risk of irreversible hepatotoxicity from 3% in low-risk populations, as reported in recent literature, to 9% in a high-risk population, as demonstrated in the current study. More frequently, however, patients with baseline hepatic dysfunction experience only transient hepatotoxicity after TACE that usually resolves within 30 days. The risk of irreversible hepatotoxicity versus the potential survival advantage for patients with unresectable HCC and baseline hepatic dysfunction must be evaluated on an individual basis. Commonly employed prognostic scales such as the CTP class and the MELD score are useful predictors of poor outcomes in this population.
The authors acknowledge the expert assistance of Nancy Hills, Ph.D., and Chengshi Jin at the Clinical and Translational Science Institute (University of California San Francisco) in performing the statistical analyses.
- 1Hepatocellular carcinoma. Lancet 2012;379:1245-1255., , .
- 2Longitudinal quality of life assessment of patients with hepatocellular carcinoma after primary transarterial chemoembolization. J Vasc Interv Radiol 2010;21:1024-1030., , , , , , et al.
- 8Evolving strategies for the management of intermediate-stage hepatocellular carcinoma: available evidence and expert opinion on the use of transarterial chemoembolization. Cancer Treat Rev 2011;37:212-220., , , , , , .
- 9National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: hepatobiliary cancers version 2.2012. http://www.nccn.org/professionals/physician_gls/pdf/hepatobiliary.pdf. Accessed August 2012.
- 12European Association for Study of Liver; European Organisation for Research and Treatment of Cancer. EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. Eur J Cancer 2012;48:599-641.
- 13Transarterial chemoembolization for unresectable hepatocellular carcinoma with portal vein tumor thrombosis: a prospective comparative study. Ann Surg Oncol 2011;18:413-420., , , , , , .
- 16Transcatheter arterial chemo-lipiodol infusion for unresectable hepatocellular carcinoma in 96 high-risk patients. Clin Radiol 2010;65:271-277., , , , , , .
- 19Periprocedural complications by Child-Pugh class in patients undergoing transcatheter arterial embolization or chemoembolization to treat unresectable hepatocellular carcinoma at a VA medical center. Am J Surg 2010;200:659-664., , , , , , .
- 20Prospective 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., , , , , , et al.; for PRECISION V Investigators.
- 21Transarterial therapy for hepatocellular carcinoma: which technique is more effective? A systematic review of cohort and randomized studies. Cardiovasc Intervent Radiol 2007;30:6-25., , , , , , et al.