|
C. S. LEUNG, C. N. TANG, K. H. FUNG* and M.K. W. LI
Department of Surgery, *Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Chai Wan, Hong Kong
Keywords: Arterial embolisation, hepatocellular, carcinoma
Background: Hepatocellular carcinoma (HCC) is a disease with a high prevalence in South East Asia. It is not uncommon
to encounter rupture of the tumour in an emergency situation. Operative measures in this situation are often associated
with high mortality rates. Transcatheter arterial embolisation (TAE) appears to be an effective alternative to surgery and
is increasingly used by many centres. In this study, we have reviewed the outcome of our patients after receiving TAE and
tried to identify prognostic indicator(s). Patients and Methods: From 1996 to 2000, we had retrospectively reviewed the
outcome of 31 patients who had undergone TAE for rupture of HCC and compared their survival with respect to
different prognostic indicators. Results: The were 31 patients with a mean age of 53 years. At the time of rupture, 19 patients
had Child’s A, 5 Child’s B and 7 of Child’s C disease. The most common presentation was abdominal pain (14 patients).
Bleeding was successfully arrested in all 31 patients. The most common complication was fever (13 patients). The overall
mean survival was 126 days. Eight patients died within 30 days of admission, the major cause of death was liver failure,
which occurred in 6 patients. In addition, we had also postulated several prognostic indicators for patients’ survival. The
results showed that only those with a bilirubin level below 50umol/L and who presented with shock had a poor outcome.
Conclusion: TAE should be considered in the initial management of patients with ruptured HCC. It is effective in
arresting tumour bleeding and allows the patient to have subsequent definitive management
J.R.Coll.Edinb., 47, October 2002, 685-688
Hepatocellular carcinoma (HCC) is a disease with a high prevalence rate in South East Asia due to the high incidence of hepatitis associated cirrhosis. The clinical presentation can be varied. Patients may present with jaundice, weight loss, abdominal mass or abdominal pain. Others may be asymptomatic and picked up incidentally on routine screening for carriers of the hepatitis virus. Also, it is not uncommon to have spontaneous rupture of a tumour as the initial presentation. In the Far East, the rupture rate is as high as 10%, while in Hong Kong the rupture rate is around 9.7%.1-4 Without any treatment, the outcome is poor and survival rate is only 10%. In view of this, a number of studies have been carried out and published in an attempt to reach a consensus on the management of this potential life-threatening condition.
Traditionally, surgeons operate on those patients who present with ruptured HCC and the treatment is varied, consisting of packing, hepatic artery ligation and hepatectomy. However, surgery is often associated with a high mortality rate; the latter has been reported to be as high as 70%.2,3,4,15 The major drawbacks of emergency hepatectomy are liver failure and tumour involvement of the resection margin due to inadequate preoperative work up. Even for the less invasive procedures like packing, argon beam coagulation or hepatic artery ligation, there is little difference in terms of the operative morbidity and mortality. Therefore, a relatively less invasive procedure should be considered for patients with rupture of HCC and transcatheter hepatic arterial embolisation (TAE) is gaining popularity in this area.
This article reviews our experience in using TAE for patients with rupture HCC. We have attempted to identify possible prognostic factor(s) that may be of use as selection criteria for TAE in the future.
Medical records of patients with ruptured HCC who had undergone TAE during the period 1996-2001, were retrospectively reviewed and the outcome was analyzed. The diagnosis of ruptured HCC was based on two major criteria: (1) clinical presentation of acute abdominal pain, distension, hypotension or shock, and (2) radiological findings of a liver tumour with evidence of bleeding - either frank haemoperitonium or a subcapsular haematoma.
All of the patients who underwent TAE also had an ultrasound (US), computed tomography (CT) scan and subsequent hepatic arteriogram (HAG), which were usually arranged within 24 hours of admission.
The agents used for embolisation were basically gelfoam, stainless steel coils or polyvinyl alcohol sponge (Ivalon); the choice was primarily depending on the size of the artery being embolised and the radiologists’ preference. In general, gelfoam was the preferred option, as the arteries could be recannalized providing an opportunity of further chemoembolisation. Immediately following embolisation, hepatic arteriogram was repeated to ascertain the success of haemostasis. Patients were monitored closely for any signs of rebleeding and complications, which included liver failure, fever, abdominal pain and wound problems.
The final outcome in terms of the procedure-related complications, survival and subsequent follow-up intervention were analyzed. Moreover, we also compared the survival of patients with respect to different possible prognostic indicators by using the survival Kaplan-Meier survival curves and the degree of significance was determined by the Mantel-cox test. Statistical significance was defined as p <0.05.
From 1996 to 2000, there were 49 patients admitted with a diagnosis of ruptured HCC. Eight of them succumbed soon after admission, despite active resuscitation. Seven patients did not have TAE performed due to thrombosis of the main portal vein, which was considered a contraindication for TAE in our unit. Three other patients had a patent portal vein but did not receive TAE as the US and CT scans showed only a small hepatic subcapsular haematoma and the patients’ condition was very stable. For the remaining 31 patients (22 males and 8 females) with a ruptured HCC, all had TAE performed.
Figure 1: Presentation of different post TAE complications
The most common presentation, for the 31 patients who underwent TAE, was abdominal pain in 14 (45%), abdominal distension in 6 (19%) and shock in 4 (13%) patients. Eleven of these (35%) patients had a significant drop of haemoglobin (more than 4g/dl) and required transfusion (Table 1).
| Presentation | Percentage of cases |
| Abdominal pain | 45% |
| Drop of haemoglobin level | 35% |
| Abdominal distension | 19% |
| Shock | 13% |
| Mass in abdomen | 19% |
Table 1: Clinical features on presentation
On admission, 19 patients had Child’s A liver cirrhosis, 5 Child’s B and 7 Child’s C disease. Thirteen patients (42%) had bilirubin levels greater than 50 umol/L and alpha-fetoprotein levels greater than 500ng/ml were documented in 16 (52%) patients.
The mean diameter of the tumour was 9.8 cm, ranging from 3 to 23 cm and 13 (43%) patients had bi-lobar disease.
Only two patients had surgery (packing) before undergoing emboillisation. However, both of these cases were readmitted again as the tumour rebled and they required subsequent embolisation. Except for these two patients, the rest of them did not receive any previous treatment - operation, radiotherapy, embolisation or chemotherapy.
Fourteen (45%) patients had the main hepatic artery embolised and more selective arterial embolisation was performed in the remaining 17 (55%) cases.
After TAE, bleeding was arrested in all 31 patients, which was confirmed by the post TAE arteriogram showing a decrease in the tumour vascularity. The clinical status of the patients subsequently stabilized and there was no further drop in haemoglobin level or need of a blood transfusion.
The most common complication after the procedure was fever, which happened in 13 (42%) patients, 8 (26%) patients suffered from persistent right upper quadrant abdominal pain.
Both clinical features resolved within a few days by low dose of nonsteroidal anti-inflammatory drugs NSAID (Volteran SR). Liver failure was encountered in 6 (19%) patients and all these 6 patients died eventually. (Figure 1) Four of these deaths had Child’s C and two had Child’s B liver cirrhosis. Three patients had undergone embolisation of the hepatic artery and three of them had selective embolisation of a branch of the hepatic artery. Three of the case presented with shock.
The overall mean survival of those after TAE was 126 days. Eight patients died within 30 days of admission, a mortality rate of 26%. Apart from liver failure, which was the major cause of death, one patient died of heart failure and the other suffered from pneumonia.
In addition to the above findings, we postulated several prognostic indicators (Table 2) and studied their effects on patients’ survival.
From our study, we found that there were only two statistically significant factors, which affected patients’ survival. The first significant prognostic factor was the level of bilirubin. Those with a bilirubin level more than 50umol/l had a mean survival of 34 days, whereas those patients with a level less than 50umol/l had a mean survival of 165 days (P=0.0046). (Figure 2) The second one was the presence of shock on admission. Those presented with shock had a worse outcome (mean survival of 87 days), compared with the group without (155 days) (P= 0.03) (Graph 3). Conversely, all the other factors (age, fall of haemoglobin, selectivity of embolisation and tumour size) were not found to influence the outcome.
Figure 2: Cumulative actuarial survival curves of patients having serum total bilirubin level above or below 50umol/l
Despite the advance in surgical technology, the management of ruptured HCC is still a challenge to surgeons. The role of TAE in the management of ruptured HCC, especially for the initial period, has been well established. The mean survival rate of those who received TAE has been reported to be from one to twenty weeks. 1,3,4,5,9,14 In a Japanese series, reported in 1991, (172 patients with ruptured HCC,) the mean survival of those treated conservatively was only 19.2 days and 39.6 days for those who received TAE. 3 Another study from Hong Kong showed a median survival rate 9 weeks in 32 patients who underwent TAE.5 In our study, the mean survival rate of those having TAE performed was 126 days. In general, the prognosis of those receiving TAE is better when compared with the group undergoing packing, argon beam coagulation or hepatic artery ligation. However, there has been no randomised control study published on this aspect.
Figure 3: Cumulative actuarial survival curves of patient presenting with and without shock.
Some centers choose emergency hepatectomy for ruptured HCC. The outcome of these patients varies between different centers.2,4,10,15 The exceptionally good outcome in some centres may be the result of selection bias. Although it is feasible, emergency hepatectomy is associated with a high mortality. Because the HCC occurs mainly in those with liver cirrhosis, with compromised liver reserve, coagulopathy and malnutrition, postoperative liver failure is not uncommon and is one of the major causes of death. In addition, emergency liver resection for ruptured HCC often results in unclear tumour resection, which is associated with a high intrahepatic recurrence rate. Comparison with the previous reported mortality of those undergone emergency hepatectomy and other operative means to control bleeding, the results of our study were much better in the group of patients being treated by TAE alone.
Although TAE has its own benefits, its application should be selective when taking the cost and complications into consideration. Therefore, one of the objectives of this article is to identify prognostic indicators. It has been shown in other studies that the level of bilirubin, size of tumour and patency of portal vein may affect the outcome.5,6,14 In our study, we have also postulated a number of possible prognostic indicators and have analysed their effects on patients’ survival. These factors include the patients’ age, the size of tumour, shock on presentation, drop in haemoglobin level, the bilirubin level and selectivity of arterial embolisation. In our studies, the survival of our patients with bilirubin levels more than 50umol/L and those without symptoms of shock was significantly better. The other factors did not alter the outcome in terms of life expectancy.
For the 6 patients who died from liver failure, four had Child’s C and two Child’s B disease. Three patients had received embolisation of the main hepatic artery and the others had selective embolisation of the branches of the hepatic arteries. The reasons for liver failure appear to be multifactorial. However, there is lack of any large-scale study to evaluate these various factors.
The adverse effects of portal vein thrombosis and selective arterial embolisation on patient’s survival have been demonstrated in some studies.8,14 Embolisation of the hepatic artery may cause further damage to the liver cell and lead to subsequent liver failure. However, the findings are not consistent in all these studies. In our patients, the survival following embolisation of the hepatic artery proper or selective embolisation of branches of the hepatic artery are comparable. The difference may result from the choice of embolic material. In the majority of cases, we used gelfoam, which is dissolvable and, therefore, the artery may recannulate after a period of time. Moreover, all our patients had a patent portal vein. Therefore, the blood supply to the normal liver parenchyma may not have been so critically affected in our patients.
Although TAE is effective in the initial management and stabilization of patients with ruptured HCC, the longterm survival depends on further definitive treatment that includes hepatectomy, transarterial chemo-embolisation or chemotherapy.3,11,12,13
| Patients age (more than 60) |
| Shock on presentation (documented hypotension with systolic blood pressure lower than 100) |
| Drop of haemoglobin level requiring transfusion |
| Bilirubin level (> 50umol/l) |
| Tumour size (> 10cm) |
| Selectivity of arterial embolisation |
Table 2: Prognostic indicators in patients with ruptured HCE
For our patients, after TAE, one patient had undergone hepatectomy and he survived for more than 18 months, six of the cases had received trans-arterial chemo-embolisation and their mean survival was 12 months, one patient was subsequently put on tamoxifen and she died 2 months after the embolisation.
In conclusion, TAE should be considered in the management of patients with ruptured HCC. It is effective in arresting tumour bleeding, so that the patients can be stabilized for subsequent definitive treatment. However, the procedure should be done selectively as not all the patients will benefit from this intervention.
1. Hirai K, Kawazoe Y, Yamashita K, Kumagai M.Transcatheter arterial embolization for spontaneous
rupture of hepatocellular carcinoma. Am J Gastro 1986; 81,
No.4: 275-279
2. Dewar GA, Griffin SM, Ku KW, Lau WY, et al Management of bleeding liver tumour in Hong Kong. Br
J Surg 1991; 78: 463-466
3. Miyamoto M, Sudo T and Kuyama T. Spontaneous
rupture of hepatocellular carcinoma: a review of 172
Japanese cases. Am J Gastro 1991; 86: 67-71
4. Zhu LX, Wang GS and S Fan ST. Spontaneous rupture
of hepatocellular carcinoma. Br J Surg 1996; 83(5):
602-607
5. Ngan H, Tso WKH, Lai CL and Fan ST. The role of
hepatic arterial embolization in the treatment of
spontaneous rupture of hepatocellular carcinoma. Clin Radiol
1998; 53: 338-341
6. Ohtomo K, Furui S, Kokubo T, Yamauchi T et al Transcatheter arterial embolization for spontaneous
rupture of hepatocellular carcinoma. Radiol Med 1988;
Jul-Aug, 6(4): 150-156
7. Hsieh JS, Huang CJ, Huang YS, Sheen PC et al Intraperitoneal haemorrhage due to spontaneous
rupture of hepatocellular carcinoma: treatment by hepatic
artery embolisation. AJR 1987; 149: 715-717
8. Sato Y, Fujiwara K, Furui S, Ogata I. et al Benefit of
transcatheter arterial embolisation for ruptured
hepatocellular carcinoma complicationg liver cirrhosis.
Gastroenterology 1985; 89: 157-159
9. Brown KT, Nevins AB, Getrajdman GI, Brody LA et
al. Particle embolization for hepatocellular carcinoma.
J Vasc Interv Radiol 1998; 9: 822-888
10. Chiappa A, Zbar A, Audisio RA, Paties C. et al. Emergency liver resection for ruptured hepatocellular
carcinoma complicating cirrhosis. Hepatogastroenterology 1999; 46:1145-1150
11. Yoshida H, Onda M, Tajiri T, Umehara M. et al. Treatment of spontaneous ruptured hepatocellular
carcinoma. Hepato-Gastroenterology 1999; 46: 2451-2453
12. Shimada R, Imamura H, Makuuchi M, Soeda J. et al. Staged hepatectomy after emergency transcatheter
arterial embolisation for ruptured hepatocellular
carcinoma. Surgery 1998; 124: 526-535
13. Shuto T, Hirohashi K, Kubo S, Tanaka H et al. Delayed
hepatic resection for ruptured hepatocellular
carcinoma. Surgery 1998; 124: 33-37
14. Boyer JC, Zins M, Vilgrain V, Sibert A. et al. Hemoperitoneum caused by spontaneous rupture of
hepatocellular carcinoma. Value and prognostic factors
of hepatic artery embolization. Radiol 1995; 76: 365-369
15. Vivarelli M, Cavallari A, Bellusci R, De Raffele E. et
al. Ruptured hepatocellular carcinoma: an important
cause of spontaneous haemoperitoneum in Italy. Eur
Surg 1995; 161:881-886
Copyright: 24 April 2002
Correspondence: C.S.F. Leung, 3 Lok Man Road, Department of Surgery, PYNEH, Chai Wan, Hong Kong