Spontaneous Rupture of Primary Hepatocellular Carcinoma: Experience of Emergency Laparotomy over a 16-Year Period

DISCUSSION

The exact mechanism of spontaneous rupture of HCC remains unknown at the present time. Some authors have proposed that superficial tumors develop bleeding due to abdominal blunt trauma^[6], either from the outside, or from the inside as a result of repeated respiratory movement, especially for tumors located under the diaphragm^[2]. In our study, 3 cases suffered from mild abdominal blunt trauma. So we can not exclude that trauma as one of the inducing factors in the patients with spontaneous rupture of a HCC. Some investigators have suspected coagulopathy as a main reason for the initiation of spontaneous rupture. However many cases with ruptured tumors have not been found in the terminal stage of HCC or with liver failure. A more convincing hypothesis is that the pressure within the tumor increases when the branches of the hepatic vein are blocked, as the cancer invades into them^[7]. Recently, researchers proposed that a preexisting vascular injury may bring about a spontaneous rupture of HCC when the blood vessel is subject to hypertension or if the blood vessel is inelastic^[8]. In short, regarding the mechanism of spontaneous rupture of HCC, all of the related literature presents only hypotheses and presumptions. In our opinion, the cause of rupture is diverse, and each case should be analyzed individually. More detailed investigation is necessary to compare the clinical features and pathological characteristics of ruptured HCC cases with that of non-ruptured HCC cases.

The incidence of spontaneous rupture in our HCC cases was 3.8%, which is significently lower than that of other reports. The difference is possibly due to the development of early detection of HCC or it maybe due to the difference in incidence of HCC in various areas and countries.

Cases with typical presentations can be diagnosed easily. However for some cases without dramatic symptoms, it is difficult to make an early correct clinical diagnosis, especially in Western countries, where the incidence of HCC is low. If the patients present with apparent abdominal pain without shock and acute anemia, the diagnosis should be distinguished from perforation of a gastroduodenal ulcer, acute cholecystitis or acute pancreatitis. The misdiagnosis of splenic rupture or gastrointestinal bleeding may be made in the patients with the presentations of significent blood loss and hypotension. In our study, 5 cases (10.2%) were misdiagnosed as a perforation of a gastroduodenal ulcer (2 cases), acute appendicitis (1case), acute cholecystitis (1 case) and rupture of an ectopic pregnancy (1 case). In summary, the following features should be considered in making the diagnosis: ①The initial symptom usually presents as a sudden right hypochondrial or epigastric pain. ② Most of the patients develop shock and show signs of peritonitis. ③Abdominal paracentesis is positive. ④ Ultrasonography may demonstrate a hepatic tumor and ascites. The rupture site appears as a hyperechoic area around the tumor. ⑤CT is valuable in showing a high attenuation close to the tumor, which represents acute blood clotting. ⑥Angiography may reveal extravasation of the radiographic contrast agent at the site of rupture. Zhu et al.^[7] reported that the positive rate of correct diagnosis was 86% with paracentesis, 66% by ultrasonography, 100% by CT, 20% by angiography, which was similar to our results: 87.8% with paracentesis, 83.7% by ultrasonography, 95.2% by CT. Sometimes it maybe helpful to examine the AFP level in the peripheral blood or to detect cancer cells in the blood-stained ascites.

Spontaneous rupture of HCC is usually a medical emergency that leads to an extremely dismal prognosis and requires immediate therapeutic approaches. Therefore it seems inappropriate simply to control bleeding. However Marini et al.^[9] reported that the majority of cases in their study had no evidence of hemodynamic instablity and therefore the patients should be initially managed conservatively. This treatment is still controversial. We hypothesize that their results may relate to the selection of patients without life-threatening conditions.

The efficient method to control intra-abdominal bleeding is considered to be laparotomy and transarterial embolization (TAE). Currently, TAE is thought to be the ideal treatment because it is simple and effective, and this method has become the universally accepted means to deal with spontaneous rupture of HCC^[10,11]. For some resectable carcinomas, a second-stage operation is recommended following TAE to obtain long-term survival^[12,13]. Using TAE, hemostasis is achieved successfully in 75%~100% of patients, but recurrent bleeding and liver failure may occur^[14]. Castells et al.^[11] reported that after application of TAE, nearly 85% of the patients developed a selflimited post-embolization syndrome, consisting of fever, abdominal pain and ALT elevation. Furthermore TAE has the limitation that an experienced interventional radiologist is essential and it is contraindicated when the main trunk of the portal vein is totally obstructed. Moreover, interventional therapy as well as conservative therapy may cause the loss of an optimal opportunity to resect the tumor because cancer cells disperse into the peritoneal cavity and grow rapidly. Hence we should also be mindful of other methods. In our study, only 6 cases with severe liver failure and poor conditions were not suitable for laparotomy, and were treated by TAE.

Non-surgical treatment of spontaneous rupture of HCC should be applied only with patients having contraindications to surgery^[15]. Baesd on our experience, laparotomy is of great value considering the following advantages: ①Some patients may receive onestage curable hepatectomy. ②Peritoneal cavity lavage can clear the cancer cells that have already dispersed into the peritoneal cavity at the time of rupture. Lin et al.^[16] reported that distilled water peritoneal lavage resulted in significantly better survival time in patients with spontaneous rupture of HCC. ③For unresectable tumors, some local ablation methods are also effective in killing cancer cells, such as microwave coagulation, argon beam coagulation or injection of absolute alcohol. ④The effect of hemostasis is efficient. ⑤The rate of recurrent bleeding is low. ⑥The incidence of liver failure is not as high as that seen in TAE.

Considering the general conditions and the local lesion of the patient, hepatectomy should be attempted to obtain a curable operation and long-term survival. Because nearly 80% of the patients with spontaneous rupture of HCC were accompanied with liver cirrhosis, to avoid the further damage to liver function, the operation should not be extensive resection. In our study, 21 cases underwent hepatectomy, and the mean survival was 9.9 months. For resectable ruptured HCC patients, emergency liver resection can achieve good early and long-term results^[17]. It has been reported that long-term results of emergency hepatectomy are similar to those receiving elective surgery^[15-18].

Hepatic artery ligation is suitable for patients with an unresectable tumor and without portal vein tumor thrombus. Since a HCC is mainly supplied by a hepatic artery, hepatic artery ligation is an effective and popular method. But at the same time it also suppresses liver function. The logical approach is the selective ligation of a branch of the hepatic artery because it results in a lower risk of postoperative liver failure. In addition, the preservation of the contralateral arterial supply allows the possibility of future definitive liver resection.

Packing and suture have a minor effect on liver function. A suture is only applicable when the bleeding site is small and easily accessible. Sometimes it is impossible for a friable tumor. Packing is feasible for a tumor located under the diaphragm. However it has the risk of rebleeding after the removal of the packs and the risk of infection if the pack is left in place longer than 72 h. Some investigators considered this method as a good procedure in hemodynamically unstable patients who require a quick damage-control laparotomy for further resuscitation and stabilization of the patient^[19].

Other methods such as microwave coagulation, argon coagulation, injection of absolute alcohol are considered to be safe, simple and efficient methods. They can not only control bleeding, but also be effective in killing cancer cells. However large-scale experience of such methods is still not available.

In our study, morbidity from the laparotomy occurred in 11 cases (22.4%). Only one case (2.0%) was readmitted for recurrent HCC rupture 6 weeks after microwave coagulation. The overall mortality during hospitalization for laparotomy was 10.2%, lower than that associated with TAE reported by some authors^[11,14]. The mean survival time of all cases was 8.8 months. We believe that emergency laparotomy possibly can be the best treatment for spontaneous rupture of HCC. But this still needs detailed and intensive future investigation.

Spontaneous rupture of HCC often leads patients into poor prognosis. Liu et al.^[4] for the first time described the adverse effect of HCC rupture on the overall survival of these patients, which is lower than that of non-ruptured HCC patients (median survival time, 8.9 weeks vs 7 months). It has been demonstrated in many studies that liver failure is the main cause of death. Miyamoto et al.^[20] reported that among 121 cases with spontaneous rupture of HCC, 42% deceased due to liver failure. Consequently any effort to avoid damage to liver function may improve survival.

In summary, spontaneous rupture of HCC is generally considered as a potentially life-threatening situation. Sometimes a correct diagnosis maybe difficult to establish. Though recently TAE followed by a second-stage resection has been the first choice of treatment, laparotomy is still a reliable method for hemostasis, and permits consideration for resection of the tumor at the same time. Especially for hospitals where TAE is unavailable or for patients for whom TAE is unsuitable, surgical operation should be the initial option.