INTRODUCTION
Hepatic cavernous hemangioma (HCH) is the most common benign tumor of the liver, its incidence in necropsy series ranges from 0.4% to 7.4%. It occurs at all ages and affects women predominantly, the sex ratio is 1.5-6:1, and may increase in size with pregnancy or with administration of estrogens. The majority of HCHs are small, single and asymptomatic. About 20% are large (> 5 cm) and 10% to 29% are multiple[1-5]. The largest HCH reported in the literature was 63 cm × 48.5 cm × 40 cm in size and 18 kg in weight[5]. They are often discovered incidentally on routine abdominal ultrasonography or at the time of laparotomy and necropsy. The widespread use of ultrasound (US) and computerized topography (CT) scanning nowadays has made the diagnosis of HCH easier. HCHs usually become symptomatic as they reach a certain size. Abdominal pain or discomfort is the most common complaint. Rarely, large HCHs sequester and destroy platelets, causing symptomatic thrombocytopenia, known as Kasabach-Merritt syndrome. Occasionally, these vascular lesions rupture spontaneously, or with minimal trauma, and present as hemoperitoneum[1-6].
The management of HCHs is still controversial. In situ radiofrequency ablation (RFA) of hepatic malignant tumors[7-21] has led us to consider using this minimally invasive technique in patients with HCH. On the basis of our successful experience in controlling hepatic malignances, we applied the percutaneous RFA (PRFA) guided by B-US to the treatment of patients with HCH and the initial results were promising[22,23]. This article describes our preliminary experiences and evaluates the efficacy, safety, and complications of PRFA therapy for HCHs.
MATERIALS AND METHODS
The study was performed at the RFA center of our department with the approval from the hospital authority. Written informed consent was obtained from each patient prior to treatment.
Twelve consecutive patients with 15 cavernous hemangiomas of the liver were treated with PRFA from May 2000 to June 2002. There were four males and eight females. Age ranged from 33 to 56 years (mean, 41.7 years). Eleven patients had a single lesion (4.8 cm to 9.5 cm), and one patient had four lesions (2.5 cm to 7.0 cm). The tumors were located in the right lobe of the liver in nine patients, left lobe in five, and both lobes in one. Nine patients had discomforts in the right upper quadrant of abdomen, while the remaining three were asymptomatic. The definite histologic diagnosis of HCHs was established by means of enhanced spiral CT scan and B-US guided percutaneous biopsy with 16-gauge biopsy needles in all cases.
The patients were fasted for 4 hours and under conscious sedation and analgesia induced by administering promethazine hydrochloride 25 mg, pethidine hydrochloride 50 mg and morphine hydrochloride 10 mg intravenously. An intercostal or subcostal approach was selected according to the location of hemangiomas in the liver. Local anesthesia was achieved by using 5 ml to 8 ml of 1% lidocaine. The 10-needle LeVeen electrode, which was connected to the RF-2000 generator (Radio Therapeutics Corp., Mountain View, CA, USA), was inserted percutaneously into and fully deployed within the target tumor under real-time B-US guidance. Grounding was achieved by attaching two dispersive pads on the thighs of patients. The power was set at 30 watts initially and increased by 10 watts per minute until 90 watts were delivered via the needle electrode. The tissue impedance was monitored automatically by using circuitry incorporated within the generator and the treatment was stopped at roll off. The ablation overlaps were intended so as to destroy HCHs completely. For tumors larger than 3 cm in diameter, the tip of LeVeen needle electrode was readjusted repeatedly under real-time B-US guidance until the entire lesion as well as a rim of normal liver tissue (approximately 0.5 cm) were encompassed and destroyed. The vital signs of patients were monitored continuously during the procedure. The patients were allowed to have a light diet six hours after PRFA therapy. Prophylactic antibiotics and liver protectives were given intravenously for 2 to 3 days. To evaluate the response of hemangiomas to PRFA therapy, B-US and enhanced spiral CT scan were performed regularly by using the same parameters as used for the pretreatment examination. Tumor size was observed, and complete tumor necrosis was considered when no focal enhancement was seen within the tumor or at its periphery on spiral CT scans obtained 3 to 6 months after the treatment.
RESULTS
All treatments were technically successful. When the patients were treated under sedation and analgesia, they experienced a mild pain during the procedure, which disappeared immediately following cessation of PRFA. As radiofrequency energy was applied to the treatment probes, a hyperechoic focus was observed by B-US around the uninsulated portion of the electrodes, which was attributed to the tissue vaporization, microbubble formation and tissue necrotic coagulation. There was a single treatment session in one tumor which was 2.5 cm in diameter, and 2-7 treatment sessions in 14 tumors which were 4.8 cm to 9.5 cm in diameter. The duration of the treatment sessions was quite different from patient to patient, and even from lesion to lesion in the same patient. The average treatment duration was 39 min (8-125 min). The liver function study showed that serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were increased two to three times above the baseline values, and normalized 2-5 days posttreatment. No change was observed in other serum element levels. No PRFA-related complications including bleeding or bile leakage occurred, and no repeated PRFA treatment was needed. The post-PRFA course was uneventful, and the patients were discharged on the third to fifth day postprocedure. The fact that the hospital stay was longer than that in Western countries was largely due to our local preferences and social reasons.
All the patients were followed up for 9 to 34 months (mean, 21 months). Nine patients’ discomforts in the right upper quadrant of abdomen were ameliorated, the symptoms were significantly or completely controlled in 7 patients for a mean of 2 months (range 1-4 months). B-US and CT scan showed that the ablated lesions were swollen within 2 to 4 weeks after PRFA and started to shrink thereafter. The shrunk range was 38%-79% (mean, 67% per 21 months). The contrast enhancement was disappeared within the tumor or at its periphery in all cases on spiral CT scans obtained 3 to 6 months after treatment.
DISCUSSION
Traditionally, surgical resection has been considered the choice of treatment for HCHs including those of symptomatic, larger than 10 cm in diameter, and manifested as Kasabach-Merritt syndrome. Steroid therapy and external radiation have provided relief for a few patients[1-5]. The transcatheter arterial embolization (TAE), an effective interventional therapy for hepatic malignancies, has also been advocated for the treatment of HCHs[24,25]. However, there are some differences of hepatic hemodynamics between hepatic malignancies and HCHs[26,27]. The value of TAE as well as its disadvantages have not been fully evaluated. Though TAE has the advantages of avoiding operative intervention and prolongation of hospitalization, TAE may result in disastrous consequences including complications of severe destructive biliary damages and ectopic embolizations[28-30].
Recent successes in local treatment of malignant tumors of the liver have led us to consider using the minimally invasive PRFA in patients with HCHs. In the procedure, the needle-electrode with an insulated shaft and an uninsulated distal tip is inserted directly into hepatic tumor through percutaneous approach under local anesthesia and conscious sedation by the guidance of ultrasonography. Energy is transferred from the uninsulated distal tip of the needle to the tissue as current rather than as direct heat. As the alternate current flows to the grounding pads, it agitates ions in the surrounding tissue of the needle tip, resulting in frictional heat exceeding 80-100 °C. The local tissue is heated and dried, creating a 3.5 cm to 5.0 cm oval or spherical thermal coagulative necrosis. Thus the tumors are ablated and destroyed in situ. The dead cells are not removed, but they eventually shrink and become scar tissues[7-17].
On the basis of our successful experience in controlling hepatic malignances with PRFA therapy, we applied this technique to the treatment of HCHs with a considerable success, and our initial experience was reported in 2001[22,23]. To our knowledge, this is the first publication on the use of percutaneous radiofrequency ablation therapy for hepatic cavernous hemangiomas. We observed that the patients who accepted PRFA recovered quickly, the ablated lesions were remarkably shrunk after treatment, and no minor or major complications occurred. PRFA has an additional advantage of being easily repeated, so large HCHs can be retreated and the benefits of it to HCHs are definite, which is quite different from the repeat PRFA for hepatic malignancies in which local tumor control is greatly influenced by its size[7-12]. In this group, we managed to adjust LeVeen needle electrode repeatedly under real-time B-US guidance, and large HCHs were completely ablated, and repeat treatment was not needed. Our study showed that this technique was safe and effective for the treatment of HCHs in selected patients. An important limitation of PRFA is related to location of HCHs, in which the near-hilum, near-gallbladder and submembranous HCHs should not be heated lest the tumor’s surrounding tissues or organs be damaged. The interference with the ribs, lung, and adjacent bowel may also affect the procedure.
It is well documented that HCHs should not be disturbed by attempts at biopsy or aspiration, lest serious bleeding occurs, so percutaneous biopsy is contraindicated[1-3]. But several studies have also shown that percutaneous fine needle biopsy with ultrasound guidance is safe for HCH diagnosis[31-35]. We performed percutaneous biopsy successfully in patients with HCHs unexpectedly in our early diagnostic liver biopsies. Along with the experience with recent practice of PRFA for hepatic malignances, we performed both B-US guided biopsy with 16-gauge biopsy needle and PRFA therapy with 15-gauge LeVeen electrode needle in patients with HCHs. There was no hemorrhage or other complications, indicating that both percutaneous biopsy and PRFA therapy under ultrasonography guidance can be performed safely for the diagnosis and treatment of HCHs.
In conclusion, the results of this study suggest that PRFA is a promising and minimally invasive technique for the treatment of HCHs. Long-term studies to assess the clinical outcome of PRFA for HCHs are ongoing. We believe that by virtue of its mini-invasion, effectiveness, simplicity, safety, and easy repeatability, PRFA therapy can be the choice of treatment for HCHs in selected patients.
Edited by Zhang JZ and Wang XL
