|
|
||||||||
1 All authors: Interventional Ultrasound Service, "D. Cotugno" Hospital, Via Quagliariello 54, Naples, 80131, Italy.
Received March 15, 2004;
accepted after revision May 6, 2004.
Address correspondence to G. Ferraioli, Viale Marconi, 41, Cava dei Tirreni
(SA), 84013 Italy
(ferraiol{at}tin.it).
Abstract
|
|
|---|
SUBJECTS AND METHODS. Between September 2000 and October 2003, 336 consecutive patients (221 men and 115 women; age range, 4478 years; mean, 67 years) with 407 malignant liver tumors were treated at our institution using radiofrequency ablation. Of these patients, 287 had hepatocellular carcinoma from cirrhosis, 47 had liver metastases (38 from colon, six from breast, two from lung, and one from cutaneous melanoma), and two had primary cholangiocarcinoma. Adverse events related to radiofrequency ablation were prospectively recorded.
RESULTS. The number of sessions performed was 375 (39 patients had two sessions). The number of patients with major complications, including death, was three (0.9%). The overall mortality rate was 0.3% (1/336). One patient died because of worsening liver decompensation. Two other major complications occurred. In one patient (0.3%), liver abscess and sepsis developed and were successfully treated with percutaneous sonography-guided needle (18-gauge) aspiration and IV antibiotics. Mild posttreatment ascites occurred in one patient (0.3%). One patient showed self-limiting subcutaneous cellulitis along the electrode-needle path that healed in 2 weeks. Fever lasting 13 days and pain lasting 1224 hr were observed in 141 patients (42%) and 211 patients (63%), respectively. So far, no cutaneous or abdominal wall seeding has been observed clinically or sonographically.
CONCLUSION. Radiofrequency ablation of liver tumors can be considered safe. Life-threatening acute liver failure can be considered a rare possible complication.
|
|
|---|
The electrode needles that have been used for local tumor ablation are single or clustered cooled-tip needles and expandable-array needles. Both types of electrode needles have been in use for more than 8 years [2, 1115].
A more recently introduced system uses a cannulated needle electrode with multiple side holes connected to a pump filled with saline solution [1619]. The targeted lesion is coagulated with a high-frequency AC produced by a radiofrequency generator. During exposure of the lesion to the radiofrequency current, saline solution can be injected into the tumor through the needle electrode, perfusing the neoplastic tissue around the needle tip. This action increases the electric conductivity of the radiofrequency fields in the targeted tissue and prevents tissue vaporization and charring, thus increasing the volume of coagulation necrosis.
The rates of and potential risk factors for complications after radiofrequency ablation of hepatic tumors using cooled-tip needles or expandable-array electrode needles have been reported for large series of treated patients [8, 2023]. In contrast, the adverse effects of the saline-perfused cannulated electrode needle have been described only for small series [18].
So far, the saline-perfused cannulated electrode needle has been shown to be safe and has been associated with fewer cases of adverse events than have the older electrode needles [18, 19].
We report the complications that occurred in a large series of patients with primary or metastatic liver tumors treated with saline-enhanced radiofrequency under sonographic guidance at a single center during 3 years of experience.
|
|
|---|
All patients either were not eligible for or refused surgery. Serum blood tests for liver function and hemocoagulation were performed. ECG and chest radiography were also performed.
The patients to be treated with radiofrequency ablation were selected according to the following criteria: presence of a single tumor or, in the case of multiple tumors, not more than three tumors; international normalized ratio less than or equal to 1.5; platelet count greater than or equal to 60,000/mm3; absence of ascites; absence of portal vein thrombosis; absence of extrahepatic hepatocellular carcinoma metastases; and liver function impairment no more severe than Child-Pugh class B-9.
In our series of patients, international normalized ratios ranged from 0.8 to 1.5 and platelet count ranged from 385,000/mm3 to 60,000/mm3.
Fifty-seven patients with tumors deep in the right lobe of the liver were treated after local anesthesia with 10 mL of 2% lidocaine (Xylocaina, AstraZeneca), given subcutaneously. Two hundred seventy-nine patients were treated while deeply sedated through premedication with atropine (Atropina Solfato, Sifra), 0.5 mg IV, and induction with propofol (Diprivan, AstraZeneca), 912 mg/kg/hr, and fentanyl (Fentanest, Pharmacia and Upjohn), 50 µg IV. The reasons for sedation included subcapsular tumors (68 patients), large tumors (> 4.0 cm) needing multiple needle insertions and therefore long exposure to thermoablation (91 patients), and refusal of patients to remain awake during percutaneous treatment (120 patients).
All patients were treated with a new generation of radiofrequency equipment (Elektrotom 106 HiTT, Berchtold) using 15-gauge needle electrodes with an exposed active tip 2.5 cm long, having eight side holes and no terminal hole.
All treatments were performed under sonographic guidance following the
procedure previously described
[10]. Briefly, under
sonographic guidance, the needle electrode was inserted into the tumor, and
then the current generator and the pump for the saline (0.9% concentration)
injection were activated for 1015 min. At the end of each insertion,
the current generator was switched off only after the needle electrode had
been retrieved to heat the needle path and prevent seeding of tumoral cells. A
single insertion was planned for tumors with diameters up to 4.0 cm. For
larger tumors, multiple needle insertions were planned. Lesions with diameters
larger than 5 cm were treated in two sessions. The second session always took
place within 2 weeks of the first session. A sonographic examination and a
blood test were performed 24 hr after the treatment; afterward, the patients
were discharged from the hospital. Treatment efficacy was evaluated by
triphasic contrast-enhanced CT 4 weeks after the treatment. All patients were
followed up for at least 4 months. For patients with hepatocellular carcinoma,
abdominal sonography and serum
-fetoprotein testing took place every 2
months and CT or MRI every year, or earlier when indicated. Patients with
liver metastases were followed up by sonography and blood testing every 2
months and helical CT or MRI every 6 months.
The treatment was approved by the institutional review board, and informed consent was obtained from all patients before the procedure.
Complications
Adverse events related to radiofrequency ablation were prospectively
recorded. The definitions of the Society for Cardiovascular and Interventional
Radiology were used to classify major and minor complications
[24]. Occurrences of events
threatening the patient's life, leading to substantial morbidity and
disability, or requiring prolonged hospitalization were considered major
complications (i.e., hemoperitoneum, pleural effusion, portal vein thrombosis,
decompensation of liver cirrhosis, renal insufficiency, marked jaundice,
abscess, or death). All other complications were considered minor.
Patients were monitored for pain and fever for 7 days after the procedure. They were instructed to call the service should pain or fever appear.
|
|
|---|
|
|
No complications occurred after local or general anesthesia. All patients were discharged from the hospital the day after the procedure.
One (0.3%) of 336 patients died. He was a 68-year-old Child B cirrhotic patient with a single 2.0-cm-diameter nodule of hepato-cellular carcinoma in segment VIII, which was treated with a single electrode insertion. The patient showed decompensation of liver cirrhosis 1 week after the radiofrequency ablation. He presented with progressively worsening jaundice (bilirubin level = 52 mg/dL), ascites, and severe hemocoagulation impairment and died 38 days after treatment. Two other major complications (0.6%) were reported. The number of major complications, including deaths, was three (0.9%). The overall mortality rate was 0.3% (1/336).
In one patient with a 3.5-cm-diameter hepato-cellular carcinoma nodule in segment VI, a liver abscess developed from Escherichia coli and sepsis 8 days after the procedure. He was treated with percutaneous sonography-guided needle (18-gauge) aspiration (Fig. 1) and IV antibiotic therapy. Fever and leukocytosis disappeared after 2 weeks. Helical CT performed 1 month later showed complete necrosis of the tumor. Mild posttreatment ascites was observed in one patient 24 hr after radiofrequency ablation. Diagnostic paracentesis yielded a yellow ascitic fluid; a blood test for RBC showed no change. Ascites was treated by diuretic administration and resolved in 3 days. In one patient with a single 2.0-cm-diameter hepatocellular carcinoma nodule in segment III, self-limiting subcutaneous cellulitis was seen along the electrode-needle path 1 week after the procedure and healed in 2 weeks with antibiotic treatment. Fever lasting 13 days and pain lasting 1224 hr after treatment were observed in 141 patients (42%) and 211 patients (63%), respectively. Pain medication (30 mg of intramuscular ketorolac tromethamine, Toradol, Recordati) was necessary for 44 (21%) of 211 patients. So far, no cutaneous or abdominal wall seeding has been observed clinically, sonographically, or by helical CT during the follow-up.
|
|
|
|---|
Two studies have concluded that, among tumor ablation techniques, radiofrequency ablation may be the superior alternative to percutaneous ethanol injection because of a higher rate of complete necrosis and fewer treatment sessions [5, 6]. However, the complication rate is higher with radiofrequency ablation than with percutaneous ethanol injection [6].
In a multicenter study that enrolled 2,320 patients who were treated with an internally cooled radiofrequency ablation technique, Livraghi et al. [21] noted six deaths (0.3%). These were due to multiorgan failure after intestinal perforation, septic shock, massive hemorrhage after tumor rupture, or liver failure after stenosis of the right bile duct, and one death was sudden and of unknown cause. Moreover, 50 instances (2.2%) of additional major complications were encountered. The most frequent of these were peritoneal hemorrhage, neoplastic seeding, intrahepatic abscesses, and intestinal perforation. In our series, the mortality rate was similar to that of Livraghi et al. The death was related to decompensation of liver cirrhosis in a patient with a small hepatocellular carcinoma. Fewer other major complications than in Livraghi's series were observed (0.6%), and they consisted of one case of liver abscess and one case of mild ascites that was due to liver decompensation. Moreover, no cases of hemoperitoneum or hemothorax were observed. The series of Livraghi et al. consisted of patients treated in different institutions by operators with a different expertise in performing radiofrequency ablation, mainly because the number of patients they had treated was quite different.
Our center has more than 20 years' experience with percutaneous ablation therapies for focal liver lesions [27]. Therefore, the lower rate of major complications we observed may be related to factors other than the radiofrequency generator and the electrode system we used. Zagoria et al. [28], in a series of 38 patients with liver metastases, reported one death related to a hepatic abscess that developed 8 days after the radiofrequency ablation procedure.
In a series of de Baère et al. [20] on 312 patients, five deaths (1.6%) and 37 other major complications (12%) were related to radiofrequency treatment. The deaths were caused by liver insufficiency, colon perforation, or portal vein thrombosis. The higher rate of complications in that series may be due to its lack of homogeneity, since the treatments included percutaneous and intraoperative radiofrequency ablation, cooled needles, and expandable arrays. In the recent series of Buscarini and Buscarini [29] on 166 patients treated by a percutaneous radiofrequency expandable system, no deaths occurred, but the rate of major complications was higher (4.6%) and included one case of peritoneal hemorrhage and one case of cutaneous seeding. The absence of deaths could be explained by the number of patients enrolledfewer than is required to have one expected death, as noted in our study and in that of Livraghi et al. [21] (one death every 300 patients).
Interestingly, during follow-up we did not observe any cases of seeding, in contrast to the findings of others [18, 19, 29], especially Llovet et al. [30]. They reported a high rate of seeding (12%), mostly in cases of superficial lesions, after treatment of hepatocellular carcinoma with internally cooled needles.
Our results showed that radiofrequency ablation using the perfused needle electrode is a safe and reliable treatment for liver malignancies, with fewer major complications than occur with internally cooled or expandable needle electrodes. However, a sudden worsening of functional liver reserve may occur with radiofrequency ablation therapy even when only a small nodule of hepatocellular carcinoma is treated. In addition, our study confirmed previous reports [20] that the complication of liver abscess may occur after radiofrequency ablation of tumors near the intestine, probably because of contamination of the area of radiofrequency-induced necrosis by pathogens from the digestive tract. In contrast, none of the 19 patients in our series who had tumors adjacent to the gallbladder experienced the major complications that were described by other authors [31]. The results of our study confirmed the report of Chopra et al. [32] that radiofrequency ablation of lesions in the gallbladder fossa is safe.
In conclusion, saline-enhanced radiofrequency of liver tumors can be considered safe. Unfortunately, liver decompensation that may rapidly worsen and lead to life-threatening acute liver failure must be considered a possible, although rare, complication.
|
|
|---|
This article has been cited by other articles:
![]() |
J. M. Willatt, H. K. Hussain, S. Adusumilli, and J. A. Marrero MR Imaging of Hepatocellular Carcinoma in the Cirrhotic Liver: Challenges and Controversies Radiology, May 1, 2008; 247(2): 311 - 330. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. E. Maturen, H. V. Nghiem, J. A. Marrero, H. K. Hussain, E. G. Higgins, G. A. Fox, and I. R. Francis Lack of tumor seeding of hepatocellular carcinoma after percutaneous needle biopsy using coaxial cutting needle technique. Am. J. Roentgenol., November 1, 2006; 187(5): 1184 - 1187. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Giorgio, G. Ferraioli, D. Choi, and H. K. Lim Liver Abscess Formation After Radiofrequency Ablation in Patients with Hepatocellular Carcinoma Am. J. Roentgenol., February 1, 2006; 186(2): 582 - 583. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |