| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, APHP-Hôpital Beaujon, 100 Boulevard du General
Leclerc, Clichy 92118, France.
2 Department of Pathology, APHP-Hôpital Beaujon, Clichy 92118,
France.
3 Department of Radiology, Ospedale Specializzato in Gastroenterologia
"Saverio de Bellis" IRCCS, Castellana Grotte (Bari) 70013,
Italia.
4 Istituto di Radiologia, Università di Palermo, Via del Vespro 127,
90127 Palermo, Italy.
5 Department of Hepatology, APHP-Hôpital Beaujon, Clichy 92118,
France.
6 Department of Digestive Surgery and Transplantation Unit, APHP-Hôpital
Beaujon, Clichy 92118, France.
7 INSERM unité 773, CRB3. Faculté Xavier-Bichat, 16 rue Henri
Huchart, 75018 Paris, France.
Received December 20, 2004;
accepted after revision April 6, 2005.
Address correspondence to P. Y. Brillet.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Twenty-six patients with 35 hepatocellular carcinomas were addressed for transplantation. Complications of the procedures were recorded. Primary and secondary technique effectiveness and causes of exclusion from the waiting list were assessed. After transplantation, tumor recurrence was evaluated for at least 1 year in all patients. Radiologic-pathologic comparison of the explant was performed.
RESULTS. Percutaneous radiofrequency ablation was performed in 21 (81%) patients for 28 tumors. Both minor and major complications occurred in three patients (10% each per session). The rates of primary technique effectiveness, secondary technique effectiveness for percutaneous radiofrequency ablation alone (seven tumors), and combined percutaneous radiofrequency ablation and transcatheter arterial chemoembolization (three tumors) were 56%, 76%, and 86%, respectively. After a mean follow-up of 11.9 months, 16 patients (76%) received transplants, whereas five patients were excluded from the waiting list because of distant tumor progression (n =3, 14%) or other causes (n = 2, 10%). After transplantation, tumor recurred in one (6%) of 16 patients. Histopathologic examinations were performed for 13 (81%) of 16 patients and showed complete necrosis and satellite nodules in, respectively, 12 (75%) and seven (44%) of 16 tumors.
CONCLUSION. Percutaneous radiofrequency ablation can be performed on hepatocellular carcinoma patients waiting for transplantation, allows most patients to undergo transplantation, and does not impair posttransplantation outcomes. The procedure produces complete necrosis of the treated tumor in most cases but is associated with a high rate of satellite nodules.
Keywords: ablation • cancer • hepatocellular carcinoma • liver transplantation • radiofrequency
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Therefore, pretransplantation treatments, including minimally invasive techniques and surgery, have been proposed to delay tumor progression. The advantages of transcatheter arterial chemoembolization [6-8], a treatment that can be combined with percutaneous therapies [9, 10], include its ability to treat large areas of hepatocellular carcinoma and to cause marked tumor necrosis [7-10]. Indeed, some authors have speculated that complete tumor necrosis could decrease tumor recurrence and improve survival outcome [8-10]. However, the severity of side effects with transcatheter arterial chemoembolization makes its use controversial, with acute hepatic decompensation occurring in 20% of patients [11]. The advantages of surgical removal of the tumor include the ability to analyze vascular invasion and satellite nodules near the tumor, two predictors of survival after transplantation [12, 13]. However, resection is not suitable in patients with compromised liver function (Child B and C) because of significant morbidity and mortality. Moreover, resection could be associated with a higher operative mortality during transplantation [14].
Percutaneous radiofrequency ablation is a minimally invasive technique that increasingly is being used for the treatment of hepatocellular carcinoma [15, 16]. Evaluation of feasibility in a large series has shown a favorable safety profile [17]. Studies have shown that percutaneous radiofrequency ablation can be performed on patients waiting for transplantation [18-20]. However, no studies have been performed to determine whether percutaneous radiofrequency ablation is applicable to all patients addressed for transplantation.
Percutaneous radiofrequency ablation produces complete necrosis of small hepatocellular carcinoma tumors in most cases [21]. However, tumor recurrence is frequent, becomes more probable over time, and was found to be 59% after a 1-year follow-up [20]. Therefore, tumor progression exceeding the transplantation criteria [18-20] has been reported. Because detailed pathologic analysis of the whole tumor and the surrounding liver is rarely available [18-20], the causes of hepatic recurrence, whether from residual nonablated tumor, seeding due to percutaneous radiofrequency ablation, or satellite nodules—that is, daughter lesions—remain unclear.
The aims of this prospective study were, first, to determine if percutaneous radiofrequency ablation is suitable for all patients with hepatocellular carcinoma waiting for liver transplantation; second, to assess treatment efficacy in patients who are waiting for transplantation; and third, to compare the radiologic and pathologic findings.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study ended in December 2004 when all 26 patients who underwent percutaneous radiofrequency ablation either had been excluded from the waiting list before transplantation or had undergone transplantation and then been followed up for a mean of more than 12 months to evaluate the short-term outcome. This study was approved by our local institutional review board, and patients were asked to give their written consent before any interventional procedure.
The diagnosis of hepatocellular carcinoma was based on the Barcelona
criteria [1]: core biopsy (18
gauge) and histologic criteria (n = 13) or imaging for tumors of at
least 2 cm (n = 13). On CT, hepatocellular carcinoma was considered
to be a well- or ill-defined lesion, hypoattenuating on unenhanced scans,
showing moderate and inhomogeneous or mixed hyperattenuation during the
hepatic arterial phase, with inhomogeneous isoattenuation or low attenuation
during the portal venous phase and lower attenuation during the delayed phase.
Additional features supporting the imaging diagnosis of hepatocellular
carcinoma were the presence of a discrete partial or complete capsule or a
mosaic appearance on portal venous and delayed-phase images. The criteria for
hepatocellular carcinoma on MRI were a mass showing heterogeneous
hyperintensity, compared with liver parenchyma, during the hepatic arterial
phase; lesion hypointensity on delayed contrast-enhanced images; and moderate
hyperintensity on T2-weighted images. After an initial investigation, 35
hepatocellular carcinoma tumors were diagnosed in these 26 patients: one
(n = 19 patients), two (n = five patients), or three
(n = two patients) per patient, with a mean diameter of 24 ±
11 (SD) mm (range, 10-50 mm). The 
-fetoprotein level was below 15 ng/mL
in 13 patients, between 15 and 200 ng/mL in eight patients, and above 200
ng/mL in five patients.
Helical CT and MRI Protocols
Helical multiphasic CT was performed on all patients using a Twin Flash
(Marconi Medical Systems) unit with a double-detector array, a scanning time
of 1 sec for a 360° rotation, 5-mm contiguous sections, and a pitch of 1.
After undergoing unenhanced scanning of the liver, all patients underwent a
helical examination including both hepatic arterial phase and portal venous
phase imaging, at 30-35 sec and 70 sec, respectively, after IV infusion of a 2
mL/kg dose of nonionic contrast material (iohexol [Omnipaque],
Nycomed-Amersham) administered at a rate of 4 mL/sec with a mechanical power
injector (Medrad). Delayed-phase imaging was performed 5 min after contrast
medium injection for all 26 patients.
MRI was performed on 22 of 26 patients. Four patients did not undergo MRI because of contraindications (n = 1) or because they refused the procedure (n = 3). Examinations were performed on a 1.5-T magnet (Gyroscan Intera, Philips Medical Systems) with a maximum gradient strength of 40 mT/m and a slew rate of 200 msec, using multiarray torso coils for reception of the signal. Pulse sequences included respiratory-triggered T2-weighted fat-suppressed turbo spin-echo imaging (TR/TE, 1,600/70; flip angle, 90°; field of view, 34 cm; reconstruction matrix, 512 x 512; number of sections, 24; section thickness, 8 mm; number of signals acquired, two) and breath-hold T1-weighted fast field-echo imaging (216/5.1; flip angle, 80°; field of view, 34 cm; matrix, 200 x 256; number of sections, 24; section thickness, 8 mm; number of signals acquired, one) with or without fat suppression. A T1-weighted fat-suppressed gradient-echo MRI sequence was performed on all patients during the hepatic arterial and portal venous phases (at 20 and 50 sec, respectively) after administration of chelates of gadolinium, 0.1 mmol/kg (Dotarem [gadoterate dimeglumine], Laboratoire Guerbet), followed by a 20-mL saline solution flush. Delayed-phase imaging was performed 5 min after contrast medium injection for all 22 patients.
Percutaneous Radiofrequency Ablation Procedure
Before percutaneous radiofrequency ablation, patients were prospectively
evaluated with sonography in B-mode to assess the visibility of tumors and of
the surrounding anatomy. Coagulation parameters were evaluated in all
patients. Patients whose tumors could not be visualized on sonography,
patients with tumors near major vessels or bile ducts, and patients with a
platelet count below 50,000/mm3 (< 50 x 109/L)
(Table 1) or a prothrombin time
less than 50% of the predicted values
[17] were given other
pretransplantation treatments.
|
Percutaneous radiofrequency ablation was performed by two experienced radiologists with 3 and 4 years of experience with that procedure. An anesthesiologist was present throughout the procedure. Patients received conscious sedation with fentanyl citrate (Sanofi-Synthelabo) and propofol (Diprivan, AstraZeneca), and vital signs were monitored continuously before, during, and after treatment. Local anesthesia was achieved by injecting 1% lidocaine hydrochloride (Xylocaine, AstraZeneca). Using sonographic guidance (Sonoline Siena, Siemens Medical Solutions), a multitined 15-gauge expandable electrode (RITA Medical Systems) with seven or nine hooks (respectively, a 3- or 5-cm diameter when deployed) was inserted into the tumors, with the nine-hook electrodes used for tumors measuring at least 30 mm. After correct positioning of the needle had been verified under real-time sonographic guidance, the electrode was attached to a 250-kHz generator (RITA Medical Systems) and ablation was started. During tumor ablation, a thermocouple embedded in the electrode tip continuously measured the local temperature, and tissue impedance was monitored continuously by circuitry in the generator.
With seven-hook needles, an application process with two 6-min phases at maximum temperature was performed according to the manufacturer's recommendations. First, the hooks were deployed at 2 and 3 cm in the distal portion of the tumor. Then, the needle was withdrawn and redeployed 1 cm closer. With nine-hook needles, a four-phase application was used for a total time of 15 min at maximum temperature, with 2-, 3-, 4-, and 5-cm deployment of the hooks in the distal portion of the tumor. Next, the needle was withdrawn and redeployed 2 cm closer. Overlapping ablations were performed for tumors exceeding the needle size. Power was gradually increased to 90 and 130 W for seven- and nine-hook needles, respectively, and applied so that the average maximum temperature was between 100°C and 110°C. At the end of the procedure, the needle track was treated by thermocoagulation.
Assessment of Treatment Response
The percutaneous radiofrequency ablation was considered technically
successful if the tumor appeared completely hyperechoic on sonography
immediately afterward. Cases in which complete coverage of the tumor was
uncertain were noted.
The safety of percutaneous radiofrequency ablation was evaluated by recording the complications. Complications were considered minor or major. Major complications were defined as life-threatening or resulting in a substantially longer hospital stay [22]. Common undesired consequences of the procedure—that is, mild pain, postablation syndrome, and minimal asymptomatic perihepatic fluid or blood collections seen on imaging—were considered side effects [22] and not reported as complications. The time at which complications occurred after percutaneous radiofrequency ablation was categorized as immediate, periprocedural, or delayed if within 6-24 hr, within 30 days, or later, respectively. Impaired coagulation parameters were evaluated as potential risk factors for complications.
Follow-up imaging was performed at 1 month after percutaneous radiofrequency ablation using CT and MRI. Afterward, follow-up imaging was performed every 3 months for 1 year and every 6 months thereafter until transplantation, using either CT or MRI because both techniques are effective in assessing the efficacy of percutaneous radiofrequency ablation [23].
The response to percutaneous radiofrequency ablation was evaluated on follow-up imaging by quantifying tumor necrosis and calculating the primary and secondary technique effectiveness rates. Necrosis was considered complete or partial when residual nonablated tumor was seen as an irregular, peripheral nodular enhancement [22-24] after the first imaging follow-up. The primary technique effectiveness rate was defined as the percentage of tumors that showed complete necrosis after one percutaneous radiofrequency ablation session. The secondary technique effectiveness rate was defined as the percentage of tumors that underwent successful repeated ablation after identification of local tumor progression. When local tumor progression could not be controlled by percutaneous radiofrequency ablation alone, additional therapy could be used— either a combination of ablation and transcatheter arterial embolization using gelatin sponge pledgets (Gelfoam, Upjohn) or transcatheter arterial chemoembolization alone if the tumor neared major vessels or bile ducts.
To evaluate the efficacy of percutaneous radiofrequency ablation in managing patients on the waiting list for transplantation, we noted the causes of exclusion from the list. Tumor progression exceeding the transplantation criteria was referred to either as local progression (not controlled by percutaneous radiofrequency ablation alone or with additional embolization, and a tumor at least 5 cm in diameter) or as distant progression (with more than three hepatocellular tumors or metastases).
To evaluate the tumor recurrence rate after transplantation, we
prospectively examined patients every 4 months during the first year and every
6 months thereafter with -fetoprotein determinations, liver sonography,
and helical multiphasic CT.
Image Interpretation
Images were interpreted by consensus by two radiologists with 6 and 15
years of experience. Before percutaneous radiofrequency ablation, the size
(mm) and location of tumors—that is, subcapsular or near large
vessels—were evaluated. During imaging follow-up after percutaneous
radiofrequency ablation, the two radiologists evaluated the size (mm) of the
ablation zone [22] (defined as
the area of low-attenuation or intensity that did not enhance after
administration of contrast medium using CT or MRI); the amount of tumor
necrosis (complete or partial); the presence of periablational enhancement
(which was considered to be benign if it was a relatively concentric,
symmetric, and uniform thin rim
[22-24]);
the number and size (mm) of new hepatocellular carcinoma tumors (defined as
satellite nodules if they were observed within 1 cm from the ablation zone or
as distant hepatocellular carcinoma if farther away); and the presence of
arterioportal shunting (defined as wedge-shaped or irregularly shaped
homogeneous enhancement observed in the arterial phase only) in the liver
parenchyma adjacent to the areas treated by percutaneous radiofrequency
ablation [23,
24].
Pathologic Evaluation
Within 24 hr after transplantation, the explanted livers were cut into 5-
to 10-mm-thick sections and examined by a pathologist with 15 years of
experience. Serial samples of the tumor nodules and background cirrhosis were
stored in 10% formalin until fixation was complete and were subsequently
embedded in paraffin. For microscopic examination, serial samples of tumoral
nodules were stained with H and E.
The pathologist assessed the size (mm) of the ablation zone; whether complete or partial tumor necrosis was present (defined as areas of thermal fixation [25] and coagulation necrosis [25-27]); the number and size (mm) of satellite nodules (defined as nodules within 1 cm of the coagulation necrosis) or of distant hepatocellular carcinoma (defined as nodules farther away); the presence of vascular invasion; and the presence of tumoral seeding along the needle track. Fibrous change on the nontumoral tissue interface was defined as the peritumoral fibrotic capsule.
Data Analysis
Pathologic data were reviewed, and imaging and pathologic findings were
compared by consensus between the pathologist and the two radiologists. The
following criteria were evaluated as potential risk factors for incomplete
necrosis, satellite nodules, and distant nodules: the size of hepatocellular
carcinoma on imaging before percutaneous radiofrequency ablation (
25 mm);
the location of tumors; the presence of arterioportal shunting on imaging
after percutaneous radiofrequency ablation; the number of percutaneous
radiofrequency ablation sessions performed before transplantation (> 1);
the number of applications per tumor (> 2); the extent of tumor necrosis on
pathologic examination; the presence of vascular invasion; and the waiting
time before transplantation (> 12 months).
Statistical Analysis
The risk for development of distant nodules was analyzed on a patient
basis; tumor necrosis and the risk for development of satellite nodules were
analyzed on a tumor basis. The Fisher exact test was used for statistical
analysis, and a p value of less than 0.05 was considered
statistically significant. To evaluate the efficacy of imaging follow-up in
diagnosing partial necrosis, we used kappa statistics to calculate the
agreement between imaging follow-up and pathologic examination. The values
obtained were interpreted as follows: 0-0.3, poor; 0.31-0.5, low; 0.51-0.7,
moderate; 0.71-0.9, good; 0.91-1, very good.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Details of complications and their time of occurrence are presented in Table 2. Minor complications were observed in three patients (10% per session): transient severe pain in the right upper abdominal quadrant, requiring analgesics (patient 2); severe pain in the shoulder, requiring analgesics, which was due to brachial plexus compression secondary to hyperextension of the upper extremity during the procedure (patient 4); and thrombosis of a segmental right portal venous branch, which regressed with anticoagulant therapy (patient 10). Major complications were observed in three patients (10% per session): a ruptured pseudoaneurysm of the hepatic artery resulting in hemobilia, which was successfully treated by an endovascular procedure (patient 13); necrosis with evidence of gas formation, which was treated by antibiotherapy in a patient who underwent percutaneous radiofrequency ablation combined with transcatheter arterial embolization (patient 18); and peritoneal bleeding with a fatal outcome secondary to treatment of a subcapsular tumor (patient 21). All complications except one were observed in patients with a platelet count below 100,000/mm3 (< 100 x 109/L). The three patients with major complications had tumors that were subcapsular or near a large vessel and were treated using nine-hook needles because of tumor size. Two of these patients (patients 13 and 21) had a platelet count below 60,000/mm3 (< 60 x 109/L). No patient's Child class was modified after percutaneous radiofrequency ablation.
Follow-Up After Percutaneous Radiofrequency Ablation
In the 21 patients who received percutaneous radiofrequency ablation, the
mean time from the initial investigation to liver transplantation (n
= 16) or exclusion from the waiting list (n = 5) was 11.9 ± 5
months (range, 3-22 months) (Fig.
1).
|
Evaluation of the extent of necrosis after the first percutaneous radiofrequency ablation was available for 25 tumors in 20 patients, because patient 21 was excluded from analysis (Table 2). Complete necrosis was observed in 14 tumors; thus, the primary technique effectiveness rate was 56%. Seven tumors underwent repeated percutaneous radiofrequency ablation, and complete necrosis was achieved in five, two with percutaneous radiofrequency ablation alone and three with combined percutaneous radiofrequency ablation and transcatheter arterial embolization. Therefore, the secondary technique effectiveness rates were 76% and 86% using percutaneous radiofrequency ablation alone and the combined treatments, respectively. The two tumors successfully treated by combined percutaneous radiofrequency ablation and transcatheter arterial embolization were adjacent residual nonablated tumors (patient 18). Conversely, partial necrosis was observed in a residual nonablated tumor near the inferior vena cava and hepatic veins (patient 15) that was treated by transcatheter arterial embolization alone.
Among the five (24%) of 21 patients excluded from the waiting list, three (14%) were excluded because of distant tumor progression exceeding transplantation criteria secondary to new hepatocellular tumors (patients 2 and 19) and lung metastasis (patient 4). These three patients were excluded from the waiting list more than 6 months after the initial workup. The two remaining patients (10%) were excluded for reasons other than tumor progression: severe portopulmonary hypertension (patient 5) and a fatal outcome after percutaneous radiofrequency ablation (patient 21).
Follow-Up After Liver Transplantation
No perioperative complication was related directly to percutaneous
radiofrequency ablation. At a mean follow-up time after transplantation of 25
± 12 months (range, 0-40 months), 11 of 16 transplantation patients
(69%) were alive and five of the 16 transplantation patients had died: two
(12%) in the perioperative period and three (19%) several months after
transplantation. Recurrence took place in one of the 16 transplantation
patients (6%), in whom adrenal gland metastasis developed.
Radiologic-Pathologic Comparison
The mean time between the last session of percutaneous radiofrequency
ablation and liver transplantation was 8.8 ± 5.3 months (range, 1-19
months), and the mean time between the last imaging follow-up and liver
transplantation was 2 ± 2.3 months (range, 0-7 months). Three of the 16
transplantation patients were excluded from histopathologic analysis because
they underwent either transcatheter arterial embolization (patients 15 and 18)
or tumorectomy (patient 1) before transplantation. Therefore, histopathologic
examination (Table 3) was
performed for 13 (81%) of 16 transplantation patients.
|
Complete necrosis after percutaneous radiofrequency ablation was observed at pathologic analysis in 12 (75%) of 16 tumors for nine (69%) of 13 patients (Figs. 2A, 2B, and 2C). In one of the four tumors with partial necrosis, residual nonablated tumor was due to bundles of tumoral cells inside a peritumoral fibrotic capsule (patient 11). The mean diameter of the ablation zone for the 16 treated tumors was 26 ± 8 mm (range, 15-45 mm) at imaging follow-up and 24 ± 10 mm (range, 10-45 mm) at pathologic examination.
|
|
|
= 0.53), with
underestimation of necrosis on imaging in two cases (12%) and failure to
detect incomplete necrosis on imaging in the case of microscopic tumor inside
a peritumoral capsule (patient 11). Small satellite nodules (all < 3 mm) and microvascular invasion were observed at histopathology in seven and four (44% and 25%, respectively) of the 16 treated tumors (Figs. 2A, 2B, 2C, 3A, 3B, 3C, and 3D). Therefore, residual tumor in 10 (62%) of 16 tumors was because of satellite nodules (n = 6), partial necrosis (n = 3), or both (n = 1). In seven patients (54%), 13 distant hepatocellular carcinoma tumors (size, < 1-25 mm) were diagnosed on histopathologic analysis. Therefore, additional tumor foci, either local or distant hepatocellular carcinoma, were observed in 11 (85%) of 13 patients. Imaging detected none of the satellite nodules and only three of the 13 distant hepatocellular carcinomas, all being at least 10 mm (patient 10). No neoplastic seeding was observed along the needle track.
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-fetoprotein level. Compared with these
studies, none of our patients were excluded from the waiting list because of
local tumor progression. Our results showed that percutaneous radiofrequency
ablation successfully controlled local tumor progression, even though a
combined treatment with transcatheter arterial embolization was necessary to
manage 10% of patients until transplantation. The rate of dropout that was
directly related to tumor progression was 14% after a waiting time of 11.9
months, and exclusion was due to distant tumors, either new hepatocellular
carcinoma (n = 2) or metastasis (n = 1). Our results are in
agreement with others [19] and
confirm the ability of percutaneous radiofrequency ablation to bridge patients
to transplantation. The second goal when treating hepatocellular carcinoma in patients waiting for liver transplantation is to obtain complete coagulation necrosis of the tumor, an event that is associated with both low tumor recurrence and improvement of disease-free survival after transplantation [8-10]. In our series, complete necrosis was seen on imaging follow-up after one session in 56% of tumors. After multiple sessions or with combined treatments, the secondary technique effectiveness rates were 76% and 86%, respectively. Combined treatment produced complete necrosis of two adjacent residual, nonablated 2-cm tumors and one residual tumor near a large vessel—known risk factors for local recurrence. As expected, tumor size > 2-3 cm [20, 21], and the proximity to a large vessel, acting as a heat sink [30], were identified as risk factors for tumor recurrence. Conversely, we did not identify the subcapsular location of tumor [21] as a risk factor. The results of our pathologic analysis confirmed the high rate of complete coagulation necrosis of hepatocellular carcinoma after percutaneous radiofrequency ablation. With complete necrosis achieved in 75% of tumors after percutaneous radiofrequency ablation—a rate similar to those previously reported [18, 20]—this treatment could improve posttransplantation outcomes.
However, percutaneous radiofrequency ablation can have serious drawbacks, especially in patients waiting for liver transplantation. First, percutaneous radiofrequency ablation is not appropriate for all patients. In this prospective study, patients with small tumors that were not detected on sonography, and patients with a coagulation deficiency that contraindicated a percutaneous approach, were not treated. A second drawback of percutaneous treatment is complications, mainly hemorrhagic. In our series, we reported 10% minor and 10% major complications per session—rates that are higher than those initially reported [15, 16] and are in accordance with the recent study by Mazzaferro et al. [20]. Because of the lengthening waiting time before transplantation in Western countries, the strategy to select patients with hepatocellular carcinoma for liver transplantation is changing [4] and patients with deteriorating liver function or active hepatitis are often selected first. Thus, patients waiting for transplantation may be more susceptible to complications. Because most complications occurred in patients with a platelet count below 100,000/mm3 (< 100 x 109/L) and two major complications occurred in patients with a platelet count below 60,000/mm3 (< 60 x 109/L), we hypothesize that a poor coagulation status is a risk factor for complications in patients waiting for transplantation and that the benefit of platelet transfusion should be questioned. Moreover, we showed that all major complications occurred in tumors that were subcapsular or near a large vessel and were treated using nine-hook needles. Therefore, our results suggest that percutaneous radiofrequency ablation is probably not suitable in all patients waiting for transplantation.
Finally, even if percutaneous radiofrequency ablation induces complete necrosis, it does not prevent recurrence or the formation of new nodules while patients are on the waiting list. In our series, we reported additional tumor foci, either local or distant from the treated hepatocellular carcinoma, in 11 (85%) of 13 patients. Locally, tumor recurrence was mainly due to satellite nodules, observed in 44% of tumors, and complete necrosis of the treated tumor was achieved in most cases. The high frequency of satellite nodules also has been reported by Pulvirenti et al. [18], and their occurrence might be more frequent after radiofrequency ablation than after transcatheter arterial chemoembolization before transplantation [9, 31].
Distant new hepatocellular carcinoma tumors were observed in 54% of patients, a finding that is in accordance with that of Cha et al. [12], who reported a 55% recurrence rate at a median follow-up of 26 months after resection of hepatocellular carcinoma. However, because of underlying active liver disease, hepatic tumor recurrence after successful treatment of hepatocellular carcinoma is frequent whether surgical or minimally invasive treatment is applied.
We failed to identify any significant risk factors for the development of distant nodules, but we noted a higher frequency of satellite nodules in patients with arterioportal shunting after percutaneous radiofrequency ablation (p = 0.04). Therefore, our study raises the question of whether mechanical lesions induced by percutaneous radiofrequency ablation could increase the frequency of satellite nodules. To our knowledge, such a relationship has not been reported before. If confirmed by other studies, this relationship may affect the management of patients undergoing percutaneous radiofrequency ablation. However, we did not report any increase in hepatocellular carcinoma recurrence after transplantation, and the value of satellite nodules in determining an unfavorable prognosis after percutaneous radiofrequency ablation remains unclear from our study.
This study evaluated helical multiphasic CT and MRI for the diagnosis of residual local tumors and additional foci of hepatocellular carcinoma. Residual tumors were diagnosed successfully on imaging in all patients except in one with benign periablational enhancement and microscopic tumor bundles. Conversely, most of the additional foci of hepatocellular carcinoma were not detected on imaging. Because most of them were small (< 10 mm in diameter), our study confirmed the low sensitivity of helical multiphasic CT and MRI in the detection of small hepatocellular carcinoma tumors (< 2 cm) before liver transplantation [32, 33].
One limitation of our study was the relatively low number of patients. However, the prospective design of our study allowed us to reach our objectives, including evaluation of the feasibility of percutaneous radiofrequency ablation in a pretransplantation population and radiologic-pathologic comparison. In addition, we reported a low dropout rate from the waiting list due to tumor progression (14%), but comparison with other treatments and historical controls should be investigated further.
Today, for a single hepatocellular carcinoma tumor smaller than 5 cm in a Child A patient waiting for transplantation, excision remains recommended because it enables pathologic analysis of the tumor for vascular invasion and satellite nodules, which can affect the transplantation strategy [13]. However, strategies based on liver resection are still being discussed, and authors have reported an increased operative mortality from transplantation [14]. Belghiti et al. [13] recommended removal of tumors that are in the upper part of the right liver, because the tumors can be removed easily through a thoracic incision but may be difficult to target through a percutaneous approach.
Compared with surgery, minimally invasive techniques may be proposed for use in more advanced cirrhotic disease. Percutaneous ethanol injection and transcatheter arterial chemoembolization induce a lower rate of complete necrosis than does radiofrequency ablation and could be less effective in control of local progression [34]. Conversely, our results show that percutaneous radiofrequency ablation induces mechanical lesions that could favor the development of satellite nodules. A multitechnique, minimally invasive approach combining transcatheter arterial chemoembolization and percutaneous radiofrequency [35] could be an ideal treatment for 3- to 5-cm tumors by increasing the percentage of tumor necrosis and minimizing the rate of additional nodules.
In conclusion, our study showed that percutaneous radiofrequency ablation can be performed on patients with hepatocellular carcinoma who are waiting for transplantation. The technique allows most patients to undergo transplantation, does not impair posttransplantation outcomes, and provides complete necrosis of the treated tumor in most cases but is associated with a high rate of satellite nodules.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
