| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Departments of Radiology (D.S.K.L., N.C.Y., S.S.R., P.L.), Pathology (C.L., K.M.), Medicine (M.J.T., R.G.A.), and Surgery (R.W.B.), UCLA School of Medicine, 10833 LeConte Ave, Los Angeles, CA 90095-1721. From the 2003 RSNA Annual Meeting. Received January 26, 2004; revision requested April 6; revision received June 9; accepted July 20. Address correspondence to D.S.K.L.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The study was approved by the medical center Institutional Review Board, with waiver of informed consent. Forty-seven HCC nodules in 24 patients (18 men, six women; age range, 33–71 years; mean age, 56 years) were treated with single or double RF ablation sessions prior to liver transplantation. Histologic data from hematoxylin-eosin staining of explanted liver specimens were retrospectively reviewed to determine treatment success, which was defined as the absence of residual or recurrent viable carcinoma cells at treatment site. Tumor size and the presence of large (
3 mm) abutting vessels that were observed during imaging were tested as potential predictors of treatment success or failure (Fisher exact test). In patients who underwent postablation computed tomographic (CT) or magnetic resonance (MR) imaging within 3 months prior to transplantation (21 patients with 44 tumors), imaging results were analyzed for sensitivity and specificity of residual or recurrent tumor by using histologic data as the reference standard.
RESULTS: Thirty-five (74%) of 47 ablated tumors, including 29 (83%) of 35 tumors less than 3 cm, were found to be successfully treated on the basis of histologic findings after a mean interval of 7.5 months between RF ablation and transplantation. Nodules that were successfully treated had mean maximal diameter of 2.0 cm, and nodules that were unsuccessfully treated had mean maximal diameter of 3.1 cm (P = .014). Seven (47%) of 15 perivascular lesions were successfully treated whereas 28 (88%) of 32 nonperivascular lesions were successfully treated (P < .01). Imaging correlation showed 100% (33 of 33) specificity and 36% (four of 11) sensitivity of postablation CT and MR imaging for the depiction of residual or recurrent tumor.
CONCLUSION: Histologic evidence directly validates RF ablation as an effective treatment of small (<3 cm) HCC.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
5 cm or multiple [up to three] lesions 3 cm each) (4). Even for patients who meet the acceptance criteria and are suitable for transplantation, scarce organ supply is limiting, and tumor progression during the prolonged waiting period results in a substantial patient dropout rate (5). In this setting, radiofrequency (RF) ablation has emerged as an alternative treatment modality with promising preliminary outcome data for nonresectable HCC (6–8). This modality may also be useful as a bridge to liver transplantation to curb disease progression and thereby decrease patient dropout rate (9). RF treatment failure may occur as a result of incomplete tumor coagulation and can be detected at follow-up imaging as residual or recurrent tissue that is abnormally enhanced (10–12). Factors previously reported to be associated with a greater likelihood of treatment failure include large tumor size and the presence of large vessels, which act as a "heat-sink," located adjacent to the tumor (13–15). Almost all prior clinical studies have used clinical and imaging follow-up to determine local treatment success rates, and the results have been variable (16–19). Thus, the purpose of our study was to retrospectively evaluate the effectiveness of RF ablation of HCC by using findings from histologic examination of the explanted liver.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
-fetoprotein levels or nodule growth over time (13 patients). In patients with multiple lesions, the smaller nodules were deemed to be HCC as long as each nodule exhibited imaging characteristics similar to those of the dominant lesion. The total patient population comprised 18 men and six women (age range, 33–71 years; mean age, 56 years). Of these patients, those with less than a 3-month interval between the last imaging follow-up and transplantation (21 patients with 44 tumors) were included in the imaging comparison portion of the study.
RF Ablation Technique
All tumors were treated percutaneously by two radiologists (D.S.K.L., S.S.R.) with 7 and 5 years experience in RF ablation. Thirty-four tumors were treated with 2.0-, 3.0-, or 3.5-cm electrodes (LeVeen; Radiotherapeutics, Mountain View, Calif). Eight tumors were treated with single or clustered cool-tip electrodes (Radionics, Cambridge, Mass). Five tumors were treated with 2–3-cm model 30 and 3-cm Starburst XL electrodes (Rita Medical, Mountain View, Calif). Fewer patients were treated with the Rita Medical device than were treated with the Radiotherapeutics or Radionics device because of the long wait that occurred during the study period between the use of early model 30 electrodes and the introduction of later Starburst models. Regardless of the device used, the same guiding technical principles were applied to all ablations. In all patients, single or overlapping ablations were performed on the basis of lesion size and geometry, as well as on findings from ultrasonography (US) or computed tomography (CT) performed during ablation. The goal was to achieve complete ablation both of the visible tumor and, when feasible, of a 5–10-mm margin of normal surrounding tissue. Three of the 47 tumors were treated with a second ablation session within 6 months as deemed necessary at follow-up.
Each RF ablation cycle consisted of energy application that was based on the manufacturer’s recommended protocols for each device. For the Radiotherapeutics device, each ablation cycle consisted of a manual increase in power output until impedance started to rise. The power setting was then kept constant, and impedance was allowed to rise until the device automatically shuts off the power. For the Radionics device, cooled water was circulated through the electrode needles to keep the temperature at the electrode tip between 10° and 25°C during active heating. Power was applied by using an automated pulsing algorithm that was based on impedance, with a maximum cycle time of up to 12 minutes. For the Rita device, power application was primarily based on reaching the target temperature (usually set between 95° and 110°C) as measured at the electrode tip.
For US-guided procedures (in 34 tumors), one of two US machines was used (HDI 3000 or 5000; ATL, Bothell, Wash). Gray-scale imaging was used to monitor US changes continuously at the site of ablation. During the procedure, an area of echogenicity typically develops, which reflects the size of the ablation zone. Single or multiple overlapping ablations were strategically placed such that the total hyperechoic zone covered the intended volume of ablation without extending to critical structures, such as the central bile ducts or liver capsule that is contiguous with the gallbladder, bowel, and pericardium. RF ablation was performed by using CT guidance when the index tumor was not adequately visualized at US (13 tumors). CT guidance was performed by using spiral CT scanners (HighSpeed, GE Medical Systems, Milwaukee, Wis; or Somatom, Siemens Medical Systems, Erlangen, Germany). Precise placement of the electrode tips was confirmed before the initiation of active heating. Intermittent scanning was also performed during ablation and often depicted gas bubbles, as well as a subtle decrease in the attenuation of the area being ablated. No discrete margin of the ablated zone, however, was typically seen. In four patients, small doses (50–80 mL) of intravenous contrast material (iohexol, Omnipaque 350; Nycomed Amersham, Princeton, NJ) were infused to better outline the coagulated zone when necessary.
Retrospective Review of Histologic Findings in the Explanted Liver
All explanted liver specimens, which consisted of sections measuring 1.0 cm or smaller through the entire liver, were processed by pathologists according to routine protocol. All lesions and suspicious areas underwent standard histologic staining with hematoxylin-eosin stain. For tumors smaller than 2.0 cm in diameter, the slides encompassed the entire lesion. For larger lesions, representative samples were taken for slide preparation; special attention was given to areas that appeared viable at gross evaluation, according to routine practice. For the purpose of this study, all slides were reviewed retrospectively by a single pathologist (C.L.) with 6 years experience in analyzing postablation HCC histologic findings to assess the completeness of coagulation and to determine the presence of any viable carcinoma cells. Treatment success was defined as the lack of any viable cancer cells at the treatment site on the basis of this histologic review.
Comparison between Imaging Follow-up and Histologic Findings
In 23 patients, posttreatment dual-phase contrast material–enhanced CT was performed within 1 month (usually within a few days) of RF ablation and every 3 months thereafter. If iodinated contrast material was contraindicated, as was the case in one patient, then multiphase gadolinium-enhanced magnetic resonance (MR) imaging was performed. At our institution, CT was performed by using single- or multi–detector row spiral CT scanners (CTi or Lightspeed, GE Medical Systems; or Sensation 16, Siemens Medical Systems) with a dual-phase protocol consisting of precontrast imaging, hepatic arterial dominant phase imaging, and portal dominant phase imaging (collimation, 5–7 mm; pitch, 1–1.5). MR imaging was performed with 1.5-T imagers (Horizon, GE Medical Systems; or Vision, Siemens Medical Systems) and consisted of precontrast T2-weighted spin-echo imaging, as well as precontrast and dynamic postcontrast fat-saturated T1-weighted gradient-echo imaging. For MR imaging, an intravenous contrast agent (gadodiamide, Omniscan; Nycomed Amersham) was used during dynamic postcontrast imaging. Postablation images were analyzed and compared with preablation images. Any preexisting tumor region not encompassed by the new coagulation zone was considered residual tumor even if distinct nodular enhancement was not clearly seen, and repeat ablation was performed when feasible. On follow-up scans, any new discrete foci of abnormally enhanced tissue located within or along the margin of the coagulation zone were considered residual tumor, and, in cases of uncertainty, additional follow-up scans were obtained to determine stability. Growth of the questioned areas was used as a criterion for diagnosing residual tumor. If clinically feasible, some of the residual tumors were retreated with a second ablation session before patients eventually underwent transplantation. The adequacy of treatment was assessed by analyzing imaging findings on follow-up scans, as is described above, after either single-session or multisession ablation treatments.
Only findings from patients who underwent imaging within 3 months of transplantation were included for comparison with histologic findings from the explanted liver (44 tumors in 21 patients). The prospective clinical readings of these imaging results were compared with the histologic findings of treatment outcome. Clinical readings were performed by one of five attending abdominal radiologists at our institution who had experience in liver imaging ranging from 5 to 35 years. Comparison was based on the presence or absence of any viable tumor at the ablation site, without strict regard for exact localization within that site.
Perivascular Classification
All tumors were retrospectively classified as either nonperivascular or perivascular, with the latter group consisting of those tumors abutting vessels greater than or equal to 3 mm in diameter. The classification was made by one of two radiologists (P.L. and D.S.K.L., with 3 and 13 years of abdominal imaging experience, respectively) who reviewed preablation CT, MR, or US findings and were blinded to patient outcome. Images were reviewed on printed film, and calipers were used for measurements. If no printed images were available, images were retrieved electronically and reviewed on a computer workstation. Any tumor with any contiguity to vessels greater than or equal to 3 mm in diameter was considered perivascular, regardless of the length of tumor involvement.
Statistical Analysis
The dependence of treatment outcome on several variables (tumor size, perivascular vs nonperivascular, sex, age, and RF device type) was tested by using the Fisher exact test for nominal variables and the t test for continuous variables. A threshold P value of .05 was chosen to indicate a statistically significant difference. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the last imaging follow-up were derived and compared with histologic findings from the explanted liver slides. Tests were performed by using a statistical software program (InStat, version 3.05; GraphPad Software, San Diego, Calif).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
Fifteen of the 47 tumors included in this study were found to be perivascular, which was defined as any tumor abutting a vessel measuring 3 mm or larger in diameter. Treatment was successful in seven (47%) of 15 perivascular lesions and in 28 (88%) of 32 nonperivascular lesions (P = .009). Conversely, seven (20%) of 35 tumors that were successfully treated and eight (67%) of 12 tumors that were unsuccessfully treated were perivascular.
Of note, a tight correlation was seen between tumor size and perivascular status. Perivascular tumors had a mean maximal diameter of 3.2 cm, while nonperivascular tumors had a mean maximal diameter of 1.8 cm (P < .001, t test). Four (13%) of 30 tumors smaller than or equal to 2.5 cm were determined to be perivascular. Among larger tumors, 11 (65%) of 17 were perivascular.
All other variables (age, sex, and RF ablation device) did not show a statistically significant difference when tested individually against treatment outcome (Table 3).
|
Intrahepatic lymphovascular invasion was noted in three of 12 tumors that had viable malignancy remaining in the explant. Five previously undetected HCC nodules in three patients were found at histologic sectioning. During the mean posttransplantation follow-up of 26.2 months, four of 24 patients had died (two died of perioperative complications, one died of recurrent hepatitis C and liver failure, and one died of graft rejection). Three patients were not followed up at our institution within the year before this writing. Of note, no patient was found to have recurrent or metastatic HCC on the basis of autopsy findings or follow-up imaging.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Liver transplantation in a patient with a history of RF ablation provides a unique opportunity to histologically examine the explanted liver to determine the efficacy of treatment in a more rigorous and accurate manner. Such studies in the literature are few and convey varying results. Fontana et al (9) observed on 14 postablation explants, of which only three showed complete necrosis. In a similar study, Pulvirenti et al (20) noted no residual disease in 10 (75%) of 14 tumors, a finding that is comparable to CT findings. In the latter study (20), however, viable satellite nodules were noted around several of the tumors that exhibited complete necrosis.
Taking advantage of the high volume of liver transplantations performed at our institution, we studied a relatively larger series of postablation explanted livers. Using histologic criteria, we determined that 35 (74%) of 47 index tumors showed no residual carcinoma at or adjacent to the ablation site after an average of 7.5 months after ablation. The lower local control rate found at histologic analysis, as opposed to the local control rates found at imaging follow-up (as described earlier), was anticipated given the expected higher sensitivity of histologic findings in demonstrating viable carcinoma. Nevertheless, this success rate is higher than the success rates reported in previous series that were based on histologic criteria, and it supports RF ablation as a robust, minimally invasive modality for management of selected HCC.
Many of the treated tumors that did not have 100% necrosis were found to be almost completely necrotic, with only small pockets of residual neoplasm. Although complete eradication is ideal, debulking ablations to reduce the size of the tumor may curb disease progression and thereby improve outcome, particularly for patients awaiting liver transplantation. Hence, defining treatment success strictly as complete coagulation may be simplistic and limiting in some cases. The benefit of pretransplantation RF ablation, however, currently remains unproved and must be carefully weighed against potential complications.
Had imminent transplantation not been anticipated, then the majority of cases in which residual or recurrent tumor was detected at CT would have been amenable to repeat ablation treatment, with the potential achieving a local cure. Even those partially viable tumors that were not observed at CT may have been detected eventually during subsequent follow-up imaging after an interim period of growth, at which time a repeat ablation session could be pursued. The ability of patients to undergo multiple treatments, which could boost the ultimate local control rates above those reported in our current study, is indeed a considerable advantage of RF ablation. Another potential factor affecting the success rate is the percutaneous ablative approach that was used in our patient population; intraoperative or laparoscopic ablation in appropriate cases may result in better rates of tumor coagulation (14,15), albeit at the cost of higher morbidity.
The discrepancy between radiologic and histologic outcomes can be attributed mostly to the insensitivity of CT in demonstrating small remnants of viable carcinoma. Of the 11 tumors with positive histologic findings that were included for imaging correlation, only four were detected at CT (36% sensitivity). Previously Dromain et al (21) reported a prospective CT sensitivity of 44% for depicting residual tumor at 2 months after ablation. In that study (21), the presence of disease was defined as eventual development of positive CT or MR imaging findings after a mean follow-up of 19 months. Though this criterion is a reasonable one, histologic examination may have revealed more tumors with incomplete response, thereby lowering the sensitivity of cross-sectional imaging further. The low sensitivity of cross-sectional imaging for the early detection of small remnants of viable tumor soon after ablation is expected. This serves as the basis for a policy of vigilant postablation imaging surveillance at short intervals to identify recurrence once the residual tumor grows to a radiologically detectable size.
Other imaging modalities have been tested for the depiction of residual tumor after RF ablation. Contrast-enhanced pulse-inversion harmonic US and power Doppler US were useful, but the sensitivity of these modalities was below that of CT or MR imaging (22,23). Nevertheless, because it provides a timely depiction of tumor remnants that require further ablation, contrast-enhanced US can provide a convenient method for monitoring the progress of the procedure. Fluorodeoxyglucose positive emission tomography was used in a small study with 100% accurate detection of liver tumor recurrence in eight of 13 treated patients (24). However, colorectal metastasis accounted for most of these cases, and only two patients had HCC. These techniques, furthermore, remain under investigation. We currently recommend postablation surveillance by using dual-phase contrast-enhanced CT or multiphase gadolinium-enhanced MR imaging at least every 3 months.
Finally, we attempted to eliminate factors affecting treatment effectiveness. Univariate analysis showed that increasing size was an independent predictor of incomplete response to treatment. This finding is consistent with many previous reports and reflects a well-known limitation of RF ablation and other local ablative therapies (9,13,15,25). In addition to tumor size, the presence of large (3-mm or larger) vessels located adjacent to the tumor is also a known risk factor for partial treatment response because of the well-described "heat-sink" effect that mediates convective heat loss into those vascular structures (15,26,27). This influence of peritumoral vessels was reproduced in our study. However, given the tight correlation between tumor size and perivascular status in the setting of a relatively small sample size, we were unable to analyze systematically the independent contribution of each variable to treatment outcome. Nevertheless, it seems reasonable to assume, on the basis of the previously mentioned data, that both factors have a direct role in determining the likelihood of achieving complete tumor coagulation.
The determination of cell viability based solely on hematoxylin-eosin staining was also a limitation in this study. Some authors advocate the use of vital histochemical stains to assess tumor kill definitively (28,29). The problem with hematoxylin-eosin staining, however, is encountered mostly when tissue is harvested immediately after ablation, before obvious cellular or structural and staining characteristics of necrosis develop over 24–72 hours (30,31). Also, even if hematoxylin-eosin staining is inadequate, it would favor an underestimation of the treatment success because the inaccuracy of hematoxylin-eosin staining arises from its low sensitivity for the detection of cell death (28). On the other hand, limited sectioning for microscopic analysis may have missed viable carcinoma because of insufficient sampling. Three subcentimeter nodules in two patients with multifocal HCC were ablated on the basis of suspicious CT characteristics and the amenability of these nodules to RF treatment given the visibility of the nodules at US. Although larger lesions from these patients were confirmed as HCC with histologic findings, true malignancy of these particular small nodules could not be proved retrospectively. Such nodules were part of only a small subset of the sample size but, nevertheless, may have contributed to a slight overestimation of RF ablation efficacy.
In summary, we describe our success with percutaneous RF treatment of HCC in patients who eventually underwent liver transplantation. While many studies have focused on postablation outcomes, such studies have largely relied on follow-up imaging to gauge local control. Here, we report treatment efficacy based on histologic outcome. As previously reported elsewhere (21) and as confirmed in this study, cross-sectional imaging modalities are fairly insensitive in demonstrating residual disease after ablation. Histologic findings provide a direct, rigorous, and powerful method of assessing the adequacy of tumor kill, and these findings validate RF ablation as a robust option for conservative management of HCC in patients who are unable to tolerate resection. Nevertheless, difficulty with larger and perivascular tumors highlights the necessity for further advances in ablation systems and techniques to produce more consistently uniform and sizable ablations when needed.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, D.S.K.L.; study concepts and design, D.S.K.L.; literature research, N.C.Y.; clinical studies, D.S.K.L., S.S.R., R.W.B., C.L., M.J.T., R.G.A.; data acquisition, N.C.Y., P.L., C.L., K.M.; data analysis/interpretation, N.C.Y., D.S.K.L.; statistical analysis, N.C.Y., P.L.; manuscript preparation, D.S.K.L., N.C.Y.; manuscript definition of intellectual content, D.S.K.L., N.C.Y., S.S.R.; manuscript editing, D.S.K.L., S.S.R.; manuscript revision/review, D.S.K.L., S.S.R, M.J.T., R.W.B., R.G.A.; manuscript final version approval, D.S.K.L., N.C.Y.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  O. Seror, G. N'Kontchou, M. Ibraheem, Y. Ajavon, C. Barrucand, N. Ganne, E. Coderc, J. Claude Trinchet, M. Beaugrand, and N. Sellier Large (>=5.0-cm) HCCs: Multipolar RF Ablation with Three Internally Cooled Bipolar Electrodes--Initial Experience in 26 Patients Radiology, July 1, 2008; 248(1): 288 - 296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M W LEE, Y J KIM, S W PARK, H J JEON, J G YI, W H CHOE, S Y KWON, and C H LEE Percutaneous radiofrequency ablation of liver dome hepatocellular carcinoma invisible on ultrasonography: a new targeting strategy Br. J. Radiol., May 1, 2008; 81(965): e130 - e134. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 A. X. Zhu and G. K. Abou-Alfa Expanding the Treatment Options for Hepatocellular Carcinoma: Combining Transarterial Chemoembolization With Radiofrequency Ablation JAMA, April 9, 2008; 299(14): 1716 - 1718. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Belghiti, B. I. Carr, P. D. Greig, R. Lencioni, and R. T. Poon Treatment before Liver Transplantation for HCC Ann. Surg. Oncol., April 1, 2008; 15(4): 993 - 1000. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. W.-T. Lam, K. K. Ng, K. S.-H. Chok, T.-T. Cheung, J. Yuen, H. Tung, W.-K. Tso, S.-T. Fan, and R. T. P. Poon Incomplete Ablation After Radiofrequency Ablation of Hepatocellular Carcinoma: Analysis of Risk Factors and Prognostic Factors Ann. Surg. Oncol., March 1, 2008; 15(3): 782 - 790. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Wong, C.-J. Lin, C.-C. Lin, W.-T. Chen, I. H. Y. Cua, and S.-M. Lin Combined Percutaneous Radiofrequency Ablation and Ethanol Injection for Hepatocellular Carcinoma in High-Risk Locations Am. J. Roentgenol., March 1, 2008; 190(3): W187 - W195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hakime, A. Hines-Peralta, H. Peddi, M. B. Atkins, V. P. Sukhatme, S. Signoretti, M. Regan, and S. N. Goldberg Combination of Radiofrequency Ablation with Antiangiogenic Therapy for Tumor Ablation Efficacy: Study in Mice Radiology, August 1, 2007; 244(2): 464 - 470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schwartz Liver Transplantation: the Preferred Treatment for Early Hepatocellular Carcinoma in the Setting of Cirrhosis? Ann. Surg. Oncol., February 1, 2007; 14(2): 548 - 552. [Full Text] [PDF]
|
|
|
|
|
|
C. J. Simon, D. E. Dupuy, D. A. Iannitti, D. S. K. Lu, N. C. Yu, B. I. Aswad, R. W. Busuttil, and C. Lassman Intraoperative Triple Antenna Hepatic Microwave Ablation Am. J. Roentgenol., October 1, 2006; 187(4): W333 - W340. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hines-Peralta, V. Sukhatme, M. Regan, S. Signoretti, Z.-j. Liu, and S. N. Goldberg Improved Tumor Destruction with Arsenic Trioxide and Radiofrequency Ablation in Three Animal Models Radiology, July 1, 2006; 240(1): 82 - 89. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. U. Hines-Peralta, N. Pirani, P. Clegg, N. Cronin, T. P. Ryan, Z. Liu, and S. N. Goldberg Microwave Ablation: Results with a 2.45-GHz Applicator in ex Vivo Bovine and in Vivo Porcine Liver Radiology, April 1, 2006; 239(1): 94 - 102. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
