contrast enhanced ultrasound for radio frequency ablation of canine prostates: initial results

Contrast Enhanced Ultrasound for RadioFrequency Ablation of Canine Prostates: Initial ResultsJi-Bin Liu,* Daniel A. Merton, Gervais Wansaicheong, Flemming Forsberg, Pamela R. Edmonds,Xue-Dong Deng, Yan Luo, Laurence Needleman, Ethan Halpern and Barry B. GoldbergFrom the Departments of Radiology (JBL, DAM, FF, LN, EH, BBG) and Pathology (PRE), Thomas Jefferson University Hospital,Philadelphia, Pennsylvania, Departments of Ultrasound, Nanjing Medical University Suzhou Hospital (XDD), Suzhou and SichuanUniversity Huaxi Hospital (YL), Chengdu, People’s Republic of China, and Department of Diagnostic Radiology, Tan Tock Seng Hospital(GW), Singapore

Purpose: We determined the feasibility of contrast enhanced ultrasound for radio frequency ablation of the entire prostateas a method of minimally invasive treatment for prostate cancer in a canine model.Materials and Methods: Approval of the Institutional Animal Use and Care committee was obtained. Initially 5 dogs (group1) were tested using variable power (5 to 30 W), time (4 to 12 minutes), bolus (0.01 to 0.04 ml/kg) and infusion (3 to 11 ml perminute at 0.015 �l/kg) injections of an ultrasound contrast agent with conventional grayscale power Doppler and pulseinversion harmonic imaging to establish optimal parameters. Subsequently 4 dogs (group 2) underwent entire prostateablation using parameters based on group 1. The size of the thermal lesions and residual viable tissue was measured withImageJ software (National Institutes of Health, Bethesda, Maryland) on ultrasound and pathological study. Linear regres-sion and Student’s t test were used for statistical analysis.Results: A bolus of 0.04 ml/kg, an infusion of 11 ml per minute at 0.015 �l/kg and the contrast enhanced pulse inversionharmonic imaging mode were ranked best for guiding ablation. Thermal lesion volume was proportional to ablation powerand time. There was no significant difference in measured thermal lesion size in group 1 between ultrasound and pathologicalfindings (mean � SD 1.51 � 0.74 and 1.46 � 0.74 cm3, p � 0.56) or in residual viable tissue in group 2 (0.43 � 0.043 and 0.41� 0.291 cm3, p � 0.21). The average volume of prostate ablation achieved in group 2 was 96.3%.Conclusions: Contrast enhanced pulse inversion harmonic imaging is able to guide, monitor and control radio frequencyablation of the entire prostate.

Key Words: ablation, ultrasonography, prostate, contrast media, neoplasms

Prostate cancer is the most commonly diagnosed cancerand the most common cause of cancer death in Amer-ican men. It is estimated that in 2004 approximately

230,110 new cases of prostate cancer were diagnosed in theUnited States.1 There is little consensus regarding treat-ment for various disease stages.2 A major reason is thebiological variability of the disease.3 Currently there are 3predominant strategies for treating localized prostate cancerconfined to the gland, including 1) clinical observation withor without hormonal therapy, 2) external beam radiation orbrachytherapy and 3) radical prostatectomy.4 These treat-ment methods target the entire gland. Complications of ra-diation therapy and radical prostatectomy include cystitis,proctitis, impotence, incontinence and infection.5,6

To minimize the complications associated with prostatecancer therapy, while eliminating the risk of multifocal tu-mors, alternative treatments for localized prostate cancerthat preserve vital structures must be explored.7 RF ther-

Submitted for publication September 1, 2005.Study received Institutional Animal Use and Care committee ap-

proval.Supported by National Institutes of Health Grant EB002794.* Correspondence: Department of Radiology, 7th Floor, Main

Building, 132 South 10th St., Thomas Jefferson University Hospi-
tal, Philadelphia, Pennsylvania 19107 (telephone: 215-955-4862;FAX: 215-955-8549; e-mail: [email protected]).
0022-5347/06/1764-1654/0THE JOURNAL OF UROLOGY®

Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION

1654

mal ablation has been used experimentally and clinically fortreating hepatic and other tumors throughout the body.8

Current imaging modes are unable to reliably distinguishviable tissue/tumor from necrotic tissue during ablation.This may lead to over or under treatment.9 Although US isused for guiding electrode needle placement, conventionalfindings on grayscale and Doppler imaging without a con-trast agent cannot identify residual viable tissue/tumor.10,11

Previous studies have shown that contrast enhanced US canreveal the prostatic vasculature.12,13

We hypothesized that contrast enhanced US imaging canbe used to facilitate detection of the vasculature and paren-chymal perfusion in viable prostate tissue for monitoring RFablation of the entire prostate gland. In this study we in-tended to optimize ablation, contrast injection and imagingparameters in contrast enhanced transrectal US for guiding,monitoring and controlling RF ablation of the prostate. Wealso investigated the effectiveness of normal periurethralblood flow as a natural heat sink effect to protect the urethrain an animal model.

MATERIALS AND METHODS

RF Ablation SystemA commercially available RF generator (Radionics, Burling-
ton, Massachusetts) was used for prostate ablation. The
Vol. 176, 1654-1660, October 2006Printed in U.S.A.

DOI:10.1016/j.juro.2006.06.090

ULTRASOUND FOR RADIO FREQUENCY ABLATION OF CANINE PROSTATES 1655

operating frequency of the system was 480 kHz with a poweroutput of 0 to 250 W at 65 � load. A Cool-Tip™ electrode wasused to deliver RF energy. The 17 gauge electrode wasinternally cooled with ice water circulated by a hydraulicPE-PM perfusion pump (Radionics). To complete the electri-cal circuit an external grounding pad was affixed to the backof the subject.

US Contrast Agent and Imaging ModesSonazoid™ was selected for this study.12,13 The protocolused for the US contrast agent was a continuous infusionthroughout the RF ablation procedure and an intravenousbolus injection before ablation, after completing each RFablation and at the end of the entire ablation procedure. Forbolus injection a dose of 0.01, 0.02, 0.03 and 0.04 ml/kgSonazoid® was given intravenously at a rate of 1 to 2 ml persecond, followed by a 5 ml saline flush to clear the injectionline. For continuous infusion the rates were varied (3, 4, 5and 11 ml per minute) with a contrast dose of 0.015 �l/kg.Transrectal US was performed using an Elegra™ scannerwith a 6.5EC10 endocavitary probe using a low mechanicalindex that was kept below 0.4 to minimize microbubbledestruction. Grayscale contrast enhanced power Doppler im-aging and contrast enhanced PIHI modes were evaluated inthe first dog.

Three researchers (JBL, FF and DAM) experienced withcontrast sonography qualitatively assessed enhancement ofthe prostatic vasculature, prostatic perfusion, the ability toperform real-time monitoring of the prostate and ablationarea, and artifacts during imaging. Based on the results inthe first dog the same imaging mode and dose of US contrastmaterial was used in each subsequent dog in groups 1 and 2.

Animal ModelThis protocol was approved by the Institutional Animal Useand Care committee. Nine adult male mongrel dogs weredivided into 2 groups, including 5 in group 1 and 4 in group2. The dogs were sedated and anesthetized before the exper-iments. Cardiac and respiratory parameters were monitoredthroughout the procedures. An intravenous fluid channelwas established through an 18 gauge angiocatheter placedin a forelimb vein.

Ablation ParametersOn baseline transrectal grayscale images prostate volumewas calculated by one of us (DAM) by measuring the diam-eter in 3 orthogonal dimensions. Cross-sectional imageswere acquired throughout the prostate at 5 mm intervalsbefore and after ablation.

In group 1, 1 thermal lesion was produced on each side ofthe prostate at the mid gland position for a total of 10thermal lesions. The diameter of the thermal lesions wasmeasured on contrast enhanced PIHI 10 to 15 minutes aftereach RF ablation. A combination of variable time (4 to 12minutes) and power output settings (5 to 30 W) on the RFgenerator was tested (table 1).

In group 2 multiple overlapping volumes of RF ablationwere used to ablate the entire gland. Ablation parameterswere adjusted, so that a larger field was used in the midportion of the gland and a smaller area was used at the baseand apex of the gland. After each RF ablation a sufficient
interval (10 to 15 minutes) was allowed for cooling of the
ablated site to avoid transient hyperechoic artifacts beforebolus contrast imaging was performed to assess residualviable tissue as areas with marked contrast perfusion. Al-though most ablation was done with an electrode with a 2 cmtip, an electrode with 1 cm exposed tip was used for precise,targeted ablation of small residual areas of enhancing, ieviable, tissue.

The amount of periurethral flow was qualitatively scoredin comparison to the beginning of the procedure by 2 of us(JBL and DAM). Additional protection was not used. Wholeprostrate ablation was considered to be achieved when therewas no more parenchymal enhancement, and only minimalresidual periurethral and periprostatic flow remained. Thedegree of enhancement of viable tissue compared to thethermal lesion was assessed qualitatively by 3 of us (JBL,DAM and FF). Residual viable tissue was measured frompost-ablation US imaging sections using the digital imaginganalysis software ImageJ (National Institutes of Health,Bethesda, Maryland) by one of us (JBL). Volume was calcu-lated by adding the cross-sectional area of each slice andmultiplying by slice thickness (5 mm). An ellipsoid formulacould not be used due to the irregular shape of the residualtissue.

Pathological AssessmentImmediately after the RF ablation procedure the dogs weresacrificed using established methods. The entire prostatewas excised and serially sectioned by freehand technique inthe same plane and intervals corresponding to the obliquecross-sectional US images obtained in vivo. A thorough grossinspection of the surrounding structures was done to evalu-ate any injury or burns.

In group 1 the visible cross-sectional areas of focal coag-ulation were mapped and measured in 3 dimensions withcalipers in the fresh tissue specimen before preservation forhistological analysis. In group 2 an additional assay fortissue viability was performed with TTC14 after sectioning.The irregularly shaped residual viable areas were measuredin the same manner as in vivo US findings.

Whole mount and routine histology with hematoxylin andeosin stain for light microscopy were performed by one of us(PRE). The urethra was divided into quadrants and eachquadrant was scored as positive or negative for thermaldamage.

Statistical AnalysisAll measured volumes are expressed as the mean � SD incm3. Calculated thermal lesion volumes on contrast en-hanced PIHI in group 1 were compared to volumes obtainedfrom the pathological specimens. Calculated residual viable

TABLE 1. Variable time and RF power output settings ingroup 1 dogs

Dog No.

Time (mins)/Power (W)

Rt Gland Lt Gland

1 4/25 8/52 5/25 8/103 6/25 8/204 8/25 8/255 12/25 8/30

tissue volumes in group 2 were compared to the volumes

ULTRASOUND FOR RADIO FREQUENCY ABLATION OF CANINE PROSTATES1656

obtained from the TTC stained gross pathological speci-mens. The 2-tailed paired Student t test was used to evalu-ate any systematic difference between US and pathologymeasurements using Stata® 8.0 software. In addition, aPearson correlation coefficient was calculated to quantifyhow closely the US estimation corresponded to pathologyfindings. Statistical significance was calculated by 2 of us(FF and EH) at p �0.05.

RESULTS

Conventional grayscale transrectal US was used success-fully to guide RF electrode placement in all dogs. No tech-

FIG. 1. Conventional grayscale US of prostate after ablation. Ther-mal lesion margin (asterisk) is poorly differentiated.

FIG. 2. Select contrast enhanced US images of guided ablation of canperfusion on contrast enhanced PIHI before RF ablation. B, thermPIHI of prostate after RF ablation. C, contrast enhanced PIHI of prolesion (asterisks) borders. Lesions appeared as hypoechoic areas in
periurethral and periprostatic flow. D, thermal lesion size and locatio(marked areas).
nical failures occurred. However, this mode cannot show UScontrast enhancement. Although the thermal lesion mayappear as a heterogeneous echogenic area, the lesion marginwas poorly differentiated on conventional US. Therefore, itwas not possible to distinguish coagulated from viable tissue(fig. 1).

Contrast enhanced PIHI mode showed normal prostatevasculature as a radial, spoke-like pattern with smallervessels, resulting in parenchymal enhancement of the entiregland. In addition, real-time monitoring showed peripheralarteries enhancing early in the arterial phase, followed bydrainage mainly in the urethral area in the venous phase.Thermal lesions were visible as nonperfused areas duringcontinuous infusion and they appeared as well-defined hy-poechoic areas relative to surrounding normal prostate pa-renchyma with bolus injection of contrast material (fig. 2).Also, electric artifacts from RF energy were visible on US.

The contrast enhanced power Doppler mode showed asimilar spoke-like pattern but could not demonstratesmaller vessels or parenchymal enhancement (fig. 3). Ther-mal lesions appeared as avascular areas. Blooming artifactsled to power Doppler signal extending inside the thermallesion (fig. 4). This can cause underestimation of thermallesion size.

Based on the ability to see the thermal lesions better oncontrast enhanced PIHI in the first dog in group 1 this wasselected as the optimal imaging mode in subsequent dogs.Also, the optimal dose of Sonazoid™ was found to be 0.04ml/kg for bolus injection and 11 ml per minute at 0.015 �l/kgfor infusion. These doses produced suitable vascular en-

rostate with gross pathological specimen. A, homogeneous prostaticion border (right) was barely perceptible on noncontrast enhancedafter second RF ablation (left) shows clear demarcation of thermal

trast enhanced hyperechoic prostate parenchyma. Note increased

ine pal lesstate

con
n matched hematoxylin and eosin stained histological specimen


hancement and allowed continuous monitoring of the area ofRF ablation (fig. 2).

Although continuous infusion of contrast medium wasuseful for real-time monitoring of ablation, bolus injection ofcontrast material provided increased enhancement and con-spicuity between the thermal lesion and viable tissue. Thus,the measurements obtained with contrast enhanced PIHIfollowing bolus injection were used for testing statisticalsignificance in the correlation of ablation and residual viabletissue sizes.

In group 1 mean prostate size was 18.36 � 10.30 cm3

(range 7.9 to 35.01). The mean size of ablated lesions oncontrast enhanced PIHI was 1.51 � 0.74 cm3 (range 0.57 to2.97). Mean size of ablated lesions based on gross patholog-ical measurements was 1.46 � 0.80 cm3 (range 0.42 to 3.0).There was no significant difference in predicted lesion sizedetermined with the contrast enhanced PIHI mode andgross pathological measurements using the 2-sample pairedt test (p � 0.56). Linear regression resulted in a Pearsoncorrelation coefficient r of 0.9 (p �0.001).

An oval RF ablation volume was observed in group 1. Thevolume of the thermal lesion was directly proportional to theduration and power of ablation (fig. 5). The range of volumesachieved by changing the duration of ablation at a power of25 to 30 W was greater than the range of volumes achievedby changing the power at a fixed time duration of 8 minutes

FIG. 3. Power Doppler US shows radial, spoke-like pattern of vas-cularity with peripheral arteries enhancing early, followed by ve-nous flow drainage mainly in urethral area. A, before US contrastmaterial injection. B, after injection blooming artifact was present.

in the dog prostate. Multiple ablations with overlapping

volumes in group 2 were performed to achieve ablation of theentire prostate using a power of 25 to 30 W and changing theduration of ablation to create smaller or larger thermallesions.

In group 2 mean prostate size was 9.99 � 5.8 cm3 (range1.37 to 13.85). Despite the smaller mean size the Student t

FIG. 4. Contrast enhanced US of canine prostate. A, thermal lesion(asterisk) could be delineated from hyperechoic prostate paren-chyma. Note increased periurethral and periprostatic flow. B, arti-facts due to blooming on contrast enhanced power Doppler USextended inside thermal lesion (asterisk).

FIG. 5. Ablation parameters in group 1 dogs. Thermal lesion volumewas calculated based on expected shape of prolate ellipsoid usingequation, volume � [4/3] � � d1 � d2 � d3, where d represents
ablated lesion diameter. Circle size is proportional to thermal lesionvolume created in prostate after 1 RF ablation.


test showed no significant difference compared to group 1(p � 0.19). Based on TTC stained specimens residual tissuevolume after ablation was 0.41 � 0.291 cm3 (table 2). Theaverage volume of prostate ablation achieved was 96.3%.Figure 6 shows an example of 97.2% ablation. There was nosignificant difference between the size of residual viabletissue measured on contrast enhanced PIHI images and onTTC stained specimens (0.43 � 0.043 cm3, p �0.2, fig. 7).The Pearson correlation coefficient showed a trend toward

FIG. 6. A, contrast enhanced PIHI did not show parenchymal pros-tate perfusion after ablating entire prostate in group 2. In addition,no periurethral flow demonstrated. Capsular flow (arrows) was seenon right side. B, cross-sectional gross pathological specimen at samelevel reveals complete coagulation of prostate tissue and damage to

TABLE 2. Residual tissue volume on TTC stain and prostateablation in group 2 dogs

Dog No.Residual Tissue

(cm3) % Ablation

1 0.634 95.52 0.326 97.23 0.649 95.34 0.036 97.3

urethra, as evidenced by lack of TTC staining for viable tissuearound urethra.

statistical significance (r � 0.9, p � 0.09). An average of 6ablations (range 3 to 9) was required to complete the entireprostate ablation. Periurethral and periprostatic blood flow,represented by greater echogenicity, increased during RFablation (figs. 2, C and 4, A).

No thermal damage to the bladder or rectum was foundon gross inspection. Thermal damage to the urethral wallwas demonstrated on contrast enhanced PIHI and patholog-ical study (fig. 6). Urethral damage involved between halfand three-quarters of the circumference of the prostatic ure-thra in group 2 dogs.

DISCUSSION

In 1995 a study by McGahan et al demonstrated the feasi-bility of percutaneous RF ablation of prostate tissue indogs.15 Subsequently groups evaluated the use of RF abla-tion for treating benign prostate hyperplasia and prostatecancer using a transurethral or transperineal approach.16,17

None of these groups attempted to ablate the entire prostatesince there was no established method of precisely control-ling the extent of ablation.

Tumor ablation and contrast US are known techniques.To our knowledge no one has combined contrast enhancedUS to guide RF ablation of the entire prostate. Based on thebiological behavior of prostate cancer and preliminary stud-ies4,6,10,12,13,18 we propose a new therapeutic strategy fortreating prostate cancer, that is RF ablation of the wholeprostate with contrast enhanced US guidance to monitorand control ablation.

The single tip RF electrode that we used created anellipsoid volume of ablation. This was confirmed by thecorrelation between calculated ablation volume on PIHI andpathology results. The size of the thermal lesions in thisstudy depended on time, power and the length of the elec-trode tip. This means that the size of the thermal lesionobserved under contrast US is accurate and the operator canuse this method to precisely control the area of intended

FIG. 7. Boxplots show residual viable tissue in group 2 dogs oncontrast enhanced PIHI (US) and in pathological specimens (path).Box represents median, and upper and lower quartiles. Whiskersrepresent upper and lower adjacent values with upper adjacentvalue defined as largest data point 1.5 or less � IQR and loweradjacent value defined as smallest data point 1.5 or greater � IQR.Outliers indicate data points more extreme than adjacent values.

ablation.


By overlapping ablation volumes it is possible to achieveablation of a larger and more complex volume of tissue, suchas the prostate gland. This requires the operator to have athorough understanding of ablation parameters. A criticalcomponent of monitoring ablation is the ability to determineresidual viable tissue and its relationship to vital structures.

The current study demonstrates that contrast enhancedPIHI is superior to contrast enhanced power Doppler imag-ing for monitoring the extent of tissue ablation and conven-tional US. PIHI is a US contrast specific imaging mode witha superior contrast-to-tissue signal-to-noise ratio and spatialresolution that is devoid of Doppler related artifacts.19 Us-ing a combination of harmonic imaging and low acousticpower continuous depiction of smaller vessels containing UScontrast material can be achieved.

Comparisons between the measurements obtained fromcontrast enhanced PIHI and pathological findings showedgood agreement and good linear correlation in groups 1 and2 (r � 0.9, p � 0.56 and 0.2, respectively). This suggests thatcontrast enhanced PIHI can show the true area of RF coag-ulated tissue in vivo. The ability to visualize the entireborder of the thermal ablation area during the proceduremakes it superior to cryosurgical ablation, on which only theborder of the ice ball facing the US transducer can be visu-alized.20 When tissue ablation is guided by contrast en-hanced PIHI, the RF electrode can be repositioned (withoutcomplete removal of the electrode from the gland) to a newtargeted area under imaging guidance. This decreases thenumber of electrode insertions through the capsule, which inturn decreases the risk of damage to surrounding structures.Direct visualization of the urethra and neurovascular bun-dles allows ablation to minimize damage to these criticalareas. The ability to ablate 96.3% of the prostate in thisstudy shows that it is possible to achieve the goal of totalprostatic ablation, similar to radical prostatectomy orbrachytherapy.

This study demonstrated increased echogenicity of peri-urethral flow on contrast enhanced PIHI during ablation(fig. 2). To our knowledge the reason for this is unknown. Itcould be a response to thermal damage in the prostate or areaction to increased periurethral temperature (hyperemia).It is possible the hyperemia could act as a natural heat sinkeffect but the presence of damage, as evidenced by nonviabletissue on TTC stain, indicates that it is insufficient to protectthe urethral wall during ablation. We believe that additionalcooling of the urethra may prevent damage and methods ofcooling are currently under investigation.

There are limitations to this study. The small number ofdogs in the groups limits the power of the study. Ablationparameters are likely to be equipment and to some degreecontrast agent specific. Also, the canine prostate is smallerthan the human prostate and its mobility during ablationmade the procedure more difficult. Translating these param-eters to the human prostate will require additional animalstudies.

CONCLUSIONS

This study demonstrates a new method using contrast en-hanced US for guiding and monitoring RF ablation of thewhole prostate. The advantage of the method is the ability toprecisely control thermal lesions, resulting in 96.3% ablation
of the prostate.
Abbreviations and Acronyms

PIHI � pulse inversion harmonic imagingRF � radio frequency

TTC � triphenyl tetrazolium chlorideUS � ultrasound

REFERENCES

1. Jemal, A., Tiwari, R. C., Murray, T., Ghafoor, A., Samuels, A.,Ward, E. et al: Cancer statistics, 2004. CA Cancer J Clin,54: 8, 2004

2. Cancer Statistics Review 1973–1992. National Institutes ofHealth Publication 95–2789. Bethesda: National CancerInstitute Division of Cancer Prevention and Control, 1995

3. Quinlan, D. M., Partin, A. W. and Walsh, P. C.: Can aggressiveprostatic carcinomas be identified and can their naturalhistory be altered by treatment? Urology, suppl., 46: 77,1995

4. Middleton, R. G., Thompson, I. M., Austenfeld, M. S., Cooner,W. H., Correa, R. J., Gibbons, R. P. et al: Prostate cancerclinical guidelines panel summary report on the manage-ment of clinically localized prostate cancer. J Urol, 154:2144, 1995

5. Lawton, C. A., Won, M., Pilipich, M. V., Asbell, S. O., Shipley,W. U., Hanks, G. E. et al: Long-term treatment sequelaefollowing external bean irradiation for adenocarcinoma ofthe prostate: analysis of RTOG studies 7506 and 7706. IntJ Radiat Oncol Biol Phys, 21: 935, 1991

6. Gerber, G. S., Thisted, R. A., Scardino, P. T., Frohmuller, H. G.,Schroder, F. H., Paulson, D. F. et al: Results of radicalprostatectomy in men with clinically localized prostate can-cer. JAMA, 276: 615, 1996

7. Trojan, L., Kiknavelidze, K., Knoll, T., Alken, P. and Michel,M. S.: Prostate cancer therapy: standard management, newoptions and experimental approaches. Anticancer Res, 26:551, 2005

8. Solbiati, L., Goldberg, S. N., Ierace, T., Livraghi, T., Meloni, F.,Dellanoce, M. et al: Hepatic metastases: percutaneousradio-frequency ablation with cooled-tip electrodes. Radiol-ogy, 205: 367, 1997

9. Boehm, T., Malich, A., Goldberg, S. N., Hauff, P., Reinhardt,M., Reichenback, J. R. et al: Radio-frequency ablation ofVX2 rabbit tumors: assessment of completeness of treat-ment by using contrast-enhanced harmonic power DopplerUS. Radiology, 225: 815, 2002

10. Liu, J. B., Goldberg, B. B., Merton, D. A., Rawool, N. M.,Needleman, L. and Forsberg, F.: The role of contrast-enhanced sonography for radiofrequency ablation of livertumors. J Ultrasound Med, 20: 517, 2001

11. Solbiati, L., Goldberg, S. N., Ierace, T., Dellanoce, M., Livraghi, T.and Gazelle, G. S.: Radio-frequency ablation of hepatic metasta-ses: postprocedural assessment with a US microbubble contrastagent—early experience. Radiology, 211: 643, 1999

12. Forsberg, F., Johnson, D. K., Merton, D. A., Liu, J. B., Losco,P. E., Hagen, E. K. et al: Contrast enhanced transrectalsonography of a novel canine prostate cancer model. J Ul-trasound Med, 21: 1003, 2002

13. Hagen, E. K., Forsberg, F., Aksnes, A. K., Merton, D. A., Liu,J. B., Tornes, A. et al: Enhanced detection of blood flow inthe normal canine prostate using an ultrasound contrastagent. Invest Radiol, 35: 118, 2000

14. Adegboyega, P. A., Adesokan, A., Haque, A. K. and Boor, P. J.:Sensitivity and specificity of triphenyl tetrazolium chloridein the gross diagnosis of acute myocardial infarcts. ArchPathol Lab Med, 121: 1063, 1997

15. McGahan, J. P., Griffey, S. M., Budenz, R. W. and Brock, J. M.:
Percutaneous ultrasound-guided radiofrequency electro-


cautery ablation of prostate tissue in dogs. Acad Radiol, 2:61, 1995

16. Schulman, C. C. and Zlotta, A. R.: Transurethral needleablation of the prostate for treatment of benign prostatichyperplasia: early clinical experience. Urology, 45: 28,1995

17. Zlotta, A. R., Djavan, B., Matos, C., Noel, J. C., Peny, M. O.,Silverman, D. E. et al: Percutaneous transperineal radio-frequency ablation of prostate tumor: safety, feasibility andpathological effects on human prostate cancer. Br J Urol,
81: 265, 1998
18. Hagen, E. K., Forsberg, F., Liu, J. B., Gomella, L. G., Aksnes,A. K., Merton, D. A. et al: Contrast-enhanced power Dopp-ler imaging of normal and decreased blood flow in canineprostates. Ultrasound Med Biol, 27: 909, 2001

19. Forsberg, F., Liu, J. B., Chiou, H. J., Rawool, N. M., Parker, L.and Goldberg, B. B.: Comparison of fundamental and wide-band harmonic contrast imaging of liver tumors. Ultrason-ics, 38: 110, 2000

20. Onik, G.: The male lumpectomy: rationale for a cancer tar-geted approach for prostate cryoablation. A review. Technol
Cancer Res Treat, 3: 365, 2004

contrast enhanced ultrasound for radio frequency ablation of canine prostates: initial results

Documents