A Hydrophobic Gel Phantom for Study of Thermochemical Ablation: Initial Results Using a Weak Acid and Weak Base

Andrew J. Misselt, MD, Theresa L. Edelman, BS, Jeung H. Choi, PhD, John C. Bischof, PhD, and Erik N.K. Cressman, PhD, MD

PURPOSE: To develop a model for study of exothermic chemical reactions potentially useful for tissue ablation. MATERIALS AND METHODS: Seven gelatins ranging from 0.5% to 30% wt/vol with and without 15% or 30% caps and several commercial gels were evaluated. Baseline temperature measurements were taken. Acetic acid and ammonium hydroxide were sequentially injected over periods of 10 15 seconds in 1-mL aliquots, forming a discrete aqueous reaction chamber. Congo red pH indicator was included to assess the reaction. A thermocouple allowed data collection at completion of injection and every 15 seconds for 5 minutes. Injections were performed in triplicate, and average temperatures for each time point were reported. RESULTS: Gelatins fractured or refluxed even at the lowest concentrations tested. Most commercial gels proved too viscous and likewise led to reflux along the needle tract. A mineral oil based gel was selected because of its ability to form a chamber without reflux or fracture and its clear colorless character, hydrophobic nature, chemical stability, viscosity, specific gravity, and cost. Temperatures during the first 60 seconds of the neutralization reaction showed an immediate increase that correlated well with concentration. CONCLUSIONS: The oil gel phantom is a safe, useful, readily available, inexpensive model to study mixing behaviors and maximum heating potentials for reactions that may prove useful in thermochemical tissue ablation for oncologic interventions. Measurable temperature changes occurred even at the lowest concentrations, and higher concentrations produced a greater release of heat energy.
J Vasc Interv Radiol 2009; 20:13521358

INTERSTITIAL or intraparenchymal injection of chemotherapy solutions (1,2) or ablative agents (3) have been investigated by researchers in many specialties (4,5) over many years, but success has been somewhat elusive. The potential for high local concentraFrom the Departments of Radiology (A.J.M., T.L.E., E.N.K.C.) and Mechanical Engineering (J.H.C., J.C.B.), University of Minnesota Medical Center, MMC 292, 420 Delaware Street Southeast, Minneapolis, MN 55455. Received October 31, 2008; final revision received June 16, 2009; accepted June 29, 2009. Address correspondence to E.N.K.C.; E-mail: cress013@umn.edu From the SIR 2007 Annual Meeting. None of the authors have identified a conflict of interest. SIR, 2009 DOI: 10.1016/j.jvir.2009.06.027

tion of a given agent with very little systemic exposure nevertheless continues to make such methods attractive (6,7). This is particularly true in cases in which, despite the systemic nature of most cancers, a localized tumor burden is a significant factor in a patients symptoms and/or prognosis. Surgical resection is useful, but in many instances patients are not suitable operative candidates. In general, there has been more success with nonspecific chemical ablation agents such as acetic acid or ethanol than with specific drugs dissolved in a vehicle or carrier. This may be a reflection of the difficulty in obtaining even distribution within tissues and also the therapeutic index for the drug being considered. Thermal therapies for tissue destruction have also become popular,

and data are now accruing to show survival benefits in some cases (8). Thermal therapies generally have the advantage that the shape of a coagulation zone is larger and more predictable in the vast majority of cases (conduction of heat through tissues vs convection via hydrostatic interstitial pressure), but they are more susceptible to heat-sink effects leading to inadequate treatment near larger blood vessels and the equipment is costly. Exothermic chemical reactions might offer a new source of heat energy that could augment a chemical ablation. This concept was first reported by Cressman et al (9) and subsequently by others (10,11). The effect of a hostile environment from the dual components of fluid convection in tissues and conduction of heat through tissues might provide ad-

1352

Volume 20

Number 10

Misselt et al

1353

vantages over existing methods. An injected solution distributed through the tissues via convection (12) could be combined with the more predictable thermal injury from conduction as heat from an exothermic reaction is released. This might offer a new way to destroy tissue that is more controlled than injection alone and could be seen as a cross between methods such as ethanol injection (13) and boiling saline solution (14), for example. One simple example of exothermic chemistry is the neutralization reaction between an acid and a base, in which the products are heat, a salt, and water. It was desirable to begin the investigations with a familiar material already in use for tumor ablation such as acetic acid and to combine it with a relatively safe counterpart, ammonium hydroxide (Fig 1). To study this area in a detailed and systematic way, there was a need for a suitable phantom for injectable therapies undergoing a chemical reaction in situ. Such a phantom would need to be more than a simple open calorimeter for reasons discussed later. Inspection of the literature showed numerous phantoms optimized for various purposes, including thermosensitive phantoms (15,16), water mimics for dosimetry (17), and some for intraparenchymal injection (18,19) or ablation in renal (20) or hepatic tissues (21). Among intraparenchymal phantoms, injection rates orders of magnitude too slow (on the order of microliters per minute) are clearly not in a useful range for clinical practice outside the calvarium because of the size of tumor to treat, the time frame required, and respiratory and cardiac motion. In the end, none of the reported phantoms were suitable for the task at hand. Given these issues, an investigation was begun to interrogate a number of materials and identify those that might be suitable for the current purpose.

CH3CO2H + NH4OH
Figure 1.

Heat + NH4Cl + H2O

Exothermic neutralization of acetic acid with NH4OH.

Figure 2. Baby oil gel is in a T25 flask and a ring stand holds a syringe of acetic acid solution and congo red indicator (blue at acidic pH). Thermocouple with probe attached to syringe containing NH4OH solution is adjacent. Note the starting temperature of the system as indicated by the thermocouple (22.4C).

MATERIALS AND METHODS

University policy did not require institutional review board approval for this in vitro study. Baby oil gel, hair styling gel, and mineral oil (ie, baby oil) were purchased at a large retail discount store and used as supplied. Hydrogenated polyisobutene gels (Versagel ME 500, ME 750, and ME 1600; Penreco, Dickinson, Texas) and silicone viscosity standards (Brookfield Engineering Laboratories, Middleboro, Massachusetts)

were donated samples from the various manufacturers and used as supplied. Ultrasound transmission gel (Aquasonic 100; Parker Laboratories, Fairfield, New Jersey) was used as supplied. Acetic acid, NH4OH, and congo red pH indicator (Thermo Fisher Scientific; Waltham, Massachusetts) were purchased and used as supplied. A series of seven different gelatins ranging from 0.5% to 30% wt/vol with and without 15% or 30% gelatin caps were prepared and cast into small test tubes and allowed to set. The caps were intended to function as thin, superficial, but very cohesive seals to prevent reflux along the needle tract. In the cases in which caps were poured, the deep layer was refrigerated before the cap was poured. This was done to prevent the second layer from melting the first during the second pouring, which could create an uneven interface. The commercially available gel products were warmed if necessary to decrease their viscosity and facilitate pouring into small test tubes or T25 flasks. After transfer, materials were allowed to cool and equilibrate at room temperature, after which baseline temperature measurements were taken. A thermocouple probe (type T MT-29/1; Physitemp Instruments, Clifton, New Jersey) was positioned within the phantom at the point of injection to

collect temperature data over time from the moment of completion of the injection and every 15 seconds for 5 minutes (Fig 2) with a T-type thermocouple thermometer (DigiSense; Cole-Parmer, Vernon Hills, Illinois). Equimolar amounts of acetic acid and ammonium hydroxide at four different concentrations (1, 5, 10, and 15 M) were sequentially hand-injected over a period of 10 15 seconds in 1-mL aliquots to permit formation of a discrete aqueous reaction chamber (Fig 3). The pH indicator, congo red, was included in the acid solution before use to assess the extent of the neutralization reaction and completeness of mixing (Fig 4). Injections were performed in triplicate, and average temperatures for each time point were reported.

RESULTS
Visual assessments of materials were made and are reported in Table 1. Viscosity and cohesiveness in various tested materials proved to be a problem. With gelatins, fracturing occurred during injection of the solutions without exception. This led to cleavage planes and irregular distribution of the injected solutions within the phantom. Without a contained volume, dis-

1354

Thermochemical Ablation in Hydrophobic Gel Phantom: Initial Results

October 2009

JVIR

Figure 3. Injection of acetic acid solution, colored blue by the congo red pH indicator. Note the rounded shape of the solution contained within the gel model, facilitating mixing of the aqueous solution within the hydrophobic gel.

Figure 4. Contained neutralization reaction within the gel model. Note the clear NH4OH solution in the syringe, which has reacted with the previously dispensed (blue-colored) acetic acid solution. Note the color change of the congo red indicator demonstrating the neutralization reaction.

Table 1 Results of Initial Injections to Assess Materials Material Gelatin Concentration/Product No. 0.5% wt/vol 0.5% wt/vol; 0.5% wt/vol; 1% wt/vol 1% wt/vol; 1% wt/vol; 2% wt/vol 2% wt/vol; 2% wt/vol; 5% wt/vol 5% wt/vol; 5% wt/vol; 10% wt/vol 10% wt/vol; 10% wt/vol; 15% wt/vol 15% wt/vol; 30% ME 500 ME 750 ME 1600 100% As supplied As supplied As supplied As supplied As supplied 15% cap 30% cap 15% cap 30% cap 15% cap 30% cap 15% cap 30% cap 15% cap 30% cap 30% cap Reflux No No No No No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No Yes No No No No No No Fracture Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No Yes No No No No Yes No Appearance Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/tinged Clear/tinged Clear/tinged Clear/tinged Clear/tinged Clear/tinged Clear/yellow Clear/yellow Clear/yellow Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear/colorless Clear Clear/blue Clear/colorless, became hazy, hydrolyzed on exposure to acid

Versagel Mineral oil Silicone viscosity standard 1 Silicone viscosity standard 2 Baby oil gel Ultrasound gel Hair styling gel

tribution and rapid and adequate mixing within the gelatin in the desired manner could not occur. To address fracturing, the concentration of the

gelatin was increased, but this did not improve the results. In addition to fracture, at the higher gelatin concentrations, an additional problem was

identified: the injected reagents refluxed along the needle tract and exited at the surface of the gel, making accurate measurements from adequate

Volume 20

Number 10

Misselt et al

1355

Table 2 Thermal Profile of Neutralization of 1 M Acetic Acid and 1 M NH4OH (1 mL Each) over a Period of 5 Minutes Temperature (C) Time (h:min) 0:00 0:15 0:30 0:45 1:00 1:15 1:30 1:45 2:00 2:15 2:30 2:45 3:00 3:15 3:30 3:45 4:00 4:15 4:30 4:45 5:00 Run 1 23.3 30.2 29.6 29.2 28.9 28.7 28.5 28.3 28.1 27.9 27.7 27.6 27.5 27.3 27.2 27.1 26.9 26.8 26.7 26.6 26.6 Run 2 23.2 29.5 29.3 28.9 28.6 28.3 28.0 27.8 27.5 27.3 27.1 27.0 26.8 26.7 26.5 26.4 26.3 26.2 26.1 26.0 25.9 Run 3 23.0 29.6 29.1 28.7 28.4 28.1 27.9 27.6 27.3 27.2 27.0 26.8 26.6 26.5 26.4 26.3 26.2 26.1 26.0 25.9 25.8 Average Temperature (C) 23.2 29.8 29.3 28.9 28.6 28.4 28.1 27.9 27.6 27.5 27.3 27.1 27.0 26.8 26.7 26.6 26.5 26.4 26.3 26.2 26.1

Table 3 Thermal Profile of Neutralization of 5 M Acetic Acid and 5 M NH4OH (1 mL Each) over a Period of 5 Minutes Temperature (C) Time (h:min) 0:00 0:15 0:30 0:45 1:00 1:15 1:30 1:45 2:00 2:15 2:30 2:45 3:00 3:15 3:30 3:45 4:00 4:15 4:30 4:45 5:00 Run 1 23.2 49.3 48.2 46.5 45.5 44.1 43.2 42.3 41.4 40.6 39.8 39.3 38.6 38.3 37.8 37.4 37.0 36.6 36.1 35.6 35.3 Run 2 22.8 48.9 48.5 46.7 45.2 43.9 42.9 41.9 41.1 40.3 39.6 39.0 38.4 37.9 37.4 36.9 36.5 36.0 35.7 35.3 34.9 Run 3 22.7 36.5 37.1 36.7 35.4 33.9 33.2 32.5 32.0 31.4 31.3 31.2 31.0 30.7 30.7 30.6 30.8 30.4 30.2 30.0 29.9 Average Temperature (C) 22.9 44.9 44.6 43.3 42.0 40.6 39.8 38.9 38.2 37.4 36.9 36.5 36.0 35.6 35.3 35.0 34.8 34.3 34.0 33.6 33.4

curred, separating the layers. As a last effort with gelatin, the concentration of gelatin was decreased to 0.5%, but fracture still resulted instead of dispersion. Several alternate materials were then evaluated, including an ultrasound transmission gel, commercial colorless viscosity standards, mineral oil, hydrocarbon gels used in pharmaceutical and personal care product formulations (ie, Versagel; Penreco), a hair styling gel, and a baby oil gel. Baby oil gel had a tendency to trap air bubbles, which, in several instances, obscured visualization of the reactions. However, of all the materials examined, it proved most satisfactory, and further experimentation was done with this model. Other materials that did not reflux or fracture were deemed unsuitable as they were of inadequate viscosity and were too prone to settling of the reagent solutions to the bottom of the containers. Tables 25 show temperature data over time for each tested concentration of acid and base. These data are summarized and compared in Fig 5, which superimposes the time and temperature data for all four concentrations tested. Temperatures during the first 60 seconds of the neutralization reaction showed an immediate increase that correlated well with concentration.

DISCUSSION
Gelatin initially seemed the obvious first choice rather than alginate, agar, or acrylamide gels simply based on cost, safety, and availability. However, there were difficulties with obtaining satisfactory measurements as noted earlier. This exercise led to separation of questions involving distribution from questions about thermodynamics. The decision was therefore made to focus first solely on the energetics of the reactions. It is worth noting at this point that it was anticipated that the vast majority of the reactions would be carried out in aqueous solution. Therefore, hydrophobic materials were interrogated with the reasoning that, as oil tends to form droplets in water, a similar effect might result in creation of the desired expandable but visible reaction chamber. What was needed, then, was a medium that was thin enough in viscosity to allow an injection to create a focal reaction chamber but not so thin as to allow

mixing all but impossible. Making gels of a lower concentration with a layer of gelatin cast at a much higher concentration as a cap or skin to prevent

reflux along the needle tract was successful to a certain extent, but in general, reflux occurred to the level of the cap layer and then dissection oc-

1356

Thermochemical Ablation in Hydrophobic Gel Phantom: Initial Results

October 2009

JVIR

Table 4 Thermal Profile of Neutralization of 10 M Acetic Acid and 10 M NH4OH (1 mL Each) over 5 Minutes Temperature (C) Time (h:min) 0:00 0:15 0:30 0:45 1:00 1:15 1:30 1:45 2:00 2:15 2:30 2:45 3:00 3:15 3:30 3:45 4:00 4:15 4:30 4:45 5:00 Run 1 23.3 88.6 84.1 78.7 75.0 71.9 68.5 66.6 64.7 63.0 61.4 60.0 58.6 57.4 56.2 55.3 54.3 53.3 52.4 51.5 50.7 Run 2 23.1 79.8 77.7 74.9 72.6 70.1 67.9 65.8 64.0 62.3 60.8 59.5 58.1 56.9 55.8 54.7 53.7 52.8 51.9 51.0 50.2 Run 3 22.8 84.6 79.5 74.2 70.7 68.0 65.4 63.3 62.0 60.4 59.0 57.8 56.6 55.7 54.6 54.3 53.6 52.8 52.0 51.3 50.6 Average Temperature (C) 23.1 84.3 80.4 75.9 72.8 70.0 67.3 65.2 63.6 61.9 60.4 59.1 57.8 56.7 55.5 54.8 53.9 53.0 52.1 51.3 50.5

Table 5 Thermal Profile of Neutralization of 15 M Acetic Acid and 15 M NH4OH (1 mL Each) over a Period of 5 Minutes Temperature (C) Time (h:min) 0:00 0:15 0:30 0:45 1:00 1:15 1:30 1:45 2:00 2:15 2:30 2:45 3:00 3:15 3:30 3:45 4:00 4:15 4:30 4:45 5:00 Run 1 23.8 101.3 91.2 85.6 81.1 75.3 70.7 68.6 66.5 63.2 61.1 60.0 58.2 56.6 55.0 53.7 52.7 51.5 50.4 49.4 48.5 Run 2 23.0 80.6 78.7 77.3 76.8 73.7 71.4 69.3 67.3 65.5 63.7 62.0 60.6 59.2 57.8 56.6 55.5 54.6 53.3 52.3 51.4 Run 1 23.0 105.9 95.0 86.1 79.8 75.8 72.7 69.9 66.8 63.9 63.0 61.4 60.4 59.4 58.2 57.2 55.6 53.8 52.9 52.0 51.2 Average Temperature (C) 23.3 95.9 88.3 83.0 79.2 74.9 71.6 69.3 66.9 64.2 62.6 61.1 59.7 58.4 57.0 55.8 54.6 53.3 52.2 51.2 50.4

any differences in density between the reagent solutions and the phantom to become manifest. This is an issue because hydrocarbons, which typically

have a density lower than water, would not support an aqueous solution. Rather, the aqueous injections would simply flow to the bottom of a

container and away from a thermocouple. This problem would be exaggerated when higher concentrations of reactants were studied because the salts formed would significantly increase solution density while, at the same time, the warmer environment would inherently decrease viscosity of the phantom medium. At the opposite extreme, a very dense hydrocarbon medium would likely cause a reaction mixture to rise to the top of the chamber, resulting in an uncontrolled release of reactants and loss of data. An ideal setup, when identified, could then be used to gather temperature data from exothermic reactions, essentially as a hybrid between an open and bomb-type calorimeter. A summary of the development of a new phantom and the requirements and reasoning are listed in Table 6. An ideal tissue phantom for better distribution studies would embody many of these characteristics with some exceptions. For instance, the hydrophobic nature of the current gel phantom and the viscosity would be actually be antagonistic. An ideal tissue phantom might have a hydraulic conductivity or intrinsic permeability and elasticity closer to the target tissue, such as cirrhotic liver and/or tumor. When a suitable phantom was identified, the data shown were obtained. Measurable temperature changes occurred even at the lowest concentrations, and higher concentrations produced a greater release of heat energy. Indeed, the reaction of 1-mL aliquots of 15 M acetic acid and NH4OH within the gel model proved quite vigorous, demonstrating potential for creating a zone of tissue coagulation based on temperature alone. The peak temperature in two of three trials was more than 100C and the average was 95C. After only a 15-second input of chemical energy, a temperature still in excess of 50C was observed at the end of the 5-minute data collection period. Incomplete mixing would be expected to cause inaccurate data, hence the need for a means of visualization with a pH indicator. Congo red was specifically chosen because of its range of transition (pH 35) and its distinct color change, although other indicators such as litmus could also be used. The congo red pH indicator is blue in solutions with a pH lower than 3 and red in solutions with a pH greater than

Volume 20

Number 10

Misselt et al

1357

AcOH and NH4OH

100 90 80 Temperature (C) 70 60 50 40 30 20 10 Time 0 0:15 0:30 0:45 1:00 1:15 15 1:45 2:00 2:15 2:30 10 2:45 3:00 3:15 3:30 5 3:45 4:00 4:15 4:30 1 4:45 5:00 Molarity

Time

Figure 5. Graphical summation of the thermal neutralization profiles (averages) of all four tested concentrations of the reaction between acetic acid and NH4OH. Tumoricidal temperatures are reached at 10-M (88.6C) and 15-M (101.3C) concentrations. Table 6 Desired Properties and Rationale for Injection Phantom Characteristic Clarity Color Viscosity Density Toxicity Cost Miscibility Chemical stability Volatility Selection Transparent Clear Medium Matched or lower than reaction solutions None Low Hydrophobic Nonreactive Very low volatility/ high flashpoint Rationale Visibility in container Allows use of pH indicators Low- or high-viscosity fluids would not form a satisfactory reaction chamber Ensures that reaction solutions remain in phantom Exposure to personnel, disposal issues after assays High throughput likely Reaction solutions aqueous and could mix with hydrophilic media Many materials hydrolyze or become opaque; potential for reprocessing and reuse Anticipated temperatures could potentially cause evaporation or even ignition of volatile materials

5, making it straightforward to determine when all the acid had reacted in the solution. In general, with acetic acid and NH4OH, mixing occurred readily and the indicator was easily seen. The only issue identified was at the higher temperatures, at which vigorous reaction may have had an adverse effect as a result of disruption of the in situ reaction chamber.

Several other observations can be made. Hyperthermia and cryoablation could be viewed for the purpose of this discussion as therapies with favorable therapeutic indices based on high local toxicity and low systemic toxicity. This is a result of the contained distribution of the input variable or treatment. Survival of the whole organism is indeed in jeopardy

if, in the case of hyperthermia, the systemic thermal dose is sufficient to increase the body temperature greater than a relatively low threshold. Fortunately, this does not occur in clinical applications. Temperatures for 10 M and 15 M concentrations were greater than the tumoricidal range, even if only hyperthermic effects are invoked. With chemical ablation, the nature of the injected material and the potential for systemic exposure from intravascular access is inherently greater and therefore must be taken into account. The salt product in the present case is NH4Cl, which is a relatively benign product if this type of reaction were to see eventual use in mammals and some portion were to result in systemic exposure. In fact, NH4Cl is used clinically as an acidifying agent in some conditions and is metabolized to urea for eventual excretion by the kidneys. In addition, the byproducts of many reactions would ensure that the local environment would be orders of magnitude greater in osmolarity than the physiologic range of 270 300 mOsm for an indeterminate length of time. The local hyperosmolarity makes this strategy even more appealing from a local cytotoxicity standpoint. This residual toxicity might be of benefit in decreasing local recurrence from dormant or hibernating tumor cells. These initial experiments were performed with use of a weak acid and a weak base. It is known that, for any given concentration of acid and base, the actual amount of energy released during neutralization depends on the strength of each of the reactants. In the case of a weak acid and weak base pairing, the expected amount of energy is approximately 30 kJ/mol. More heat should be released with the use of a strong acid and base (approximately 55 kJ/mol) as such reagents are already fully ionized and therefore no energy is lost for ionization before reaction. Further investigation with the gel model will involve reactions with strong acids, strong bases, various combinations, and other exothermic reactions beyond neutralization. Other avenues of investigation might involve variations in the injection technique such as concomitant rather than sequential injections or the use of alternating fractionated boluses of each reagent. Such techniques may offer better reaction control with an even greater exotherm

1:30

1358

Thermochemical Ablation in Hydrophobic Gel Phantom: Initial Results

October 2009

JVIR

and potential for tissue ablation. In summary, the oil gel phantom is a useful, readily available, inexpensive model to study mixing behaviors and calorimetry (ie, maximum heating potentials) for exothermic reactions. Further study with the gel model and the exothermic potential of neutralization reactions will be reported in due course.
References 1. Emerich DF, Snodgrass P, Lafreniere D, et al. Sustained release chemotherapeutic microspheres provide superior efficacy over systemic therapy and local bolus infusions. Pharm Res 2002; 19:10521060. 2. Caton T, Thibeault SL, Klemuk S, et al. Viscoelasticity of hyaluronan and nonhyaluronan based vocal fold injectables: implications for mucosal versus muscle use. Laryngoscope 2007; 117:516 521. 3. Clark TWI. Chemical ablation of liver cancer. Tech Vasc Interv Radiol 2007; 10: 58 63. 4. Vogl TJ, Engelmann K, Mack MG, et al. CT-guided intratumoural administration of cisplatin/epinephrine gel for treatment of malignant liver tumours. Br J Cancer 2002; 86:524529. 5. Celikoglu F, Celikoglu SI, Goldberg EP. Bronchoscopic intratumoral chemotherapy of lung cancer. Lung Cancer 2008; 61:112. 6. Plante MK, Gross AL, Kliment J, Kida M, Zvara P. Intraprostatic ethanol chemoa-

7.

8.

9.

10.

11.

12.

13.

14.

blation via transurethral and transperineal injection. BJU Int 2003; 91:9498. Chowning SL, Susil RC, Krieger A, Fichtinger G, Whitcomb LL, Atalar E. A preliminary analysis and model of prostate injection distributions. Prostate 2006; 66:344 357. Lencioni R, Crocetti L. Image-guided thermal ablation of hepatocellular carcinoma. Crit Rev Oncol Hematol 2008; 66: 200 207. Cressman ENK, Misselt AJ, Brix TL, Choi JH, Bischof JC. A new hydrophobic gel phantom for study of thermochemical ablation: initial results using a weak acid and weak base (abstr.). J Vasc Interv Radiol 2007; 18(suppl 1):S154. Deng Z, Liu J. Minimally invasive thermotherapy method for tumor treatment based on an exothermic chemical reaction. Minim Invas Ther Allied Technol 2007; 16:341346. Rao W, Liu J. Tumor thermal ablation therapy using alkali metals as powerful self-heating seeds. Minim Invas Ther 2008; 17:43 49. Zhang XY, Luck J, Dewhirst MW, Yuan F. Interstitial hydraulic conductivity in a fibrosarcoma. Am J Physiol Heart Circ Physiol 2000; 279:H2726 H2734. Huang GT, Lee PH, Tsang YM, et al. Percutaneous ethanol injection versus surgical resection for the treatment of small hepatocellular carcinoma a prospective study. Ann Surg 2005; 242: 36 42. Honda N, Guo Q, Uchida H, Ohishi H, Hiasa Y. Percutaneous hot saline injec-

15.

16.

17.

18.

19.

20.

21.

tion therapy for hepatic tumors: an alternative to percutaneous ethanol injection therapy. Radiology 1994; 190:5357. Kuroda M, Kato H, Hanamoto K, et al. Development of a new hybrid gel phantom using carrageenan and Gellan gum for visualizing three-dimensional temperature distribution during hyperthermia and radiofrequency ablation. Int J Oncol 2005; 27:175184. Lafon C, Zderic V, Noble M, et al. Gel phantom for use in high-intensity focused ultrasound dosimetry. Ultrasound Med Biol 2005; 31:13831389. Homolka P, Gahleitner A, Nowotny R. Temperature dependence of HU values for various water equivalent phantom materials. Phys Med Biol 2002; 47:2917 2923. Chen Z, Gillies G, Broaddus W, et al. A realistic brain tissue phantom for intraparenchymal infusion studies. J Neurosurg 2004; 101:314 322. Wientjes MG, Zheng JH, Hu L, Gan Y, Au JL. Intraprostatic chemotherapy: distribution and transport mechanisms. Clin Cancer Res 2005; 11:4204 4211. Taylor GD, Johnson DB, Hogg DC, Cadeddu JA. Development of a renal tumor mimic model for learning minimally invasive nephron sparing surgical techniques. J Urol 2004; 172:382385. Scott DJ, Young WN, Watumull LM, et al. Development of an in vivo tumor-mimic model for learning radiofrequency ablation. J Gastrointest Surg 2000; 4:620625.