| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of Wisconsin, 600 Highland Ave., Madison,
WI 53792-3252.
2 Department of Biostatistics and Medical Informatics, University of Wisconsin,
Madison, WI 53792.
Received August 5, 2004;
accepted after revision January 24, 2005.
Address correspondence to F. T. Lee, Jr.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Ten female domestic swine (mean weight, 45 kg) underwent radiofrequency ablation at open surgery. Group 1 (n = 12 lesions) was pretreated with peritoneal D5 before radiofrequency ablation. Group 2 (n = 11 lesions) was pretreated with peritoneal 0.9% saline. A 2.7-mm spacer was placed between the liver surface and diaphragm in groups 1 and 2. Group 3 (n = seven lesions) served as a control group with no pretreatment regimen. Group 4, an additional control group (n = eight lesions), consisted of animals pretreated with D5 in which a larger spacer was used. After radiofrequency ablation, the animals were sacrificed and the liver, diaphragm, and lung were removed. The extent of thermal injury to the surface of each organ was recorded.
RESULTS. The animals in the D5 and saline pretreatment groups experienced fewer diaphragm injuries than the control animals (D5, p = 0.02). The smallest lesions in the lung and diaphragm were in the D5 group, followed by the saline and control groups (diaphragm, p = 0.0001; lung, p = 0.13). Diaphragm lesions were significantly smaller in the D5 and saline groups than in the control group (p = 0.0001 and 0.01, respectively).
CONCLUSION. Instillation of D5 into the peritoneal cavity before hepatic radiofrequency ablation decreases the risk and severity of diaphragm and lung injuries compared with no pretreatment or pretreatment with 0.9% saline in this animal model. Pretreatment with D5 may increase both the safety of and the number of patients eligible for treatment with thermal therapies.
Keywords: ablation • animal studies • radiofrequency • thermal injury
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although radiofrequency ablation is considered safe, several complications are associated with all thermal therapies [3-7]. In particular, bowel injuries caused by radiofrequency ablation are one of the most feared complications and have been associated with mortality due to sepsis and abscess formation [3]. The risk of colon or small-intestine injury is increased when the bowel is near the targeted tumor [8]. Therefore, most physicians will defer radiofrequency ablation in tumors adjacent to bowel. Body wall injuries can cause severe pain, and diaphragm injury manifests as referred right shoulder pain that can last up to 2 weeks after ablation [3]. Although these injuries are rarely life threatening, they can be quite distressing to both patients and referring physicians.
The ability to protect perihepatic structures during radiofrequency ablation would increase both the safety and the number of patients who undergo thermal ablation. Ideally, agents and techniques to provide thermal protection would be simple to use, safe, and inexpensive and would provide both thermal and electric insulation. Instillation of fluids into spaces around target organs is one method that has been successfully used in various anatomic areas. Injection of 0.9% saline into Denonvilliers' fascia to increase the space between the prostate and rectum has been shown to lessen the risk of rectal freezing during prostatic cryoablation [9]. Likewise, sterile water instilled into the pararenal space has helped prevent bowel injury during radiofrequency ablation of the kidney [10]. Intraperitoneal injection of saline before radiofrequency ablation has been shown to decrease the frequency and severity of associated diaphragm injury in an animal model [11]. However, the effectiveness of this approach may be limited by the inherent ability of saline to conduct electricity.
Recently, Gillams and Lees [12] introduced the use of 5% dextrose in water (D5) injected into the peritoneal space as a method to protect the body wall and bowel during hepatic radiofrequency ablation. D5 is an isotonic fluid that does not conduct electricity and potentially provides a thermal barrier when surrounding the target organ. It is well tolerated when injected IV or into most body cavities [13]. Preliminary human data showed decreased postprocedural pain and the ability to displace bowel away from the liver before radiofrequency ablation (Chen EA, et al. unpublished data). The purpose of this study was to compare the protective effects of D5 and normal saline for the prevention of diaphragm and lung injury during radiofrequency ablation at the hepatic dome.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals, Anesthesia, and Procedures
Preapproval was obtained from the research animal care and use committee of
our institution, and all husbandry and experimental studies were compliant
with the National Institutes of Health Guide for the Care and Use of
Laboratory Animals [14].
Anesthesia was induced with intramuscular tiletamine hydrochloride, zolazepam
hydrochloride (Telazol, Fort Dodge Laboratories), atropine (Phoenix
Pharmaceutical), and xylazine hydrochloride (Xyla-Ject, Phoenix
Pharmaceutical). Animals were intubated, and anesthesia was maintained with
inhaled isoflurane (Halocarbon Laboratories). Each pig received lactated
Ringer's solution IV at a rate of 20 mL/kg per hour during anesthesia. The
abdomen was shaved, scrubbed with 2% chloroxylenol (Vet Solutions), and
prepared with a 10% povidone-iodine solution (Purdue Frederick). The animal
was draped in the standard surgical fashion, and a midline incision was made
to expose the liver.
Once the liver had been exposed, target areas were selected in the hepatic dome adjacent to the diaphragm. A cluster radiofrequency electrode (Cool-tip Cluster, Valleylab) was placed into either the right lateral, right medial, left lateral, or left medial lobe of the liver (multiple lesions per pig). The radiofrequency probes were placed just below the liver capsule (approximately 1.0 cm below the liver surface) with the intent of accentuating the degree of thermal injury to the liver surface, diaphragm, and underlying lung. In D5 and saline animals, a linoleum spacer was used to standardize the distance between liver and diaphragm. The spacer used for groups 1 and 2 was a ring approximately 6.5 cm in outer diameter, 2.7 cm in inner diameter, and 2.7 mm thick (Fig. 1). It was placed so that the inner area was directly over the active tips of the radiofrequency electrode. To prevent liver tissue from pushing through the spacer and contacting the diaphragm, 3-0 coated Vicryl sutures (Ethicon) were placed in approximately a 1-cm grid pattern over the internal exposed region. A spacer similar to that used for groups 1 and 2 was used for group 4. However, the spacer used in group 4 had an inner diameter of 4.5 cm and an outer diameter of 8.5 cm. A 1-L solution of either 0.9% sodium chloride or D5 was then added to the peritoneal cavity surrounding the liver in the experimental groups. For control animals, the spacer and protective solutions were not used.
|
). The generator uses a feedback algorithm that
monitors tissue impedance and automatically adjusts output to maximize energy
delivery. The radiofrequency generator cuts back the output to 0.10 A for 15
sec whenever the currently measured impedance increases above the baseline
impedance (defined as the minimum impedance recorded during the radiofrequency
power delivery) by 25 .. Radiofrequency energy was applied for a total
of 12 min per lesion according to manufacturer recommendations. An infusion
pump (Cool-tip, model PE-PM, Valleylab) was used to infuse cold sterile water
through the probes during the ablation. Typical temperatures for the
pretreatment radiofrequency probe tip ranged from approximately 18°C to
20°C.
Animal Recovery
The midline abdominal incision was closed in a three-layer technique using
number 1 Dexon II (Tyco Healthcare Group) and 3-0 coated Vicryl sutures. For
added protection, the skin layer was sealed with VetBond (3M) tissue adhesive
and covered with Tegaderm (3M Health Care). Postprocedural pain was controlled
with intramuscular buprenorphine (Reckitt Benckiser Pharmaceutical).
Animal Sacrifice and Lesion Analysis
The day after radiofrequency ablation, the pigs were reanesthetized and
then sacrificed with IV pentobarbital sodium and phenytoin sodium
(Beuthanasia-D, Schering-Plough). The liver, diaphragm, and affected lung were
removed en bloc. Each lesion was digitally imaged (Coolpix 995, Nikon) and
analyzed using ImageJ freeware
(rsb.info.nih.gov/ij).
Lesions were preserved in 10% buffered formalin.
Statistical Analysis
The true proportion and 95% confidence intervals of radiofrequency
ablations resulting in injury to the lung or diaphragm were determined using
the Clopper-Pearson method. The Fisher exact test was used to check for
overall differences between groups and for pairwise comparison of groups.
Analysis of variance and pairwise t tests (assuming unequal variances
in the two groups) were used to check for differences in lesion surface area
between the saline, D5, and control groups. Analysis of variance was also used
to test for differences in the depth of lung injury between groups.
Statistical significance was defined as p < 0.05, and p
values between 0.05 and 0.10 were considered suggestive of significance.
Unless otherwise stated, results given for D5 correspond to group 1. Values
for group 4 are listed in the tables. However, further analysis of group 4 was
limited to a comparison with group 1 because the purpose of this group was to
ensure that the spacer was not a limiting factor.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Lesion Size
The size of thermal injuries of the liver, diaphragm, and lung are
summarized in Table 2 and
depicted in Figure 2. Thermal
injury of the liver surface was largest for the control group, followed by the
D5 and saline groups (saline vs control and D5 vs control, p = 0.07;
saline vs D5, p > 0.10). Animals pretreated with D5 had smaller
thermal injuries to both the diaphragm and the lung than did animals
pretreated with saline or the control animals (overall test, D5 vs control,
and saline vs control for the diaphragm, p < 0.05; D5 vs saline
for the diaphragm, p = 0.07; D5 vs saline for the lung, p =
0.10; all other comparisons, p > 0.10). The mean depth of lung
injury is also given in Table
2. No significant differences were seen between groups. Finally,
no statistically significant differences in diaphragm or lung injury were seen
between the two D5 groups (diaphragm, p = 0.63; lung, p =
0.89).
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Infusion of D5 into the abdomen has several potential advantages over two other materials that have been used for the same purpose: normal saline and sterile water. Normal saline (0.9%) is an isoosmolar solution (308 mosmol/L) that is well tolerated in virtually every body cavity. A saline buffer surrounding an organ provides some degree of thermal protection because of the ability to conduct heat away from the ablation zone. However, saline is an ionic fluid and is able to conduct electricity. Therefore, current applied for the ablation procedure can be carried through the saline into an adjacent structure, resulting in frictional heating (Figs. 3A, 3B, and 3C). In this study, saline appeared to have a protective effect, likely because of thermal buffering, but the overall protective effect was not as great as that found with D5. Sterile water has the advantage of being nonionic and therefore does not conduct electricity. However, the hypoosmolar nature of water can also lead to large shifts in systemic fluid, particularly when the water is infused into the peritoneal space. Peritoneal dialysis is based on the ability of the peritoneum to exchange ions of different concentrations across the peritoneal membrane [15]. Ultimately, this exchange will severely limit the volume of sterile water that can be infused safely into the abdomen. The use of sterile water may be safer in the retroperitoneum than in the peritoneum because the former lacks a visceral peritoneal lining [10].
Because of its isoosmolarity (252 mosmol/L) and nonionic composition, D5 is a nearly ideal protective buffer to infuse into the abdomen before thermal ablation. The ability to infuse large amounts of fluid safely into the peritoneum increases the protective potential of D5, because large volumes are more likely to surround and isolate peritoneal organs such as the liver. Even small amounts of D5 between the liver and diaphragm (in this study, 2.7 mm) clearly had a protective effect. Without the D5 infusion, radiofrequency burns invariably extended unimpeded into the diaphragm and often into the underlying lung (Figs. 3A, 3B, 3C, and 4). To our knowledge, a systematic comparison of D5 and saline in humans has not been attempted. On the basis of the results of this study, such a comparison is not warranted, and any other fluids used for protection should be compared with D5, not saline.
Potential disadvantages of the use of D5 for hepatic radiofrequency ablation include the added time necessary to infuse the material into the abdomen, the potential for fluid overload if the hydration status of the patient is not carefully considered, and the necessity for caution in diabetic patients. The potential that D5 will cause bacterial peritonitis is unknown but is considered slight because of the rapid reabsorption of D5 across the peritoneal membrane. Introduction of dextrose-containing fluids into the peritoneal cavity is common. For example, peritoneal dialysate contains up to 4.25% dextrose and is well tolerated [16]. Another consideration in the use of peritoneal D5 is the potential to isolate, electrically, the liver from the body wall, reducing the surface area for the return current. This complication could potentially lead to burns of the ligamentous attachments of the liver. No such burns were seen during this exploration, and we expect that large amounts of D5 would be necessary for this complication to be possible.
Another material that has been used successfully for protection against thermal injury is intraperitoneal CO2. This has the advantage of being widely available, cheap, and easy to infuse. In addition, the increased intraabdominal pressure associated with CO2 introduction may help decrease bleeding complications from hepatic puncture by decreasing portal blood flow [17]. Infusion of CO2 for laparoscopy has been associated with substantial postprocedural cramping, abdominal pain, and bloating if the gas cannot be removed completely [18]. However, unlike an infusion of liquids into the peritoneal space, the use of CO2 virtually excludes application of sonography for guidance and monitoring. Because sonography is the predominant method used to guide and monitor radiofrequency ablation worldwide, we elected not to include CO2 in this study.
The concept of D5 infusion into the abdomen originated from Gillams and Lees [12]. One important fact raised by those authors was that a high proportion of liver tumors reside on the surface because the centrifugal portal blood supply causes metastases to lodge near the surface and because a disproportionate amount of hepatic parenchyma is in the outer half of the organ [19]. Therefore, most colonic metastases are found in peripheral liver, and treatment of these tumors is associated with diaphragm, body wall, and bowel injuries. The preliminary clinical results of those authors showed a low incidence of postprocedural pain and complications. An additional benefit noted by Gillams and Lees was the excellent sonographic window provided by the D5 solution, similar to that of ascitic fluid. The fluid over the diaphragm caudally displaces the liver, increasing the amount of hepatic parenchyma that can be visualized and accessed by radiofrequency probes from a subcostal approach [20].
The use of peritoneal D5 may increase the acceptance of radiofrequency ablation by patients and referring physicians by decreasing both postprocedural pain and complications. In addition, this technique may increase the number of patients who are eligible for radiofrequency ablation. Currently, most centers are not performing radiofrequency ablation on patients with tumors adjacent to bowel because of the risk of thermal bowel injury. Pretreatment with peritoneal D5 will likely decrease the risk of bowel injury, thus increasing the number of patients eligible for radiofrequency ablation (Chen EA, et al., unpublished data).
This study had several limitations. The first was that it was performed on an open-surgery animal model rather than percutaneously on humans. This choice was a necessary compromise because of our desire to keep the distance between the liver and diaphragm constant and to place the radiofrequency probes in a consistent and superficial location beneath the hepatic capsule. In the D5 and saline groups, the spacer was 2.7 cm in diameter. This size could theoretically limit the size of the diaphragm injury to 5.73 cm2 in these groups. Group 4 consisted of two animals that were pretreated with D5 and in which a larger spacer was used to verify that the spacer was not limiting the transmission of energy. The spacer used in group 4 had an inner area (15.9 cm2) greater than the mean hepatic surface area of that group. The energy transmission should have increased—that is, the protective effect should have decreased—in this group if the spacer was in fact contributing to the protective effect in the other D5 group. The lack of any significant difference in diaphragm or lung injury, or protective effect, in either indicates that the spacer was not a contributing factor. The animal model limited our ability to obtain subjective patient data such as pain scores, which are obtained best in a human clinical study. We also did not attempt to determine the ability of D5 or saline to protect from bowel injuries after radiofrequency ablation. We speculate that the protective effect shown for the diaphragm and lung in this study would also extend to bowel. One note of caution: Adhesions from prior surgery or other causes could limit the ability to displace the diaphragm and bowel away from the liver, potentially resulting in inadvertent thermal injury across the tethered organs if physicians assume that D5 has produced a safe degree of displacement. An additional limitation of the animal model was that this study was performed on healthy liver tissue. The effect of altered liver composition such as seen with cirrhosis or hepatic steatosis may alter the electric and thermal transmission of radiofrequency energy. In addition, different organs and tumors have unique electric and thermal conductivities, likely affecting the size of ablations in these structures.
Intraperitoneal D5 substantially protected the diaphragm during application of radiofrequency ablation near the hepatic dome. Less protection was afforded by peritoneal normal saline. Consideration should be given to the use of peritoneal D5 in patients who have tumors near the diaphragm, body wall, or bowel.
Acknowledgments
We gratefully acknowledge Carrie E. Poole for providing general assistance
and Debra L. Chicks and William D. Lewis for helping to develop this technique
in our clinical population.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  I. H. Lee, Y. C. Yoon, E. Y. Cho, J. W. Kwon, and S. T. Kwon Perineural Air Injection as a Means of Prevention of Thermal Injury of the Sciatic Nerve During Radio Frequency Ablation: A Preliminary Experimental Study in Rabbits J. Ultrasound Med., August 1, 2008; 27(8): 1221 - 1227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Lee, H. Rhim, H. K. Lim, D. Choi, W. J. Lee, and K. S. Min Effect of Artificial Ascites on Thermal Injury to the Diaphragm and Stomach in Radiofrequency Ablation of the Liver: Experimental Study with a Porcine Model Am. J. Roentgenol., June 1, 2008; 190(6): 1659 - 1664. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Rhim, H. K. Lim, Y.-s. Kim, and D. Choi Percutaneous Radiofrequency Ablation with Artificial Ascites for Hepatocellular Carcinoma in the Hepatic Dome: Initial Experience Am. J. Roentgenol., January 1, 2008; 190(1): 91 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. Laeseke, T. M. Frey, C. L. Brace, L. A. Sampson, T. C. Winter III, J. R. Ketzler, and F. T. Lee Jr. Multiple-Electrode Radiofrequency Ablation of Hepatic Malignancies: Initial Clinical Experience Am. J. Roentgenol., June 1, 2007; 188(6): 1485 - 1494. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Lucey Radiofrequency ablation: the future is now. Am. J. Roentgenol., May 1, 2006; 186(5 Suppl): S237 - S240. [Full Text] [PDF]
|
|
|
|
|
|
J. L. Hinshaw, P. F. Laeseke, T. C. Winter III, M. A. Kliewer, J. P. Fine, and F. T. Lee Jr. Radiofrequency ablation of peripheral liver tumors: intraperitoneal 5% dextrose in water decreases postprocedural pain. Am. J. Roentgenol., May 1, 2006; 186(5 Suppl): S306 - S310. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
