| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Radiology, University of Wisconsin, E3/311 CSC, 600 Highland Ave., Madison, WI 53792.
Received January 27, 2005;
accepted after revision May 19, 2005.
Address correspondence to F. T. Lee, Jr.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. This case-control study compared the hospital course of the last 20 patients to undergo radiofrequency ablation of peripheral liver tumors at our institution. Ten consecutive subjects in the D5W group (seven women and three men; age range, 37-78 years; mean, 58 years) were pretreated with intraperitoneal D5W (average, 963 ± 436 [SD] mL) before the ablation. This group was compared with a control group of 10 patients (five men and five women; age range, 36-73 years; mean, 54 years) who did not receive intraperitoneal D5W. The size of the ablation zone, degree of liver capsule involvement, number of burns, patient-reported pain, patient-controlled analgesia use, and length of hospital stay were recorded and compared between the groups.
RESULTS. The patients in the D5W group reported significantly less pain in the first 24 hr after the procedure than did the controls (2.3 vs 6.3, p = 0.003) despite a significant decrease in patient-controlled analgesia use (17.3 vs 125.1 mg of morphine, p = 0.003). The differences in pain and patient-controlled analgesia use were greatest in the first 16 hr and decreased over time in both groups. The total volume of ablation (61.9 vs 45.2 cm3, p = 0.44), liver surface involvement (6.2 vs 6.0 cm, p = 0.73), and number of burns (1.8 vs 1.1, p = 0.14) were slightly higher in the D5W group than in the control group, but this difference was not statistically significant.
CONCLUSION. Pretreatment with intraperitoneal D5W before radiofrequency ablation of peripheral liver tumors decreased pain, narcotic use, and length of hospital stay. The effect of pretreatment with D5W is greatest in the first 16 hr after the procedure.
Keywords: interventional radiology • liver • percutaneous ablation • radiofrequency ablation
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although percutaneous radiofrequency ablation has an excellent overall safety profile, inadvertent thermal injury to adjacent structures such as bowel, body wall, and diaphragm can be associated with significant morbidity and postprocedural pain [6-8]. The risk of thermal injury is directly related to the proximity of perihepatic organs to the ablation zone. As a result, peripheral liver tumors close to the bowel generally are not treated with percutaneous ablation, and peripheral liver tumors adjacent to the body wall or diaphragm often are associated with severe postprocedural pain that can last several weeks [6]. Various spacer materials and techniques have been used experimentally and clinically to displace and insulate at-risk tissue before ablation. These include injection of fluids and gases such as 5% dextrose in water (D5W), normal saline, sterile water, and carbon dioxide [9-11]. Esophageal dilator balloons also have been used to displace tissue mechanically [12]. The use of intraperitoneal D5W to protect the body wall during hepatic radiofrequency ablation was recently described by Gillams and Lees [13]. On the basis of that work, we began investigating the efficiency of various fluids as insulators during radiofrequency ablation. Experimental work by Laeseke et al. [14] showed the potential advantages of D5W over saline for protecting the diaphragm during radiofrequency ablation at the hepatic dome.
In our initial clinical experience with the intraperitoneal instillation of D5W, we noted a dramatic decrease in the severity of periprocedural pain reported by the patients. On the strength of these observations, we began using the technique for all patients undergoing radiofrequency ablation of peripheral liver tumors. The purpose of this study was to determine the potential clinical efficacy of intraperitoneal D5W in insulating the body wall and diaphragm, thus decreasing both postprocedural pain and postprocedural narcotic use.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Twenty consecutive patients (12 women and eight men) who underwent percutaneous radiofrequency ablation of a peripheral liver tumor at our institution were included in the study. For the purposes of this study, a peripheral liver tumor was defined as a tumor that was adjacent to the surface of the liver such that the ablation zone itself would be expected to involve the capsule. The procedures took place over a 28-month period. One patient who had a significant complication related to the ablation (bleeding requiring an emergent laparotomy) was not included in the study because the patient underwent surgery approximately 6 hr after the procedure, precluding 24-hr follow-up. On the basis of the promising results of earlier work done at our institution and others, all ablations of peripheral liver tumors after December 2003 were performed with the instillation of intraperitoneal D5W. Thus, the D5W group comprised the 10 most recent patients, and the control group comprised the 10 consecutive patients who underwent percutaneous radiofrequency ablation of a peripheral liver tumor immediately before the introduction of intraperitoneal D5W. This control group consisted of five women and five men between the ages of 36 and 73 years (mean, 54 years), whereas the D5W group consisted of seven women and three men between the ages of 37 and 78 years (mean, 58 years). The most common tumor type for the 20 patients—hepatocellular carcinoma—was present in five patients in the control group and three in the D5W group. Other tumor types included breast cancer (one control patient), colon cancer (one control patient and two D5W patients), cholangiocarcinoma (one control patient and one D5W patient), leiomyosarcoma (one control patient and one D5W patient), focal nodular hyperplasia (one control patient), carcinoid (one D5W patient), neurofibrosarcoma (one D5W patient), and squamous cell carcinoma of the rectum (one control patient).
All ablations were performed with real-time sonographic guidance, with
confirmation of electrode position by CT fluoroscopy as necessary.
Contrast-enhanced CT was performed immediately after treatment with
radiofrequency to confirm the size and location of the ablation zone and
evaluate for immediate complications. These immediate-postablation
contrast-enhanced CT scans were the basis for our assessment of the procedural
results. Ablation zone size (length, width, and height), liver capsular
involvement (length of ablation zone involving the liver capsule), number of
ablations performed, tumor type, and type of ablation electrode were recorded.
The axial CT image showing the greatest linear involvement of the liver
surface was chosen for measuring liver surface involvement, and the axial CT
image showing the greatest ablation zone diameter was chosen for measuring the
ablation zone size (Fig. 1).
The volume of the ablation was calculated using the standard formula for a
prolate ellipse: volume = (length x width x height) x 
/
6. If more than one ablation was performed, the total linear liver surface
involvement of all the ablation zones was used. These data were evaluated
primarily to establish any intergroup differences that might affect
postprocedural pain and narcotic use.
|
D5W was instilled under real-time sonographic guidance using the setup shown in Figure 2. An 18- or 20-gauge needle was advanced into the left lobe of the liver in the region of the epigastrum. Although 18-gauge needles were used initially, 20-gauge needles proved adequate and now are our standard. After the position of the needle in the hepatic parenchyma had been confirmed, continuous and gentle forward pressure was applied to the syringe plunger and the needle was slowly retracted. A release of resistance could be detected when the needle tip entered the peritoneal space. The needle position and location of the expanding fluid collection were confirmed by sonography. D5W then was actively injected through the needle using a three-way stopcock until at least a 2- to 3-mm layer of fluid was identified between the target tumor and the adjacent hemidiaphragm or body wall (Fig. 3). On the basis of the physician's preference, the ablation electrode was placed either before or after the D5W infusion. The needle was then left in place during the ablation to allow intermittent infusions of additional D5W as necessary to maintain an adequate layer of fluid. Re-infusions were not necessary in most patients. The volume of D5W infused and the maximal thickness of the fluid layer on sonography or CT (depending on the primary image guidance method) were recorded. Once an adequate layer of D5W was in place, the ablation was performed with either a Cool-tip single electrode or a Cool-tip cluster electrode (Valleylab) with standard 12-min ablation cycles.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
At all time points, the self-reported pain level was higher for patients not receiving D5W, with absolute pain decreasing over time in both groups (Fig. 4A) from 6.3 to 1.7 and from 2.3 to 0.6 over the first 24 hr without and with D5W, respectively. The mean difference in pain levels between the groups decreased over time from 3.8 immediately after the ablation to 1.1 at 24 hr. The difference in pain between the D5W and control groups was statistically significant immediately after the procedure and at 8 hr after the procedure (p < 0.05). The lack of statistical significance at the later time points may have been due to the small sample sizes (10 per group) and decrease in overall magnitude of the pain in both groups.
|
|
The difference in length of hospital stay was not statistically significant (p = 0.58, Fisher's exact test), with three patients remaining in the hospital for more than 24 hr in the control group versus one patient in the D5W group. However, two of the three patients in the control group who were kept in the hospital for more than 24 hr required prolongation of the hospitalization because of significant residual pain at 24 hr. The one patient in the D5W group who remained in the hospital longer than 24 hr had urinary retention. No patients in the D5W group had clinically significant pain at 24 hr.
The volume of infused D5W varied between patients (range, 350-1,500 mL; average, 963 mL). The average maximal thickness of the fluid layer varied as well (range, 3-17 mm; average, 8.5 mm).
Two patients in the control group and one patient in the D5W group had an epidural catheter in place for pain control after the procedure. Epidural catheters were used to provide additional pain control during the procedure for patients in whom general anesthesia was contraindicated. This difference in treatment was not statistically significant (p = 0.99, Fisher's exact test).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The length of hospital stay was not significantly shorter for the D5W group. However, the reason may well be related to the relatively small sample size and the overall decrease in pain by 24 hr in both groups. Most patients in both groups were discharged at 24 hr, though there were two patients in the control group who were kept in the hospital more than 24 hr, primarily for pain control, and none in the D5W group. Given the minimal pain experienced by most patients in the D5W group, it may become possible to discharge some of these patients even more rapidly.
The success of this technique may derive from the constitution of the burn site. Radiofrequency ablation uses alternating electrical current to create ionic agitation in the tissues adjacent to the applicator, thus producing molecular friction and heat [16]. When tissue temperature exceeds 50°C, tissue coagulation and necrosis result. The area that undergoes tissue coagulation varies, depending on the local environment of the tissue being ablated and the tissue type. Histologic studies have identified two distinct zones of tissue damage that occur during radiofrequency ablation [17]. A white zone of complete tissue necrosis occurs centrally around the electrode. Surrounding this zone is a zone of partial ablation, referred to as the red zone, which contains areas of partial necrosis and inflammation. This zone likely is responsible for much of the pain and inflammatory reaction that occurs after thermal ablation. Because the red zone is the most peripheral area of the thermal lesion, the temperatures are relatively low in this region. Therefore, we postulate that a relatively thin thermal and electrical insulator may be sufficient to provide the necessary protection. This hypothesis has been confirmed in prior animal studies in which only 2.7 mm of D5W between the liver and diaphragm substantially mitigated diaphragmatic damage during hepatic radiofrequency ablation at the liver dome [14].
Apart from pain management, intraperitoneal D5W also might serve to reduce the complication rate of radiofrequency ablation. One of the most serious complications of radiofrequency ablation is bowel perforation because of thermal injury [6, 7]. We suspect that intraperitoneal D5W can decrease the risk of bowel injuries by the same mechanisms as those for the decreased pain seen in this study. Sterile water has been used clinically in the paranephric space during percutaneous radiofrequency ablation of a renal tumor to prevent bowel injury [18], and intraperitoneal D5W might have a similar effect. This possibility certainly warrants further investigation. However, a protective effect of D5W would be a difficult hypothesis to prove in a clinical trial because of the low incidence of bowel injuries. We have been fortunate in avoiding bowel perforation at our institution, primarily because our selection criteria exclude patients with tumors adjacent to bowel and these patients usually undergo either surgical resection or ablation at open surgery or laparoscopy. This policy, of course, commits the patient to surgery or intraoperative ablation, which results in additional morbidity and recovery time.
The strategy behind D5W use has a precedent in the ablation literature. Early in vivo animal and clinical studies have described using various materials and devices to act as insulators during radiofrequency ablation to protect tissues adjacent to the ablation zone. Ideally, these techniques should serve two separate purposes: displacing bowel or other critical tissues from the target organ to create a safety margin, and providing an electrical and thermal barrier. Saline, D5W, carbon dioxide gas, sterile water, and esophageal dilatation balloons have been used for this purpose [9-11]. We believe that D5W is particularly well suited for several reasons to serve as a protective material. Unlike saline, D5W is nonionic and thus does not conduct electricity or produce heating because of ionic agitation. In fact, the ionic properties of saline are exploited in perfused electrodes to enlarge the zone of thermal ablation [19]. D5W is also isoosmolar and thus is well tolerated in virtually every body cavity [20]. Because sterile water is nonosmolar, tissue damage from a severe osmotic gradient resulting in fluid shifts and cellular dehydration is possible when water is in direct contact with tissue. D5W is easy to instill and distribute, unlike esophageal dilatation balloons or other mechanical devices, which can be technically difficult and time-consuming to position. The distribution of fluid in the abdomen creates an anechoic sonic window that enhances sonographic imaging during radiofrequency. Lastly, D5W is inexpensive and widely available.
Although D5W infusion into the peritoneal cavity is technically simple, we initially were concerned that the fluid would disperse throughout the abdomen and pelvis, preventing sufficient accumulation around the liver and limiting thermal and electrical insulation. In the 10 patients we studied, the liver and diaphragm could be insulated via an anterior, epigastric infusion of D5W. Even when the maximal fluid layer was not directly adjacent to the tumor, the protective effect did not seem to be compromised. In vivo animal work at our institution has shown that a 2- to 3-mm layer of D5W substantially decreases thermal damage to the diaphragm or body wall, and this thickness of fluid was easily obtained using our methods in all patients [14]. A potential limitation of D5W infusion would be in patients who have undergone prior hepatic or abdominal surgery. Adhesions may prevent the displacement of bowel or body wall from the radiofrequency site. These patients may be treated best at laparoscopy or, if feasible, at conventional surgery.
The primary limitation of our study was the relatively small sample size. Even so, no statistically significant differences existed between the group characteristics, other than the clinical course. A second limitation of this study was the lack of comparison between D5W and the other available insulators because of the limited scope of the study. Another limitation was the use of a subjective measurement of pain on a linear scale. The use of a pain scale has been validated and is widely applied in clinics and hospitals [21]. Moreover, the subjective assessment of pain was corroborated by the objective use of PCA. Our patient cohort did not include insulin-dependent diabetics. However, we would recommend careful monitoring of serum glucose in such subjects because D5W is absorbed rapidly and completely from the peritoneal space, making these patients at risk for hyperglycemia. Finally, although the procedural time necessary to infuse D5W was not accounted for in this study, that time was minimal (approximately 5-10 min).
On the basis of these findings, we now infuse D5W into the peritoneum for all radiofrequency ablation of peripheral hepatic tumors. Considering that the mean postprocedural pain was less than 3 (on a 10-point scale) for all time points in the D5W group, oral analgesia may be sufficient for this group. In contrast, the mean pain in the control group was initially greater than 6 (on a 10-point scale) despite relatively heavy PCA use, indicating that oral analgesia alone would not have been adequate.
Ablation technology is advancing toward the production of larger thermal lesions, making protection of perihepatic tissues even more important. This study shows that intraperitoneal D5W is an effective method for decreasing postprocedural pain and analgesia use in the setting of percutaneous radiofrequency ablation of peripheral liver tumors. Techniques such as this promise to increase the safety and expand the scope of percutaneous radiofrequency ablation while further decreasing the associated complications and costs.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Sartori, P. Tombesi, F. Macario, I. Nielsen, D. Tassinari, M. Catellani, and V. Abbasciano Subcapsular Liver Tumors Treated with Percutaneous Radiofrequency Ablation: A Prospective Comparison with Nonsubcapsular Liver Tumors for Safety and Effectiveness Radiology, August 1, 2008; 248(2): 670 - 679. [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]
|
|
|
|
|
|
J. L. Hinshaw, P. J. Laeseke, S. M. Weber, and F. T. Lee Jr. Multiple-Electrode Radiofrequency Ablation of Symptomatic Hepatic Cavernous Hemangioma Am. J. Roentgenol., September 1, 2007; 189(3): W146 - W149. [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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
= no D5W, 
= D5W.