ORIGINAL ARTICLE
Radiofrequency Ablation in the Treatment of Large Hepatic Hemangiomas A Comparison of Multitined and Internally Cooled Electrodes Jun Gao, MD,* Xuemei Ding, PhD,* Shan Ke, MD,* Zonghai Xin, PhD,w Chunmin Ning, PhD,z Qi Sha, PhD,y and Wenbing Sun, MD*
Objective: Radiofrequency (RF) ablation is an accepted nonsurgical treatment of hepatic hemangiomas, but with an appreciable complication rate. Our study aimed to assess the safety and efficacy of RF ablation, administered with either multitined expandable electrodes or with internally cooled cluster electrodes, in the treatment of large (Z10 cm diameter) hepatic hemangiomas. Methods: We retrospectively reviewed our sequential experience of treating 43 large hepatic hemangiomas in 42 patients with RF ablation/multitined expandable electrodes or with RF ablation/internally cooled electrodes. Twenty-two hemangiomas in 21 patients were treated with expandable electrode (multitined electrode group), and 21 hemangiomas in 21 patients were treated with internally cooled cluster electrode (internally cooled electrode group). Results: Among the 43 large hepatic hemangiomas, 27 subcapsular lesions were treated by a laparoscopic approach, and 16 lesions located in liver parenchyma were treated by a computed tomography-guided percutaneous approach. In the multitined electrode group, RF ablation treatment was performed in all 21 patients in 1 session. In the internally cooled electrode group, 18 patients were treated by RF ablation in 1 session, and 3 patients, with Z14.0-cm single hemangioma, were treated with RF ablation in 2 sessions. Complete ablation was achieved in 81.8% (18/22) and 90.5% (19/21) in the multitined electrode group and the internally cooled electrode group, respectively (P > 0.05). Ablation time for single hemangioma was shorter with the internally cooled electrode than with the multitined electrode (P < 0.05). There were 79 complications related to ablation (2 major and 77 minor) in 31 patients. All 21 patients in the multitined electrode group experienced complications, compared with 10 of 21 patients (47.6%) in the internally cooled electrode group (P < 0.05). Both of the 2 major complications occurred in the multitined electrode group. All the complications were treated successfully with conservative measures. Conclusions: RF ablation is a safe and effective treatment for large hepatic hemangiomas. Use of the internally cooled cluster electrodes and a more defensive treatment algorithm can reduce the complications.
Received for publication March 7, 2013; accepted June 3, 2013. From the *Department of Hepatobiliary Surgery, Beijing Chao-yang Hospital Affiliated to Capital Medical University, Beijing; wDepartment of General Surgery, Zhanhua People’s Hospital; yDepartment of Gastroenterology, Dezhou People’s Hospital, Shandong; and zDepartment of General Surgery, Chaoyang Central Hospital, Liaoning, China. Supported by grants from the National Natural Science Foundation of China (No. 30872490; 81172320), and from Dr Jieping Wu Medical Foundation (No. 32067501207). The authors declare that they have nothing to disclose. Reprints: Wenbing Sun, MD, Department of Hepatobiliary Surgery, Beijing Chao-yang Hospital Affiliated to Capital Medical University, No. 5 Jingyuan Street, Beijing 100043, China (e-mail: [email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
540 | www.jcge.com
Key Words: hepatic hemangioma, radiofrequency ablation, internally cooled cluster electrode, multitined expandable electrode, ablation-related complication, hemoglobulinuria
(J Clin Gastroenterol 2014;48:540–547)
H
epatic hemangioma is the most common form of benign tumor of the liver. Fortunately, most of them are small and asymptomatic, and therefore medical or surgical intervention is not necessary.1–3 However, if the hemangioma reaches 5 cm or larger, it likely will continue to grow and become symptomatic, and intervention may be needed.4 Surgical resection is the most effective treatment for symptomatic, enlarging hepatic hemangiomas, but it is a highly invasive procedure, with morbidity and mortality rates of up to 27% and 3%, respectively.5–8 Alternatively, minimally invasive procedures such as transcatheter arterial embolization or radiation therapy may be used, but these treatments are not curative.9–12 Radiofrequency (RF) ablation has been proved successful in the treatment of hepatic hemangiomas in the range of 5 to 9.9 cm.13–19 However, treatment of larger hepatic hemangiomas (> 10 cm diameter) with RF ablation is controversial because of a low technical success rate17 and high incidence of ablation-related complications.15 Unfortunately, it is these large hepatic hemangiomas for which nonsurgical treatment is most needed because the risks associated with operative resection are the highest.5,20,21 Almost all severe ablation-related complications in the treatment of hemangiomas are due to the abundant blood supply of the tumors and the relatively long time required for ablation. The most common complications are the direct results of hemolysis, host’s response to heat, and puncture injury associated with the placement of electrode arrays.13,15–19 We have hypothesized that the risk of ablation-related complications would be reduced by the use of highly efficient, internally cooled cluster electrodes to shorten the ablation time and to avoid the puncture injuries associated with the placement of array electrodes. We have reasoned that internally cooled cluster electrodes might be most suitable for RF ablation of large hemangiomas because of their efficiency and more concentrated release of heat,22–26 which are features that can reduce the occurrence of hyperthermia and hemoglobulinuria. Also, we have proposed to use multiple treatment sessions if needed in order to help further prevent complications such as hyperthermia and hemoglobulinuria. In this study, we have evaluated our recent results in RF ablation treatment of large (Z10 cm) hepatic hemangiomas with the use of internally cooled cluster electrodes. J Clin Gastroenterol
Volume 48, Number 6, July 2014
J Clin Gastroenterol
Volume 48, Number 6, July 2014
We have compared these results with our previous experience in which we used multitined expandable electrodes. Our current strategies include using internally cooled cluster electrodes and multiple ablation sessions if the patients’ body temperature exceeds 391C or hemoglobinuria develops during the procedure. The comparative effectiveness and complication rate associated with these 2 treatment modalities are the focus of this research.
PATIENTS AND METHODS Study Design We retrospectively reviewed the records of patients with large hepatic hemangiomas (Z10 cm in diameter) whom we had treated by RF ablation from October 2007 to May 2012. The indications for ablation are described in our previously published article.15 The staff team consisted of hepatobiliary surgeons, anesthesiologists, and radiologists. All procedures were performed by the same experienced operator (W.S.). Between October 2007 and September 2011, we treated hemangiomas in 1 session using multitined expandable electrodes (multitined electrode group). From October 2011 to May 2012, we used internally cooled cluster electrodes (internally cooled electrode group). The study was approved by the Institutional Review Board of our hospital.
Strategies of Ablation We used different treatment strategies of RF ablation in the 2 patient groups. In the multitined electrode group, we used the approach commonly used in ablation treatment of liver malignancies, with emphasis placed on attaining complete ablation in 1 ablation session using multitined expandable electrodes; no attention was paid to high temperature or hemoglobinuria intraoperatively. In the internally cooled electrode group, we introduced 2 measures that would result in few complications, while still maintaining a successful rate of ablation: (a) cool-tip cluster electrodes were used; (b) during the procedure, if the patient’s body temperature exceeded 391C or if hemoglobinuria developed, we stopped the procedure and scheduled a repeat session. Common strategies of ablation used in the 2 patient groups were as follows: (a) subcapsular hemangiomas were treated by laparoscopy and ultrasonic guidance of electrode placement, whereas hemangiomas located in liver parenchyma were treated by computed tomography (CT)-guided percutaneous placement of electrodes; (b) probes were advanced through some normal liver parenchyma before entering the target area in order to prevent bleeding from the hemangiomas; (c) ablation was started at the tumor edge, where the heat-sink effect is relatively small and bleeding is relatively easy to control; (d) if needed, the Pringle maneuver was used in the laparoscopic approach to decrease the heatsink effect; (e) the RF probe was repositioned to target areas in the overlapping ablation method; (f ) the entire area of hemangioma was ablated without injuring the surrounding liver tissue; and (g) track ablation was performed to prevent bleeding as the RF electrode was withdrawn.
RF Ablation Procedures All patients underwent a thorough preoperative evaluation, including laboratory tests and radiological assessment by ultrasonography, helical CT, and/or magnetic resonance imaging (MRI). During the ablative procedure, all patients were on mechanical ventilation, with tracheal r
2013 Lippincott Williams & Wilkins
RF Ablation in the Treatment of Hemangiomas
tube or laryngeal mask airway under intravenous anesthesia for respiratory control. Furthermore, all patients were catheterized and nasopharyngeal temperature was measured for real-time monitoring of hemoglobinuria and body temperature during the procedure. Electrical grounding was achieved by attaching 2 pads to the patient’s thighs, and iced saline packs were used to protect the skin. For patients treated with multitined expandable electrode, the RITA StarBurst Xli-enhanced RF electrode with RF generator (Radiofrequency Interstitial Thermal Ablation Medical System) was used. The RITA system can achieve maximal ablation zones of 7 cm with a single placement of electrodes, with a maximum power of 250 W. When arrays of the needles are introduced into the tumor and positioned satisfactorily, the RF generator achieves RF energy and maintains an average temperature of 1051C. A series of arrays, radiating from the central hollow probe, are pushed forward and unfolded gradually to 3, 4, 5, 6, or 7 cm until they reach the borders of the hemangioma. RF energy is delivered at 5-minute intervals until the output power drops below 30 W in the final step of the procedure. For patients treated with internally cooled cluster electrodes, Cool-tip ACTC2025 or ACTC1525 electrodes and RF generator (Covidien Healthcare, Ireland) were used. Unlike the RITA electrode, the Cool-tip electrode is straight, without arrays. With a 2.5 cm exposed tip, the Cool-tip electrodes can produce ablation zones of 4.5 cm with a single placement of electrodes and a maximum power of 200 W. Also, the Cool-tip RF generator continuously monitors tissue impedance throughout the procedure and adjusts the output accordingly. For the first application, the ablation protocol was preset to the automatic mode, and ablation usually was carried out for 40 to 50 min, which was much longer than that suggested in the manufacturer’s protocol, with the intent of attaining a satisfactory extent of tumor collapse. When a subsequent treatment was given, the duration of ablation was 20 min or shorter, depending on the volume of the tumor. Cold saline continuous irrigation of the needle was provided with an external pump. In the multitined electrode group, our strategy was to pursue a complete ablation at one time; thus, the treatment was terminated mainly on the basis of imaging findings. In the internally cooled electrode group, we followed a more defensive strategy: treatment was stopped not only on the basis of imaging findings, but also on the basis of patient’s body temperature and the presence of hemoglobinuria.
Postoperative Evaluation On the ablation day and the following 3 days, we monitored patient’s vital signs, examined their urine for hemoglobinuria, and monitored blood tests of blood routine, liver and kidney function. All patients were followed up with enhanced CT or MRI 1 month after ablation. Complete ablation was defined as no nodular or irregular enhancement adjacent to the ablated zone, as shown by enhanced CT or MRI images. Incomplete ablation was defined as irregular, peripheral-enhanced foci in the ablated zone. In the cases of complete ablation, subsequent CT or MRI examinations were repeated at 6-month intervals. In the cases of incomplete ablation, repeat RF ablation procedures were not performed unless the residual tumor had progressed during follow-up at 6-month intervals. www.jcge.com |
541
J Clin Gastroenterol
Gao et al
Study Endpoints Primary endpoints of the study were technical success, safety (complications related to RF ablation), mean hospital stay, and confirmed complete ablation. Secondary endpoints were improvement of symptoms, change in size of ablation zone, recurrence of the residual tumor, and quality of life. The endpoints of the study were defined at 6 months after RF ablation treatment.
Statistical Analysis Values were expressed as means ± SD. Continuous variables between groups were compared using the Student’s t test and analysis of variance. Differences in the categorical data were analyzed using the w2 test or Fisher exact test. Two-tailed P values of 0.05). Ablation times for single lesion were significantly shorter in the internally cooled electrode group than in the multitined electrode group (107.3 ± 22.5 vs. 126.7 ± 16.9 min, P < 0.05) (ablation times of the 3 patients whose intervention was interrupted were calculated as the sum of the time of the 2 ablation sessions) (Table 3).
Safety of RF Ablation: Ablation-related Complications As shown in Table 4, there were more ablation-related complications in the multitined electrode group than in the
542 | www.jcge.com
Volume 48, Number 6, July 2014
cooled electrode group (100% vs. 47.6%, P < 0.05). Among individual complications, patients in the cooled electrode group had significantly less hemoglobinuria than the patients in the multitined electrode group (47.6% vs. 100%, P < 0.001); there was also a trend for less hemolytic hyperbilirubinemia (total bilirubin >34.2 mmol/L), anemia (hemoglobin 80 U/L) in the cooled electrode group, but these differences did not reach statistical significance. Complications in the multitined group included 2 major complications [lower esophageal fistula and acute respiratory distress syndrome (ARDS) (Fig. 1), grade III and grade IV, respectively] according to the Dindo-Clavien classification27; others were minor complications (grade I). All the 10 patients with complications in the cooled electrode group were those who had received only a single ablation session; the 3 patients who were treated with 2 ablation sessions, for 14 cm or larger hemangiomas had no ablation-related complications. Thirty-one patients developed hemoglobinuria, as manifested by macroscopically bloody appearance, and were managed with hydration to maintain at least 100 mL per hour urine output; all hemoglobinuria subsided within 72 hours. Of the 31 patients with hemoglobinuria, 12 had hemolytic hyperbilirubinemia, 7 developed anemia, and 1 developed an elevated serum creatinine (>200 mmol/L). All these abnormalities resolved spontaneously within 1 month and did not require major pharmacological intervention or blood transfusion. Fourteen patients experienced moderate fever, from 38 to 391C, which persisted for 3 to 11 days; the fevers resolved spontaneously and did not require antibiotics treatment. Elevated serum transaminase developed in 10 patients; it spontaneously resolved within 2 to 3 weeks after ablation. Superficial second-degree skin burns occurred at the edge of grounding pads in 2 patients in the multitined electrode group; these healed spontaneously. One patient in the multitined electrode group, with an 11.0 cm hemangioma in the left lateral lobe, developed a lower esophageal fistula after laparoscopic ablation. The fistula evidently was caused by the direct puncture of one of the radiated arrays and the subsequent ablation. Another patient in the multitined electrode group, with 2 large hemangiomas in the right lobe, treated with percutaneous RF ablation as a 1-stage treatment, developed ARDS immediately after an ablative time of total 250 minutes (Fig. 1). The 2 major complications were resolved by conservative treatment. Even if rehospitalization for second-stage RF ablation is included, the total hospital stay was significantly shorter in the cooled electrode group than in the multitined electrode group (6.4 ± 2.5 vs. 10.2 ± 4.0 d, P < 0.05), a difference that was directly related to postoperative complications in the multitined electrode group.
Efficacy of RF Ablation Complete ablation was achieved in 81.8% (18/22) and 90.5% (19/21) in the multitined electrode group and the internally cooled electrode group, respectively (P > 0.05) (Table 3 and Fig. 2). Three large hemangiomas (Z14 cm) in the internally cooled electrode group were completely ablated after 2 RF ablation sessions (Fig. 3). Four hemangiomas (18.2%) in the multitined electrode group and 2 (9.5%) in the internally cooled electrode group were incompletely ablated, r
2013 Lippincott Williams & Wilkins
J Clin Gastroenterol
Volume 48, Number 6, July 2014
RF Ablation in the Treatment of Hemangiomas
FIGURE 1. A 46-year-old man in the multitined electrode group had 3 large hepatic hemangiomas in the left lobe (35.0 cm), right anterior lobe (16.0 cm), and right posterior lobe (11.0 cm), as revealed by abdominal CT scans (A–C). Immediately after CT-guided percutaneous RF ablation treatment of the hemangiomas in the right lobe, CT scans showed diffuse edema and inflammatory consolidation of both lungs (D). Thirty-two hours later and after a series of conservative medical treatments, a CT scan showed the pathological changes of the lungs had markedly improved (E). Three months after ablation treatment, CT scans in the arterial phase showed 2 hepatic hemangiomas in the right lobe had been completely ablated; the left liver lobe had been resected (F and G). CT indicates computed tomography; RF, radiofrequency.
r
2013 Lippincott Williams & Wilkins
www.jcge.com |
543
J Clin Gastroenterol
Gao et al
TABLE 1. Demographic Characteristics of Patients in the Study
Group
Multitined Electrode (n = 21)
Internally Cooled Electrode (n = 21)
Age, mean (SD) 53.6 (12.0) 51.3 (9.7) Gender (male:female) 8:13 6:15 Single hemangioma 20 (95.2) 21 (100) lesion, N (%) Comorbidities, N (%) Gallbladder stones 2 (9.5) 3 (14.3) Type 2 diabetes 1 (4.8) 2 (9.5) mellitus History of open 0 (0) 1 (4.8) cholecystectomy Chronic hepatitis B 1 (4.8) 0 (0) History of previous 0 (0) 1 (4.8) liver surgery Hepatic cysts 0 (0) 1 (4.8) Reasons for radiofrequency ablation, N (%) Abdominal 3 (14.3) 4 (19.1) discomfort only Enlargement of 2 (9.5) 2 (9.5) hemangioma only Abdominal 16 (76.2) 15 (71.4) discomfort and enlargement
TABLE 3. Outcome of Radiofrequency (RF) Ablation on Large Hepatic Hemangiomas
Multitined Electrode (n = 22)
0.501 0.513 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Group
Approach of RF ablation, N (%) Laparoscopic approach 13 Percutaneous approach 9 RF ablation sessions, N (%) 1 RF ablation session 22 2 RF ablation sessions 0 Complete ablation, N (%) 18 No. punctures for per lesion, N (%) 6.1 Time of ablation per lesion, mean 126.7 (SD) (min) Diameter of ablated zone 1 mo 10.5 after ablation, mean (SD) (cm) Diameter of ablated zone 6 mo 7.9 after ablation, mean (SD) (cm)
1.000 0.726
After RF ablation, there was no perioperative mortality or delayed complications, such as local tumor progression, destructive biliary damages, or liver abscess. The 6 residual lesions shrunk lightly during the follow-up period and necessitated no further treatment. Of the 38 patients who had pretreatment symptoms related to hemangioma, TABLE 2. Characteristics of 43 Hepatic Hemangiomas Treated by Radiofrequency Ablation
544 | www.jcge.com
Multitined Electrode (n = 22)
Internally Cooled Electrode (n = 21)
P
12.4 (1.6)
12.5 (1.5)
0.855
10 16
10 15 0.663
14 (63.6) 8 (36.4)
12 (57.1) 9 (42.9)
13 (59.1) 9 (40.9)
14 (66.7) 7 (33.3)
Internally Cooled Electrode (n = 21)
P 0.607
(59.1) (40.9)
14 (66.7) 7 (33.3)
(100) 18 (0) 3 (81.8) 19 (1.1) 6.0 (16.9) 107.3
0.233 (85.7) (14.3) (90.5) 0.664 (1.0) 0.740 (22.5) 0.007
(2.4)
10.7 (2.5)
0.840
(2.1)
8.2 (3.2)
0.694
1.000
35 had complete resolution of symptoms and 3 had amelioration of symptoms without any therapy after ablation. During follow-up in 6 months, no patient developed new symptoms attributed to hemangiomas. Both the subjective health status and the quality of life were rated as good or better in 100% at follow-up. After RF ablation of the hemangioma, all patients were able to work full-time or part-time.
DISCUSSION The major finding of our study was that the use of internally cooled electrodes in the RF ablation treatment of large hepatic hemangiomas was successful and had fewer complications (47.6% vs. 100%) compared with ablation TABLE 4. Complications of Radiofrequency Ablation
Follow-up Results
Maximal size of hemangioma, mean (SD) (cm) Minimum (cm) Maximum (cm) Distribution of lesion, N (%) Right lobe Left lobe Location of lesion, N (%) Subcapsular Deep-sited
Volume 48, Number 6, July 2014
P
showing subtle enhancement on the peripheral rim of the ablated tumors in follow-up CT or MRI. The mean diameter of ablation zone was reduced from 12.4 ± 1.6 to 10.5 ± 2.4 cm in the multitined electrode group, and from 12.5 ± 1.5 to 10.7 ± 2.5 cm in the internally cooled electrode group after 1 month, an insignificant difference between the groups (P > 0.05). At 6 months after ablation, the mean diameter of ablation zone had decreased further in both groups (to 7.9 ± 2.1 cm in the multitined electrode group and to 8.2 ± 3.2 cm in the internally cooled electrode group), also an insignificant difference.
Group
0.607
Internally Cooled Electrode (n = 21)
P
(100)
10 (47.6)
0.000
(100) (47.6)
10 (47.6) 4 (19.0)
0.000 0.100
(42.9)
3 (14.3)
0.085
(28.6)
1 (4.8)
0.093
(28.6)
4 (19.0)
0.719
(4.8)
0 (0.0)
1.000
(9.5) (4.8) (4.8)
0 (0.0) 0 (0.0) 0 (0.0)
0.488 1.000 1.000
(4.0)
6.4 (2.5)
0.000
Multitined Electrode (n = 21)
Group
Total no. patients with 21 complication, N (%) Incidence of complication, N (%) Hemoglobinuria 21 Fever (body temperature 10 Z381C) Hemolytic 9 hyperbilirubinemia (total bilirubin >34.2 mmol/L) Anemia (hemoglobin 6 80 U/L) Elevated creatinine 1 (>200 mmol/L) Skin burns 2 Lower esophageal fistula 1 Acute respiratory 1 distress syndrome Hospital stay, mean 10.2 (SD)
r
2013 Lippincott Williams & Wilkins
J Clin Gastroenterol
Volume 48, Number 6, July 2014
RF Ablation in the Treatment of Hemangiomas
FIGURE 2. A 53-year-old woman in the internally cooled electrode group had an 11.0 cm hemangioma in the right lobe, as illustrated by CT scans (A and B). One month after percutaneous RF ablation in 1 session, CT scans showed that the hemangioma was completely ablated and remarkably smaller (C). After 6 months, the hemangioma was smaller yet, as illustrated in an abdominal CT scan (D). CT indicates computed tomography; RF, radiofrequency.
using multitined electrodes. A similarly high frequency of complete ablation (>80%) of these very challenging tumors was attained with both forms of treatment. Also, the immediate and sustained reduction in size of the tumors was similar in the 2 groups. We also found that the length of hospital stay was significantly shorter for patients treated with the internally cooled electrode approach, most likely because of a lower incidence of complications in them. Several series have reported a morbidity rate of 10% to 27% and a mortality rate of 0% to 3% after resection or enucleation of hepatic hemangiomas, respectively.5–8 The size of hepatic hemangiomas is a significant risk factor for hepatectomy, mainly because of the massive intraoperative blood loss and blood transfusion requirement associated with major hepatic resection.5 Compared with the considerable risk and invasiveness of surgical resection, we consider a complication rate of 47.6% in patients with hemangiomas Z10 cm in the internally cooled electrode group to be acceptable, especially because all the complications were minor (grade I). The lower rate of complications associated with the internally cooled electrodes in the treatment of large hemangiomas is a significant improvement in RF ablation treatment as this form of treatment has been beset by complications. Indeed, some investigators have doubted that RF ablation is an acceptable approach for large hemangiomas because of the long ablation time required. r
2013 Lippincott Williams & Wilkins
Lengthy ablation time unavoidably causes hemolysis, which can lead to the complications of hemoglobinuria, hemolytic jaundice, anemia, or even renal failure. The high incidence of complications associated with ablation is directly proportional to the size of hemangiomas and the longer ablation time.17 We find it reassuring that complete and sustained resolution of large hepatic hemangiomas was attained through use of the internally cooled electrode approach. Although RF ablation has been shown effective and reasonably safe in the treatment of hemangiomas of 5 to 9.9 cm in size,13–19 this form of treatment for larger hemangiomas remains controversial. For example, Park et al17 reported a technical failure rate of 40% with percutaneous RF ablation of hemangiomas Z10 cm in diameter. In our earlier experience with multitined electrodes we also encountered difficulties, including significant systemic inflammatory responses and ARDS, likely resulting from long ablation times.15,28 It is desirable that there be safe and effective noninvasive treatment of hepatic hemangiomas because surgical resection is highly risky, especially for large tumors, such as those that occupy major portion of a lobe or are located deep within the hepatic parenchyma.5,20,21 Although transcatheter arterial embolization and radiation have been used in the treatment of hepatic hemangiomas, they have not been proved uniformly effective.9,10,12 www.jcge.com |
545
Gao et al
J Clin Gastroenterol
Volume 48, Number 6, July 2014
FIGURE 3. A 45-year-old woman in the internally cooled electrode group had a 15.0 cm hemangioma in the right lobe, as illustrated in an abdominal CT scan (A). In laparoscopic views, the tumor is evident on the superior surface (B) and on the posterior and inferior surfaces of the liver (C). During the ablation procedure, hemoglobinuria developed and the ablation was stopped (D). The lesion became a depressed mass with hard texture after the first RF ablation session. The abdomen CT at 1 month showed a partially enhancing area in the hemangioma (E). The hemangioma was completely ablated after the second RF ablation session and 1 month later was smaller, as illustrated in an abdominal CT scan (F).
Currently available RF electrodes vary in their design and mechanism of energy distribution. Multitined expandable electrodes, for example RITA provide even tissue penetration and heat delivery and produce maximal ablated zones of up to 7 cm with a single application in the temperature-controlled ablation mode. In normal liver tissue, use of
546 | www.jcge.com
5, 6, and 7-cm array extensions take about 18.5, 23.3, and 30.4 minutes to achieve a 4.9, 5.4, and 6.1 cm diameter zone of tissue coagulation. In contrast, internally cooled cluster electrodes, for example Cool-tip, are straight electrodes without arrays, and the tip is internally cooled by continuous infusion of cold saline. Thus, the temperature of the electrode r
2013 Lippincott Williams & Wilkins
J Clin Gastroenterol
Volume 48, Number 6, July 2014
is kept low, which helps prevent charring at the tissue-probe interface and reduces tissue impedance. The efficient heat deposition creates a larger ablated zone in shorter time (a 4.5 cm zone in about 12 min in normal liver tissue). Several studies have shown that internally cooled cluster electrodes are more efficient than multitined expandable electrodes in the ablation of solid tumors.22–26 Another advantage of internally cooled electrodes, such as Cool-tip, is that they keep a steady high temperature in the tumor while limiting vascular cooling. This characteristic increases the effectiveness of perivascular ablation.29 Moreover, the design of Cooltip electrodes permits their ready placement into target lesions without risk of accidental injury to adjacent organs; the full length of the electrode is readily visible with CT or ultrasound guidance during the procedure. In contrast, localization of each tip of the multitined expandable electrodes requires verification one by one, with attendant increased effort, time, and expense. We have had no accidental damage of adjacent organs using Cool-tip electrodes. The major limitations of our study include its retrospective nature, the nonrandomized selection of patients, and a relatively small number of patients evaluated. We feel it important, though, that all the RF procedures were performed by 1 surgeon, thus minimizing the chance of bias that might have occurred if multiple surgeons had been involved. Despite possible shortcomings in this study, we feel that our results favoring the use of internally cooled cluster electrodes in RF treatment of large hepatic hemangiomas are persuasive and justify further use of this technology in this challenging disease. However, a controlled, randomized study to verify this conclusion would be advisable. In conclusion, RF ablation is an effective therapeutic modality for treatment of large hepatic hemangiomas. Use of the internally cooled cluster electrodes and a more defensive treatment algorithm can reduce the complications.
REFERENCES 1. Farges O, Daradkeh S, Bismuth H. Cavernous hemangiomas of the liver: are there any indications for resection? World J Surg. 1995;19:19–24. 2. Schnelldorfer T, Ware AL, Smoot R, et al. Management of giant hemangioma of the liver: resection versus observation. J Am Coll Surg. 2010;211:724–730. 3. Yeh WC, Yang PM, Huang GT, et al. Long-term follow-up of hepatic hemangiomas by ultrasonography: with emphasis on the growth rate of the tumor. Hepatogastroenterology. 2007; 54:475–479. 4. Hamaloglu E, Altun H, Ozdemir A, et al. Giant liver hemangioma: therapy by enucleation or liver resection. World J Surg. 2005;29:890–893. 5. Hanazaki K, Kajikawa S, Matsushita A, et al. Giant cavernous hemangioma of the liver: is tumor size a risk factor for hepatectomy? J Hepatobiliary Pancreat Surg. 1999;6:410–413. 6. Lerner SM, Hiatt JR, Salamandra J, et al. Giant cavernous liver hemangiomas: effect of operative approach on outcome. Arch Surg. 2004;139:818–821; discussion 821-813. 7. Terkivatan T, de Wilt JH, de Man RA, et al. Indications and long-term outcome of treatment for benign hepatic tumors: a critical appraisal. Arch Surg. 2001;136:1033–1038. 8. Yoon SS, Charny CK, Fong Y, et al. Diagnosis, management, and outcomes of 115 patients with hepatic hemangioma. J Am Coll Surg. 2003;197:392–402.
r
2013 Lippincott Williams & Wilkins
RF Ablation in the Treatment of Hemangiomas
9. Gaspar L, Mascarenhas F, da Costa MS, et al. Radiation therapy in the unresectable cavernous hemangioma of the liver. Radiother Oncol. 1993;29:45–50. 10. Giavroglou C, Economou H, Ioannidis I. Arterial embolization of giant hepatic hemangiomas. Cardiovasc Intervent Radiol. 2003;26:92–96. 11. Huang XQ, Huang ZQ, Duan WD, et al. Severe biliary complications after hepatic artery embolization. World J Gastroenterol. 2002;8:119–123. 12. Park WC, Rhillips R. The role of radiation therapy in the management of hemangiomas of the liver. JAMA. 1970;212:1496–1498. 13. Cui Y, Zhou LY, Dong MK, et al. Ultrasonography guided percutaneous radiofrequency ablation for hepatic cavernous hemangioma. World J Gastroenterol. 2003;9:2132–2134. 14. Fan RF, Chai FL, He GX, et al. Laparoscopic radiofrequency ablation of hepatic cavernous hemangioma. A preliminary experience with 27 patients. Surg Endosc. 2006;20:281–285. 15. Gao J, Ke S, Ding XM, et al. Radiofrequency ablation for large hepatic hemangiomas: Initial experience and lessons. Surgery. 2012;153:78–85. 16. Hinshaw JL, Laeseke PJ, Weber SM, et al. Multiple-electrode radiofrequency ablation of symptomatic hepatic cavernous hemangioma. Am J Roentgenol. 2007;189:W146–W149. 17. Park SY, Tak WY, Jung MK, et al. Symptomatic-enlarging hepatic hemangiomas are effectively treated by percutaneous ultrasonography-guided radiofrequency ablation. J Hepatol. 2011;54:559–565. 18. Tak WY, Park SY, Jeon SW, et al. Ultrasonography-guided percutaneous radiofrequency ablation for treatment of a huge symptomatic hepatic cavernous hemangioma. J Clin Gastroenterol. 2006;40:167–170. 19. Zagoria RJ, Roth TJ, Levine EA, et al. Radiofrequency ablation of a symptomatic hepatic cavernous hemangioma. Am J Roentgenol. 2004;182:210–212. 20. Hering J, Garrean S, Saied A, et al. Use of radiofrequency hepatic parenchymal transection device in hepatic hemangioma resection: early experience and lessons learned. HPB (Oxford). 2007;9:319–323. 21. Li MH, Yan LN, Wang SR. Autologous transfusion with modified total hepatic vascular exclusion for extracapsular resection of giant hepatic cavernous hemangioma. Hepatobiliary Pancreat Dis Int. 2007;6:43–48. 22. Burdio F, Tobajas P, Quesada-Diez R, et al. Distant infusion of saline may enlarge coagulation volume during radiofrequency ablation of liver tissue using cool-tip electrodes without impairing predictability. Am J Roentgenol. 2011;196:W837–W843. 23. Hung WK, Mak KL, Ying M, et al. Radiofrequency ablation of breast cancer: a comparative study of two needle designs. Breast Cancer. 2011;18:124–128. 24. Manenti G, Bolacchi F, Perretta T, et al. Small breast cancers: in vivo percutaneous US-guided radiofrequency ablation with dedicated cool-tip radiofrequency system. Radiology. 2009;251: 339–346. 25. Martel J, Bueno A, Ortiz E. Percutaneous radiofrequency treatment of osteoid osteoma using cool-tip electrodes. Eur J Radiol. 2005;56:403–408. 26. Wakui N, Iida K, Takayama R, et al. Cool-tip 10-mm electrode useful for radiofrequency ablation of hepatocellular carcinoma. Hepatogastroenterology. 2009;56:1585–1591. 27. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–213. 28. Ng KK, Lam CM, Poon RT, et al. Safety limit of large-volume hepatic radiofrequency ablation in a rat model. Arch Surg. 2006;141:252–258. 29. Haemmerich D, Lee FT Jr, Schutt DJ, et al. Large-volume radiofrequency ablation of ex vivo bovine liver with multiple cooled cluster electrodes. Radiology. 2005;234:563–568.
www.jcge.com |
547
Radiofrequency Ablation in the Treatment of Large Hepatic Hemangiomas A Comparison of Multitined and Internally Cooled Electrodes Jun Gao, MD,* Xuemei Ding, PhD,* Shan Ke, MD,* Zonghai Xin, PhD,w Chunmin Ning, PhD,z Qi Sha, PhD,y and Wenbing Sun, MD*
Objective: Radiofrequency (RF) ablation is an accepted nonsurgical treatment of hepatic hemangiomas, but with an appreciable complication rate. Our study aimed to assess the safety and efficacy of RF ablation, administered with either multitined expandable electrodes or with internally cooled cluster electrodes, in the treatment of large (Z10 cm diameter) hepatic hemangiomas. Methods: We retrospectively reviewed our sequential experience of treating 43 large hepatic hemangiomas in 42 patients with RF ablation/multitined expandable electrodes or with RF ablation/internally cooled electrodes. Twenty-two hemangiomas in 21 patients were treated with expandable electrode (multitined electrode group), and 21 hemangiomas in 21 patients were treated with internally cooled cluster electrode (internally cooled electrode group). Results: Among the 43 large hepatic hemangiomas, 27 subcapsular lesions were treated by a laparoscopic approach, and 16 lesions located in liver parenchyma were treated by a computed tomography-guided percutaneous approach. In the multitined electrode group, RF ablation treatment was performed in all 21 patients in 1 session. In the internally cooled electrode group, 18 patients were treated by RF ablation in 1 session, and 3 patients, with Z14.0-cm single hemangioma, were treated with RF ablation in 2 sessions. Complete ablation was achieved in 81.8% (18/22) and 90.5% (19/21) in the multitined electrode group and the internally cooled electrode group, respectively (P > 0.05). Ablation time for single hemangioma was shorter with the internally cooled electrode than with the multitined electrode (P < 0.05). There were 79 complications related to ablation (2 major and 77 minor) in 31 patients. All 21 patients in the multitined electrode group experienced complications, compared with 10 of 21 patients (47.6%) in the internally cooled electrode group (P < 0.05). Both of the 2 major complications occurred in the multitined electrode group. All the complications were treated successfully with conservative measures. Conclusions: RF ablation is a safe and effective treatment for large hepatic hemangiomas. Use of the internally cooled cluster electrodes and a more defensive treatment algorithm can reduce the complications.
Received for publication March 7, 2013; accepted June 3, 2013. From the *Department of Hepatobiliary Surgery, Beijing Chao-yang Hospital Affiliated to Capital Medical University, Beijing; wDepartment of General Surgery, Zhanhua People’s Hospital; yDepartment of Gastroenterology, Dezhou People’s Hospital, Shandong; and zDepartment of General Surgery, Chaoyang Central Hospital, Liaoning, China. Supported by grants from the National Natural Science Foundation of China (No. 30872490; 81172320), and from Dr Jieping Wu Medical Foundation (No. 32067501207). The authors declare that they have nothing to disclose. Reprints: Wenbing Sun, MD, Department of Hepatobiliary Surgery, Beijing Chao-yang Hospital Affiliated to Capital Medical University, No. 5 Jingyuan Street, Beijing 100043, China (e-mail: [email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
540 | www.jcge.com
Key Words: hepatic hemangioma, radiofrequency ablation, internally cooled cluster electrode, multitined expandable electrode, ablation-related complication, hemoglobulinuria
(J Clin Gastroenterol 2014;48:540–547)
H
epatic hemangioma is the most common form of benign tumor of the liver. Fortunately, most of them are small and asymptomatic, and therefore medical or surgical intervention is not necessary.1–3 However, if the hemangioma reaches 5 cm or larger, it likely will continue to grow and become symptomatic, and intervention may be needed.4 Surgical resection is the most effective treatment for symptomatic, enlarging hepatic hemangiomas, but it is a highly invasive procedure, with morbidity and mortality rates of up to 27% and 3%, respectively.5–8 Alternatively, minimally invasive procedures such as transcatheter arterial embolization or radiation therapy may be used, but these treatments are not curative.9–12 Radiofrequency (RF) ablation has been proved successful in the treatment of hepatic hemangiomas in the range of 5 to 9.9 cm.13–19 However, treatment of larger hepatic hemangiomas (> 10 cm diameter) with RF ablation is controversial because of a low technical success rate17 and high incidence of ablation-related complications.15 Unfortunately, it is these large hepatic hemangiomas for which nonsurgical treatment is most needed because the risks associated with operative resection are the highest.5,20,21 Almost all severe ablation-related complications in the treatment of hemangiomas are due to the abundant blood supply of the tumors and the relatively long time required for ablation. The most common complications are the direct results of hemolysis, host’s response to heat, and puncture injury associated with the placement of electrode arrays.13,15–19 We have hypothesized that the risk of ablation-related complications would be reduced by the use of highly efficient, internally cooled cluster electrodes to shorten the ablation time and to avoid the puncture injuries associated with the placement of array electrodes. We have reasoned that internally cooled cluster electrodes might be most suitable for RF ablation of large hemangiomas because of their efficiency and more concentrated release of heat,22–26 which are features that can reduce the occurrence of hyperthermia and hemoglobulinuria. Also, we have proposed to use multiple treatment sessions if needed in order to help further prevent complications such as hyperthermia and hemoglobulinuria. In this study, we have evaluated our recent results in RF ablation treatment of large (Z10 cm) hepatic hemangiomas with the use of internally cooled cluster electrodes. J Clin Gastroenterol
Volume 48, Number 6, July 2014
J Clin Gastroenterol
Volume 48, Number 6, July 2014
We have compared these results with our previous experience in which we used multitined expandable electrodes. Our current strategies include using internally cooled cluster electrodes and multiple ablation sessions if the patients’ body temperature exceeds 391C or hemoglobinuria develops during the procedure. The comparative effectiveness and complication rate associated with these 2 treatment modalities are the focus of this research.
PATIENTS AND METHODS Study Design We retrospectively reviewed the records of patients with large hepatic hemangiomas (Z10 cm in diameter) whom we had treated by RF ablation from October 2007 to May 2012. The indications for ablation are described in our previously published article.15 The staff team consisted of hepatobiliary surgeons, anesthesiologists, and radiologists. All procedures were performed by the same experienced operator (W.S.). Between October 2007 and September 2011, we treated hemangiomas in 1 session using multitined expandable electrodes (multitined electrode group). From October 2011 to May 2012, we used internally cooled cluster electrodes (internally cooled electrode group). The study was approved by the Institutional Review Board of our hospital.
Strategies of Ablation We used different treatment strategies of RF ablation in the 2 patient groups. In the multitined electrode group, we used the approach commonly used in ablation treatment of liver malignancies, with emphasis placed on attaining complete ablation in 1 ablation session using multitined expandable electrodes; no attention was paid to high temperature or hemoglobinuria intraoperatively. In the internally cooled electrode group, we introduced 2 measures that would result in few complications, while still maintaining a successful rate of ablation: (a) cool-tip cluster electrodes were used; (b) during the procedure, if the patient’s body temperature exceeded 391C or if hemoglobinuria developed, we stopped the procedure and scheduled a repeat session. Common strategies of ablation used in the 2 patient groups were as follows: (a) subcapsular hemangiomas were treated by laparoscopy and ultrasonic guidance of electrode placement, whereas hemangiomas located in liver parenchyma were treated by computed tomography (CT)-guided percutaneous placement of electrodes; (b) probes were advanced through some normal liver parenchyma before entering the target area in order to prevent bleeding from the hemangiomas; (c) ablation was started at the tumor edge, where the heat-sink effect is relatively small and bleeding is relatively easy to control; (d) if needed, the Pringle maneuver was used in the laparoscopic approach to decrease the heatsink effect; (e) the RF probe was repositioned to target areas in the overlapping ablation method; (f ) the entire area of hemangioma was ablated without injuring the surrounding liver tissue; and (g) track ablation was performed to prevent bleeding as the RF electrode was withdrawn.
RF Ablation Procedures All patients underwent a thorough preoperative evaluation, including laboratory tests and radiological assessment by ultrasonography, helical CT, and/or magnetic resonance imaging (MRI). During the ablative procedure, all patients were on mechanical ventilation, with tracheal r
2013 Lippincott Williams & Wilkins
RF Ablation in the Treatment of Hemangiomas
tube or laryngeal mask airway under intravenous anesthesia for respiratory control. Furthermore, all patients were catheterized and nasopharyngeal temperature was measured for real-time monitoring of hemoglobinuria and body temperature during the procedure. Electrical grounding was achieved by attaching 2 pads to the patient’s thighs, and iced saline packs were used to protect the skin. For patients treated with multitined expandable electrode, the RITA StarBurst Xli-enhanced RF electrode with RF generator (Radiofrequency Interstitial Thermal Ablation Medical System) was used. The RITA system can achieve maximal ablation zones of 7 cm with a single placement of electrodes, with a maximum power of 250 W. When arrays of the needles are introduced into the tumor and positioned satisfactorily, the RF generator achieves RF energy and maintains an average temperature of 1051C. A series of arrays, radiating from the central hollow probe, are pushed forward and unfolded gradually to 3, 4, 5, 6, or 7 cm until they reach the borders of the hemangioma. RF energy is delivered at 5-minute intervals until the output power drops below 30 W in the final step of the procedure. For patients treated with internally cooled cluster electrodes, Cool-tip ACTC2025 or ACTC1525 electrodes and RF generator (Covidien Healthcare, Ireland) were used. Unlike the RITA electrode, the Cool-tip electrode is straight, without arrays. With a 2.5 cm exposed tip, the Cool-tip electrodes can produce ablation zones of 4.5 cm with a single placement of electrodes and a maximum power of 200 W. Also, the Cool-tip RF generator continuously monitors tissue impedance throughout the procedure and adjusts the output accordingly. For the first application, the ablation protocol was preset to the automatic mode, and ablation usually was carried out for 40 to 50 min, which was much longer than that suggested in the manufacturer’s protocol, with the intent of attaining a satisfactory extent of tumor collapse. When a subsequent treatment was given, the duration of ablation was 20 min or shorter, depending on the volume of the tumor. Cold saline continuous irrigation of the needle was provided with an external pump. In the multitined electrode group, our strategy was to pursue a complete ablation at one time; thus, the treatment was terminated mainly on the basis of imaging findings. In the internally cooled electrode group, we followed a more defensive strategy: treatment was stopped not only on the basis of imaging findings, but also on the basis of patient’s body temperature and the presence of hemoglobinuria.
Postoperative Evaluation On the ablation day and the following 3 days, we monitored patient’s vital signs, examined their urine for hemoglobinuria, and monitored blood tests of blood routine, liver and kidney function. All patients were followed up with enhanced CT or MRI 1 month after ablation. Complete ablation was defined as no nodular or irregular enhancement adjacent to the ablated zone, as shown by enhanced CT or MRI images. Incomplete ablation was defined as irregular, peripheral-enhanced foci in the ablated zone. In the cases of complete ablation, subsequent CT or MRI examinations were repeated at 6-month intervals. In the cases of incomplete ablation, repeat RF ablation procedures were not performed unless the residual tumor had progressed during follow-up at 6-month intervals. www.jcge.com |
541
J Clin Gastroenterol
Gao et al
Study Endpoints Primary endpoints of the study were technical success, safety (complications related to RF ablation), mean hospital stay, and confirmed complete ablation. Secondary endpoints were improvement of symptoms, change in size of ablation zone, recurrence of the residual tumor, and quality of life. The endpoints of the study were defined at 6 months after RF ablation treatment.
Statistical Analysis Values were expressed as means ± SD. Continuous variables between groups were compared using the Student’s t test and analysis of variance. Differences in the categorical data were analyzed using the w2 test or Fisher exact test. Two-tailed P values of 0.05). Ablation times for single lesion were significantly shorter in the internally cooled electrode group than in the multitined electrode group (107.3 ± 22.5 vs. 126.7 ± 16.9 min, P < 0.05) (ablation times of the 3 patients whose intervention was interrupted were calculated as the sum of the time of the 2 ablation sessions) (Table 3).
Safety of RF Ablation: Ablation-related Complications As shown in Table 4, there were more ablation-related complications in the multitined electrode group than in the
542 | www.jcge.com
Volume 48, Number 6, July 2014
cooled electrode group (100% vs. 47.6%, P < 0.05). Among individual complications, patients in the cooled electrode group had significantly less hemoglobinuria than the patients in the multitined electrode group (47.6% vs. 100%, P < 0.001); there was also a trend for less hemolytic hyperbilirubinemia (total bilirubin >34.2 mmol/L), anemia (hemoglobin 80 U/L) in the cooled electrode group, but these differences did not reach statistical significance. Complications in the multitined group included 2 major complications [lower esophageal fistula and acute respiratory distress syndrome (ARDS) (Fig. 1), grade III and grade IV, respectively] according to the Dindo-Clavien classification27; others were minor complications (grade I). All the 10 patients with complications in the cooled electrode group were those who had received only a single ablation session; the 3 patients who were treated with 2 ablation sessions, for 14 cm or larger hemangiomas had no ablation-related complications. Thirty-one patients developed hemoglobinuria, as manifested by macroscopically bloody appearance, and were managed with hydration to maintain at least 100 mL per hour urine output; all hemoglobinuria subsided within 72 hours. Of the 31 patients with hemoglobinuria, 12 had hemolytic hyperbilirubinemia, 7 developed anemia, and 1 developed an elevated serum creatinine (>200 mmol/L). All these abnormalities resolved spontaneously within 1 month and did not require major pharmacological intervention or blood transfusion. Fourteen patients experienced moderate fever, from 38 to 391C, which persisted for 3 to 11 days; the fevers resolved spontaneously and did not require antibiotics treatment. Elevated serum transaminase developed in 10 patients; it spontaneously resolved within 2 to 3 weeks after ablation. Superficial second-degree skin burns occurred at the edge of grounding pads in 2 patients in the multitined electrode group; these healed spontaneously. One patient in the multitined electrode group, with an 11.0 cm hemangioma in the left lateral lobe, developed a lower esophageal fistula after laparoscopic ablation. The fistula evidently was caused by the direct puncture of one of the radiated arrays and the subsequent ablation. Another patient in the multitined electrode group, with 2 large hemangiomas in the right lobe, treated with percutaneous RF ablation as a 1-stage treatment, developed ARDS immediately after an ablative time of total 250 minutes (Fig. 1). The 2 major complications were resolved by conservative treatment. Even if rehospitalization for second-stage RF ablation is included, the total hospital stay was significantly shorter in the cooled electrode group than in the multitined electrode group (6.4 ± 2.5 vs. 10.2 ± 4.0 d, P < 0.05), a difference that was directly related to postoperative complications in the multitined electrode group.
Efficacy of RF Ablation Complete ablation was achieved in 81.8% (18/22) and 90.5% (19/21) in the multitined electrode group and the internally cooled electrode group, respectively (P > 0.05) (Table 3 and Fig. 2). Three large hemangiomas (Z14 cm) in the internally cooled electrode group were completely ablated after 2 RF ablation sessions (Fig. 3). Four hemangiomas (18.2%) in the multitined electrode group and 2 (9.5%) in the internally cooled electrode group were incompletely ablated, r
2013 Lippincott Williams & Wilkins
J Clin Gastroenterol
Volume 48, Number 6, July 2014
RF Ablation in the Treatment of Hemangiomas
FIGURE 1. A 46-year-old man in the multitined electrode group had 3 large hepatic hemangiomas in the left lobe (35.0 cm), right anterior lobe (16.0 cm), and right posterior lobe (11.0 cm), as revealed by abdominal CT scans (A–C). Immediately after CT-guided percutaneous RF ablation treatment of the hemangiomas in the right lobe, CT scans showed diffuse edema and inflammatory consolidation of both lungs (D). Thirty-two hours later and after a series of conservative medical treatments, a CT scan showed the pathological changes of the lungs had markedly improved (E). Three months after ablation treatment, CT scans in the arterial phase showed 2 hepatic hemangiomas in the right lobe had been completely ablated; the left liver lobe had been resected (F and G). CT indicates computed tomography; RF, radiofrequency.
r
2013 Lippincott Williams & Wilkins
www.jcge.com |
543
J Clin Gastroenterol
Gao et al
TABLE 1. Demographic Characteristics of Patients in the Study
Group
Multitined Electrode (n = 21)
Internally Cooled Electrode (n = 21)
Age, mean (SD) 53.6 (12.0) 51.3 (9.7) Gender (male:female) 8:13 6:15 Single hemangioma 20 (95.2) 21 (100) lesion, N (%) Comorbidities, N (%) Gallbladder stones 2 (9.5) 3 (14.3) Type 2 diabetes 1 (4.8) 2 (9.5) mellitus History of open 0 (0) 1 (4.8) cholecystectomy Chronic hepatitis B 1 (4.8) 0 (0) History of previous 0 (0) 1 (4.8) liver surgery Hepatic cysts 0 (0) 1 (4.8) Reasons for radiofrequency ablation, N (%) Abdominal 3 (14.3) 4 (19.1) discomfort only Enlargement of 2 (9.5) 2 (9.5) hemangioma only Abdominal 16 (76.2) 15 (71.4) discomfort and enlargement
TABLE 3. Outcome of Radiofrequency (RF) Ablation on Large Hepatic Hemangiomas
Multitined Electrode (n = 22)
0.501 0.513 1.000 1.000 1.000 1.000 1.000 1.000 1.000
Group
Approach of RF ablation, N (%) Laparoscopic approach 13 Percutaneous approach 9 RF ablation sessions, N (%) 1 RF ablation session 22 2 RF ablation sessions 0 Complete ablation, N (%) 18 No. punctures for per lesion, N (%) 6.1 Time of ablation per lesion, mean 126.7 (SD) (min) Diameter of ablated zone 1 mo 10.5 after ablation, mean (SD) (cm) Diameter of ablated zone 6 mo 7.9 after ablation, mean (SD) (cm)
1.000 0.726
After RF ablation, there was no perioperative mortality or delayed complications, such as local tumor progression, destructive biliary damages, or liver abscess. The 6 residual lesions shrunk lightly during the follow-up period and necessitated no further treatment. Of the 38 patients who had pretreatment symptoms related to hemangioma, TABLE 2. Characteristics of 43 Hepatic Hemangiomas Treated by Radiofrequency Ablation
544 | www.jcge.com
Multitined Electrode (n = 22)
Internally Cooled Electrode (n = 21)
P
12.4 (1.6)
12.5 (1.5)
0.855
10 16
10 15 0.663
14 (63.6) 8 (36.4)
12 (57.1) 9 (42.9)
13 (59.1) 9 (40.9)
14 (66.7) 7 (33.3)
Internally Cooled Electrode (n = 21)
P 0.607
(59.1) (40.9)
14 (66.7) 7 (33.3)
(100) 18 (0) 3 (81.8) 19 (1.1) 6.0 (16.9) 107.3
0.233 (85.7) (14.3) (90.5) 0.664 (1.0) 0.740 (22.5) 0.007
(2.4)
10.7 (2.5)
0.840
(2.1)
8.2 (3.2)
0.694
1.000
35 had complete resolution of symptoms and 3 had amelioration of symptoms without any therapy after ablation. During follow-up in 6 months, no patient developed new symptoms attributed to hemangiomas. Both the subjective health status and the quality of life were rated as good or better in 100% at follow-up. After RF ablation of the hemangioma, all patients were able to work full-time or part-time.
DISCUSSION The major finding of our study was that the use of internally cooled electrodes in the RF ablation treatment of large hepatic hemangiomas was successful and had fewer complications (47.6% vs. 100%) compared with ablation TABLE 4. Complications of Radiofrequency Ablation
Follow-up Results
Maximal size of hemangioma, mean (SD) (cm) Minimum (cm) Maximum (cm) Distribution of lesion, N (%) Right lobe Left lobe Location of lesion, N (%) Subcapsular Deep-sited
Volume 48, Number 6, July 2014
P
showing subtle enhancement on the peripheral rim of the ablated tumors in follow-up CT or MRI. The mean diameter of ablation zone was reduced from 12.4 ± 1.6 to 10.5 ± 2.4 cm in the multitined electrode group, and from 12.5 ± 1.5 to 10.7 ± 2.5 cm in the internally cooled electrode group after 1 month, an insignificant difference between the groups (P > 0.05). At 6 months after ablation, the mean diameter of ablation zone had decreased further in both groups (to 7.9 ± 2.1 cm in the multitined electrode group and to 8.2 ± 3.2 cm in the internally cooled electrode group), also an insignificant difference.
Group
0.607
Internally Cooled Electrode (n = 21)
P
(100)
10 (47.6)
0.000
(100) (47.6)
10 (47.6) 4 (19.0)
0.000 0.100
(42.9)
3 (14.3)
0.085
(28.6)
1 (4.8)
0.093
(28.6)
4 (19.0)
0.719
(4.8)
0 (0.0)
1.000
(9.5) (4.8) (4.8)
0 (0.0) 0 (0.0) 0 (0.0)
0.488 1.000 1.000
(4.0)
6.4 (2.5)
0.000
Multitined Electrode (n = 21)
Group
Total no. patients with 21 complication, N (%) Incidence of complication, N (%) Hemoglobinuria 21 Fever (body temperature 10 Z381C) Hemolytic 9 hyperbilirubinemia (total bilirubin >34.2 mmol/L) Anemia (hemoglobin 6 80 U/L) Elevated creatinine 1 (>200 mmol/L) Skin burns 2 Lower esophageal fistula 1 Acute respiratory 1 distress syndrome Hospital stay, mean 10.2 (SD)
r
2013 Lippincott Williams & Wilkins
J Clin Gastroenterol
Volume 48, Number 6, July 2014
RF Ablation in the Treatment of Hemangiomas
FIGURE 2. A 53-year-old woman in the internally cooled electrode group had an 11.0 cm hemangioma in the right lobe, as illustrated by CT scans (A and B). One month after percutaneous RF ablation in 1 session, CT scans showed that the hemangioma was completely ablated and remarkably smaller (C). After 6 months, the hemangioma was smaller yet, as illustrated in an abdominal CT scan (D). CT indicates computed tomography; RF, radiofrequency.
using multitined electrodes. A similarly high frequency of complete ablation (>80%) of these very challenging tumors was attained with both forms of treatment. Also, the immediate and sustained reduction in size of the tumors was similar in the 2 groups. We also found that the length of hospital stay was significantly shorter for patients treated with the internally cooled electrode approach, most likely because of a lower incidence of complications in them. Several series have reported a morbidity rate of 10% to 27% and a mortality rate of 0% to 3% after resection or enucleation of hepatic hemangiomas, respectively.5–8 The size of hepatic hemangiomas is a significant risk factor for hepatectomy, mainly because of the massive intraoperative blood loss and blood transfusion requirement associated with major hepatic resection.5 Compared with the considerable risk and invasiveness of surgical resection, we consider a complication rate of 47.6% in patients with hemangiomas Z10 cm in the internally cooled electrode group to be acceptable, especially because all the complications were minor (grade I). The lower rate of complications associated with the internally cooled electrodes in the treatment of large hemangiomas is a significant improvement in RF ablation treatment as this form of treatment has been beset by complications. Indeed, some investigators have doubted that RF ablation is an acceptable approach for large hemangiomas because of the long ablation time required. r
2013 Lippincott Williams & Wilkins
Lengthy ablation time unavoidably causes hemolysis, which can lead to the complications of hemoglobinuria, hemolytic jaundice, anemia, or even renal failure. The high incidence of complications associated with ablation is directly proportional to the size of hemangiomas and the longer ablation time.17 We find it reassuring that complete and sustained resolution of large hepatic hemangiomas was attained through use of the internally cooled electrode approach. Although RF ablation has been shown effective and reasonably safe in the treatment of hemangiomas of 5 to 9.9 cm in size,13–19 this form of treatment for larger hemangiomas remains controversial. For example, Park et al17 reported a technical failure rate of 40% with percutaneous RF ablation of hemangiomas Z10 cm in diameter. In our earlier experience with multitined electrodes we also encountered difficulties, including significant systemic inflammatory responses and ARDS, likely resulting from long ablation times.15,28 It is desirable that there be safe and effective noninvasive treatment of hepatic hemangiomas because surgical resection is highly risky, especially for large tumors, such as those that occupy major portion of a lobe or are located deep within the hepatic parenchyma.5,20,21 Although transcatheter arterial embolization and radiation have been used in the treatment of hepatic hemangiomas, they have not been proved uniformly effective.9,10,12 www.jcge.com |
545
Gao et al
J Clin Gastroenterol
Volume 48, Number 6, July 2014
FIGURE 3. A 45-year-old woman in the internally cooled electrode group had a 15.0 cm hemangioma in the right lobe, as illustrated in an abdominal CT scan (A). In laparoscopic views, the tumor is evident on the superior surface (B) and on the posterior and inferior surfaces of the liver (C). During the ablation procedure, hemoglobinuria developed and the ablation was stopped (D). The lesion became a depressed mass with hard texture after the first RF ablation session. The abdomen CT at 1 month showed a partially enhancing area in the hemangioma (E). The hemangioma was completely ablated after the second RF ablation session and 1 month later was smaller, as illustrated in an abdominal CT scan (F).
Currently available RF electrodes vary in their design and mechanism of energy distribution. Multitined expandable electrodes, for example RITA provide even tissue penetration and heat delivery and produce maximal ablated zones of up to 7 cm with a single application in the temperature-controlled ablation mode. In normal liver tissue, use of
546 | www.jcge.com
5, 6, and 7-cm array extensions take about 18.5, 23.3, and 30.4 minutes to achieve a 4.9, 5.4, and 6.1 cm diameter zone of tissue coagulation. In contrast, internally cooled cluster electrodes, for example Cool-tip, are straight electrodes without arrays, and the tip is internally cooled by continuous infusion of cold saline. Thus, the temperature of the electrode r
2013 Lippincott Williams & Wilkins
J Clin Gastroenterol
Volume 48, Number 6, July 2014
is kept low, which helps prevent charring at the tissue-probe interface and reduces tissue impedance. The efficient heat deposition creates a larger ablated zone in shorter time (a 4.5 cm zone in about 12 min in normal liver tissue). Several studies have shown that internally cooled cluster electrodes are more efficient than multitined expandable electrodes in the ablation of solid tumors.22–26 Another advantage of internally cooled electrodes, such as Cool-tip, is that they keep a steady high temperature in the tumor while limiting vascular cooling. This characteristic increases the effectiveness of perivascular ablation.29 Moreover, the design of Cooltip electrodes permits their ready placement into target lesions without risk of accidental injury to adjacent organs; the full length of the electrode is readily visible with CT or ultrasound guidance during the procedure. In contrast, localization of each tip of the multitined expandable electrodes requires verification one by one, with attendant increased effort, time, and expense. We have had no accidental damage of adjacent organs using Cool-tip electrodes. The major limitations of our study include its retrospective nature, the nonrandomized selection of patients, and a relatively small number of patients evaluated. We feel it important, though, that all the RF procedures were performed by 1 surgeon, thus minimizing the chance of bias that might have occurred if multiple surgeons had been involved. Despite possible shortcomings in this study, we feel that our results favoring the use of internally cooled cluster electrodes in RF treatment of large hepatic hemangiomas are persuasive and justify further use of this technology in this challenging disease. However, a controlled, randomized study to verify this conclusion would be advisable. In conclusion, RF ablation is an effective therapeutic modality for treatment of large hepatic hemangiomas. Use of the internally cooled cluster electrodes and a more defensive treatment algorithm can reduce the complications.
REFERENCES 1. Farges O, Daradkeh S, Bismuth H. Cavernous hemangiomas of the liver: are there any indications for resection? World J Surg. 1995;19:19–24. 2. Schnelldorfer T, Ware AL, Smoot R, et al. Management of giant hemangioma of the liver: resection versus observation. J Am Coll Surg. 2010;211:724–730. 3. Yeh WC, Yang PM, Huang GT, et al. Long-term follow-up of hepatic hemangiomas by ultrasonography: with emphasis on the growth rate of the tumor. Hepatogastroenterology. 2007; 54:475–479. 4. Hamaloglu E, Altun H, Ozdemir A, et al. Giant liver hemangioma: therapy by enucleation or liver resection. World J Surg. 2005;29:890–893. 5. Hanazaki K, Kajikawa S, Matsushita A, et al. Giant cavernous hemangioma of the liver: is tumor size a risk factor for hepatectomy? J Hepatobiliary Pancreat Surg. 1999;6:410–413. 6. Lerner SM, Hiatt JR, Salamandra J, et al. Giant cavernous liver hemangiomas: effect of operative approach on outcome. Arch Surg. 2004;139:818–821; discussion 821-813. 7. Terkivatan T, de Wilt JH, de Man RA, et al. Indications and long-term outcome of treatment for benign hepatic tumors: a critical appraisal. Arch Surg. 2001;136:1033–1038. 8. Yoon SS, Charny CK, Fong Y, et al. Diagnosis, management, and outcomes of 115 patients with hepatic hemangioma. J Am Coll Surg. 2003;197:392–402.
r
2013 Lippincott Williams & Wilkins
RF Ablation in the Treatment of Hemangiomas
9. Gaspar L, Mascarenhas F, da Costa MS, et al. Radiation therapy in the unresectable cavernous hemangioma of the liver. Radiother Oncol. 1993;29:45–50. 10. Giavroglou C, Economou H, Ioannidis I. Arterial embolization of giant hepatic hemangiomas. Cardiovasc Intervent Radiol. 2003;26:92–96. 11. Huang XQ, Huang ZQ, Duan WD, et al. Severe biliary complications after hepatic artery embolization. World J Gastroenterol. 2002;8:119–123. 12. Park WC, Rhillips R. The role of radiation therapy in the management of hemangiomas of the liver. JAMA. 1970;212:1496–1498. 13. Cui Y, Zhou LY, Dong MK, et al. Ultrasonography guided percutaneous radiofrequency ablation for hepatic cavernous hemangioma. World J Gastroenterol. 2003;9:2132–2134. 14. Fan RF, Chai FL, He GX, et al. Laparoscopic radiofrequency ablation of hepatic cavernous hemangioma. A preliminary experience with 27 patients. Surg Endosc. 2006;20:281–285. 15. Gao J, Ke S, Ding XM, et al. Radiofrequency ablation for large hepatic hemangiomas: Initial experience and lessons. Surgery. 2012;153:78–85. 16. Hinshaw JL, Laeseke PJ, Weber SM, et al. Multiple-electrode radiofrequency ablation of symptomatic hepatic cavernous hemangioma. Am J Roentgenol. 2007;189:W146–W149. 17. Park SY, Tak WY, Jung MK, et al. Symptomatic-enlarging hepatic hemangiomas are effectively treated by percutaneous ultrasonography-guided radiofrequency ablation. J Hepatol. 2011;54:559–565. 18. Tak WY, Park SY, Jeon SW, et al. Ultrasonography-guided percutaneous radiofrequency ablation for treatment of a huge symptomatic hepatic cavernous hemangioma. J Clin Gastroenterol. 2006;40:167–170. 19. Zagoria RJ, Roth TJ, Levine EA, et al. Radiofrequency ablation of a symptomatic hepatic cavernous hemangioma. Am J Roentgenol. 2004;182:210–212. 20. Hering J, Garrean S, Saied A, et al. Use of radiofrequency hepatic parenchymal transection device in hepatic hemangioma resection: early experience and lessons learned. HPB (Oxford). 2007;9:319–323. 21. Li MH, Yan LN, Wang SR. Autologous transfusion with modified total hepatic vascular exclusion for extracapsular resection of giant hepatic cavernous hemangioma. Hepatobiliary Pancreat Dis Int. 2007;6:43–48. 22. Burdio F, Tobajas P, Quesada-Diez R, et al. Distant infusion of saline may enlarge coagulation volume during radiofrequency ablation of liver tissue using cool-tip electrodes without impairing predictability. Am J Roentgenol. 2011;196:W837–W843. 23. Hung WK, Mak KL, Ying M, et al. Radiofrequency ablation of breast cancer: a comparative study of two needle designs. Breast Cancer. 2011;18:124–128. 24. Manenti G, Bolacchi F, Perretta T, et al. Small breast cancers: in vivo percutaneous US-guided radiofrequency ablation with dedicated cool-tip radiofrequency system. Radiology. 2009;251: 339–346. 25. Martel J, Bueno A, Ortiz E. Percutaneous radiofrequency treatment of osteoid osteoma using cool-tip electrodes. Eur J Radiol. 2005;56:403–408. 26. Wakui N, Iida K, Takayama R, et al. Cool-tip 10-mm electrode useful for radiofrequency ablation of hepatocellular carcinoma. Hepatogastroenterology. 2009;56:1585–1591. 27. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–213. 28. Ng KK, Lam CM, Poon RT, et al. Safety limit of large-volume hepatic radiofrequency ablation in a rat model. Arch Surg. 2006;141:252–258. 29. Haemmerich D, Lee FT Jr, Schutt DJ, et al. Large-volume radiofrequency ablation of ex vivo bovine liver with multiple cooled cluster electrodes. Radiology. 2005;234:563–568.
www.jcge.com |
547
