Radiofrequency ablation-assisted liver resection: review of the literature and our experience

Abstract

Background: Surgical resection is the best established treatment known to provide long-term survival and possibility of cure for liver malignancy. Intraoperative blood loss has been the major concern during major liver resections, and mortality and morbidity of surgery are clearly associated with the amount of blood loss. Different techniques have been developed to minimize intraoperative blood loss during liver resection. The radiofrequency ablation (RFA) technique has been used widely in the treatment of unresectable liver tumors. This review concentrates on the use of RFA to provide an avascular liver resection plane. Methods and results: The following review is based on two types of RFA device during liver resection: single needle probe RFA and the In-Line RFA device. Conclusion: Liver resection assisted by RFA is safe and is associated with very limited blood loss.

Introduction

Both primary and metastatic cancers of the liver are common and lethal diseases. In Western countries, primary liver cancer is comparatively rare, but is rapidly increasing due to the increased prevalence of hepatitis C, and metastatic colorectal carcinoma is the major indication for liver resection ¹,²,³. Primary liver cancer is, however, much more common in other parts of the world, representing 10–50% of malignancies in Africa and parts of Asia ⁴,⁵,⁶.

Currently, surgical resection is the optimal treatment known to provide long-term survival and possibility of cure for both primary and metastatic liver cancer ⁷,⁸,⁹. Unfortunately, intraoperative blood loss has been a major concern during major liver resections, and mortality and morbidity are clearly associated with the amount of blood loss ¹⁰,¹¹. The mean blood loss has been reported to be between 600 ml and 1300 ml ¹²,¹³, with 28–47% of the patients requiring blood transfusion ¹²,¹⁴. Several studies have shown that blood loss correlates adversely with length of hospital stay, complication rate, and patient survival ¹⁵,¹⁶. It also has been shown that patients requiring more than 10 units of blood after liver resection for colorectal cancer metastasis have an increased risk of tumor recurrence and poor survival, probably due to immunosuppression ¹⁷,¹⁸.

Therefore, over the past decade many techniques have been developed to minimize intraoperative blood loss during liver resection ¹⁹,²⁰,²¹,²². Surgeons can decrease intraoperative blood loss by using hypotensive anesthesia to reduce the central venous pressure ²³, Pringle's maneuver ²⁴, or total vascular exclusion ²⁵. Parenchymal transection can be performed with the finger fracture or Kelly clamp/crush, using the ultrasonic dissector ²⁰, the WaterJet dissector ²⁶, or stapling devices ²⁷.

Radiofrequency ablation (RFA) has been broadly used for unsectable liver tumors and produces coagulative necrosis of the liver parenchyma and thrombosis and coagulation of small blood vessels ²⁸,²⁹,³⁰,³¹,³²,³³. Several recent studies have shown that the use of RFA to assist liver resection decreased blood loss dramatically ³⁴,³⁵,³⁶,³⁷,³⁸,³⁹,⁴⁰,⁴¹,⁴²,⁴³,⁴⁴. This new technique employs the heat produced by the radiofrequency device to coagulate the liver tissue before resecting it. A plane of tissue is coagulated by the RFA device, after which resection is performed along the coagulated tissue plane. By using this technique, blood loss has significantly decreased to a mean of 30 ml ³⁸.

Currently, there are two types of RFA device used in liver resection: single needle probe RFA and In-Line RFA. Both were successfully used in animal and patient experiments to reduce blood loss in liver surgery. This review will attempt to characterize the unique clinical features of RFA-assisted liver resection.

Methods

Single needle probe RFA

RFA-assisted liver resections using a single needle probe have been done in different medical centres (Table I). All patients underwent careful preoperative assessment for their disease, including spiral computed tomography (CT) scanning or magnetic resonance imaging (MRI), and showed no evidence of unresectable extrahepatic disease. Surgical resections ranged from multiple metastasectomies to bisegmentectomies.

Table I. Clinical details of RFA-assisted liver resection.

First author	Medical center, city, country	Year	No.	Age (years)	Types of resection	Operation time (min)	Blood loss	Transfusion	Complication	Hospital stay (days)
Navarra 34	Department of Human Pathology, University of Messina, Messina, Italy	01/2001–01/2002	27		Open hepatectomy	47.5 (30–110)	30 ml (15–992)	Nil	Nil	8 (5–86)
Tepel 36	Klinik fur Allgemeine Chirurgie und Thoraxchirurgie, Universitatsklinikum Schleswig-Holstein, Campus Kiel, Germany		7	74 (41–78)	1 segment (2); 2 segments (3); > 2 segment (2)			1 case	Second degree burn to the thigh (1); bile leak (1)
Navarra 37	Department of Surgical Oncology and Technology, Imperial College School of Medicine, Hammersmith Hospital Campus, London, UK	01/2001–07/2002	42	57.5 (25–79)	1 segment (13); 2 segments (16); > 2 segment (13)	50 (30–110)	30 ml (15–992)	Nil	Subphrenic abscess (1); chest infection (1); biliary leak (1)
Weber 38	Department of Surgical Oncology and Technology, Faculty of Medicine, Imperial College of Science, Technology and Medicine, Hammersmith Hospital Campus, London, UK	01/2000–06/2001	15	59 (36–75)	1 segment (8); 2 segments (6); > 2 segments (1)	205 (95–300)	30±10 ml	Nil	Nil	8 (5–9)
Beppu 39	Department of Gastroenterological Surgery, Graduate School of Medical Sciences, Kumamoto University, Kumamoto, Japan	2004	7	64	Laparoscopic hepatectomy	256	96 g	Nil	Nil	11
Weber 40	Centre de Chirurgie Viscerale et Transplantation, Hopitaux Universitaires de Strasbourg, Avenue Moliere, Strasbourg, France	2002	1	43	Laparoscopic hepatectomy	300	75 ml	Nil	Nil	6
Stella 42	Department of Surgery, Santa Corona Hospital, Pietra Ligure, Savona, Italy	06/2002–11/2002	8	64.8 (55–72)	1 segment (3); 2 segments (3); > 2 segments (2)	220 (170–420)	46 ml (5–150)	Nil	Abscess (1)	9.4 (5–20)
Navarra 43	Department of Surgical Oncology and Technology, Faculty of Medicine, Imperial College of Science, Technology and Medicine, Hammersmith Hospital Campus, London, UK	2002	1		Right hepatectomy	80	30 ml	Nil	Nil	9

Liver resections were carried out under general anesthesia. The peritoneal cavity was examined for evidence of extrahepatic disease. After dividing intra-abdominal adhesions and the falciform ligament, the liver was then mobilized according to the size and site of the lesion to be resected. An intraoperative ultrasonogram was always performed prior to liver resection.

Habib's group has previously described the technique of RFA-assisted liver resection ³⁸. In step 1 (Figure 1), an inner line is made on the liver surface with diathermy to mark the periphery of the tumor, assisted with bimanual palpation and intraoperative ultrasound. It is important to do this first, because after RFA is used, the parenchyma is hardened and it becomes difficult to feel the tumor edge. Also, after RFA, it is impossible to visualize the tumor edge with intraoperative ultrasound; this is a real and important difference which must be understood with all RFA-assisted liver resections.

Figure 1. .

Single needle probe RFA-assisted liver resection. T, tumor; IL, inner line; OL, outer line; RP, resection plane.

In step 2, an outer line, again using diathermy, is made on the liver capsule 2 cm outside (away from) the inner line to mark the site where the probe is going to achieve coagulative necrosis.

In step 3, coagulative necrosis is produced along the outer line using a RFA probe, whilst Habib's group have used the saline-cooled RFA probe; essentially all ablation devices can be used for this purpose.

In step 4, further probe applications are deployed to obtain a zone of necrosis according to the depth of the liver parenchyma to be resected. Each application of RFA energy will need to be applied for about 60 s. The surgeon should check that each probe is correctly positioned by means of ultrasound.

In step 5, the liver parenchyma is transected by using the scalpel. The plane of division should be situated midway between the inner and outer line so as to leave a 1 cm resection margin away from the tumor and leave in situ 1 cm of burned coagulated surface.

All of these groups achieved very good results ³⁴,³⁶,³⁷,³⁸,³⁹,⁴⁰,⁴²,⁴³ (Table I). Habib et al. demonstrated that blood loss could be reduced to <30±10 ml for a right hepatectomy, and practically zero blood loss on most occasions ³⁸. In all of their 108 patients, only one case needed blood transfusion ³⁶. Other complications included: one second degree burn to the thigh ³⁶, two bile leaks ³⁶,³⁷, one subphrenic abscess ³⁷, and one chest infection ³⁷. These complications were from two studies ³⁶,³⁷. Patients in other studies underwent an uneventful recovery; no significant bile leak, abscess, or requirement for transfusion was encountered.

In-Line RFA

The In-Line RFA device was developed by our group ³⁵,⁴¹. The principle of the In-Line RFA device is that a very precise RFA area is created between electrodes of alternating polarity leading to localized heating and coagulation of tissue including blood vessels up to approximately 3 mm. This transection plane is then dissected almost bloodlessly and the larger vessels/biliary structures are tied, sewn or otherwise secured.

The In-Line radiofrequency ablation probe (Figure 2) is a 5 cm long plastic device and it comprises six metal electrodes spaced along the device, each 6 cm long. The electrodes can be deployed to any depth up to a maximum of 4 or 6 cm depending on the model used (Figures 3 and 4). The effective length of the coagulation is 5 cm and its width is 1 cm (Figure 5). The In-Line RFA device is bipolar and does not require a patient return plate.

Figure 2. .

Prototype In-Line RFA probe.

Figure 3. .

Prototype In-Line RFA deployment.

Figure 4. .

Prototype In-Line RFA deployment.

Figure 5. .

Avascular liver resection plane.

Before use, the resection line should be marked. In nonanatomic resection, we transect the liver along this line; in anatomic resection, after dividing the landmark vessels on the cutting surface, we deploy the In-Line RFA along this resection line. We strongly recommend that intraoperative ultrasound is used to ensure that there are no important structures in the chosen resection line. This is particularly important when a vascular biliary sheath crosses the resection plane and must be avoided. The In-Line RFA device will coagulate almost any structure that it encounters and this should not be deployed through or within 1 cm of a structure that must not be damaged.

During coagulation, the device will become very hot and the operator should hold the handle rather than the body of the device. A start out coagulation time of 3 min is used, after which the electrodes are twisted half a turn and pulled back individually. The probe is removed. The electrodes should be redeployed and cleaned if further coagulation is required. The device is re-used until the resection line has been completely treated – typically three applications are required for a right hepatectomy.

The parenchymal dissection of the coagulated transaction line is now completed using either the Kelly crush or a more sophisticated means of transection, such as the ultrasonic dissector or WaterJet. Large vessels and sheaths which are revealed by this dissection may then be ligated, sewn, clipped, or otherwise secured.

In our earlier experiments, we demonstrated that In-Line RFA-assisted liver resection could reduce blood loss dramatically ³⁵,⁴¹. In a pilot study of eight patients where half of each resection was done with In-Line RFA and the remainder with an ultrasonic aspirator alone, we also reported a significant difference in blood loss (p=0.004) ⁴⁴. From March 2002 to November 2004, a total of 38 patients underwent In-Line RFA-assisted liver resection, 12 were anatomic resections, the majority of the remaining resections were bisegmentectomy (Table II). The mean blood loss was 50±12.64 ml, and all of 38 patients had an uneventful recovery. During the postoperative period, no significant increase in bile leaks, abscess, or secondary hemorrhage was found ⁴⁵; however, we accept that this is undoubtedly an issue that requires further monitoring.

Table II. Clinical details of In-Line RFA-assisted liver resection.

No.	Sex	Date of treatment	Resection segments	Blood loss (ml)	Liver quality
1	F	3/12/2002	Left hepatectomy	60	Normal
2	M	8/24/2002	Right hepatectomy	80	Mild steatosis
3	M	8/31/2002	S 6,7	50	Normal
4	M	8/1/2003	S 6,7	50	Severe steatosis
5	M	8/2/2003	S 6	50	Normal
6	F	8/2/2003	S 6,7	0	Normal
7	F	8/6/2003	S 5,6	50	Normal
8	F	8/23/2003	S 3	20	Normal
9	M	10/9/2003	S 8	0	Normal
10	M	10/12/2003	S 6,7	50	Cirrhotic
11	F	10/12/2003	Right hepatectomy	80	Normal
12	F	10/17/2003	S 2,3	0	Normal
13	F	10/17/2003	S 4a	0	Normal
14	F	12/4/2003	S 7,8	60	Mild steatosis
15	M	12/6/2003	S 4,5,8	70	Mild steatosis
16	F	12/10/2003	S 5,6	40	Moderate steatosis
17	M	12/11/2003	S 4b	0	Mild steatosis
18	M	12/12/2003	S 4	0	Mild steatosis
19	F	12/20/2003	Right hepatectomy	90	Normal
20	F	12/21/2003	S 5,6,7	50	Moderate steatosis
21	M	5/11/2004	S 4b, 5,6	50	Normal
22	F	5/12/2004	S 6	0	Cirrhotic
23	F	5/15/2004	Extended left hepatectomy	100	Mild steatosis
24	M	5/15/2004	S 2,3	0	Mild steatosis
25	M	6/19/2004	Left hepatectomy	80	Normal
26	M	6/25/2004	S 5,6,7	60	Normal
27	F	6/26/2004	S 7-subs	40	Normal
28	F	7/9/2004	Right hepatectomy	100	Mild steatosis
29	M	7/24/2004	Left hepatectomy	80	Cirrhotic
30	M	7/28/2004	Right hepatectomy	100	Mild steatosis
31	M	8/31/2004	Right hepatectomy	90	Mild steatosis
32	M	9/10/2004	S 6,7	60	Normal
33	F	10/6/2004	S 2, 3	0	Mild steatosis
34	M	10/7/2004	S 2,3	0	Normal
35	M	10/14/2004	S 7,6	50	Mild steatosis
36	M	10/21/2004	Extended left hepatectomy	150	Mild steatosis
37	F	11/2/2004	Right hepatectomy	100	Normal
38	F	11/7/2004	S 2,3	40	Normal

Conclusion

RFA has been used to treat liver tumors for many years. In 1990 McGahan et al. and Rossi et al. used RFA to destroy hepatic tumors in animal livers ⁴⁶,⁴⁷. Since then, RFA has been used widely in the treatment of unresectable liver tumors ⁴⁸,⁴⁹,⁵⁰,⁵¹. Furthermore, RFA of liver tumors is currently being performed percutaneously and laparoscopically ⁵²,⁵³,⁵⁴,⁵⁵.

The first description of RFA-assisted liver resection was published by Habib's group ³⁸. This technique was clearly a major advance, and no morbidity and mortality were observed. It was shown that anesthetic time, operative time, hospital stay, and amount of blood loss could be decreased. Liver resection became a comparatively safer procedure. Subsequent studies demonstrated the merit of this technique ³⁴,³⁶,³⁷,³⁹,⁴⁰,⁴²,⁴³.

Our In-Line RFA device is bipolar which has many advantages ⁵⁶,⁵⁷. The monopolar system is very slow approach, requiring many ablations to cover the designed resection plane. On the contrary, our In-Line device allows much more rapid (3 min for 6×5×1 cm) and precise treatment. The resection plane then can be dissected almost bloodlessly with an ultrasonic aspirator and only a few larger vessels and biliary ducts need to be ligated. Secondly, the bipolar device does not require grounding pads, because both electrodes are located on the probe and the alternating current circuit is confined within the target tissue.

Before use, the resection line should be marked. We recommend that intraoperative ultrasound is used to ensure that there are no important structures in the chosen resection line. This is particularly important when a vascular biliary sheath crosses the resection plane and must be avoided. The In-Line RFA device will coagulate almost any structure that it encounters and this should not be deployed through or within 1 cm of a structure that must not be damaged.

The In-Line device can similarly be used in segmental liver resection and nonanatomical resections. It is easy to learn to use the device. We believe the ability to sculpt a resection will be a useful tool in parenchymal sparing hepatic surgery, particularly in cirrhotics. The technique of RFA-assisted liver resection has a promising future.