Reply:
We thank Dr. Machi et al for their comments.1 Radiofrequency (RF) ablation has been widely used in the past few years ablating unresectable liver tumors.2 Using the same principle, we have developed and reported a new liver resection technique with radiofrequency by creating a zone of necrosis along a line of intended division of liver parenchyma.1
Unfortunately, the RF ablation system used by Dr. Machi (RITA Medical Systems, Mountain View, CA) has been designed for ablation of unresectable liver tumors only which is totally ineffective for liver resection and is completely different from the system we have described (Radionics). The RITA system is known to generate a spherical zone of coagulation necrosis rather than a cylindrical zone.3 We found this system was both slow to create coagulation in normal liver parenchyma and inadequate for liver resection. Furthermore, application of its probe for liver resection is also difficult and imprecise because of its specially designed umbrella tip for tumor ablation. On the contrary, the Radionics system can create a zone of coagulative necrosis around 1 cm radius and 3 cm in depth. This allows occlusion of blood vessels and bile duct in ablated areas.
In the original description of this technique based on our initial experience, the coagulated area left behind following transection of liver parenchyma was around 2 cm to make sure a sufficient amount of ablated tissues at the resection margin to prevent possible postoperative bleeding. Having performed a large number of resections with this technique, we realized this was in fact not necessary, and we have since extended the resection line to the probe entry points to leave behind less than 1 cm of ablated tissue, which is more than enough to ensure tissues with obliterated vessels and bile duct. In addition, this also gives at least 1 cm safely ablated resection margin for oncologic reasons. When tumor is close to a major vessel (ie, major branches of hepatic vein or hepatic pedicle), RF should be avoided. But when possible, a RF probe cannot usually be applied 2 cm outside tumor but 1 cm close to the tumor to leave 1 cm behind after transection.
With regards to the resection time raised by Dr. Machi, we included the length of time required to complete all 5 steps described in the original article consisting of the time for RF coagulation and parenchymal division. In our experience, ablation of normal liver parenchyma is much quicker than that of tumor or abnormal liver. Intratumoral fibrosis can influence heat diffusion.4 For ablation of solid and firm tumor, it takes considerably longer to achieve coagulative necrosis to destroy viable tumor cells and requires several applications for tumors more than 2 cm in diameter, which are extremely time-consuming. Dodd et al5 reported that, for tumors between 2 to 3 cm, it would require 6 overlapping ablations; and for tumors greater than 3 cm, at least 12 overlapping ablations to achieve a complete destruction of tumor. In case of RF-assisted hepatectomy, ablation is performed away from the tumor to obtain occlusion of blood vessels and bile duct of normal liver parenchyma, which is much quicker. Technically, there is no real need for major overlapping so long as the entry points of each RF application are kept close to each other around 1 cm. To create coagulative necrosis in a different plane, the transaction of the parenchyma is achieved step by step, and the probe is introduced centimeter by centimeter to alter the direction of transection plane.
Since description of RF-assisted liver resection technique,1 we have applied this to all hepatic resections performed in our unit, including cases with more than 5 tumorectomies and tumors up to 125 mm in size. More than 30 major resections of more than 3 segments based on Couinaud's classification have been performed with the RF-assisted technique in the past few years. We shall be shortly submitting the result of RF-assisted liver resection for publication. For major liver resection, the hilus was dissected to ligate arterial and portal branches of the resected part. Hepatic veins were then either coagulated with RF or ligated pending on the site of tumor and resection. The median resection time was 90 minutes, with a median blood loss of less than 200 mL. There has been no postoperative intensive care admission, and hospital stay has also been shortened considerably. Very few cases had postoperative bile leakage or liver failure. We think that RF-assisted liver resection is safe, effective, and reasonably rapid, and the technique may be applicable to other solid and vascular organs such as the spleen and kidney. Based on this technique, a “Habib sealer” has recently been developed and used clinically, which has been shown to be more effective in reduction of ablation and resection time than conventional RF.
Ahmet Ayav, MD
Giuseppe Navarra, MD
Nagy A. Habib, ChM, FRCS
Long R. Jiao, MD, FRCS
Department of Surgical Oncology and Technology
Imperial College School of Medicine
Hammersmith Hospital Campus
London, UK
l.jiao@imperial.ac.uk
REFERENCES
- 1.Weber JC, Navarra G, Jiao LR, et al. New technique for liver resection using heat coagulative necrosis. Ann Surg. 2002;236:560–563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Curley SA. Radiofrequency ablation of malignant liver tumors. Ann Surg Oncol. 2003;10:338–347. [DOI] [PubMed] [Google Scholar]
- 3.de Baere T, Denys A, Wood BJ, et al. Radiofrequency liver ablation: experimental comparative study of water-cooled versus expandable systems. AJR Am J Roentgenol. 2001;176:187–192. [DOI] [PubMed] [Google Scholar]
- 4.Livraghi T, Goldberg SN, Lazzaroni S, et al. Small hepatocellular carcinoma: treatment with radio-frequency ablation versus ethanol injection. Radiology. 1999;210:655–661. [DOI] [PubMed] [Google Scholar]
- 5.Dodd GO, Soulen MC, Kane RA, et al. Minimally invasive treatment of malignant hepatic tumors: at the threshold of a major breakthrough. Radiographies. 2000;20:9–27. [DOI] [PubMed] [Google Scholar]