Surgery is regarded as the “gold standard” of therapy for patients with primary or secondary liver cancer as well as selected patients with benign liver disease. Results of liver resection have improved dramatically during the past decade with the National Surgical Quality Improvement Program’s (NSQUIP) methodology reporting a 30-day morbidity of 22.6% and mortality of 2.6% [1]. Despite these acceptable outcome measures, a number of challenges remain, including the need to minimize blood loss occurring during transection of the hepatic parenchyma, as numerous studies have linked blood loss and the need for transfusion with postoperative outcome and disease recurrence [2–7].

Several techniques have been utilized to reduce blood loss during parenchymal dissection, such as low central venous pressure anesthesia [8], the Pringle maneuver [9], and total vascular occlusion [10], all of which have shown some benefits. A further novel concept is that of precoagulation, which aims to create a zone of coagulative necrosis before transecting the liver. The technique was first described by Weber and colleagues in 2002 who used a monopolar radiofrequency (RF) ablation device in open hepatic resection reporting a significant reduction of blood loss in patients in whom the device was used [11]. The group subsequently developed a bipolar device [12], and in their most recent series reported on 236 patients who underwent precoagulation with monopolar and bipolar RF devices, confirming the safety and efficacy of both devices [13]. Several other groups also have verified the efficacy of the devices for precoagulation in open hepatic resection [14–18].

A variant instrument was developed by Yao et al. [19]; this was validated in a multicenter study that included 108 patients, again showing a reduced blood loss compared with transitional techniques of tissue transection [20]. Reuter and Martin have reported an alternative device utilizing microwave technology and it too would appear to be beneficial for the purpose of reducing blood loss [21]. More recently, Habib and colleagues developed a bipolar laparoscopic RF device with initial results suggesting that it too is effective in reducing blood loss [22, 23], although there are far limited confirmatory studies [24].

The purpose of this study is to describe the technique and to provide an algorithm for the use of monopolar and bipolar RF devices for precoagulation-assisted laparoscopic liver resection.

Materials and methods

Patients who underwent laparoscopic liver resection during the period 2007–2009 were identified from a prospectively maintained, institutional review board-approved database. The database was interrogated to identify those patients in who precoagulation was utilized before transection of the parenchyma using monopolar or bipolar RF devices (Figs. 1, 2, 3). Data collected included: patient demographics (age, gender); tumor size; type of resection performed; type of device used; time taken for precoagulation; total operative time; intraoperative blood loss; need for a blood transfusion; and development of postoperative complications.

Fig. 1
figure 1

Intraoperative photo shows bipolar precoagulation for left lateral sectionectomy. Because the transection line is straight, there was preference to use bipolar Habib device for this type of resection

Full size image
Fig. 2
figure 2

Intraoperative photo showing monopolar precoagulation for a segment 5 tumor requiring circumferential coverage

Full size image
Fig. 3
figure 3

Intraoperative photo showing the parenchymal resection after precoagulation. Blood loss is minimal after 20–30 min of precoagulation

Full size image

RF-assisted precoagulation was performed using a second-generation monopolar RF ablation catheter (Starburst XL, AngioDynamics) or bipolar RF device (Habib 4X, AngioDynamics). The monopolar RF instrument was introduced percutaneously under direct visualization, whereas the bipolar device was inserted via a pre-placed 12-mm port. Vascular control was not utilized for any patients in the study. Patients with malignant tumors were followed up with 3-monthly liver imaging to detect recurrence whilst those with benign disease were seen in the clinic for a routine postoperative visit and then discharged.

Data relating to procedure timings and blood loss were tested for normality and found to be normally distributed and were therefore expressed as mean ± standard deviation (SD). Categorical data were analyzed using Pearson’s chi-squared test and continuous data evaluated using the Student’s t test. All statistical analyses were assessed by using JMP statistical software 5.1.2.

Results

Precoagulation was utilized in 31 patients, consisting of 18 women and 13 men with a mean age of 61.4 ± 12.7 years. The monopolar RF device was used in 19 cases and the bipolar in 12 patients.

Within the monopolar RF group, there were 11 women and 8 men with a mean age of 59.9 ± 14.7 years. There was no difference in gender distribution (p = 0.9) or age (p = 0.5) between the monopolar and bipolar groups. Tumors in the monopolar group consisted of colorectal liver metastases (n = 9), hepatocellular carcinoma (n = 3), cholangiocarcinoma (n = 2), gastrointestinal stromal tumor metastasis (n = 1), neuroendocrine carcinoma metastasis (n = 1), focal nodular hyperplasia (n = 1), adenoma (n = 1), and hemangioma (n = 1). The bipolar RF group consisted of 7 women and 5 men with a mean age of 63.8 ± 8.6 years. Histopathological diagnoses within the bipolar group consisted of colorectal liver metastases (n = 5), hepatocellular carcinoma (n = 4), cholangiocarcinoma (n = 1), breast metastasis (n = 1), and retroperitoneal sarcoma metastasis (n = 1). Tumor histopathologies treated by each modality were comparable. Within the monopolar group, 18 of the resections were segmental/wedge resections and a single left lateral sectionectomy was performed, whereas in the bipolar RF group, there were 3 segmental/wedge resections and 9 left lateral sectionectomies. The bipolar RF device was therefore used for 90% of left lateral sectionectomies in which there was linear transection. The monopolar RF catheter was used mainly for nonlinear segmental and wedge resections mainly involving the right lobe.

Tumor size was slightly larger at 3.4 ± 1.5 cm in the bipolar compared with 2.8 ± 1.5 cm in the monopolar group, but this difference did not reach statistical significance (p = 0.06). There was no significant difference in the time required for precoagulation at 22.5 ± 7.2 min for the bipolar device versus 33.6 ± 15.9 min for the monopolar RF (p = not significant (NS)). Furthermore, there was no difference in the overall operative time between the groups at 242 ± 75 min in the bipolar and 224 ± 79 min in the monopolar group (p = 0.3). The estimated blood loss was 89 ± 77 cc in the bipolar and 224 ± 79 cc in the monopolar group (p = NS) and no perioperative or postoperative transfusions were required. One patient in the bipolar group required conversion for intraoperative bleeding from the liver bed. There were no postoperative complications. There were no episodes of postoperative bleeding, bile duct injury, or liver abscess. No transfusions were required during the perioperative or postoperative period. At follow-up, one (6.2%) patient in the monopolar group had developed disease recurrence consisting of hepatic recurrence within a previously unaffected segment and also extrahepatic disease. This individual had undergone resection of a hepatocellular cancer in segment 5 arising on a background of cirrhosis. All margins of the resected specimen were negative.

Discussion

To our knowledge, this is the first study to evaluate critically the use of RF-energy sources for precoagulation in laparoscopic liver surgery. Our objective was to illustrate an algorithm of precoagulation based on the type of laparoscopic resection planned. We have had favorable results using this algorithm and demonstrated that parenchymal transections performed following precoagulation, with either device, are associated with low levels of blood loss with negligible complications, whilst adding only an additional 20–30 min to the length of the procedure.

Furthermore, our results would suggest that monopolar and bipolar have complementary roles, thus allowing development of an algorithm to choose the appropriate device. Our clinical experience would suggest that monopolar RF device is ideal in the case of small centrally located lesions in which the tumor has to be encircled and allows resection of tumors without having to resect a large volume of nontumorous parenchyma. Because the instrument is small and introduced percutaneously, it may be repositioned without the need for additional ports. The bipolar RF instrument on the other hand is ideal for straight-line transections, such as left lateral sectionectomies, because it can be easily replaced within the parenchyma; however, due to its size and the need for a port it is less maneuverable than monopolar RF. In the current series, 90% of the left lateral sectionectomies were performed using the bipolar device. The bipolar RF was used before five wedge resections on the periphery of the liver, whereas 95% of precoagulation performed using monopolar RF energy were for lesions away from the peripheral margins of the liver and thus required circumferential precoagulation. Monopolar RF precoagulation was of particular benefit for lesions in segment 8 where the angle of approach was limited by the costal margin.

Intraoperative blood loss and the need for perioperative transfusion is well recognized as important in terms of the development of complications and for malignant lesions, and the need also correlates with tumor recurrence and patient survival [2–7]. Evidence from published literature would suggest that precoagulation in both open and laparoscopic surgery is beneficial in reducing blood loss, thus making the transection itself quicker. Due to the design of the study, we cannot comment whether precoagulation decreases intraoperative blood loss compared with the other laparoscopic techniques; however, blood loss has been low, with no patients requiring blood transfusions despite the fact that inflow occlusion was not used in any patient. Overall, estimated blood loss was higher and time spent for precoagulation longer with the monopolar device versus the bipolar device. However, due to our preference to use the two devices in different types of resections, there is no practical value of the statistical comparison.

One important aspect of application of new technologies, such as RF precoagulation, is that they should not compromise oncological principles. When resecting a malignant neoplasm, it is imperative to achieve a disease-free resection margin, which is difficult during laparoscopic liver resection because the loss of tactile sensation. In the current series, no patients had positive margins on their resected specimens. The precoagulation process also ensures that there is a further ablated margin left behind, thus providing improved oncological safety compared with traditional resection techniques. The oncological safety of the approach has been supported by the follow-up CT scans as no recurrence has been seen in relation to the resection lines. Only one patient developed hepatic recurrence, in a previously tumor-free segment, which was associated with tumor dissemination.

Although precoagulation possesses many advantages, it has some adverse effects, too. During manipulation, the area of coagulative necrosis allows little traction, which can break precoagulated liver tissue and cause bleeding to be difficult to control [25]. In patients with malignant tumors, precoagulation in the close vicinity of a tumor might hamper the microscopic evaluation of the specimen. More importantly, it necessitates careful utilization close to the vicinity of biliary tree to prevent postoperative structures. The proper use of intraoperative ultrasonography is invaluable for this purpose.

Conclusions

The results of this study support the use of complementary RF precoagulation in laparoscopic liver resection to achieve low blood loss and local recurrence with minimal morbidity. It highlights that both techniques appear to be valid and safe and that each of them may be superior to other in selected cases depending on the resection type. This study has demonstrated that laparoscopic RF precoagulation can be performed safely and efficiently. Precoagulation adds approximately 30 min to the operative time but leads to a minimal blood loss and is not associated with an increase in perioperative complications. We have suggested a rational algorithm to select a given device based on tumor location and type of resection, with monopolar devices ideal for centrally placed lesions where circumferential coverage is required, and bipolar devices for peripheral wedge resections and segmental/sectional resections with straight transection lines. The precoagulation process also could provide an oncologic benefit due to an additional functional margin obtained with the RF effect.