Introduction

Laparoscopic liver resection is progressively gaining popularity. For minor liver resections, the minimally invasive approach has been shown to be feasible, safe and efficient when appropriate criteria are applied to patient selection.1–12 In contrast, expansion of laparoscopic major hepatectomy (e.g. right or left hepatectomy) is still limited. This is mainly due to: (1) the well-recognised technical difficulties compared to open surgery; (2) doubts regarding the oncological efficiency when major resections are required. Evidence available from a few case series supports the role of laparoscopic major hepatectomy as a safe and efficient procedure when performed in selected patients and in centres with extensive experience in hepatic and laparoscopic surgery.10,13 However, encouraging results from comparative and prospective randomized studies are needed before promoting laparoscopic major hepatectomies on a large scale.

Open right hepatectomy (ORH) is a well-standardized surgical procedure and represents an optimal group to compare with the laparoscopic counterpart for assessing limits and advantages of the minimally invasive approach for major hepatectomy. To date, one comparative study is available in the English literature, analysing short-term results in 22 patients undergoing laparoscopic right hepatectomy (LRH).13 The authors concluded that laparoscopy improves surgical and postsurgical outcomes compared with ORH in selected patients, with similar operative time. More evidence is certainly needed, including assessment of oncological efficiency when the laparoscopic approach is adopted for patient with malignant disease.

The aim of this study was to compare short-term postoperative outcomes of LRH with ORH in a case–control study from a high-volume laparoscopic liver centre. In addition, we discussed the oncological validity of the laparoscopic approach in terms of tumour-free resection margins and mid-term overall survival in patients operated for colorectal carcinoma liver metastasis.

Patients and Methods

We compared patients undergoing LRH with matched patients undergoing ORH between 2006 and 2009 at Southampton University Hospitals NHS Trust. The 4-year period was chosen for this case-controlled study as LRH was started in our institution in 2006.

Patients requiring right hepatectomy in whom the procedure appeared possible via either open or laparoscopic surgery were identified by the hepatobiliary multidisciplinary team, including surgeons, pathologists, oncologists, gastroenterologists and radiologists in our centre.

Exclusion criteria for the comparison were: tumours near the hilum; tumours near the planned resection margin; very large fixed tumours; patients with cirrhosis Child-Pugh category B and C; patients who underwent non-anatomic resection or additional procedures.

Patients in the open-surgery group were matched according to sex, age and liver disease.

Patients undergoing laparoscopic resection were under the care of two surgeons (MAH and NWP). Most of the patients undergoing open resection (85%) were under the care of another surgeon (JNP). The remaining 15% of the open resections were performed by MAH or NWP. These 15% were not converted cases, but planned open resections. The three surgeons were all working in the same hospital and the two laparoscopic surgeons received their basic training in liver surgery at Southampton University Hospitals NHS Trust.

The variables considered for the comparison were: demographics, conversion rate, number of portal triad clamping, intra-operative blood loss (calculated by measuring the volume of blood in the suction bottles, after subtracting wash fluid, at the end of surgery with the addition of weighed swabs), patients requiring transfusion, operation time, rate of benign/malignant lesions, weight of the resected specimen, resection margins, high dependency unit/intensive care unit length of stay, postoperative length of stay, postoperative complications and mortality (within 30 days from surgery).

Complications were classified into specific hepatectomy related (e.g. hepatic failure, bile leak and bleeding) and general complications.

Overall survival analysis was limited to patients with colorectal carcinoma liver metastases. In this group, resections margins were classified into R0 (microscopically more than 1 mm from resection margin) and R1 (microscopically less than 1 mm from resection margin).

When performing LRH, a pure laparoscopic approach was attempted in all patients.14 Standard nomenclature was used to describe the resection performed.15 Our technique for LRH has previously been described in details elsewhere.16

Statistical Analysis

The analyses were performed using the statistical software Strata for Windows (Strata Corporation; College Station, TX, USA). Median values and range were considered for continuous variables as their values’ distribution was skewed. The nonparametric Mann–Whitney U test was used to compare continuous variables. Chi-square or Fisher’s exact test was applied for analysis of categorical variables. When conversion to open or laparoscopic-assisted surgery was required in the LRH group, patients were analysed in the laparoscopic group on an intention-to-treat basis. Survival analysis was limited to patients with colorectal carcinoma liver metastases, excluding patients who died within 30 days from surgery. Overall survival was analysed by the Kaplan–Meier method with log-rank comparison between groups. Survival was calculated from the date of surgery until the date of death or the time of manuscript preparation for those patients known to be alive. The level of statistical significance was set at p < 0.05.

Results

During the study period 95 patients underwent right hepatectomy at Southampton University Hospitals NHS Trust. Laparoscopic resection was attempted in 36 patients (38%). From the 59 patients who underwent ORH, 34 were selected for matching with the laparoscopic group. The remaining 15 cases did not fit the inclusion criteria for comparison with the laparoscopic group. Indication for surgery is summarized in Table 1.

Table 1 Indication for surgery
Full size table

A detailed comparison of demographic variables, surgical results and postoperative course is summarized in Table 2. Operative time was significantly longer for LRH (median 300 vs. 180 min for ORH; p < 0.0001, Mann–Whitney U test). No significant improvement in operative time was observed in the laparoscopic group by comparing the first half of the cases with the second half (p = 0.656, Mann–Whitney U test). Intensive care unit/high dependency unit length of stay (median, 2 days for LRH vs. 4 days for ORH; p < 0.0001; Mann–Whitney U test) and postoperative length of stay (5 days for LRH vs. 9 days for ORH; p < 0.0001; Mann–Whitney U test) were significantly shorter for LRH.

Table 2 Demographics, surgical results and postoperative course
Full size table

A total of four patients in the laparoscopic group required conversion to formal open surgery and four other patients required conversion to a laparoscopic-assisted surgery. This procedure involves a mini-laparotomy in the right upper quadrant to complete the procedure. Most of the conversions (six of eight) happened in the first half of the patients undergoing LRH. The causes for conversion to open procedure were: failure to locate tumour with intra-operative ultrasound in one case, failed hilar dissection in two cases, and difficulty manipulating a large necrotic tumour in one case. The causes of conversion to laparoscopic-assisted technique were: difficult control of bleeding in two cases, bile leak in one case, and closure of diaphragm in one case. Conversion was significantly associated with longer high dependency unit/intensive care unit stay (median, 2.5 for converted cases vs. 2 days for pure laparoscopic cases; p = 0.025; Mann–Whitney U test), and longer postoperative stay (median, 9 for converted cases vs. 4 days for pure laparoscopic cases; p = 0.0003; Mann–Whitney U test).

No significant difference in postoperative complications between the two groups was observed. Complications occurred in five patients undergoing LRH (14%) and in five patients undergoing ORH (15%). Two patients from the open group died in the postoperative period. Details on postoperative complications are shown in Table 3.

Table 3 Description of postoperative complications and outcome
Full size table

Surgical Margins and Mid-term Survival in Patients with Colorectal Carcinoma Liver Metastases

Twenty-one patients in the laparoscopic group and 25 in the open group were considered for surgical margins and survival analysis. R0 resection was obtained in 20/21 (95%) patients after laparoscopic surgery, and in 20/25 (80%) in the open group (p = 0.198, Fisher’s exact test). When considering R0 resections in patients with colorectal carcinoma liver metastases, a median tumour-free resection margin of 20 mm (2–50 mm) was achieved in LRH and 10 mm (2–60 mm) in ORH. To date, we have a median follow-up of 14 months (range, 6–51 months) for the laparoscopic group and median overall survival has not yet been reached. This compares with a median follow-up of 13 months (range, 7–50) and a median overall survival of 27 months in the open group. We recorded a 61% 2-year survival in the open group and 73% in the laparoscopic group (p = 0.283, log-rank test) (Figure 1). There have been 12 deaths during follow-up (eight among ORH and four among LRH).

Fig. 1
figure 1

Mid-term overall survival comparison (Kaplan–Meier method) between laparoscopic right hepatectomy (LRH) and open right hepatectomy (ORH) for colorectal carcinoma liver metastases (p = 0.283; log-rank test)

Full size image

Discussion

Laparoscopic major hepatectomies are technically demanding and require great expertise in open liver surgery and minimally invasive techniques.10,17 Several series have confirmed the advantages of the laparoscopic approach in minor liver resections in terms of less pain and analgesic drug consumption, shorter hospital stay, less transfusion requirements, faster recovery, less postoperative adhesion, reduction of abdominal wall damage and improved cosmetic results compared to open surgery.1,6,9 However, it is unclear whether these benefits are maintained in laparoscopic major hepatectomy. Furthermore, the oncological validity of laparoscopic major liver resection for malignant diseases is still a matter of discussion.18 Encouraging and solid results are needed before advocating this approach on a large scale.

In this observational case–control study, we compared two well-matched groups of patients undergoing open and laparoscopic right hepatectomy. We analysed both short-term results and oncological validity of both approaches showing that LRH can be a safe, effective, and oncologically efficient alternative to ORH in selected cases.

Operation time was longer in the laparoscopic group. This observation is confirmed by most previous series comparing laparoscopic vs. open minor liver resection.19–21 Interestingly, the only available case–control study comparing laparoscopic and open right hepatectomy reports similar operation times in the two groups. The significantly shorter operation time we noticed in the open group remained constant throughout our whole laparoscopic experience. Therefore, it would not be significantly affected by the learning curve so far.

There is universal agreement that conversion should not be viewed as a complication of laparoscopic liver surgery, but as prudent care when continuation of the procedure is no longer safe for the patient.14,17 However, the fact that most of the conversions (six of eight) occurred in the first half of the patients undergoing LRH is the positive result of improved skills, better management of the liver parenchyma, and standardization of laparoscopic instruments and techniques.

Despite all the available preventive measures, bleeding during liver resection was still a frequent occurrence in the laparoscopic group, although no significant difference with the open group was found. Bleeding can obscure views, making surgery difficult, and occasionally causing conversion to open procedure.12 Bleeding during a LRH is initially controlled by application of pressure to the transection surface. This usually stops minor bleeding or oozing and permits the surgeon to gain time while a more definitive method, such as intracorporeal sutures, bipolar diathermy, or an appropriate clip is used for more serious bleeding.12

Our data clearly showed that laparoscopy drastically reduced intensive care and postoperative length of stay in patients requiring right hepatectomy, without any increase in postoperative morbidity and mortality. A median postoperative length of stay of 5 days (with a minimum of 3 days) after a right hepatectomy is a major achievement which contributes to lowering the cost of patients’ hospitalization and favouring early return of patients to family and social life. Predictably, subgroup analysis in the laparoscopic group showed that conversion adversely affected intensive care unit and postoperative length of stay.

Analysis of postoperative complications did not revealed significant differences between the laparoscopic and the open group. The two fatalities reported in our series occurred in the open group and they potentially affected the median length of stay in the open group. Studies comparing laparoscopic with open surgery for minor liver resections have suggested that the frequency of postoperative complications is lower for laparoscopy than for open surgery.7,20,22 This may be the result of the fact that challenging liver lesions in unfit patients are more commonly selected for an open approach. Our observation that operative time was significantly shorter in the open group may, however, reflect fair criteria adopted for group selection in the matching process.

Oncological validity of LRH was confirmed by showing favourably comparable free resection margins to the open group. Furthermore, no peritoneal or wound seeding was observed in this series. We showed that LRH for colorectal carcinoma liver metastases is associated on the mid-term with an overall survival similar to the open group. Our data are comparable with other series including minor and major liver resections, showing that the laparoscopic approach is associated with adequate medium-term survival. O’Rourke at al.23 recorded a 75% 2-year survival rate after laparoscopic resections for colorectal carcinoma metastases, suggesting that the adequacy of resection does not suffer using the laparoscopic approach.

Conclusions

This case-controlled study showed that laparoscopy drastically reduced intensive care and postoperative length of stay after right hepatectomy. In addition, we observed that LRH can be an oncologically efficient alternative to open resection in selected cases treated by experienced laparoscopic liver surgeons. The main limitation of this study is the retrospective design. The ideal setting for the comparison between open and laparoscopic right hepatectomy would be within a large randomized controlled clinical trial, which is still lacking. Meanwhile, large observational studies are needed to provide relevant evidence useful in clinical practice.