- Email: [email protected]
Laparoscopic vs. open liver resection for malignant liver disease. A systematic review
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
available at www.sciencedirect.com
The Surgeon, Journal of the Royal Colleges of Surgeons of Edinburgh and Ireland www.thesurgeon.net
Laparoscopic vs. open liver resection for malignant liver disease. A systematic review Ahsan Rao*, Ghaus Rao a, Irfan Ahmed b Department of Surgery, Ward 31, Foresterhill, Aberdeen Royal Infirmary, Aberdeen AB25 2ZA, United Kingdom
article info
abstract
Article history:
Introduction: Since the introduction of minimally invasive techniques, there is little agree-
Received 13 May 2011
ment about use of laparoscopic surgery for malignant liver lesions as compared to open
Accepted 28 June 2011
resection. We aim to analyse all available data comparing both these groups.
Available online 15 September 2011
Methods: All the studies that compared laparoscopic and open liver resections for malignant lesions were searched on various databases. Data were collected and analysed in
Keywords:
Review Manager RevMan (version 5.0).
Laparoscopic
Results: There were total of 10 studies (n ¼ 700) that compared laparoscopic (296/700) and
Open liver resection
open (404/700) hepatic resections for malignant lesions. Laparoscopic group was associated
Hepatocellular carcinoma
with reduced number of patients requiring blood transfusion [Odds ratio 0.35 CI 0.20, 0.60
Hepatic malignancy
P<0.001 HG 0.85], decreased number of positive resection margin [Odds ratio 0.34 CI 0.16,
Meta-analysis
P0.006 HG 0.73] and decrease in overall complication rate [Odds ratio 0.43, CI 0.26, 0.73 P0.002 HG 0.22]. Laparoscopic group was associated with less operative blood loss [WMD 162.6 ml CI 261.79, 73.45 P<0.001] and reduced hospital stay [WMD 4.28 days CI 6.33, 2.23 P<0.001]; however, there was significant heterogeneity [HG <0.001] between the studies for these parameters. Conclusion: The laparoscopic group was associated with reduce overall complication rate, positive resection margins and number of patients requiring blood transfusion. There is still need for level I and II data to compare laparoscopic versus open hepatic resection in malignant lesions. Crown Copyright ª 2011 Royal College of Surgeons of Edinburgh (Scottish charity number SC005317) and Royal College of Surgeons in Ireland. Published by Elsevier Ltd. All rights reserved.
Introduction Liver cancer is the fifth most common cancer in the world1,2 and it is associated with poor prognosis.2,3 If the tumour is not removed completely after surgical resection, the survival rate is usually between 3 and 6 months.1,2 In 1990, it was estimated that there were more than 400,000 new cases of
primary liver cancer worldwide, and a similar number of patients died as a result of this disease.4,5 Hepatic tumours of malignant origin are commonly secondary to metastatic cancer.1 The most common type of primary hepatic cancer is hepatocellular carcinoma.1 The treatment options depend on the stage of liver cancer and the overall condition of the patient.
* Corresponding author. Tel.: þ44 (0) 1224 323224. E-mail addresses: [email protected], [email protected] (A. Rao), [email protected] (G. Rao), [email protected] (I. Ahmed). a Tel.: þ44 (0) 1224 323224. b Tel.: þ44 (0) 1224 555056, Secretary: þ44 (0) 1224 551050; fax: þ44 (0) 1224 551236. 1479-666X/$ e see front matter Crown Copyright ª 2011 Royal College of Surgeons of Edinburgh (Scottish charity number SC005317) and Royal College of Surgeons in Ireland. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.surge.2011.06.007
195
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
First open liver resection was carried out in 1949.6 After the explanation of functional anatomy of liver by Couinaud,7 its progress became worldwide. The description of segmental anatomy immensely assisted in improving surgical techniques. The concept of minimally invasive surgery gradually crept into liver surgery after the first laparoscopic resection of liver was carried out in 1992.8,9 It was initially performed for benign and cystic lesions that were located at easy accessible locations. As the surgical techniques became better defined, the laparoscopic liver resection surgery was carried out for malignant lesions. With on going technological advancements in laparoscopic techniques, more hepatic resections of malignant tumours are being carried out. The evidence of comparison between laparoscopic versus open hepatic resection of malignant tumours is still scarce. Some observational studies have directly compared the periand post-operative outcomes of laparoscopic and open liver resection for hepatic lesions. In the recent years there have been few comparative studies that investigated the shortterm and long-term effects of laparoscopic and open hepatic resections for malignant lesions only. There is no level 1 or II evidence comparing laparoscopic and open liver resection for malignant lesions. In this study, we will analyse all the available data of observational studies that compare laparoscopic and open liver resections for malignant lesions.
All the searched abstracts, studies, and citations were analyzed. All the potential articles were cross-referenced. There were no language restrictions. The latest date for the search was 30th January 2010. Two independent researchers, AR and IA reviewed the selected studies separately.
Inclusion criteria Studies meeting the following criteria were included in the systemic review: (1) Studies that compared peri- and post-operative outcomes in patients undergoing laparoscopic and open hepatic resection (2) Studies reporting at least one of the peri-operative parameters, post-operative outcome measures or pathological measures (3) Studies in which all resected lesions were malignant.
Exclusion criteria Studies were excluded from the analysis if (1) The outcomes of interest were not reported (2) It was impossible to extract or calculate the appropriate data from the published results; (3) The resected lesions were benign
Methods
Outcomes of interest and definitions
All the studies that compared laparoscopic vs. open liver resections for malignant lesions were searched on various databases that included Medline, Ovid, Embase, and Pubmed. The following Mesh search terms were used: “laparoscopy,” “hepatectomy,” “liver resection,” “Open liver resection,” “Hepatic resection,” “laparoscopic liver resection,” “segmentectomy,” “sectionectomy,” “comparative study.” Further, the combinations of these terms were used. The term like “vs” was used to find comparative studies in particular.
The following parameters were identified and reviewed for each study. Basic Demographics: first author, year of publication, total number of patients in laparoscopic and open resection group in each study, study design, matching criteria, inclusion and exclusion criteria, and male to female ratio. Peri-operative parameters: operative time, operative blood loss, number of patients requiring blood transfusion, use of portal triad clamping and duration of portal triad clamping.
Table 1 e Demographics of the studies included. Author
Mala et al. Laurent et al. Kaneko et al. Belli Cai et al. Belli et al. Tranchart et al. Sarpel et al. Endo et al. Castaing et al.
Year
2002 2003 2005 2007 2008 2009 2009 2009 2009 2009
Design
R, P(L) RM(O) P(L), R(O) RM RM RM, P(L), RM (O) RM RM RM
No of patients Lap
Open
13 13 30 23 31 54 42 20 10 60
14 14 28 23 31 125 42 56 11 60
Matchinga
1e6, 9, 11, 12 1, 2, 4e8, 10, 12, 1, 2, 13 1e3, 5e8 1, 2, 5, 6, 8 1e3, 5e8 1e3, 5, 7, 13 1, 2, 5, 8 1e7, 1 1e6, 9, 13
Inclusion criteriab
1 1, 1, 2, e 3, 1, 3, 3, 1,
3, 4 3 4, 6 4e6 4, 5 5 5 2, 5e7
Exclusion criteriac
1, 1, 1, 1, 1 1, 1 1 1 1
3 2, 4 4 5, 6 5, 6
L, Laparoscopic; max, maximum; O, open; n/c, not commented; P, prospective; PM, prospective matched; R, retrospective; RM, retrospective matched. a 1 age; 2 gender; 3 American Society of Anaesthesiologists (ASA) classification; 4 malignancy; 5 mean size of lesion; 6 location of neoplasm; 7 resection type; 8 cirrhosis; 9 liver metastasis; 10 primary malignancy; 11 previous operations; 12 neoplasm histology; 13 Child-Pugh grading. b 1 malignancy; 2 tumour location; 3 primary hepatic cancer; 4 only patients with chronic liver disease; 5 ASA classification; 6 tumour size; 7 resection type. c 1 benign disease; 2 tumour location; 3 primary hepatic cancer; 4 metastatic cancer, 5 ASA classification; 6 Child-Pugh grading.
196
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Table 2 e Characteristics of the studies included. Author
Conversions n (%)
Mala et al. Laurent et al. Kaneko et al. Belli Cai et al. Belli et al. Tranchart et al. Sarpel et al. Endo et al. Castaing et al.
0 2 1 1 1 4 2 4 0 6
(15.4) (3.3) (4.3) (3.2) (7) (4.7) (17) (10)
Mean age (median)
Female (n [%])
Lap
Open
Lap
Open
68 62.6 59 59.5 54.2 63.6 63.7 63.8 72 62
59 65.9 61 62.4 51.7 61.5 65.7 58.3 64 62
4 (31) 3 (23) 12 (40) 10 (43.5) 7 (22.5) 23 (42.5) 15 (35.7) 5 (25) 2 (20) 23 (28.3)
4 (29) 4 (24) 18 (64) 9 (39.1) 5 (16.1) 47 (37.6) 14 (33.3) 11(20) 3 (29) 23 (28.3)
Postoperative parameter: time to first oral Intake, duration of post operative hospital stay and post-operative analgesia requirement. Early post operative adverse events: post operative complications divided into liver resectionerelated (cirrhotic decompensation/ascites, hepatic hemorrhage, biliary leakage, liver failure) and general complications (chest infection, bowel perforation, urinary infection, wound infection, cardiac complications, intra-abdominal abscesses, Clostridium difficile infection, pleural effusion). Oncological clearance: Conversion rate, pathologic resection margin size, positive resection margins, resection margins <1 cm, and resection margins >1 cm. Long-term outcomes: Short-term and long-term cost effectiveness, overall complication rate (includes all liver-specific and general complications), incisional hernia, mortality rate, 5 year and 3 year survival outcome, and recurrence rate.
was used to analyse difference among continuous variables. Odds Ratio (OR) was used to analyse difference in dichotomous variables. 95% confidence interval (CI) was reported for each analysed value. To standardize the data on continuous variables standard deviation (SD) were included. For the studies that provided the range values for each variable, their range values were converted to SD and analysed. Continuous variables without SD were not included in the analysis. Studies with no events in a particular outcome for laparoscopic and open groups were not included. To assess publication bias, graphical Exploration with funnel plots were used.
Results There were total of 10 studies10e19 (n ¼ 700) that compared laparoscopic (296/700) and open (404/700) hepatic resections for malignant lesions.
Statistical analysis Basic demographics Statistical software Review Manager, version 5.0 (The Cochrane Collaboration, Software Update, Oxford, United Kingdom) was used to perform the analysis. Weighted mean difference (WMD)
Tables 1 and 2: The average age for laparoscopic and open liver resection was 62.84 and 61.15 respectively. The proportion of
Table 3 e Pathological characteristics of laparoscopic group. Author
No. of procedures n (%)
Mala et al. Laurent et al. Kaneko et al. Belli et al. Cai et al. Belli et al. Tranchart et al.
15 13 30 23 31 54 42
(52) (48) (52) (50) (50) (30) (50)
Sarpel et al. Endo et al. Castaing et al.
20 (24) 10 (49) 60 (50)
Pathology of lesions resecteda
Mean size (mm)
Procedures performedb
C ¼ 13 E ¼ 13 E ¼ 30 E ¼ 23 E ¼ 24, G ¼ 4, C ¼ 3 E ¼ 54 E ¼ 42
26 33.5 30 31 39.9 36 35.8
E ¼ 20 E ¼ 10 C ¼ 60
43 30 33
S ¼ 6, B ¼ 7, T ¼ 2 W ¼ 3, S ¼ 7, B ¼ 3 LLS ¼ 10, PH ¼ 20 W ¼ 15, S ¼ 3, LLS ¼ 5 PH ¼ 17, S ¼ 8, LLS ¼ 3, LH ¼ 3 W ¼ 21, S ¼ 16, LLS ¼ 14 RHL ¼ 3, LH ¼ 2, LLS ¼ 9, B ¼ 3, S ¼ 15, W ¼ 10 n/c LLS ¼ 10 RHL ¼ 20, S ¼ 5, B ¼ 5, W ¼ 30
L, Laparoscopic; n/c, not commented; O, open. T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented. a C, malignant metastatic; E, hepatocellular carcinoma; G, gallbladder cancer. b W, Wedge resections; S, segmentectomies; B, bisegmentectomies; T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented.
197
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Table 4 e Pathological characteristics of open group. Number of procedures n (%)
Pathology of lesions resecteda
Mean size (mm)
Procedures performedb
Mala et al. Laurent et al. Kaneko et al. Belli et al. Cai et al. Belli et al. Tranchart et al.
14 (48) 14 (52) 28 (48) 23 (50) 31 (50) 125 (70) 42 (50)
C ¼ 14 E ¼ 14 E ¼ 28 E ¼ 23 E ¼ 26, G ¼ 3, C ¼ 2 E ¼ 125 E ¼ 42
30 31 31 32.4 36.2 60 36.8
Sarpel et al. Endo et al. Castaing et al.
5 (76) 11 (51) 60 (50)
E ¼ 56 E ¼ 11 C ¼ 60
43 41 44
S ¼ 9, B ¼ 3, T ¼ 2 W ¼ 4, S ¼ 7, B ¼ 3 LLS ¼ 8, PH ¼ 20 W ¼ 12, S ¼ 5, LLS ¼ 6 PH ¼ 17, S ¼ 8, LLS ¼ 3, LH ¼ 3 T ¼ 39 RHL ¼ 3, LH ¼ 2, LLS ¼ 7, B ¼ 7, S ¼ 13, W ¼ 10 n/c LLS ¼ 11 RHL ¼ 7, S ¼ 10, B ¼ 8 W ¼ 35
Author
L, Laparoscopic; n/c, not commented; O, open. T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented. a C, malignant metastatic; E, hepatocellular carcinoma; G, gallbladder cancer. b W, Wedge resections; S, segmentectomies; B, bisegmentectomies; T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented.
females in the laparoscopic and open groups was 31.15% and 32.04% respectively. Six studies10e14,18 were retrospectively matched and one study17 was retrospectively unmatched. Two studies16,19 included prospective selection of laparoscopic group, which was matched with retrospective data for open group. Twenty one patients from laparoscopic group were converted to open (7%). Tables 3 and 4: Seven studies10,11,14e16,18,19 included only metastatic malignant lesions, 2 studies13,17 only included hepatocellular carcinoma (HCC) lesions and one study12 included a combination of metastatic malignant lesion, HCC and gallbladder cancer.
Peri-operative parameters Eight studies10e12,14e17,19 reported that laparoscopic group was associated with less operative blood loss by 162.6 ml [CI 261.79, 73.45] than open group ( p < 0.001) but there was significant heterogeneity between the studies ( p < 0.001). Six studies10,11,13,16,17,19 reported that laparoscopic group was associated with reduced number of patients requiring blood transfusion by 0.35 [CI 0.20, 0.60] than open group ( p < 0.001) and it was not associated with heterogeneity between studies (p0.85) (Figs. 1 and 2). Five studies10,11,13,16,19 reported that laparoscopic group was associated with less use of portal triad clamping by 0.08
Study or Subgroup Mala et al Laurent et al Belli et al Belli et al. Castaing et al Tranchart et al Total (95% CI)
Laparoscopic Open Events Total Events Total Weight 1 1 0 6 9 4
13 13 23 54 60 42 205
1 4 4 32 22 7
[CI 0.01, 0.46] than open group (p0.005) but it was associated with significant heterogeneity between the studies (p0.01). There was no significant difference between the two groups for operative time (p0.71). Although laparoscopic group was associated with reduced duration of portal triad clamping ( p < 0.001) but it was only reported by one study.16
Post operative parameters Eight studies10e12,14e17,19 reported that laparoscopic group was associated with reduced hospital stay by 4.28 days [CI 6.33, 2.23] than open group ( p < 0.001) but there was significant heterogeneity between the studies ( p < 0.001). Three studies12,14,15 reported that laparoscopic group was associated with reduced time to oral intake by 1.29 days [CI 2.23, 0.35] than open group (p0.007) but there was significant heterogeneity between the studies (p0.007).
Early post-operative adverse outcomes Liver related adverse outcomes:There was no significant difference between the two groups for cirrhotic decompression/ascites (p 0.39), hepatic haemorrhage (p 0.08), biliary leakage (p 0.25), and liver failure (p 0.07). General complications: There was no significant difference between the two groups for chest infection (p 0.56), wound infection (p 0.79) and mortality (p 0.58). Odds Ratio M-H, Random, 95% CI Year
3.5% 5.3% 3.3% 33.2% 37.6% 17.0%
1.08 [0.06, 19.31] 0.21 [0.02, 2.18] 0.09 [0.00, 1.82] 0.36 [0.14, 0.93] 0.30 [0.13, 0.74] 0.53 [0.14, 1.95]
278 100.0%
0.35 [0.20, 0.60]
14 14 23 125 60 42
70 21 Total events Heterogeneity: Tau² = 0.00; Chi² = 2.03, df = 5 (P = 0.85); I² = 0% Test for overall effect: Z = 3.81 (P = 0.0001)
Odds Ratio M-H, Random, 95% CI
2002 2003 2007 2009 2009 2009
0.01 0.1 1 10 Laparoscopic Open
100
Fig. 1 e Laparoscopic vs open resection: Number of patients requiring blood transfusion (n).
198
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
0
SE(log[OR])
0.5
1
1.5
OR
2 0.01
0.1
1
10
100
Fig. 2 e Laparoscopic vs open resection: Funnel plot of number of patients requiring blood transfusion (n).
Oncological clearance: Three studies11,13,18 reported that laparoscopic group was associated with reduced number of positive resection margin by 0.34 [CI 0.16, 0.73] than open group (p0.006) and it was not associated with heterogeneity between studies (p0.75) (Fig. 3). There was no significant difference between the two groups for size of pathological resection margin (p0.84).
Long term outcomes Nine studies10e15,17e19 reported that laparoscopic group was associated with reduced overall complication rate by 0.43 [CI 0.26, 0.73] than open group (p0.002) and it was not associated with heterogeneity between studies (p0.22) (Figs. 4 and 5). There was no significant difference between two groups for recurrence rate (p0.17), 5-year survival (p 0.10 and 0.07) and 3year survival (p 0.17 and 0.26).
Discussion Overall, laparoscopic group has been favoured over open group for a few post-operative parameters in our analysis (Table 5). The laparoscopic group was associated with less number of patients requiring blood transfusion and reduce
Study or Subgroup Castaing et al Belli et al. Sarpel et al Total (95% CI)
Laparoscopic Open Events Total Events Total Weight 8 0 2
60 54 20 134
17 8 15
60 125 56
68.7% 7.3% 24.0%
241 100.0%
10 40 Total events Heterogeneity: Tau² = 0.00; Chi² = 0.58, df = 2 (P = 0.75); I² = 0% Test for overall effect: Z = 2.75 (P = 0.006)
positive resection margins. Similarly, laparoscopic group had decreased overall complication rate than open group. All these results were significant and not associated with heterogeneity between the studies. Although, laparoscopic group was also associated with reduced operative blood loss, use of portal triad clamping, duration of hospital stay and time to oral intake, however, this was linked to significant heterogeneity between the studies. There was no significant difference between the two groups for other peri- and postoperative parameters. This study is the one of the first meta-analysis to be conducted that compared laparoscopic and open hepatic resection for malignant lesions. To our knowledge, there has been no randomised controlled trial or systemic review evaluating this topic before. The study included 10 observational studies with a good participants’ size (n ¼ 700). It investigated most of the peri- and post-operative parameters along with short and long term adverse outcomes. The inclusion and exclusion criteria of majority of the studies were clearly indicated to delineate any discrepancies in the results and its analysis. Most studies had laparoscopic and open groups matched for demographics and characteristics of the patients. It included the patients with cirrhosis and moderate to severe co-morbidity (ASA grade II and above) to take into account common factors associated with
Odds Ratio M-H, Random, 95% CI Year
Odds Ratio M-H, Random, 95% CI
0.39 [0.15, 0.99] 2009 0.13 [0.01, 2.24] 2009 0.30 [0.06, 1.47] 2009 0.34 [0.16, 0.73]
0.01 0.1 1 10 Laparoscopic Open
Fig. 3 e Laparoscopic vs open resection: Number of positive resection margins (n).
100
199
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Study or Subgroup
Laparoscopic Open Total Events Total Weight Events 2 3 5 0 10 1 5 16 3
Mala et al Kaneko et al Belli et al Cai et al Belli et al. Sarpel et al Tranchart et al Castaing et al Endo et al
13 30 23 31 54 20 42 60 10 283
Total (95% CI)
4 5 17 5 45 4 12 17 3
Odds Ratio M-H, Random, 95% CI Year
6.4% 9.2% 11.1% 2.9% 22.8% 4.8% 14.2% 22.1% 6.5%
0.45 [0.07, 3.04] 0.51 [0.11, 2.37] 0.10 [0.03, 0.38] 0.08 [0.00, 1.45] 0.40 [0.19, 0.88] 0.68 [0.07, 6.51] 0.34 [0.11, 1.07] 0.92 [0.41, 2.05] 1.14 [0.17, 7.60]
390 100.0%
0.43 [0.26, 0.73]
14 28 23 31 125 56 42 60 11
Odds Ratio M-H, Random, 95% CI
2002 2005 2007 2008 2009 2009 2009 2009 2009
112 45 Total events Heterogeneity: Tau² = 0.15; Chi² = 10.71, df = 8 (P = 0.22); I² = 25% Test for overall effect: Z = 3.15 (P = 0.002)
0.01 0.1 1 10 Laparoscopic Open
100
Fig. 4 e Laparoscopic vs open resection: Overall complication rate (n).
resection of malignant lesions. Most of the studies included in the analysis were conducted within the last 5 years. It means that most of the operative measures for laparoscopic technique indicate advancement in the technique and takes into account surgical experience with the technology. There was previous meta-analysis conducted in 2006, which included 6 studies.20 It was performed on the studies whose pathological resections were both malignant and benign. The individual studies15,16 have shown reduced operative time for open group as compared to laparoscopic group but the analysis of all the studies showed no difference between the groups. The analyses have taken into account sample size and standard deviation. With the advent of minimally invasive surgery, it was hoped that the blood loss would be reduced. Our compiled data suggested less blood loss with laparoscopic technique but it was associated with heterogeneity between the studies. On the other hand, number of patients requiring blood transfusion was reduced in laparoscopic technique without heterogeneity between the studies.
0
More evidence is required to justify blood loss associated with each technique. Similarly, previous studies indicated reduction in hospital stay and time to oral intake.12,14,15 Our analysis showed similar results, but there was marked heterogeneity between studies when standard deviation for each parameter was taken into consideration. There was no significant difference between two groups for early post-operative adverse outcomes. As there were few studies that reported early adverse outcomes for each group, the results of these parameters are not fully justified. For that reason, when all the complications were taken into account, which increased the sample size, the laparoscopic group was associated with reduced overall complication rate. Because of the limited access through laparoscopic technique, it was assumed that this technique was prone to reduced resection margin size and increased positive resection sampling. However, our analysis showed reduced positive resection associated with laparoscopic group and no
SE(log[OR])
0.5
1
1.5
2 0.01
OR 0.1
1
10
100
Fig. 5 e Laparoscopic vs open resection: Funnel plot of overall complication rate (n).
200
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Table 5 e Overall summary of analysis of outcomes. Outcome of interest Peri-operative parameters Operative time (min) Operative blood loss (mls) No of Patients requiring blood transfusion Use of portal triad clamping Duration of portal triad clamping (min) Post operative parameter Duration of hospital stay (days) Time to first oral intake (days) Early post operative adverse outcomes Liver resection related Cirrhotic decompression/ascites hepatic haemorrhage biliary leakage Liver failure General complications Chest infection Wound infection Mortality Oncological clearance Pathological resection margin size (mm) Positive resection margins Long term outcomes Overall complications 3 year survival rate 3 year survival without recurrence 5 year survival rate 5 year survival without recurrence Recurrence rate
No of studies
No of resections
OR/WMD
95% CI
P value
HG P value
10 8 6 5 1
700 504 91 140 27
3.31 162.62 0.35 0.08 43
[14.05, 20.66] [-251.79, 73.45] [0.20, 0.60] [0.01, 0.46] [26.59, 59.41]
0.71 <0.001 <0.001 0.005 <0.001
<0.001 <0.001 0.85 0.01 Not estimated
8 3
504 141
4.28 1.29
[6.33, 2.33] 2.23, 0.35]
<0.001 0.007
<0.001 0.007
5 3 4 3
43 4 8 8
0.46 5.31 0.43 0.22
[0.08, [0.84, [0.10, [0.04,
5 3 5
13 4 13
0.7 0.74 0.71
4 3
152 50
9 3 3 6 6 4
157 166 89 215 112 202
difference in the size of margin size between the two groups. The possible reason being the camera used with laparoscopy, which magnifies area of resection and provides detailed visibility. In the long-term post-operative outcome, however, there was no difference between the recurrence and 5 and 3 year survival rate between the two groups. Since the analysis was carried out on the observational studies, there was an inherent selection bias as the analysis was unable to assess the differences between open and laparoscopic groups within each study. For example, the rate of major hepatectomies was higher in open group than laparoscopic group in the study conducted by Belli et al.11 Likewise, the open group included resection of larger tumours in the study by Castaing et al.13 We aimed to analyse all the parameters as mentioned in the methodology but there were not enough studies to document all the parameters, for example, cost effectiveness and post op analgesia. There is still a need for level I and II data to compare laparoscopic versus open hepatic resection in malignant lesions. With more data available, the results will be reliable, promising and significant with reduced heterogeneity between the studies. Although we feel that it will be difficult to plan RCTs and recruit patients as minimally invasive techniques are now well established in most institutions that will make it difficult for researchers to recruit patients in the open surgery arm.
Conflicts of interest None declared.
2.73] 33.55] 1.82] 1.11]
0.39 0.08 0.25 0.07
0.004 0.81 0.88 0.91
[0.21, 2.32] [0.08, 6.87] [0.21, 2.41]
0.56 0.79 0.58
0.7 0.24 0.77
0.13 0.34
[1.13, 1.40] [0.16, 0.73]
0.84 0.006
1 0.75
0.43 3.21 2.38 2.33 1.97 0.75
[0.26, [0.61, [0.52, [0.85, [0.93, [0.50,
0.002 0.17 0.26 0.1 0.07 0.17
0.22 0.005 0.007 <0.001 0.04 0.75
0.73] 16.90] 10.91] 6.42] 4.16] 1.13]
references
1. Bosch FX, Ribes J. Epidemiology of liver cancer in Europe. Can J Gastroenterol 2000 Jul-Aug;14(7):621e30. 2. Srivatanakul P, Sriplung H, Deerasamee S. Epidemiology of liver cancer: an overview. Asian Pac J Cancer Prev Apjcp 2004 Apr-Jun;5(2):118e25. 3. Llovet JM, Wurmbach E. Gene expression profiles in hepatocellular carcinoma: not yet there. J. Hepatol 2004 Aug; 41(2):336e9. 4. Bosch FX, Ribes J, Borra`s J. Epidemiology of primary liver cancer. Semin Liver Dis 1999;19(3):271e85. 5. Bosch FX, Ribes J, Cle´ries R, Dı´az M. Epidemiology of hepatocellular carcinoma. Clin Liver Dis 2005 May;9(2): 191e211. 6. Tsuchiya R. A tribute to Dr. Ischio Honjo. Surg Today 1989; 19(1):1. 7. Couinaud C. Contribution of anatomical research to liver surgery. France Medecine 1956 May;19(5):5e12. 8. Gagner M, Rogula T, Selzer D. Laparoscopic liver resection: benefits and controversies. Surg.Clin North Am 2004 Apr;84(2): 451e62. 9. Topal B, Fieuws S, Aerts R, Vandeweyer H, Penninckx F. Laparoscopic versus open liver resection of hepatic neoplasms: comparative analysis of short-term results. Surg. Endosc 2008 Oct;22(10):2208e13. 10. Belli G, Fantini C, D’Agostino A, Cioffi L, Langella S, Russolillo N, et al. Laparoscopic versus open liver resection for hepatocellular carcinoma in patients with histologically proven cirrhosis: short- and middle-term results. Surg Endosc 2007;21(11):2004e11. 11. Belli G, Limongelli P, Fantini C, D’Agostino A, Cioffi L, Belli A, et al. Laparoscopic and open treatment of hepatocellular
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
12.
13.
14.
15.
16.
carcinoma in patients with cirrhosis. Br.J.Surg 2009 Sep;96(9): 1041e8. Cai XJ, Yang J, Yu H, Liang X, Wang YF, Zhu ZY, et al. Clinical study of laparoscopic versus open hepatectomy for malignant liver tumors. Surg.Endosc 2008 Nov;22(11):2350e6. Castaing D, Vibert E, Ricca L, Azoulay D, Adam R, Gayet B. Oncologic results of laparoscopic versus open hepatectomy for colorectal liver metastases in two specialized centers. Ann.Surg 2009 Nov;250(5):849e55. Endo Y, Ohta M, Sasaki A, Kai S, Eguchi H, Iwaki K, et al. A comparative study of the long-term outcomes after laparoscopy-assisted and open left lateral hepatectomy for hepatocellular carcinoma. Surg Laparosc Endosc Percutan Tech 2009;19(5):171e4. Kaneko H, Takagi S, Otsuka Y, Tsuchiya M, Tamura A, Katagiri T, et al. Laparoscopic liver resection of hepatocellular carcinoma. Am.J.Surg 2005 Feb;189(2):190e4. Laurent A, Cherqui D, Lesurtel M, Brunetti F, Tayar C, Fagniez PL. Laparoscopic liver resection for subcapsular
17.
18.
19.
20.
201
hepatocellular carcinoma complicating chronic liver disease. Arch Surg 2003;138(7):763e9. discussion 769; Jul. Mala T, Edwin B, Gladhaug I, Fosse E, Soreide O, Bergan A, et al. A comparative study of the short-term outcome following open and laparoscopic liver resection of colorectal metastases. Surg.Endosc 2002 Jul;16(7):1059e63. Sarpel U, Hefti MM, Wisnievsky JP, Roayaie S, Schwartz ME, Labow DM. Outcome for patients treated with laparoscopic versus open resection of hepatocellular carcinoma: case-matched analysis. Ann Surg Oncol 2009 Jun;16(6):1572e7. Tranchart H, Di Giuro G, Lainas P, Roudie J, Agostini H, Franco D, et al. Laparoscopic resection for hepatocellular carcinoma: a matched-pair comparative study. J Surg Endosc; 2009. doi:10.1007/s00464-009-0745-3 (epub). Simillis C, Constantinides VA, Tekkis PP, Darzi A, Lovegrove R, Jiao L, et al. Laparoscopic versus open hepatic resections for benign and malignant neoplasmsea metaanalysis. Surgery 2007 Feb;141(2):203e11.
available at www.sciencedirect.com
The Surgeon, Journal of the Royal Colleges of Surgeons of Edinburgh and Ireland www.thesurgeon.net
Laparoscopic vs. open liver resection for malignant liver disease. A systematic review Ahsan Rao*, Ghaus Rao a, Irfan Ahmed b Department of Surgery, Ward 31, Foresterhill, Aberdeen Royal Infirmary, Aberdeen AB25 2ZA, United Kingdom
article info
abstract
Article history:
Introduction: Since the introduction of minimally invasive techniques, there is little agree-
Received 13 May 2011
ment about use of laparoscopic surgery for malignant liver lesions as compared to open
Accepted 28 June 2011
resection. We aim to analyse all available data comparing both these groups.
Available online 15 September 2011
Methods: All the studies that compared laparoscopic and open liver resections for malignant lesions were searched on various databases. Data were collected and analysed in
Keywords:
Review Manager RevMan (version 5.0).
Laparoscopic
Results: There were total of 10 studies (n ¼ 700) that compared laparoscopic (296/700) and
Open liver resection
open (404/700) hepatic resections for malignant lesions. Laparoscopic group was associated
Hepatocellular carcinoma
with reduced number of patients requiring blood transfusion [Odds ratio 0.35 CI 0.20, 0.60
Hepatic malignancy
P<0.001 HG 0.85], decreased number of positive resection margin [Odds ratio 0.34 CI 0.16,
Meta-analysis
P0.006 HG 0.73] and decrease in overall complication rate [Odds ratio 0.43, CI 0.26, 0.73 P0.002 HG 0.22]. Laparoscopic group was associated with less operative blood loss [WMD 162.6 ml CI 261.79, 73.45 P<0.001] and reduced hospital stay [WMD 4.28 days CI 6.33, 2.23 P<0.001]; however, there was significant heterogeneity [HG <0.001] between the studies for these parameters. Conclusion: The laparoscopic group was associated with reduce overall complication rate, positive resection margins and number of patients requiring blood transfusion. There is still need for level I and II data to compare laparoscopic versus open hepatic resection in malignant lesions. Crown Copyright ª 2011 Royal College of Surgeons of Edinburgh (Scottish charity number SC005317) and Royal College of Surgeons in Ireland. Published by Elsevier Ltd. All rights reserved.
Introduction Liver cancer is the fifth most common cancer in the world1,2 and it is associated with poor prognosis.2,3 If the tumour is not removed completely after surgical resection, the survival rate is usually between 3 and 6 months.1,2 In 1990, it was estimated that there were more than 400,000 new cases of
primary liver cancer worldwide, and a similar number of patients died as a result of this disease.4,5 Hepatic tumours of malignant origin are commonly secondary to metastatic cancer.1 The most common type of primary hepatic cancer is hepatocellular carcinoma.1 The treatment options depend on the stage of liver cancer and the overall condition of the patient.
* Corresponding author. Tel.: þ44 (0) 1224 323224. E-mail addresses: [email protected], [email protected] (A. Rao), [email protected] (G. Rao), [email protected] (I. Ahmed). a Tel.: þ44 (0) 1224 323224. b Tel.: þ44 (0) 1224 555056, Secretary: þ44 (0) 1224 551050; fax: þ44 (0) 1224 551236. 1479-666X/$ e see front matter Crown Copyright ª 2011 Royal College of Surgeons of Edinburgh (Scottish charity number SC005317) and Royal College of Surgeons in Ireland. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.surge.2011.06.007
195
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
First open liver resection was carried out in 1949.6 After the explanation of functional anatomy of liver by Couinaud,7 its progress became worldwide. The description of segmental anatomy immensely assisted in improving surgical techniques. The concept of minimally invasive surgery gradually crept into liver surgery after the first laparoscopic resection of liver was carried out in 1992.8,9 It was initially performed for benign and cystic lesions that were located at easy accessible locations. As the surgical techniques became better defined, the laparoscopic liver resection surgery was carried out for malignant lesions. With on going technological advancements in laparoscopic techniques, more hepatic resections of malignant tumours are being carried out. The evidence of comparison between laparoscopic versus open hepatic resection of malignant tumours is still scarce. Some observational studies have directly compared the periand post-operative outcomes of laparoscopic and open liver resection for hepatic lesions. In the recent years there have been few comparative studies that investigated the shortterm and long-term effects of laparoscopic and open hepatic resections for malignant lesions only. There is no level 1 or II evidence comparing laparoscopic and open liver resection for malignant lesions. In this study, we will analyse all the available data of observational studies that compare laparoscopic and open liver resections for malignant lesions.
All the searched abstracts, studies, and citations were analyzed. All the potential articles were cross-referenced. There were no language restrictions. The latest date for the search was 30th January 2010. Two independent researchers, AR and IA reviewed the selected studies separately.
Inclusion criteria Studies meeting the following criteria were included in the systemic review: (1) Studies that compared peri- and post-operative outcomes in patients undergoing laparoscopic and open hepatic resection (2) Studies reporting at least one of the peri-operative parameters, post-operative outcome measures or pathological measures (3) Studies in which all resected lesions were malignant.
Exclusion criteria Studies were excluded from the analysis if (1) The outcomes of interest were not reported (2) It was impossible to extract or calculate the appropriate data from the published results; (3) The resected lesions were benign
Methods
Outcomes of interest and definitions
All the studies that compared laparoscopic vs. open liver resections for malignant lesions were searched on various databases that included Medline, Ovid, Embase, and Pubmed. The following Mesh search terms were used: “laparoscopy,” “hepatectomy,” “liver resection,” “Open liver resection,” “Hepatic resection,” “laparoscopic liver resection,” “segmentectomy,” “sectionectomy,” “comparative study.” Further, the combinations of these terms were used. The term like “vs” was used to find comparative studies in particular.
The following parameters were identified and reviewed for each study. Basic Demographics: first author, year of publication, total number of patients in laparoscopic and open resection group in each study, study design, matching criteria, inclusion and exclusion criteria, and male to female ratio. Peri-operative parameters: operative time, operative blood loss, number of patients requiring blood transfusion, use of portal triad clamping and duration of portal triad clamping.
Table 1 e Demographics of the studies included. Author
Mala et al. Laurent et al. Kaneko et al. Belli Cai et al. Belli et al. Tranchart et al. Sarpel et al. Endo et al. Castaing et al.
Year
2002 2003 2005 2007 2008 2009 2009 2009 2009 2009
Design
R, P(L) RM(O) P(L), R(O) RM RM RM, P(L), RM (O) RM RM RM
No of patients Lap
Open
13 13 30 23 31 54 42 20 10 60
14 14 28 23 31 125 42 56 11 60
Matchinga
1e6, 9, 11, 12 1, 2, 4e8, 10, 12, 1, 2, 13 1e3, 5e8 1, 2, 5, 6, 8 1e3, 5e8 1e3, 5, 7, 13 1, 2, 5, 8 1e7, 1 1e6, 9, 13
Inclusion criteriab
1 1, 1, 2, e 3, 1, 3, 3, 1,
3, 4 3 4, 6 4e6 4, 5 5 5 2, 5e7
Exclusion criteriac
1, 1, 1, 1, 1 1, 1 1 1 1
3 2, 4 4 5, 6 5, 6
L, Laparoscopic; max, maximum; O, open; n/c, not commented; P, prospective; PM, prospective matched; R, retrospective; RM, retrospective matched. a 1 age; 2 gender; 3 American Society of Anaesthesiologists (ASA) classification; 4 malignancy; 5 mean size of lesion; 6 location of neoplasm; 7 resection type; 8 cirrhosis; 9 liver metastasis; 10 primary malignancy; 11 previous operations; 12 neoplasm histology; 13 Child-Pugh grading. b 1 malignancy; 2 tumour location; 3 primary hepatic cancer; 4 only patients with chronic liver disease; 5 ASA classification; 6 tumour size; 7 resection type. c 1 benign disease; 2 tumour location; 3 primary hepatic cancer; 4 metastatic cancer, 5 ASA classification; 6 Child-Pugh grading.
196
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Table 2 e Characteristics of the studies included. Author
Conversions n (%)
Mala et al. Laurent et al. Kaneko et al. Belli Cai et al. Belli et al. Tranchart et al. Sarpel et al. Endo et al. Castaing et al.
0 2 1 1 1 4 2 4 0 6
(15.4) (3.3) (4.3) (3.2) (7) (4.7) (17) (10)
Mean age (median)
Female (n [%])
Lap
Open
Lap
Open
68 62.6 59 59.5 54.2 63.6 63.7 63.8 72 62
59 65.9 61 62.4 51.7 61.5 65.7 58.3 64 62
4 (31) 3 (23) 12 (40) 10 (43.5) 7 (22.5) 23 (42.5) 15 (35.7) 5 (25) 2 (20) 23 (28.3)
4 (29) 4 (24) 18 (64) 9 (39.1) 5 (16.1) 47 (37.6) 14 (33.3) 11(20) 3 (29) 23 (28.3)
Postoperative parameter: time to first oral Intake, duration of post operative hospital stay and post-operative analgesia requirement. Early post operative adverse events: post operative complications divided into liver resectionerelated (cirrhotic decompensation/ascites, hepatic hemorrhage, biliary leakage, liver failure) and general complications (chest infection, bowel perforation, urinary infection, wound infection, cardiac complications, intra-abdominal abscesses, Clostridium difficile infection, pleural effusion). Oncological clearance: Conversion rate, pathologic resection margin size, positive resection margins, resection margins <1 cm, and resection margins >1 cm. Long-term outcomes: Short-term and long-term cost effectiveness, overall complication rate (includes all liver-specific and general complications), incisional hernia, mortality rate, 5 year and 3 year survival outcome, and recurrence rate.
was used to analyse difference among continuous variables. Odds Ratio (OR) was used to analyse difference in dichotomous variables. 95% confidence interval (CI) was reported for each analysed value. To standardize the data on continuous variables standard deviation (SD) were included. For the studies that provided the range values for each variable, their range values were converted to SD and analysed. Continuous variables without SD were not included in the analysis. Studies with no events in a particular outcome for laparoscopic and open groups were not included. To assess publication bias, graphical Exploration with funnel plots were used.
Results There were total of 10 studies10e19 (n ¼ 700) that compared laparoscopic (296/700) and open (404/700) hepatic resections for malignant lesions.
Statistical analysis Basic demographics Statistical software Review Manager, version 5.0 (The Cochrane Collaboration, Software Update, Oxford, United Kingdom) was used to perform the analysis. Weighted mean difference (WMD)
Tables 1 and 2: The average age for laparoscopic and open liver resection was 62.84 and 61.15 respectively. The proportion of
Table 3 e Pathological characteristics of laparoscopic group. Author
No. of procedures n (%)
Mala et al. Laurent et al. Kaneko et al. Belli et al. Cai et al. Belli et al. Tranchart et al.
15 13 30 23 31 54 42
(52) (48) (52) (50) (50) (30) (50)
Sarpel et al. Endo et al. Castaing et al.
20 (24) 10 (49) 60 (50)
Pathology of lesions resecteda
Mean size (mm)
Procedures performedb
C ¼ 13 E ¼ 13 E ¼ 30 E ¼ 23 E ¼ 24, G ¼ 4, C ¼ 3 E ¼ 54 E ¼ 42
26 33.5 30 31 39.9 36 35.8
E ¼ 20 E ¼ 10 C ¼ 60
43 30 33
S ¼ 6, B ¼ 7, T ¼ 2 W ¼ 3, S ¼ 7, B ¼ 3 LLS ¼ 10, PH ¼ 20 W ¼ 15, S ¼ 3, LLS ¼ 5 PH ¼ 17, S ¼ 8, LLS ¼ 3, LH ¼ 3 W ¼ 21, S ¼ 16, LLS ¼ 14 RHL ¼ 3, LH ¼ 2, LLS ¼ 9, B ¼ 3, S ¼ 15, W ¼ 10 n/c LLS ¼ 10 RHL ¼ 20, S ¼ 5, B ¼ 5, W ¼ 30
L, Laparoscopic; n/c, not commented; O, open. T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented. a C, malignant metastatic; E, hepatocellular carcinoma; G, gallbladder cancer. b W, Wedge resections; S, segmentectomies; B, bisegmentectomies; T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented.
197
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Table 4 e Pathological characteristics of open group. Number of procedures n (%)
Pathology of lesions resecteda
Mean size (mm)
Procedures performedb
Mala et al. Laurent et al. Kaneko et al. Belli et al. Cai et al. Belli et al. Tranchart et al.
14 (48) 14 (52) 28 (48) 23 (50) 31 (50) 125 (70) 42 (50)
C ¼ 14 E ¼ 14 E ¼ 28 E ¼ 23 E ¼ 26, G ¼ 3, C ¼ 2 E ¼ 125 E ¼ 42
30 31 31 32.4 36.2 60 36.8
Sarpel et al. Endo et al. Castaing et al.
5 (76) 11 (51) 60 (50)
E ¼ 56 E ¼ 11 C ¼ 60
43 41 44
S ¼ 9, B ¼ 3, T ¼ 2 W ¼ 4, S ¼ 7, B ¼ 3 LLS ¼ 8, PH ¼ 20 W ¼ 12, S ¼ 5, LLS ¼ 6 PH ¼ 17, S ¼ 8, LLS ¼ 3, LH ¼ 3 T ¼ 39 RHL ¼ 3, LH ¼ 2, LLS ¼ 7, B ¼ 7, S ¼ 13, W ¼ 10 n/c LLS ¼ 11 RHL ¼ 7, S ¼ 10, B ¼ 8 W ¼ 35
Author
L, Laparoscopic; n/c, not commented; O, open. T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented. a C, malignant metastatic; E, hepatocellular carcinoma; G, gallbladder cancer. b W, Wedge resections; S, segmentectomies; B, bisegmentectomies; T, three segments resection; LLS, left lateral segmentectomy; RHL, right hepatic lobectomy; LH, Left hemihepatectomy; PH, partial hepatectomy; n/c not commented.
females in the laparoscopic and open groups was 31.15% and 32.04% respectively. Six studies10e14,18 were retrospectively matched and one study17 was retrospectively unmatched. Two studies16,19 included prospective selection of laparoscopic group, which was matched with retrospective data for open group. Twenty one patients from laparoscopic group were converted to open (7%). Tables 3 and 4: Seven studies10,11,14e16,18,19 included only metastatic malignant lesions, 2 studies13,17 only included hepatocellular carcinoma (HCC) lesions and one study12 included a combination of metastatic malignant lesion, HCC and gallbladder cancer.
Peri-operative parameters Eight studies10e12,14e17,19 reported that laparoscopic group was associated with less operative blood loss by 162.6 ml [CI 261.79, 73.45] than open group ( p < 0.001) but there was significant heterogeneity between the studies ( p < 0.001). Six studies10,11,13,16,17,19 reported that laparoscopic group was associated with reduced number of patients requiring blood transfusion by 0.35 [CI 0.20, 0.60] than open group ( p < 0.001) and it was not associated with heterogeneity between studies (p0.85) (Figs. 1 and 2). Five studies10,11,13,16,19 reported that laparoscopic group was associated with less use of portal triad clamping by 0.08
Study or Subgroup Mala et al Laurent et al Belli et al Belli et al. Castaing et al Tranchart et al Total (95% CI)
Laparoscopic Open Events Total Events Total Weight 1 1 0 6 9 4
13 13 23 54 60 42 205
1 4 4 32 22 7
[CI 0.01, 0.46] than open group (p0.005) but it was associated with significant heterogeneity between the studies (p0.01). There was no significant difference between the two groups for operative time (p0.71). Although laparoscopic group was associated with reduced duration of portal triad clamping ( p < 0.001) but it was only reported by one study.16
Post operative parameters Eight studies10e12,14e17,19 reported that laparoscopic group was associated with reduced hospital stay by 4.28 days [CI 6.33, 2.23] than open group ( p < 0.001) but there was significant heterogeneity between the studies ( p < 0.001). Three studies12,14,15 reported that laparoscopic group was associated with reduced time to oral intake by 1.29 days [CI 2.23, 0.35] than open group (p0.007) but there was significant heterogeneity between the studies (p0.007).
Early post-operative adverse outcomes Liver related adverse outcomes:There was no significant difference between the two groups for cirrhotic decompression/ascites (p 0.39), hepatic haemorrhage (p 0.08), biliary leakage (p 0.25), and liver failure (p 0.07). General complications: There was no significant difference between the two groups for chest infection (p 0.56), wound infection (p 0.79) and mortality (p 0.58). Odds Ratio M-H, Random, 95% CI Year
3.5% 5.3% 3.3% 33.2% 37.6% 17.0%
1.08 [0.06, 19.31] 0.21 [0.02, 2.18] 0.09 [0.00, 1.82] 0.36 [0.14, 0.93] 0.30 [0.13, 0.74] 0.53 [0.14, 1.95]
278 100.0%
0.35 [0.20, 0.60]
14 14 23 125 60 42
70 21 Total events Heterogeneity: Tau² = 0.00; Chi² = 2.03, df = 5 (P = 0.85); I² = 0% Test for overall effect: Z = 3.81 (P = 0.0001)
Odds Ratio M-H, Random, 95% CI
2002 2003 2007 2009 2009 2009
0.01 0.1 1 10 Laparoscopic Open
100
Fig. 1 e Laparoscopic vs open resection: Number of patients requiring blood transfusion (n).
198
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
0
SE(log[OR])
0.5
1
1.5
OR
2 0.01
0.1
1
10
100
Fig. 2 e Laparoscopic vs open resection: Funnel plot of number of patients requiring blood transfusion (n).
Oncological clearance: Three studies11,13,18 reported that laparoscopic group was associated with reduced number of positive resection margin by 0.34 [CI 0.16, 0.73] than open group (p0.006) and it was not associated with heterogeneity between studies (p0.75) (Fig. 3). There was no significant difference between the two groups for size of pathological resection margin (p0.84).
Long term outcomes Nine studies10e15,17e19 reported that laparoscopic group was associated with reduced overall complication rate by 0.43 [CI 0.26, 0.73] than open group (p0.002) and it was not associated with heterogeneity between studies (p0.22) (Figs. 4 and 5). There was no significant difference between two groups for recurrence rate (p0.17), 5-year survival (p 0.10 and 0.07) and 3year survival (p 0.17 and 0.26).
Discussion Overall, laparoscopic group has been favoured over open group for a few post-operative parameters in our analysis (Table 5). The laparoscopic group was associated with less number of patients requiring blood transfusion and reduce
Study or Subgroup Castaing et al Belli et al. Sarpel et al Total (95% CI)
Laparoscopic Open Events Total Events Total Weight 8 0 2
60 54 20 134
17 8 15
60 125 56
68.7% 7.3% 24.0%
241 100.0%
10 40 Total events Heterogeneity: Tau² = 0.00; Chi² = 0.58, df = 2 (P = 0.75); I² = 0% Test for overall effect: Z = 2.75 (P = 0.006)
positive resection margins. Similarly, laparoscopic group had decreased overall complication rate than open group. All these results were significant and not associated with heterogeneity between the studies. Although, laparoscopic group was also associated with reduced operative blood loss, use of portal triad clamping, duration of hospital stay and time to oral intake, however, this was linked to significant heterogeneity between the studies. There was no significant difference between the two groups for other peri- and postoperative parameters. This study is the one of the first meta-analysis to be conducted that compared laparoscopic and open hepatic resection for malignant lesions. To our knowledge, there has been no randomised controlled trial or systemic review evaluating this topic before. The study included 10 observational studies with a good participants’ size (n ¼ 700). It investigated most of the peri- and post-operative parameters along with short and long term adverse outcomes. The inclusion and exclusion criteria of majority of the studies were clearly indicated to delineate any discrepancies in the results and its analysis. Most studies had laparoscopic and open groups matched for demographics and characteristics of the patients. It included the patients with cirrhosis and moderate to severe co-morbidity (ASA grade II and above) to take into account common factors associated with
Odds Ratio M-H, Random, 95% CI Year
Odds Ratio M-H, Random, 95% CI
0.39 [0.15, 0.99] 2009 0.13 [0.01, 2.24] 2009 0.30 [0.06, 1.47] 2009 0.34 [0.16, 0.73]
0.01 0.1 1 10 Laparoscopic Open
Fig. 3 e Laparoscopic vs open resection: Number of positive resection margins (n).
100
199
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Study or Subgroup
Laparoscopic Open Total Events Total Weight Events 2 3 5 0 10 1 5 16 3
Mala et al Kaneko et al Belli et al Cai et al Belli et al. Sarpel et al Tranchart et al Castaing et al Endo et al
13 30 23 31 54 20 42 60 10 283
Total (95% CI)
4 5 17 5 45 4 12 17 3
Odds Ratio M-H, Random, 95% CI Year
6.4% 9.2% 11.1% 2.9% 22.8% 4.8% 14.2% 22.1% 6.5%
0.45 [0.07, 3.04] 0.51 [0.11, 2.37] 0.10 [0.03, 0.38] 0.08 [0.00, 1.45] 0.40 [0.19, 0.88] 0.68 [0.07, 6.51] 0.34 [0.11, 1.07] 0.92 [0.41, 2.05] 1.14 [0.17, 7.60]
390 100.0%
0.43 [0.26, 0.73]
14 28 23 31 125 56 42 60 11
Odds Ratio M-H, Random, 95% CI
2002 2005 2007 2008 2009 2009 2009 2009 2009
112 45 Total events Heterogeneity: Tau² = 0.15; Chi² = 10.71, df = 8 (P = 0.22); I² = 25% Test for overall effect: Z = 3.15 (P = 0.002)
0.01 0.1 1 10 Laparoscopic Open
100
Fig. 4 e Laparoscopic vs open resection: Overall complication rate (n).
resection of malignant lesions. Most of the studies included in the analysis were conducted within the last 5 years. It means that most of the operative measures for laparoscopic technique indicate advancement in the technique and takes into account surgical experience with the technology. There was previous meta-analysis conducted in 2006, which included 6 studies.20 It was performed on the studies whose pathological resections were both malignant and benign. The individual studies15,16 have shown reduced operative time for open group as compared to laparoscopic group but the analysis of all the studies showed no difference between the groups. The analyses have taken into account sample size and standard deviation. With the advent of minimally invasive surgery, it was hoped that the blood loss would be reduced. Our compiled data suggested less blood loss with laparoscopic technique but it was associated with heterogeneity between the studies. On the other hand, number of patients requiring blood transfusion was reduced in laparoscopic technique without heterogeneity between the studies.
0
More evidence is required to justify blood loss associated with each technique. Similarly, previous studies indicated reduction in hospital stay and time to oral intake.12,14,15 Our analysis showed similar results, but there was marked heterogeneity between studies when standard deviation for each parameter was taken into consideration. There was no significant difference between two groups for early post-operative adverse outcomes. As there were few studies that reported early adverse outcomes for each group, the results of these parameters are not fully justified. For that reason, when all the complications were taken into account, which increased the sample size, the laparoscopic group was associated with reduced overall complication rate. Because of the limited access through laparoscopic technique, it was assumed that this technique was prone to reduced resection margin size and increased positive resection sampling. However, our analysis showed reduced positive resection associated with laparoscopic group and no
SE(log[OR])
0.5
1
1.5
2 0.01
OR 0.1
1
10
100
Fig. 5 e Laparoscopic vs open resection: Funnel plot of overall complication rate (n).
200
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
Table 5 e Overall summary of analysis of outcomes. Outcome of interest Peri-operative parameters Operative time (min) Operative blood loss (mls) No of Patients requiring blood transfusion Use of portal triad clamping Duration of portal triad clamping (min) Post operative parameter Duration of hospital stay (days) Time to first oral intake (days) Early post operative adverse outcomes Liver resection related Cirrhotic decompression/ascites hepatic haemorrhage biliary leakage Liver failure General complications Chest infection Wound infection Mortality Oncological clearance Pathological resection margin size (mm) Positive resection margins Long term outcomes Overall complications 3 year survival rate 3 year survival without recurrence 5 year survival rate 5 year survival without recurrence Recurrence rate
No of studies
No of resections
OR/WMD
95% CI
P value
HG P value
10 8 6 5 1
700 504 91 140 27
3.31 162.62 0.35 0.08 43
[14.05, 20.66] [-251.79, 73.45] [0.20, 0.60] [0.01, 0.46] [26.59, 59.41]
0.71 <0.001 <0.001 0.005 <0.001
<0.001 <0.001 0.85 0.01 Not estimated
8 3
504 141
4.28 1.29
[6.33, 2.33] 2.23, 0.35]
<0.001 0.007
<0.001 0.007
5 3 4 3
43 4 8 8
0.46 5.31 0.43 0.22
[0.08, [0.84, [0.10, [0.04,
5 3 5
13 4 13
0.7 0.74 0.71
4 3
152 50
9 3 3 6 6 4
157 166 89 215 112 202
difference in the size of margin size between the two groups. The possible reason being the camera used with laparoscopy, which magnifies area of resection and provides detailed visibility. In the long-term post-operative outcome, however, there was no difference between the recurrence and 5 and 3 year survival rate between the two groups. Since the analysis was carried out on the observational studies, there was an inherent selection bias as the analysis was unable to assess the differences between open and laparoscopic groups within each study. For example, the rate of major hepatectomies was higher in open group than laparoscopic group in the study conducted by Belli et al.11 Likewise, the open group included resection of larger tumours in the study by Castaing et al.13 We aimed to analyse all the parameters as mentioned in the methodology but there were not enough studies to document all the parameters, for example, cost effectiveness and post op analgesia. There is still a need for level I and II data to compare laparoscopic versus open hepatic resection in malignant lesions. With more data available, the results will be reliable, promising and significant with reduced heterogeneity between the studies. Although we feel that it will be difficult to plan RCTs and recruit patients as minimally invasive techniques are now well established in most institutions that will make it difficult for researchers to recruit patients in the open surgery arm.
Conflicts of interest None declared.
2.73] 33.55] 1.82] 1.11]
0.39 0.08 0.25 0.07
0.004 0.81 0.88 0.91
[0.21, 2.32] [0.08, 6.87] [0.21, 2.41]
0.56 0.79 0.58
0.7 0.24 0.77
0.13 0.34
[1.13, 1.40] [0.16, 0.73]
0.84 0.006
1 0.75
0.43 3.21 2.38 2.33 1.97 0.75
[0.26, [0.61, [0.52, [0.85, [0.93, [0.50,
0.002 0.17 0.26 0.1 0.07 0.17
0.22 0.005 0.007 <0.001 0.04 0.75
0.73] 16.90] 10.91] 6.42] 4.16] 1.13]
references
1. Bosch FX, Ribes J. Epidemiology of liver cancer in Europe. Can J Gastroenterol 2000 Jul-Aug;14(7):621e30. 2. Srivatanakul P, Sriplung H, Deerasamee S. Epidemiology of liver cancer: an overview. Asian Pac J Cancer Prev Apjcp 2004 Apr-Jun;5(2):118e25. 3. Llovet JM, Wurmbach E. Gene expression profiles in hepatocellular carcinoma: not yet there. J. Hepatol 2004 Aug; 41(2):336e9. 4. Bosch FX, Ribes J, Borra`s J. Epidemiology of primary liver cancer. Semin Liver Dis 1999;19(3):271e85. 5. Bosch FX, Ribes J, Cle´ries R, Dı´az M. Epidemiology of hepatocellular carcinoma. Clin Liver Dis 2005 May;9(2): 191e211. 6. Tsuchiya R. A tribute to Dr. Ischio Honjo. Surg Today 1989; 19(1):1. 7. Couinaud C. Contribution of anatomical research to liver surgery. France Medecine 1956 May;19(5):5e12. 8. Gagner M, Rogula T, Selzer D. Laparoscopic liver resection: benefits and controversies. Surg.Clin North Am 2004 Apr;84(2): 451e62. 9. Topal B, Fieuws S, Aerts R, Vandeweyer H, Penninckx F. Laparoscopic versus open liver resection of hepatic neoplasms: comparative analysis of short-term results. Surg. Endosc 2008 Oct;22(10):2208e13. 10. Belli G, Fantini C, D’Agostino A, Cioffi L, Langella S, Russolillo N, et al. Laparoscopic versus open liver resection for hepatocellular carcinoma in patients with histologically proven cirrhosis: short- and middle-term results. Surg Endosc 2007;21(11):2004e11. 11. Belli G, Limongelli P, Fantini C, D’Agostino A, Cioffi L, Belli A, et al. Laparoscopic and open treatment of hepatocellular
t h e s u r g e o n 1 0 ( 2 0 1 2 ) 1 9 4 e2 0 1
12.
13.
14.
15.
16.
carcinoma in patients with cirrhosis. Br.J.Surg 2009 Sep;96(9): 1041e8. Cai XJ, Yang J, Yu H, Liang X, Wang YF, Zhu ZY, et al. Clinical study of laparoscopic versus open hepatectomy for malignant liver tumors. Surg.Endosc 2008 Nov;22(11):2350e6. Castaing D, Vibert E, Ricca L, Azoulay D, Adam R, Gayet B. Oncologic results of laparoscopic versus open hepatectomy for colorectal liver metastases in two specialized centers. Ann.Surg 2009 Nov;250(5):849e55. Endo Y, Ohta M, Sasaki A, Kai S, Eguchi H, Iwaki K, et al. A comparative study of the long-term outcomes after laparoscopy-assisted and open left lateral hepatectomy for hepatocellular carcinoma. Surg Laparosc Endosc Percutan Tech 2009;19(5):171e4. Kaneko H, Takagi S, Otsuka Y, Tsuchiya M, Tamura A, Katagiri T, et al. Laparoscopic liver resection of hepatocellular carcinoma. Am.J.Surg 2005 Feb;189(2):190e4. Laurent A, Cherqui D, Lesurtel M, Brunetti F, Tayar C, Fagniez PL. Laparoscopic liver resection for subcapsular
17.
18.
19.
20.
201
hepatocellular carcinoma complicating chronic liver disease. Arch Surg 2003;138(7):763e9. discussion 769; Jul. Mala T, Edwin B, Gladhaug I, Fosse E, Soreide O, Bergan A, et al. A comparative study of the short-term outcome following open and laparoscopic liver resection of colorectal metastases. Surg.Endosc 2002 Jul;16(7):1059e63. Sarpel U, Hefti MM, Wisnievsky JP, Roayaie S, Schwartz ME, Labow DM. Outcome for patients treated with laparoscopic versus open resection of hepatocellular carcinoma: case-matched analysis. Ann Surg Oncol 2009 Jun;16(6):1572e7. Tranchart H, Di Giuro G, Lainas P, Roudie J, Agostini H, Franco D, et al. Laparoscopic resection for hepatocellular carcinoma: a matched-pair comparative study. J Surg Endosc; 2009. doi:10.1007/s00464-009-0745-3 (epub). Simillis C, Constantinides VA, Tekkis PP, Darzi A, Lovegrove R, Jiao L, et al. Laparoscopic versus open hepatic resections for benign and malignant neoplasmsea metaanalysis. Surgery 2007 Feb;141(2):203e11.