This case-control study compares the surgical outcomes of robotic vs laparoscopic right hepatectomy/extended right hepatectomy.
Key Points
Question
In patients undergoing major hepatectomy procedures, is either the robotic or laparoscopic technique superior?
Findings
In this case-control study of 989 patients, robotic right or extended right hepatectomy (RH/ERH) was associated with a significantly lower open conversion rate and shorter postoperative hospital stay when compared with laparoscopic RH/ERH. The difference in open conversion rate was associated with a significant decrease for laparoscopic but not robotic RH/ERH after a center had mounted the learning curve.
Meaning
Use of robotic platform may help to overcome the initial challenges of minimally invasive RH/ERH.
Abstract
Importance
Laparoscopic and robotic techniques have both been well adopted as safe options in selected patients undergoing hepatectomy. However, it is unknown whether either approach is superior, especially for major hepatectomy such as right hepatectomy or extended right hepatectomy (RH/ERH).
Objective
To compare the outcomes of robotic vs laparoscopic RH/ERH.
Design, Setting, and Participants
In this case-control study, propensity score matching analysis was performed to minimize selection bias. Patients undergoing robotic or laparoscopic RH/EHR at 29 international centers from 2008 to 2020 were included.
Interventions
Robotic vs laparoscopic RH/ERH.
Main Outcomes and Measures
Data on patient demographics, tumor characteristics, and short-term perioperative outcomes were collected and analyzed.
Results
Of 989 individuals who met study criteria, 220 underwent robotic and 769 underwent laparoscopic surgery. The median (IQR) age in the robotic RH/ERH group was 61.00 (51.86-69.00) years and in the laparoscopic RH/ERH group was 62.00 (52.03-70.00) years. Propensity score matching resulted in 220 matched pairs for further analysis. Patients’ demographics and tumor characteristics were comparable in the matched cohorts. Robotic RH/ERH was associated with a lower open conversion rate (19 of 220 [8.6%] vs 39 of 220 [17.1%]; P = .01) and a shorter postoperative hospital stay (median [IQR], 7.0 [5.0-10.0] days; mean [SD], 9.11 [7.52] days vs median [IQR], 7.0 [5.75-10.0] days; mean [SD], 9.94 [8.99] days; P = .048). On subset analysis of cases performed between 2015 and 2020 after a center’s learning curve (50 cases), robotic RH/ERH was associated with a shorter postoperative hospital stay (median [IQR], 6.0 [5.0-9.0] days vs 7.0 [6.0-9.75] days; P = .04) with a similar conversion rate (12 of 220 [7.6%] vs 17 of 220 [10.8%]; P = .46).
Conclusion and Relevance
Robotic RH/ERH was associated with a lower open conversion rate and shorter postoperative hospital stay compared with laparoscopic RH/ERH. The difference in open conversion rate was associated with a significant decrease for laparoscopic but not robotic RH/ERH after a center had mounted the learning curve. Use of robotic platform may help to overcome the initial challenges of minimally invasive RH/ERH.
Introduction
Minimally invasive approach has been applied in liver resection (LR) for more than 2 decades.1 While minimally invasive hepatectomy has become standard practice according to various consensus conferences, its role in major LR remains in the exploration phase.2,3,4,5 Several meta-analyses have been performed to evaluate the effectiveness, safety, and efficacy of robotic vs laparoscopic hepatectomy.6,7,8,9,10,11 However, most analyzed results from retrospective cases series and studied minor LR or LR at low difficulty level. These studies also included various types of LR resulting in additional confounding factors. Studies focusing on major hepatectomy eg, right hepatectomy (RH) or extended RH (ERH), are limited to date.12
Major hepatectomy is technically demanding, which hinders the development of minimally invasive surgery in LR. An analysis of the American College of Surgeons National Surgical Quality Improvement Program data showed that major LR was one of the risk factors for open conversion during minimally invasive hepatectomy.13 The robotic system was developed to overcome some of the challenges in laparoscopic surgery. With the magnified view and the 7 degrees of movement of the EndoWrist (Intuitive Surgical), the robotic system allows fine dissection of individual vessels at the hilar region and the hepatocaval region during RH/ERH. It also allows easy suturing and knot tying in case of bleeding. In theory, the advantages of the robotic system might be more prominent in major LR at high/expert difficulty level. It has been reported that the use of robotic system allowed for an increased percentage of major hepatectomies to be performed in a purely minimally invasive manner.14,15,16
To study the true merit of the robotic system in a minimally invasive major hepatectomy, we performed the current study and focused specifically on RH/ERH for a more direct comparison. The objective of this study is to compare the surgical outcomes of robotic RH/ERH (R-RH/ERH) vs laparoscopic RH/ERH (L-RH/ERH).
Methods
This is an international multicenter retrospective case-control analysis of patients undergoing L-RH/ERH or R-RH/ERH at 29 centers between January 2008 and December 2020. All institutions obtained their respective approvals according to their local center’s requirements. This study was approved by the Singapore General Hospital institution review board, and the need for patient consent was waived. The deidentified data were collected in the individual centers. These were collated and analyzed centrally at the Singapore General Hospital.
In this study, only patients who underwent pure laparoscopic or robotic-assisted laparoscopic surgery were included. Laparoscopic-assisted (hybrid) and hand-assisted laparoscopic resections were excluded. Similarly, patients undergoing donor hepatectomy for transplant, associating liver partition and portal vin ligation for staged hepatectomy, and hepatectomy with bilio-enteric anastomoses were excluded. Data on race and ethnicity were not collected.
Definitions
RH was defined according to the 2000 Brisbane classification as resection of segments 5, 6, 7, and 8.17 ERH included individuals who underwent RH and part or all of segment 4 and did not necessarily entail a complete trisectionectomy. Diameter of the largest lesion was used in cases of multiple tumors. Postoperative complications were classified according to the Clavien-Dindo classification and recorded for up to 30 days or during the same hospitalization and included 30-day readmissions.18,19 The difficulty of resections was graded according to the Iwate score.20,21
Statistical Analysis
Propensity score matching was performed using a 1:1 nearest neighbor matching algorithm without replacement with distances determined by logistic regression. Propensity score matching was performed based on the following variables: sex, year of resection, age at operation, American Society of Anesthesiologists status, size of tumor, single or multiple tumors, malignant or benign tumors, Child-Pugh score, presence of portal hypertension, previous liver or abdominal operation, whether the hepatectomy was extended, multiple resections, concomitant noncholecystectomy operation, presence of cirrhosis, histological diagnosis, and Iwate difficulty grade. The optimal matching algorithm was used as a sensitivity analysis.
Absolute standardized mean difference in distances between patients undergoing robotic or laparoscopic RH were found to be balanced after conditioning on the propensity score, where a difference of less than 0.1 in absolute standard mean difference after matching was considered to indicate a good balance.
In the unmatched cohort, comparisons of patient characteristics and perioperative and postoperative details between patients undergoing robotic or laparoscopic RH were performed using the Mann-Whitney U test for continuous variables, while the Fisher exact test and Pearson χ2 test were used for categorical variables. Comparisons in the matched cohorts took into account the paired nature of the data; hence, paired analyses such as Wilcoxon signed rank test, McNemar test and McNemar-Bowker test were respectively used for continuous, 2-by-2 categorical variables, and 3-by-3 categorical variables.
To reduce the confounding effect of the learning curve, a subset analysis of cases performed between January 2015 and December 2020 was performed with nearest neighbor matching of distances derived by logistic regression. Apart from omitting the year of resection, the logistic regression model used variables similar to that of the primary analysis. Furthermore, only cases performed after a center had acquired a cumulative experience of more than 50 minimally invasive liver resections was included. This conservative number of 50 cases was chosen based on the recently published systematic review,22 which reported that the overall learning curve of L-LR was 50 cases and for R-LR was 25 cases. Furthermore, the learning curve for minimally invasive liver resection was reported to be decreasing from 48 cases in the 1990s to 24 cases in 2015. All analyses were done in R version 4.0.0 (R Foundation) with package ‘MatchIt’ and ‘optmatch’, and 2-sided P < .05 was regarded to indicate statistical significance.
Results
Comparison Between R-RH/ERH and L-RH/ERH in the Unmatched and Matched Cohorts
A total of 989 cases met study criteria, of which 220 underwent robotic and 769 underwent laparoscopic surgery during the study period. Nine patients (4.1%) and 54 patients (7.0%) received ERH in the robotic and laparoscopic arms, respectively (P = .12). The R-RH/ERH group was associated with a higher proportion of patients with American Society of Anesthesiologists score III/IV (87 of 220 [39.5%] vs 184 of 769 [24.0%]; P < .001), a lower proportion of patients with previous abdominal (75 of 220 [34.1%] vs 325 of 769 [42.3%]; P = .03) or liver surgery (8 of 220 [3.6%] vs 93 of 769 [12.1%]; P < .001), higher proportion of cirrhotic liver (56 of 220 [25.5%] vs 174 of 769 [22.6%]; P = .38), and a lower proportion of patients with multiple tumors (58 of 220 [26.4%] vs 307 of 769 [40.0%]; P < .001) and less multiple resections (6 of 220 [2.7%] vs 63 of 769 [8.2%]; P = .005) (Table 1). Regarding the perioperative outcomes, there were statistically significant differences in postoperative stay (median [IQR], 7.00 [5.00-10.00] vs 7.00 [6.00-11.00]; P = .01), overall postoperative morbidity (68 of 220 [30.9%] vs 295 of 769 [38.4%]; P = .04), and major morbidity (24 of 220 [10.9%] vs 129 of 769 [16.8%]; P = .03) in favor of robotic surgery.
Table 1. Comparison Between Baseline Clinicopathological Characteristics of R-RH/ERH vs L-RH/ERH.
Characteristic | No. (%) | ||||||||
---|---|---|---|---|---|---|---|---|---|
Unmatched cohort (n = 989) | 1:1 Propensity score matching (nearest neighbor matching; n = 440) | 1:1 Propensity score matching (optimal matching; n = 440) | |||||||
R-RH/ERH (n = 220) | L-RH/ERH (n = 769) | P value | R-RH/ERH (n = 220) | L-RH/ERH (n = 220) | P value | R-RH/ERH (n = 220) | L-RH/ERH (n = 220) | P value | |
Age, median (IQR), y | 61.00 (51.86-69.00) | 62.00 (52.03-70.00) | .33 | 61.00 (51.86-69.00) | 61.00 (52.00-68.00) | .79 | 61.00 (52.00-69.00) | 63.00 (55.00-71.00) | .07 |
Male | 139 (63.2) | 493 (64.2) | .78 | 139 (63.2) | 148 (67.3) | .43 | 139 (63.2) | 144 (65.5) | .71 |
Female | 81 (36.8) | 276 (35.9) | 81 (36.8) | 72 (32.7) | 81 (36.8) | 76 (34.5) | |||
Year of resection | |||||||||
2008-2014 | 62 (28.2) | 223 (29.0) | .81 | 62 (28.2) | 58 (26.4) | .75 | 62 (28.2) | 67 (30.5) | .69 |
2015-2020 | 158 (71.8) | 546 (71.0) | 158 (71.8) | 162 (73.6) | 158 (71.8) | 153 (69.5) | |||
American Society of Anesthesiologists score | |||||||||
I/II | 133 (60.5) | 584 (76.0) | <.001 | 133 (60.5) | 133 (60.5) | >.99 | 133 (60.5) | 128 (58.2) | .57 |
II/IV | 87 (39.5) | 184 (24.0) | 87 (39.5) | 87 (39.5) | 87 (39.5) | 92 (41.8) | |||
Previous abdominal surgery | 75 (34.1) | 325 (42.3) | .03 | 75 (34.1) | 76 (34.5) | >.99 | 75 (34.1) | 72 (32.7) | .84 |
Previous liver surgery | 8 (3.6) | 93 (12.1) | <.001 | 8 (3.6) | 12 (5.5) | .39 | 8 (3.6) | 11 (5.0) | .55 |
Malignant pathology | 190 (86.4) | 678 (88.2) | .47 | 190 (86.4) | 190 (86.4) | >.99 | 190 (86.4) | 194 (88.2) | .67 |
Pathology type | |||||||||
HCC | 106 (48.2) | 312 (40.6) | .049 | 106 (48.2) | 117 (53.2) | .69 | 106 (48.2) | 104 (47.3) | >.99 |
CRM | 57 (25.9) | 263 (34.2) | 57 (25.9) | 48 (21.8) | 57 (25.9) | 59 (26.8) | |||
Other | 57 (25.9) | 193 (25.1) | 57 (25.9) | 55 (25.0) | 57 (25.9) | 57 (25.9) | |||
Cirrhosis | 56 (25.5) | 174 (22.6) | .38 | 56 (25.5) | 58 (26.4) | .92 | 56 (25.5) | 50 (22.7) | .59 |
Child-Pugh score | |||||||||
No cirrhosis | 164 (74.5) | 595 (77.4) | .008 | 164 (74.5) | 160 (72.7) | .22 | 164 (74.5) | 169 (76.8) | .17 |
A | 49 (22.3) | 170 (22.1) | 49 (22.3) | 58 (26.4) | 49 (22.3) | 49 (22.3) | |||
B | 7 (3.2) | 4 (0.5) | 7 (3.2) | 2 (0.9) | 7 (3.2) | 2 (0.9) | |||
Portal hypertension | 9 (4.1) | 25 (3.3) | .55 | 9 (4.1) | 6 (2.7) | .61 | 9 (4.1) | 3 (1.4) | .15 |
Tumor size, median (IQR), mm | 50.00 (30.00-70.00) | 45.00 (28.00-75.00) | .75 | 50.00 (30.00-70.00) | 46.50 (28.75-70.00) | .62 | 50.00 (30.00-70.00) | 50.00 (30.00-75.00) | .88 |
Multiple tumors | 58 (26.4) | 307 (40.0) | <.001 | 58 (26.4) | 48 (21.8) | .28 | 58 (26.4) | 51 (23.2) | .46 |
Multiple resections | 6 (2.7) | 63 (8.2) | .005 | 6 (2.7) | 6 (2.7) | >.99 | 6 (2.7) | 4 (1.8) | .72 |
Concomitant operation noncholecystectomy | 31 (14.1) | 82 (10.7) | .16 | 31 (14.1) | 28 (12.7) | .77 | 31 (14.1) | 37 (16.8) | .48 |
Right hepatectomy | 211 (95.9) | 715 (93.0) | .12 | 211 (95.9) | 209 (95.0) | .81 | 211 (95.9) | 212 (96.4) | >.99 |
Extended right hepatectomy | 9 (4.1) | 54 (7.0) | 9 (4.1) | 11 (5.0) | 9 (4.1) | 8 (3.6) | |||
Iwate score | |||||||||
Intermediate | 0 | 6 (0.8) | .60 | 0 | 1 (0.5) | .97 | 0 | 1 (0.5) | .79 |
High | 47 (21.4) | 165 (21.5) | 47 (21.4) | 49 (22.3) | 47 (21.4) | 42 (19.1) | |||
Expert | 173 (78.6) | 598 (77.8) | 173 (78.6) | 170 (77.3) | 173 (78.6) | 177 (80.5) |
Abbreviations: CRM, colorectal metastases; HCC, hepatocellular carcinoma; L-RH/ERH, laparoscopic right hepatectomy/extended right hepatectomy; R-RH/ERH, robotic right hepatectomy/extended right hepatectomy.
Propensity score matching with 1:1 ratio resulted in 220 matched pairs for further analysis. Both groups were well balanced in all baseline clinicopathological characteristics in the matched cohort by nearest neighbor matching (Table 1; eFigures 1-3 in Supplement 1). Regarding the perioperative outcomes, R-RH/ERH was associated with a significantly lower open conversion rate (19 of 220 [8.6%] vs 39 of 220 [17.7%]; P = .01) (Table 2). With optimal matching, the baseline clinicopathological characteristics were comparable (Table 1; eFigures 4-6 in Supplement 1). The R-RH/ERH group had a lower conversion rate (19 of 220 [8.6%] vs 35 of 220 [15.9%]; P = .03) and a shorter postoperative hospital stay (median [IQR], 7.0 [5.0-10.0]; mean [SD], 9.11 [7.52] days vs 7.0 [5.75-10.0]; mean [SD], 9.94 [8.99] days; P = .048) compared with L-RH/ERH.
Table 2. Comparison Between Perioperative Outcomes of R-RH/ERH vs L-RH/ERH.
Characteristic | No. (%) | ||||||||
---|---|---|---|---|---|---|---|---|---|
Unmatched cohort (n = 989) | 1:1 Propensity score matching (nearest neighbor matching; n = 440) | 1:1 Propensity score matching (optimal matching; n = 440) | |||||||
R-RH/ERH (n = 220) | L-RH/ERH (n = 769) | P value | R-RH/ERH (n = 220) | L-RH/ERH (n = 220) | P value | R-RH/ERH (n = 220) | L-RH/ERH (n = 220) | P value | |
Operating time, median (IQR), min | 315.00 (241.50-461.25) | 345.00 (260.00-431.00) | .70 | 315.00 (241.50-461.25) | 346.50 (260.00-452.00) | .78 | 315.00 (241.50-461.25) | 330.00 (260.00-418.50) | .35 |
Blood loss, median (IQR), mL | 300.00 (100.00-600.00) | 300.00 (150.00-500.00) | .80 | 300.00 (100.00-600.00) | 300.00 (186.00-500.00) | .69 | 300.00 (100.00-600.00) | 250.00 (150.00-500.00) | .29 |
Blood loss, mL | |||||||||
<500 | 143 (65.0) | 513 (70.2) | .15 | 143 (65.0) | 147 (70.3) | .14 | 143 (65.0) | 152 (72.0) | .21 |
≥500 | 77 (35.0) | 218 (29.8) | 77 (35.0) | 62 (29.7) | 77 (35.0) | 59 (28.0) | |||
Intraoperative blood transfusion | 35 (15.9) | 100 (13.0) | .27 | 35 (15.9) | 31 (14.1) | .68 | 35 (15.9) | 31 (14.1) | .69 |
Pringle maneuver applied | 93 (42.3) | 343 (45.7) | .37 | 93 (42.3) | 106 (49.8) | .16 | 93 (42.3) | 108 (50.5) | .07 |
Open conversion | 19 (8.6) | 100 (13.0) | .08 | 19 (8.6) | 39 (17.7) | .01 | 19 (8.6) | 35 (15.9) | .03 |
Postoperative stay, median (IQR), d | 7.00 (5.00-10.00) | 7.00 (6.00-11.00) | .01 | 7.00 (5.00-10.00) | 7.00 (5.50-10.00) | .15 | 7.00 (5.00-10.00) | 7.00 (5.75-10.00) | .048 |
30-d Readmission | 12 (5.5) | 42 (5.6) | .91 | 12 (5.5) | 12 (5.7) | >.99 | 12 (5.5) | 13 (6.2) | >.99 |
Postoperative morbidity | 68 (30.9) | 295 (38.4) | .04 | 68 (30.9) | 70 (31.8) | .91 | 68 (30.9) | 75 (34.1) | .56 |
Major morbidity (Clavien-Dindo grade >2) | 24 (10.9) | 129 (16.8) | .03 | 24 (10.9) | 33 (15.0) | .24 | 24 (10.9) | 30 (13.6) | .46 |
Reoperation | 4 (1.8) | 33 (4.3) | .11 | 4 (1.8) | 8 (3.6) | .39 | 4 (1.8) | 5 (2.3) | >.99 |
Mortality | |||||||||
In-hospital | 3 (1.4) | 11 (1.4) | >.99 | 3 (1.4) | 0 | .25 | 3 (1.4) | 2 (0.9) | >.99 |
30-d | 4 (1.8) | 7 (0.9) | .28 | 4 (1.8) | 0 | .13 | 4 (1.8) | 1 (0.5) | .37 |
90-d | 6 (2.7) | 14 (1.8) | .40 | 6 (2.7) | 1 (0.5) | .13 | 6 (2.7) | 2 (0.9) | .29 |
Abbreviations: L-RH/ERH, laparoscopic right hepatectomy/extended right hepatectomy; R-RH/ERH, robotic right hepatectomy/extended right hepatectomy.
Comparison Between R-RH and L-RH/ERH in the Subset of Cases Performed Between 2015 and 2020
To minimize the confounder effect of the learning curve, we further analyzed a subset of 704 cases performed between 2015 and 2020 after a center had acquired a cumulative experience of more than 50 minimally invasive liver resection procedures. In the matched cohort, R-RH/ERH and L-RH/ERH groups had a comparable baseline clinicopathological characteristics (Table 3; eFigures 7-9 in Supplement 1). There was no longer a significant difference in open conversion rate (12 of 220 [7.6%] vs 17 of 220 [10.8]; P = .46) after matching. R-RH/ERH was associated with a significantly shorter median postoperative stay (median [IQR], 6.0 [5.0-9.0] days vs 7.0 [6.0-9.75] days; P = .04). There was no significant difference in other perioperative outcomes between the 2 groups (Table 4).
Table 3. Comparison Between Baseline Clinicopathological Characteristics of R-RH/ERH vs L-RH/ERH Performed Between 2015-2020 After a Center’s Learning Curve (50 Cases).
Characteristic | No. (%) | |||||
---|---|---|---|---|---|---|
Unmatched cohort (n = 704) | 1:1 Propensity matched cohort (n = 316) | |||||
R-RH/ERH (n = 158) | L-RH/ERH (n = 546) | P value | R-RH/ERH (n = 158) | L-RH/ERH (n = 158) | P value | |
Age, median (IQR), y | 61.00 (51.00-68.75) | 61.00 (52.00-70.00) | .30 | 61.00 (51.00-68.75) | 61.00 (47.25-69.14) | .80 |
Male | 98 (62.0) | 347 (63.7) | .71 | 98 (62.0) | 93 (58.9) | .63 |
Female | 60 (38.0) | 199 (36.4) | 60 (38.0) | 65 (41.1) | ||
American Society of Anesthesiologists score | ||||||
I/II | 101 (63.9) | 406 (74.5) | .009 | 101 (63.9) | 98 (62.0) | .79 |
III/IV | 57 (36.1) | 139 (25.5) | 57 (36.1) | 60 (38.0) | ||
Previous abdominal surgery | 49 (31.0) | 213 (39.0) | .07 | 49 (31.0) | 42 (26.6) | .45 |
Previous liver surgery | 7 (4.4) | 60 (11.0) | .01 | 7 (4.4) | 7 (4.4) | >.99 |
Malignant pathology | 130 (82.3) | 483 (88.5) | .04 | 130 (82.3) | 124 (78.5) | .45 |
Pathology type | ||||||
HCC | 74 (46.8) | 241 (44.2) | .002 | 74 (46.8) | 71 (44.9) | .31 |
CRM | 31 (19.6) | 177 (32.5) | 31 (19.6) | 30 (19.0) | ||
Other | 53 (33.5) | 127 (23.3) | 53 (33.5) | 57 (36.1) | ||
Cirrhosis | 41 (25.9) | 136 (24.9) | .79 | 41 (25.9) | 37 (23.4) | .67 |
Child-Pugh score | ||||||
No cirrhosis | 117 (74.1) | 410 (75.1) | .003 | 117 (74.1) | 119 (75.3) | .07 |
A | 34 (21.5) | 133 (24.4) | 34 (21.5) | 39 (24.7) | ||
B | 7 (4.4) | 3 (0.5) | 7 (4.4) | 0 (0.0) | ||
Portal hypertension | 8 (5.1) | 21 (3.9) | .50 | 8 (5.1) | 3 (1.9) | .23 |
Tumor size, median (IQR), mm | 50.00 (30.00-75.00) | 50.00 (30.00-78.00) | .63 | 50.00 (30.00-75.00) | 60.00 (33.25-80.75) | .21 |
Multiple tumors | 39 (24.7) | 216 (39.6) | .001 | 39 (24.7) | 45 (28.5) | .47 |
Multiple resections | 5 (3.2) | 42 (7.7) | .045 | 5 (3.2) | 4 (2.5) | >.99 |
Concomitant operation noncholecystectomy | 26 (16.5) | 48 (8.8) | .006 | 26 (16.5) | 25 (15.8) | >.99 |
Right hepatectomy | 154 (97.5) | 506 (92.7) | .03 | 154 (97.5) | 156 (98.7) | .68 |
Extended right hepatectomy | 4 (2.5) | 40 (7.3) | 4 (2.5) | 2 (1.3) | ||
Iwate score | ||||||
Intermediate | 0 | 5 (0.9) | .32 | 0 | 2 (1.3) | .55 |
High | 30 (19.0) | 125 (22.9) | 30 (19.0) | 27 (17.1) | ||
Expert | 128 (81.0) | 416 (76.2) | 128 (81.0) | 129 (81.6) |
Abbreviations: CRM, colorectal metastases; HCC, hepatocellular carcinoma; L-RH/ERH, laparoscopic right hepatectomy/extended right hepatectomy; R-RH/ERH, robotic right hepatectomy/extended right hepatectomy.
Table 4. Comparison Between Perioperative Outcomes of R-RH/ERH vs L-RH/ERH Between 2015-2020 (After Center’s First 50 Cases).
Characteristic | No. (%) | |||||
---|---|---|---|---|---|---|
Unmatched cohort (n = 704) | 1:1 Propensity score matching (n = 316) | |||||
R-RH/ERH (n = 158) | L-RH/ERH (n = 546) | P value | R-RH/ERH (n = 158) | L-RH/ERH (n = 158) | P value | |
Operating time, median (IQR), min | 317.50 (236.00-435.25) | 349.00 (260.00-434.00) | .48 | 317.50 (236.00-435.25) | 347.50 (262.75-443.25) | .85 |
Blood loss, median (IQR), mL | 300.00 (100.00-600.00) | 270.00 (150.00-500.00) | .98 | 300.00 (100.00-600.00) | 300.00 (200.00-450.00) | .30 |
Blood loss, mL | ||||||
<500 | 107 (67.7) | 377 (71.8) | .32 | 107 (67.7) | 116 (75.8) | .18 |
≥500 | 51 (32.3) | 148 (28.2) | 51 (32.3) | 37 (24.2) | ||
Intraoperative blood transfusion | 24 (15.2) | 64 (11.7) | .25 | 24 (15.2) | 16 (10.1) | .24 |
Pringle maneuver applied | 76 (48.1) | 288 (54.0) | .19 | 76 (48.1) | 78 (51.0) | .65 |
Open conversion | 12 (7.6) | 59 (10.8) | .24 | 12 (7.6) | 17 (10.8) | .46 |
Postoperative stay, median (IQR), d | 6.00 (5.00-9.00) | 7.00 (5.00-10.00) | .02 | 6.00 (5.00-9.00) | 7.00 (6.00-9.75) | .04 |
30-d Readmission | 11 (7.0) | 33 (6.0) | .68 | 11 (7.0) | 15 (9.5) | .56 |
Postoperative morbidity | 50 (31.6) | 186 (34.1) | .57 | 50 (31.6) | 45 (28.5) | .63 |
Major morbidity (Clavien-Dindo grade >2) | 15 (9.5) | 85 (15.6) | .05 | 15 (9.5) | 23 (14.6) | .22 |
Reoperation | 2 (1.3) | 18 (3.3) | .28 | 2 (1.3) | 6 (3.8) | .29 |
Mortality | ||||||
In-hospital | 3 (1.9) | 6 (1.1) | .43 | 3 (1.9) | 1 (0.6) | .62 |
30-d | 4 (2.5) | 3 (0.5) | .048 | 4 (2.5) | 0 | .13 |
90-d | 6 (3.8) | 9 (1.6) | .10 | 6 (3.8) | 1 (0.6) | .13 |
Abbreviations: L-RH/ERH, laparoscopic right hepatectomy/extended right hepatectomy; R-RH/ERH, robotic right hepatectomy/extended right hepatectomy.
We further analyzed the open conversion rates of R-RH/ERH and L-RH/ERH over time. When we compared the open conversion rates of R-RH/ERH between 2008 and 2014 vs 2015 and 2020, there was no significant difference in the open conversion rates between the 2 time periods (7 of 62 [11.3%] vs 12 of 158 [7.6%]; P = .38). On the other hand, the open conversion rate of L-RH/ERH was significantly lower in the more recent time period compared with the earlier period (59 of 546 [10.8%] vs 41 of 223 [18.4%]; P = .005).
Discussion
With a large sample size from multiple international expert centers, this study demonstrated that R-RH/ERH was associated with a lower conversion rate and shorter hospital stay compared with L-RH/ERH. After reduction of the learning curve effect, R-RH/ERH was associated with a shorter hospital stay and similar perioperative outcomes. Our results imply that a robotic approach may help to overcome the learning curve in RH/ERH at the initial phase. To our knowledge, this study represents the largest series of minimally invasive RH/ERH reported to date and also the first to focus on the outcomes of R-RH/ERH vs L-RH/ERH.
Major hepatectomy remains a hurdle to the development of minimally invasive surgery in hepatectomy owing to its technical difficulties. Whether robotic or laparoscopic is a better approach for major hepatectomy remains unknown. Although numerous studies have compared the operative outcomes of robotic vs laparoscopic hepatectomy, most reports were dominated by minor hepatectomies.14,23,24,25,26,27 Previous study demonstrated no significant difference in perioperative outcomes for hepatectomies at low and intermediate difficulty levels.16 Two meta-analyses on this topic found no significant difference between 2 approaches.11,28 However, the proportion of major hepatectomy was only 10% to 20% in the included studies. Data comparing robotic vs laparoscopic approaches in major hepatectomy are limited. Therefore, we conducted this study to compare the outcomes of R-RH/ERH and L-RH/ERH. RH/ERH was chosen for analysis because it allows standardized and comparable analysis.
Several authors had postulated that robotic LR may be associated with a shorter learning curve compared with the long learning curve reported for laparoscopic approach,22,29,30 and it may also allow an increased proportion of hepatectomies at advanced difficulty level to be performed in a purely minimally invasive approach.14,15,16 In this study, the open conversion rate was significantly lower for R-RH/ERH, particularly during the initial period of the study. Because the reported learning curves for laparoscopic and robotic LR were 50 and 25 cases, respectively,22 we performed a subset analyses that only included cases after the center had performed 50 minimally invasive liver resection procedures to reduce the confounding effect of learning curve. The difference in conversion rate was not apparent in the subset analyses. This observation implied that the robotic approach may be associated with a shorter learning curve. Therefore, the conversion rate was shorter for robotic approach overall, but little improvement was observed once the center has passed the learning curve.
Concerning major hepatectomy, a recent systemic review and meta-analysis of 525 patients found no significant difference between robotic and laparoscopic major LR regarding overall morbidity and mortality rates, conversion to open hepatectomy, blood loss and blood transfusion rate, operative time, as well as length of hospital stay.31 The authors concluded that both techniques were equivalent when performed in selected patients and high-volume centers. Similarly, Marino et al,32 in a small retrospective study, compared the outcomes of patients undergoing R-RH with L-RH and showed no significant difference between the 2 approaches in terms of operative time, estimated blood loss, conversion to open surgery, and overall morbidity. Our study demonstrated similar results when the analysis was restricted to cases performed after the center had passed the learning curve. These findings suggest that the robotic platform may offer some benefits especially at the initial period of the learning curve.
In a meta-analysis of 6 studies with 1093 patients, Hu et al9 found that robotic approach was more commonly used for RH and is associated with longer operative time, larger tumor size, and lower open conversion rate. RH was more frequently done in the robotic group, while there was no significant difference in left hepatectomy and left lateral sectionectomy. Notably, the Iwate difficulty system does not distinguish between RH and LH as both will have a resection score of 4. However, it is universally accepted by liver surgeons that RH is more difficult than left hepatectomy owing to the wider transaction surface area and deep-seated location of the right lobe.33 Our results demonstrated that the use of robotic approach was associated with a lower conversion rate and shorter hospital stay for RH/ERH.
The robotic approach has several potential advantages over the laparoscopic approach for RH/ERH.34,35,36,37,38 The 7 degrees of freedom of the robotic system may theoretically overcome the problems of difficult access by rigid laparoscopic instruments. Robotic instruments may also facilitate the fine extrahepatic hilar dissection of individual hepatic artery and portal vein in RH/ERH. It may also enhance the dissection and control of short hepatic veins in the hepatocaval region and allow easier suture plication of bleeders during parenchymal transection. Moreover, the operation time for RH/ERH is relatively long (approximately 5 hours in this series), and use of robotic system can alleviate a surgeon’s fatigue during surgery.
The main drawbacks associated with robotic surgery compared with laparoscopy include the limited access to the platform and high costs of the procedure.34,35 Results from this multicenter study found that robotic approach was associated with a shorter hospital stay without increase in operating time and other morbidities. It suggests that hospitalization costs may be possibly saved with robotic approach if the cost for robotic instruments can also be reduced. With several new models of surgical robots coming into the market, the barrier to access and cost of the robotic platform is likely to decrease.
It is also important to highlight that the advantages seen with robotic surgery may not be entirely attributable to the technical advantages of the platform. The length of stay after surgery is influenced by multiple confounding factors especially in this international multicenter study. It is well-known that global variation in factors such as local culture and health care systems have a major influence on postoperative stay. Similarly, although the study attempted to control for the institutional learning curve, detailed information of each individual surgeon’s experience and hence learning curve was not available. Furthermore, many surgeons who perform robotic LR would have prior experience with LLR before embarking on robotic surgery or occasionally vice versa. Notably, it was not possible for us to control for this complex interplay of factors between individual surgeon and institution experience with both R-LR and L-LR as in real-world practice, many surgeons frequently perform both procedures.
Limitations
The major limitation of this study is its retrospective nature, which would be associated with information and selection bias. Although propensity score matching analysis was performed to minimize the selection bias, residual selection bias from unmeasured/unknown confounders was inevitable in the absence of randomization. As an international multicenter study, our study was also associated with differing experiences between the contributing centers and heterogeneity in the surgical technique adopted by the different centers. Nevertheless, it enhances the external validity and generalizability of this study as it reflects the current real-world practice. The large sample size generated from this multicenter study also allowed robust statistical analysis, which is important when we are studying a particular procedure whereby the expected difference is small as in this study.
Conclusions
In conclusion, R-RH/ERH was associated with a lower open conversion rate and shorter postoperative hospital stay compared with L-RH/ERH. However, there was no difference in open conversion rate after a center had mounted the learning curve. Use of the robotic platform may facilitate overcoming the challenges in minimally invasive RH/ERH especially during the learning curve.
References
- 1.Goh BKP, Lee SY, Teo JY, et al. Changing trends and outcomes associated with the adoption of minimally invasive hepatectomy: a contemporary single-institution experience with 400 consecutive resections. Surg Endosc. 2018;32(11):4658-4665. doi: 10.1007/s00464-018-6310-1 [DOI] [PubMed] [Google Scholar]
- 2.Wakabayashi G, Cherqui D, Geller DA, et al. Recommendations for laparoscopic liver resection: a report from the second international consensus conference held in Morioka. Ann Surg. 2015;261(4):619-629. [DOI] [PubMed] [Google Scholar]
- 3.Cheung TT, Han HS, She WH, et al. The Asia Pacific consensus statement on laparoscopic liver resection for hepatocellular carcinoma: a report from the 7th Asia-Pacific Primary Liver Cancer Expert Meeting held in Hong Kong. Liver Cancer. 2018;7(1):28-39. doi: 10.1159/000481834 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Abu Hilal M, Aldrighetti L, Dagher I, et al. The Southampton Consensus Guidelines for Laparoscopic Liver Surgery: from indication to implementation. Ann Surg. 2018;268(1):11-18. doi: 10.1097/SLA.0000000000002524 [DOI] [PubMed] [Google Scholar]
- 5.Ciria R, Berardi G, Nishino H, et al. ; Study group of Precision Anatomy for Minimally Invasive Hepato-Biliary-Pancreatic surgery (PAM-HBP surgery) . A snapshot of the 2020 conception of anatomic liver resections and their applicability on minimally invasive liver surgery: a preparatory survey for the Expert Consensus Meeting on Precision Anatomy for Minimally Invasive HBP Surgery. J Hepatobiliary Pancreat Sci. 2021. doi: 10.1002/jhbp.959 [DOI] [PubMed] [Google Scholar]
- 6.Hu L, Yao L, Li X, Jin P, Yang K, Guo T. Effectiveness and safety of robotic-assisted versus laparoscopic hepatectomy for liver neoplasms: a meta-analysis of retrospective studies. Asian J Surg. 2018;41(5):401-416. doi: 10.1016/j.asjsur.2017.07.001 [DOI] [PubMed] [Google Scholar]
- 7.Gavriilidis P, Roberts KJ, Aldrighetti L, Sutcliffe RP. A comparison between robotic, laparoscopic and open hepatectomy: a systematic review and network meta-analysis. Eur J Surg Oncol. 2020;46(7):1214-1224. doi: 10.1016/j.ejso.2020.03.227 [DOI] [PubMed] [Google Scholar]
- 8.Guan R, Chen Y, Yang K, et al. Clinical efficacy of robot-assisted versus laparoscopic liver resection: a meta analysis. Asian J Surg. 2019;42(1):19-31. doi: 10.1016/j.asjsur.2018.05.008 [DOI] [PubMed] [Google Scholar]
- 9.Hu Y, Guo K, Xu J, et al. Robotic versus laparoscopic hepatectomy for malignancy: a systematic review and meta-analysis. Asian J Surg. 2021;44(4):615-628. doi: 10.1016/j.asjsur.2020.12.016 [DOI] [PubMed] [Google Scholar]
- 10.Qiu J, Chen S, Chengyou D. A systematic review of robotic-assisted liver resection and meta-analysis of robotic versus laparoscopic hepatectomy for hepatic neoplasms. Surg Endosc. 2016;30(3):862-875. doi: 10.1007/s00464-015-4306-7 [DOI] [PubMed] [Google Scholar]
- 11.Montalti R, Berardi G, Patriti A, Vivarelli M, Troisi RI. Outcomes of robotic vs laparoscopic hepatectomy: a systematic review and meta-analysis. World J Gastroenterol. 2015;21(27):8441-8451. doi: 10.3748/wjg.v21.i27.8441 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Goh BKP, Lee SY, Koh YX, Kam JH, Chan CY. Minimally invasive major hepatectomies: a Southeast Asian single institution contemporary experience with its first 120 consecutive cases. ANZ J Surg. 2020;90(4):553-557. doi: 10.1111/ans.15563 [DOI] [PubMed] [Google Scholar]
- 13.Fagenson AM, Gleeson EM, Pitt HA, Lau KN. Minimally invasive hepatectomy in North America: laparoscopic versus robotic. J Gastrointest Surg. 2021;25(1):85-93. doi: 10.1007/s11605-020-04703-6 [DOI] [PubMed] [Google Scholar]
- 14.Lai EC, Tang CN. Long-term survival analysis of robotic versus conventional laparoscopic hepatectomy for hepatocellular carcinoma: a comparative study. Surg Laparosc Endosc Percutan Tech. 2016;26(2):162-166. doi: 10.1097/SLE.0000000000000254 [DOI] [PubMed] [Google Scholar]
- 15.Tsung A, Geller DA, Sukato DC, et al. Robotic versus laparoscopic hepatectomy: a matched comparison. Ann Surg. 2014;259(3):549-555. doi: 10.1097/SLA.0000000000000250 [DOI] [PubMed] [Google Scholar]
- 16.Chong CCN, Lok HT, Fung AKY, et al. Robotic versus laparoscopic hepatectomy: application of the difficulty scoring system. Surg Endosc. 2020;34(5):2000-2006. doi: 10.1007/s00464-019-06976-8 [DOI] [PubMed] [Google Scholar]
- 17.Strasberg SM. Nomenclature of hepatic anatomy and resections: a review of the Brisbane 2000 system. J Hepatobiliary Pancreat Surg. 2005;12(5):351-355. doi: 10.1007/s00534-005-0999-7 [DOI] [PubMed] [Google Scholar]
- 18.Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205-213. doi: 10.1097/01.sla.0000133083.54934.ae [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rahbari NN, Garden OJ, Padbury R, et al. Posthepatectomy liver failure: a definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery. 2011;149(5):713-724. doi: 10.1016/j.surg.2010.10.001 [DOI] [PubMed] [Google Scholar]
- 20.Wakabayashi G. What has changed after the Morioka consensus conference 2014 on laparoscopic liver resection? Hepatobiliary Surg Nutr. 2016;5(4):281-289. doi: 10.21037/hbsn.2016.03.03 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ban D, Tanabe M, Ito H, et al. A novel difficulty scoring system for laparoscopic liver resection. J Hepatobiliary Pancreat Sci. 2014;21(10):745-753. doi: 10.1002/jhbp.166 [DOI] [PubMed] [Google Scholar]
- 22.Chua D, Syn N, Koh YX, Goh BKP. Learning curves in minimally invasive hepatectomy: systematic review and meta-regression analysis. Br J Surg. 2021;108(4):351-358. doi: 10.1093/bjs/znaa118 [DOI] [PubMed] [Google Scholar]
- 23.Spampinato MG, Coratti A, Bianco L, et al. Perioperative outcomes of laparoscopic and robot-assisted major hepatectomies: an Italian multi-institutional comparative study. Surg Endosc. 2014;28(10):2973-2979. doi: 10.1007/s00464-014-3560-4 [DOI] [PubMed] [Google Scholar]
- 24.Fruscione M, Pickens R, Baker EH, et al. Robotic-assisted versus laparoscopic major liver resection: analysis of outcomes from a single center. HPB (Oxford). 2019;21(7):906-911. doi: 10.1016/j.hpb.2018.11.011 [DOI] [PubMed] [Google Scholar]
- 25.Kim JK, Park JS, Han DH, et al. Robotic versus laparoscopic left lateral sectionectomy of liver. Surg Endosc. 2016;30(11):4756-4764. doi: 10.1007/s00464-016-4803-3 [DOI] [PubMed] [Google Scholar]
- 26.Salloum C, Lim C, Lahat E, et al. Robotic-assisted versus laparoscopic left lateral sectionectomy: analysis of surgical outcomes and costs by a propensity score matched cohort study. World J Surg. 2017;41(2):516-524. doi: 10.1007/s00268-016-3736-2 [DOI] [PubMed] [Google Scholar]
- 27.Montalti R, Scuderi V, Patriti A, Vivarelli M, Troisi RI. Robotic versus laparoscopic resections of posterosuperior segments of the liver: a propensity score-matched comparison. Surg Endosc. 2016;30(3):1004-1013. doi: 10.1007/s00464-015-4284-9 [DOI] [PubMed] [Google Scholar]
- 28.Kamarajah SK, Bundred J, Manas D, Jiao L, Hilal MA, White SA. Robotic versus conventional laparoscopic liver resections: a systematic review and meta-analysis. Scand J Surg. 2021;110(3):290-300. [DOI] [PubMed] [Google Scholar]
- 29.Tomassini F, Scuderi V, Colman R, Vivarelli M, Montalti R, Troisi RI. The single surgeon learning curve of laparoscopic liver resection: a continuous evolving process through stepwise difficulties. Medicine (Baltimore). 2016;95(43):e5138. doi: 10.1097/MD.0000000000005138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Broering DC, Elsheikh Y, Alnemary Y, et al. Robotic versus open right lobe donor hepatectomy for adult living donor liver transplantation: a propensity score-matched analysis. Liver Transpl. 2020;26(11):1455-1464. doi: 10.1002/lt.25820 [DOI] [PubMed] [Google Scholar]
- 31.Ziogas IA, Giannis D, Esagian SM, Economopoulos KP, Tohme S, Geller DA. Laparoscopic versus robotic major hepatectomy: a systematic review and meta-analysis. Surg Endosc. 2021;35(2):524-535. doi: 10.1007/s00464-020-08008-2 [DOI] [PubMed] [Google Scholar]
- 32.Marino MV, Shabat G, Guarrasi D, Gulotta G, Komorowski AL. Comparative study of the initial experience in performing robotic and laparoscopic right hepatectomy with technical description of the robotic technique. Dig Surg. 2019;36(3):241-250. doi: 10.1159/000487686 [DOI] [PubMed] [Google Scholar]
- 33.Goh BKP, Prieto M, Syn N, et al. Validation and comparison of the Iwate, IMM, Southampton and Hasegawa difficulty scoring systems for primary laparoscopic hepatectomies. HPB (Oxford). 2021;23(5):770-776. doi: 10.1016/j.hpb.2020.09.015 [DOI] [PubMed] [Google Scholar]
- 34.Goh BK, Low TY, Teo JY, et al. Adoption of robotic liver, pancreatic and biliary surgery in Singapore: a single institution experience with its first 100 consecutive Cases. Ann Acad Med Singap. 2020;49(10):742-748. doi: 10.47102/annals-acadmedsg.202036 [DOI] [PubMed] [Google Scholar]
- 35.Wong DJ, Wong MJ, Choi GH, Wu YM, Lai PB, Goh BKP. Systematic review and meta-analysis of robotic versus open hepatectomy. ANZ J Surg. 2019;89(3):165-170. doi: 10.1111/ans.14690 [DOI] [PubMed] [Google Scholar]
- 36.Choi SH, Han DH, Lee JH, Choi Y, Lee JH, Choi GH. Safety and feasibility of robotic major hepatectomy for novice surgeons in robotic liver surgery: a prospective multicenter pilot study. Surg Oncol. 2020;35:39-46. doi: 10.1016/j.suronc.2020.07.003 [DOI] [PubMed] [Google Scholar]
- 37.Lee SJ, Lee JH, Lee YJ, et al. The feasibility of robotic left-side hepatectomy with comparison of laparoscopic and open approach: consecutive series of single surgeon. Int J Med Robot. 2019;15(2):e1982. doi: 10.1002/rcs.1982 [DOI] [PubMed] [Google Scholar]
- 38.Cipriani F, Fiorentini G, Magistri P, et al. Pure laparoscopic versus robotic liver resections: multicentric propensity score-based analysis with stratification according to difficulty scores. J Hepatobiliary Pancreat Sci. 2021. doi: 10.1002/jhbp.1022 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.