Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/wolters-kluwer-health/implementation-and-outcome-of-robotic-liver-surgery-in-the-netherlands-fe76RvXi5H
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Wolters Kluwer Health
- Copyright
- Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc.
- ISSN
- 0003-4932
- eISSN
- 1528-1140
- DOI
- 10.1097/sla.0000000000005600
- Publisher site
- See Article on Publisher Site
Abstract
ORIGINAL ARTICLE Implementation and Outcome of Robotic Liver Surgery in the Netherlands A Nationwide Analysis Burak Görgec, MD,*† Maurice Zwart, MD,*† Carolijn L. Nota, MD, PhD,‡ Okker D. Bijlstra, MD,§ Koop Bosscha, MD, PhD,∥ Marieke T. de Boer, MD, PhD,¶ Roeland F. de Wilde, MD, PhD,# Werner A. Draaisma, MD, PhD,∥ Michael F. Gerhards, MD, PhD,** Mike S. Liem, MD, PhD,†† Daan J. Lips, MD, PhD,†† Hendrik A. Marsman, MD, PhD,** J. Sven D. Mieog, MD, PhD,§ Quintus I. Molenaar, MD, PhD,‡ Maarten Nijkamp, MD, PhD,¶ Wouter W. Te Riele, MD, PhD,‡‡ Türkan Terkivatan, MD, PhD,# Alexander L. Vahrmeijer, MD, PhD,§ Marc G. Besselink, MD, MSc, PhD,*† Rutger-Jan Swijnenburg, MD, PhD,*†§§✉ Jeroen Hagendoorn, MD, PhD,‡✉ and for the Dutch Liver Collaborative Group complications was 7.0% (n = 27), median length of hospital stay 4 days Objective: To determine the nationwide implementation and surgical (interquartile range 2–5) and 30-day/in-hospital mortality 0.8% (n = 3). outcome of minor and major robotic liver surgery (RLS) and assess the The R0 resection rate was 83.2% (n = 263). Cumulative sum analysis first phase of implementation of RLS during the learning curve. for blood loss found a learning curve of at least 33 major RLS Background: RLS may be a valuable alternative to laparoscopic liver procedures. surgery. Nationwide population-based studies with data on imple- Conclusions: The nationwide use of RLS in the Netherlands has increased mentation and outcome of RLS are lacking. rapidly with currently one-tenth of all liver resections and one-fourth of Methods: Multicenter retrospective cohort study including consecutive all minimally invasive liver resections being performed robotically. patients who underwent RLS for all indications in 9 Dutch centers Although surgical outcomes of RLS in selected patient seem favorable, (August 2014–March 2021). Data on all liver resections were obtained future prospective studies should determine its added value. from the mandatory nationwide Dutch Hepato Biliary Audit (DHBA) Keywords: hepatic resection, implementation; liver resection, liver sur- including data from all 27 centers for liver surgery in the Netherlands. Outcomes were stratified for minor, technically major, and anatomically gery, minimally invasive liver resection; minimally invasive liver surgery, major RLS. Learning curve effect was assessed using cumulative sum nationwide analysis; robotic liver resection; robotic liver surgery analysis for blood loss. (Ann Surg 2023;277:e1269–e1277) Results: Of 9437 liver resections, 400 were RLS (4.2%) procedures including 207 minor (52.2%), 141 technically major (35.3%), and 52 anatomically major (13%). The nationwide use of RLS increased from 0.2% in 2014 to 11.9% in 2020. The proportion of RLS among all he use of minimally invasive liver surgery increased gradu- minimally invasive liver resections increased from 2% to 28%. Median T ally in the last 3 decades. Concerns about technical diffi- blood loss was 150 mL (interquartile range 50–350 mL] and the con- culties combined with long learning curves have hampered the version rate 6.3% (n = 25). The rate of Clavien-Dindo grade ≥ III adoption of minimally invasive liver surgery. A recent study From the *Amsterdam UMC, location University of Amsterdam, Department A.L.V., M.G.B., R.J.S., J.H. Analysis and interpretation of data: B.G., of Surgery, Amsterdam, the Netherlands; †Cancer Center Amsterdam, M.Z., M.G.B., R.J.S., J.H. Authors participated in drafting the article or Amsterdam, the Netherlands; ‡Department of Surgery, University Medical revising it critically for important intellectual content: B.G., M.Z., C.L.N., Center Utrecht, Utrecht, the Netherlands; §Department of Surgery, Leiden O.D.B., K.B., M.T.d.B., R.F.d.W., W.A.D., M.F.G., M.S.L., D.J.L., H. University Medical Center, Leiden, the Netherlands; ∥Department of A.M., J.S.D.M., Q.I.M., M.N., W.W.T.R., T.T., A.L.V., M.G.B., R.J.S., Surgery, Jeroen Bosch Hospital, ‘s-Hertogenbosch, the Netherlands; J.H. Authors given final approval of the version to be published: B.G., ¶Department of Surgery, University Medical Center Groningen, Gronin- M.Z., C.L.N., O.D.B., K.B., M.T.d.B., R.F.d.W., W.A.D., M.F.G., M.S. gen, the Netherlands; #Department of Surgery, Erasmus MC Cancer L., D.J.L., H.A.M., J.S.D.M., Q.I.M., M.N., W.W.T.R., T.T., A.L.V., M. Institute, Erasmus University Medical Center, Rotterdam, the Netherlands; G.B., R.J.S., J.H. **Department of Surgery, OLVG, Amsterdam, the Netherlands; The authors report no conflicts of interest. ††Department of Surgery, Medical Spectrum Twente, Enschede, the Supplemental Digital Content is available for this article. Direct URL citations Netherlands; ‡‡Department of Surgery, St. Antonius Hospital, Nieuwe- are provided in the HTML and PDF versions of this article on the journal’s gein, the Netherlands; and §§Amsterdam UMC, location Vrije Universiteit website, www.annalsofsurgery.com. Amsterdam, Department of Surgery, Amsterdam, the Netherlands. This is an open access article distributed under the Creative Commons ✉J.Hagendoorn-3@umcutrecht.nl; r.j.swijnenburg@amsterdamUMC.nl. Attribution License 4.0 (CCBY), which permits unrestricted use, dis- B.G. and M.Z. shared first authorship. tribution, and reproduction in any medium, provided the original work is R.J.S. and J.H. shared senior authorship. properly cited. Study conception and design: B.G., M.Z., M.G.B., R.J.S., J.H. Acquisition of Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. data: B.G., M.Z., C.L.N., O.D.B., K.B., M.T.d.B., R.F.d.W., W.A.D., M. ISSN: 0003-4932/23/27706-e1269 F.G., M.S.L., D.J.L., H.A.M., J.S.D.M., Q.I.M., M.N., W.W.T.R., T.T., DOI: 10.1097/SLA.0000000000005600 Annals of Surgery Volume 277, Number 6, June 2023 www.annalsofsurgery.com e1269 Görgec et al Annals of Surgery Volume 277, Number 6, June 2023 proposed a 3 phase model of learning curves including a com- Audit (DHBA) to assess nationwide implementation rates of petency (operative time, blood loss, and conversion), proficiency RLS. The DHBA is a nationwide prospective registry in which (morbidity, mortality, and hospital stay), and mastery phase all Dutch hospitals performing liver surgery are obliged to record (textbook- or benchmark outcomes). Nevertheless, laparoscopic all types of liver resections performed. liver surgery is currently available in most centers, with several Patients after RLS were initially identified using center reports showing its advantages compared with open liver sur- specific liver surgery databases and individual patient data were gery, including reduced intraoperative blood loss, less trans- extracted from the DHBA. fusions, fewer complications, and a shorter hospital length of Data of all consecutive patients who underwent RLS for 1,3–10 stay. According to the international Southampton guide- all indications between January 2014 and March 2021 were lines, laparoscopy is now seen as the standard for minor liver included. Patients were excluded when no formal resection was resections. Major laparoscopic liver resections should be performed (such as fenestration/deroofing of cysts and biopsies) implemented in a stepwise manner and combined with structured or in case of emergency surgery. training in centers who have completed the learning curve for 7,8,11–14 minor laparoscopic liver resections. A recent nationwide Definitions and Outcomes study in the Netherlands showed an overall good adherence to Minor liver resection was defined as any resection from this concept with a steady increase of the proportion of techni- the anterolateral segments, that is, 2, 3, 4b, 5, and 6. Anatomi- cally major and anatomically major laparoscopic liver resections cally major liver resection was defined as resection of 3 or more over the years. Couinaud’s segments. Technically major liver resection was Robotic liver surgery (RLS) represents the most recent defined as any resection from the posterosuperior segments, that evolution in the field of minimally invasive liver surgery and has is, 7, 8, 4a, and 1. The Kawaguchi difficulty scoring system was been suggested as a valuable alternative to laparoscopic liver calculated per patient and defines 3 groups of difficulty based on surgery. Perceived benefits of RLS include a better magnified the type of resection: group I (low) includes wedge resection and 3-dimensional view, articulating instruments, tremor filtration, left lateral sectionectomy, group II (intermediate) includes platform for image-guided surgery, ease of suturing, improved anterolateral segmentectomy and left hepatectomy, and group ergonomics, and better motion scaling, as compared with the III (high) includes posterosuperior segmentectomy, right poste- 16–19 laparoscopic approach. Still, the widespread diffusion of rior sectionectomy, right hepatectomy, central hepatectomy, and RLS is limited, potentially due to higher cost and suboptimal 27 extended left/right hepatectomy. availability of robotic systems. Several high-volume expert cen- Baseline characteristics consisted of age, sex, body mass ters have shown the potential advantages of RLS as compared 2 index (kg/m ), American Society of Anesthesiologists (ASA) with open liver surgery, whereas other centers are still exploring grade, Charlson Comorbidity Index, cirrhosis, neoadjuvant 20–24 the use of robotics for liver surgery. chemotherapy, previous extrahepatic abdominal surgery, pre- Recently, several Dutch centers have implemented RLS vious liver surgery, histologic diagnosis, number of lesions, size into their daily surgical practice. Nationwide population-based of largest lesion, distribution of lesions (ie, uni- or bilobar), and studies with data on both implementation and surgical outcome extent of resection. of RLS are lacking. The aim of this study is to determine the Surgical outcomes included intraoperative blood loss rate of implementation and surgical outcome of minor and (defined by the measured amount of suctioned blood with 20 mL major RLS on a nationwide scale and assess the first phase of increments), conversion to laparotomy, 30-day overall post- implementation of RLS including the learning curve. operative complications (defined according to the Clavien-Dindo 28 29 classification and Comprehensive Complication Index ), severe postoperative complications (defined as Clavien-Dindo METHODS ≥ 3 ), 30-day readmission, 30-day reoperation, postoperative A multicenter retrospective cohort study was performed length of hospital stay (LOS), R0 resection margin (ie, 1 mm or to provide insights in the implementation rates and surgical more tumor free margin), and 30-day or in-hospital mortality. outcome of RLS in the Netherlands. Data were gathered from all 9 liver surgical centers in the Netherlands with an RLS Survey program: Amsterdam UMC, Amsterdam; Erasmus University A short survey was developed using Google Forms Survey Medical Center, Rotterdam; Leiden University Medical Cen- (Google; Mountain View, CA) and was disseminated by email to ter, Leiden; University Medical Center Groningen, Groningen; all local study investigators of the nine participating centers OLVG, Amsterdam; Medical Spectrum Twente, Enschede; (Supplement 1, Supplemental Digital Content 1, http://links.lww. Jeroen Bosch Hospital, ‘s-Hertogenbosch; St. Antonius Hos- com/SLA/E50). The survey included questions with regards to pital, Nieuwegein; and University Medical Center Utrecht, form of individual training surgeons completed before start with Utrecht. their RLS program, surgical technique, intraoperative manage- The study was initiated by the Dutch Liver Collaborative ment, and case selection for RLS. Group (DLCG) and reported in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. All data were handled anonymously. Statistical Analysis Hence, the ethics committee of the Amsterdam UMC assessed that Data were analyzed using IBM SPSS Statistics for Win- the current study was not subject to the Medical Research Involving dows version 25.0 (SPSS Inc., Chicago, IL). Continuous, not Human Subjects Act and waived the need for informed consent. normally distributed variables were expressed as median with interquartile range (IQR). In case variables were normally dis- Data Source and Patient Selection tributed, they were reported as mean with SD. A Mann-Whitney Data on all liver resections including laparoscopic and U test was used to compare continuous, not normally distributed open liver resections in all 27 centers performing liver surgery in variables between groups. Normally distributed, continuous the Netherlands were obtained from the Dutch Hepato Biliary variables were compared using an independent samples t test. e1270 www.annalsofsurgery.com Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. Annals of Surgery Volume 277, Number 6, June 2023 Nationwide Implementation of Robotic Liver Surgery Categorical variables were reported as frequencies and pro- 7 (range 0–12), which increased to 18 (range 7–31) in 2020. The portions and compared between groups using a χ test. annual use of RLS in the nine centers increased from 0.5% in Outcomes were stratified into type of resection (minor, 2014 to 26% in 2020 (P < 0.001), whereas the annual use of technically major, and anatomically major RLS) to better laparoscopic liver surgery increased from 14.9% in 2014 to 26% understand outcome distribution. Trends in time were explored in 2017 before reducing to 13% in 2020 (Supplement 2, Sup- by dividing patients into the first 200 and second 200 RLS plemental Digital Content 1, http://links.lww.com/SLA/E50). procedures. A sensitivity analysis was performed by stratifying Overall, the use of minimally invasive liver surgery (combining patients according to the Kawaguchi difficulty scoring system RLS and laparoscopic liver surgery) in the 9 centers increased into 3 groups to be able to compare outcomes with previous from 15.4% (n = 95) in 2014 to 29.0% (n = 241) in 2020 literature. A subgroup analysis included stratification for centers (P < 0.001). where the leading console surgeon did nor did not complete a fellowship in minimally invasive liver surgery. A second sub- Surgeon Experience and Training group analysis included stratification for liver cirrhosis. Corre- In 7 of the 9 centers, experience was obtained with robotic lations were expressed in Spearman’s Rho with P value. abdominal surgery before (n = 4; 44%) or in parallel with (n = 3; The learning curve effects on blood loss and conversion 33%) the start of performing RLS. Experience in previous (phase of competency), and major morbidity and hospital stay robotic abdominal surgery in the 4 centers included robotic (phase of proficiency) were assessed with trends over consecutive cholecystectomy and colorectal resections. procedures per center with cumulative sum (CUSUM) analyses. The 9 leading robotic console surgeons from the partic- The learning curves were assessed overall and, if a significant ipating centers indicated that their training in RLS consisted of correlation was found, on multivariate analysis taking center and the Intuitive Surgical basic robotic surgery course (n = 7; 78%), minor or major resection into account. First, the patients were observership programs in an expert robotic liver surgical center ranked consecutively according to the date of their procedure (n = 7; 78%) ranging from 1 day to 6 weeks, proctorship pro- and the difference of the data to the mean per center was cal- grams in own center provided by an international proctor culated per case. Hereafter, the data were aggregated for all robotic liver surgeon (n = 5; 56%), fellowships in minimally centers with a weighting for the volume of resections and a invasive HPB surgery (n = 3; 33%) ranging from 1 to 2.5 years, CUSUM was presented on the y axis per case. The magnitude by and hands-on courses of 1 or 2 days in minimally invasive liver which the line ascends or descends is determined by the differ- surgery (n = 4; 44%). ence between the observed and expected outcome. For example, the line ascends when blood loss in that case was above average Patient Selection for RLS for that center by an amount relative to the SD, and for a case The nine leading console surgeons described several where blood loss was below average, the line descends. The top patient-related factors used by the individual center to select of the CUSUM graph thus represented the total blood loss in SD patients for RLS. Overall, reported selection factors for RLS above average up to that case. were absence of centrally located tumors (3 of 9 centers; 33%), no indication for technically or anatomically major liver resec- tion (3 of 9 centers; 33%), absence of major vascular or biliary RESULTS duct involvement (2 of 9 centers; 22%), small lesions (2 of 9 centers; 22%), absence of perihilar cholangiocarcinoma (3 of 9 Nationwide Implementation of RLS centers; 33%) or gallbladder carcinoma (1 of 9 centers; 11%), During the 7-year study period, a total of 9437 liver absence of liver cirrhosis (1 of 9 centers; 11%), no indication for resections were performed in 27 centers in the Netherlands an extended hemihepatectomy (2 of 9 centers; 22%), and absence including 2320 laparoscopic liver resections (24.6%) and 400 of extensive previous abdominal surgery (1 of 9 centers; 11%). RLS (4.2%) procedures. The RLS procedures were performed in The selection factors for RLS per center are presented in Sup- 9 of the 27 centers. The nationwide use of RLS per year plement 3, Supplemental Digital Content 1, http://links.lww. increased from 3 patients (0.2%) in 2014 to 158 patients (11.9%) com/SLA/E50. in 2020 (Fig. 1) (P < 0.001). Within the group of minimally invasive liver resections, the annual use of RLS increased from Surgical Technique 2% in 2014 to 28% in 2020 (P < 0.001) (Fig. 2). Figure 3 shows Surgical techniques among the participating centers were the annual volume of RLS categorized by type of resection. largely comparable for minor and major RLS and are displayed in Supplement 4, Supplemental Digital Content 1, http://links. Center Characteristics lww.com/SLA/E50. In 7 centers (78%), RLS was performed Five out of the 9 centers that performed RLS were uni- using the da Vinci Xi Robotic Surgical System (Intuitive versity medical centers, whereas 4 centers were large teaching Surgical , Inc., Sunnyvale, CA), whereas in 2 centers (22%), the hospitals. A total of 19 surgeons performed RLS with a median da Vinci X Robotic Surgical System was used. Specimens were of 2 surgeons (IQR 2-3) per center. A gradual implementation of extracted in a plastic endoscopic bag through a widened trocar RLS was observed. In 2014, the first Dutch RLS procedure was incision in case of small lesions, or a Pfannenstiel incision in case performed. In 2015, 2 more centers started performing RLS, of larger lesions or anatomically major resections. whereas the remaining 6 centers initiated an RLS program in 2018 (n = 4) and 2019 (n = 2). The overall median volume of liver Baseline Characteristics surgery per center (including open, laparoscopic, and robotic A total of 400 patients after RLS met the inclusion criteria resections) during the study period was 453 (IQR 255–545) liver and were included in the analysis. Baseline characteristics resections. The overall median volume of RLS per center was 35 including a stratification for type of resection are shown in (IQR 23–52) resections with an overall median implementation Supplement 5, Supplemental Digital Content 1, http://links.lww. rate per center of 10.2% ranging from 2.1% to 30.8% between com/SLA/E50. Median age was 64 years (IQR 53.0–72.0) and centers. The mean annual volume of RLS per center in 2018 was 179 patients (44.7%) were women. Most patients were an Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. www.annalsofsurgery.com e1271 Görgec et al Annals of Surgery Volume 277, Number 6, June 2023 FIGURE 1. Annual rate of open, laparoscopic, and robotic liver surgery in the Netherlands (2014–2020). American Society of Anesthesiologists 1 or 2 (n = 286; 72.2%) malignant lesions were observed in 333 patients (84.5%). Median with a median Charlson Comorbidity Index of 3 (IQR 2–5). lesion size was 27.0 mm (IQR, 17.0–43.0). Most patients Neoadjuvant chemotherapy was applied in 66 patients (16.5%), underwent a minor RLS (n = 207; 51.8%) followed by 141 whereas the majority of patients had undergone previous patients (35.3%) who underwent a technically major RLS and 52 abdominal surgery (n = 239; 61.4%). On histologic diagnosis, patients (13%) who underwent an anatomically major RLS. FIGURE 2. Annual proportion of robotic and laparoscopic liver surgery in all patients after minimally invasive liver surgery in the Netherlands (2014–2020). e1272 www.annalsofsurgery.com Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. Annals of Surgery Volume 277, Number 6, June 2023 Nationwide Implementation of Robotic Liver Surgery for fellowship training showed that although more technically and anatomically major resections were performed by fellow- ship-trained surgeons, outcomes were largely comparable except the rate of overall complications (10.8% vs 23.0%; P = 0.005) (Supplement 7, Supplemental Digital Content 1, http://links.lww. com/SLA/E50). Stratifying outcomes for liver cirrhosis showed that although outcomes tended to be inferior in patients with cirrhosis as compared with patients without cirrhosis, there we no significant differences between outcomes except mortality (6.9% vs 0.3%; P < 0.001) (Supplement 8, Supplemental Digital Content 1, http://links.lww.com/SLA/E50). Trends in Time Baseline characteristics and operative outcomes of the first and second 200 consecutive RLS procedures across the study period are shown in Tables 2 and 3, respectively. In the second 200 RLS procedure more patients received neoadjuvant chemotherapy (12.5% vs 20.5%; P = 0.031) and more patients had previous extrahepatic abdominal surgery (44.9% vs 55.5%; P = 0.038) as compared with the first 200 procedures. There was no significant difference in histologic diagnosis between both periods (P = 0.100). In the second 200 procedures, more technically major FIGURE 3. Annual volume of robotic liver resections stratified RLS procedures were performed (28.5% vs. 42.0%; P = 0.005). for type of resection. Comparing outcomes between the 2 periods stratified for type of resection showed that for all 3 difficulty groups, hospital stay was significantly shorter in the second period as compared Surgical Outcome with the first period. The remaining outcomes did not differ for Table 1 shows the surgical outcome after all RLS and the 3 difficulty groups. There was a decreasing trend for hospital stratified for minor, technically major, and anatomically major stay in each difficulty group (see Supplement 9, Supplemental RLS. The overall conversion rate was 6.3% and ranged from Digital Content 1, http://links.lww.com/SLA/E50). 3.9% in the minor RLS group to 9.6% in the anatomically major RLS group. Severe postoperative complications occurred in 27 patients (7.0%) of the total cohort (5.4% after minor, 7.4% after Learning Curve technically major, and 12.5% after anatomically major RLS). The blood loss and conversion learning curve was assessed The 3 main severe postoperative complications were bile leakage separately for minor and major resections. In the maturation of (1.3%), intra-abdominal abscess (1.3%) and hemorrhage (0.8%). experience for minor resections, blood loss remained consistent The overall Comprehensive Complication Index was 4.6 (12.3) and no decrease of outliers occurred (Rho =−0.031, P = 0.493). with 3.9 (10.2) in the minor RLS group, 4.2 (12.5) in the tech- In the maturation of the experience for major resections, blood nically major RLS group and 8.6 (18.2) in the anatomically loss diminished (Rho =−0.231, P < 0.001). Consequently, major RLS group. Overall hospital stay was 4 days (IQR 2–5d) CUSUM analysis of blood loss in major resections revealed an with a 30-day mortality rate of 0.8%. Radical resection (R0 inflection point at 33 procedures to a plateau phase till 44 RLS resection margin) in case of malignancy was achieved in 264 procedures, whereafter blood loss was consistently lower than patients (83.3%) varying from 136 patients (86.6%) in the minor average (see supplement 10, Supplemental Digital Content 1, RLS group to 26 patients (70.3%) in the anatomically major http://links.lww.com/SLA/E50). The rate of conversion dimin- RLS group. Outcomes stratified for the Kawaguchi difficulty ished from 8.3% over the first 10 procedures to 3.2% after the scoring system are shown in Supplement 6, Supplemental Digital first 30 consecutive procedures of all centers combined (Rho = Content 1, http://links.lww.com/SLA/E50. Stratifying outcomes −0.06, P = 0.184). The CUSUM curve showed a turning point at TABLE 1. Surgical Outcome After Robotic Liver Resection, Stratified for Type of Surgery All RLS Minor RLS Technically Major RLS Anatomically Major N= 400 N= 242 N= 141 RLS N= 52 Blood loss (mL), median (IQR) 150 (50–350) 100 (24–200) 200 (100–500) 300 (150–1200) Conversion to laparotomy 25 (6.3) 8 (3.9) 12 (8.5) 5 (9.6) Postoperative complications 76 (19.1) 37 (17.9) 23 (16.3) 16 (31.4) Severe postoperative complications 27 (7.0) 11 (5.4) 10 (7.4) 6 (12.5) CCI, mean (SD) 4.6 (12.3) 3.9 (10.2) 4.2 (12.5) 8.6 (18.2) Postoperative hospital stay (d), median 4(2–5) 3 (2–3) 3 (2–5) 6 (4–10) (IQR) Reoperation within 30 d 10 (2.6) 5 (2.5) 4 (2.9) 1 (2.1) Readmission within 30 d 12 (3.0) 4 (1.9) 5 (3.6) 3 (6.0) R0 resection in case of malignancy 264 (83.3) 136 (86.6) 102 (82.9) 26 (70.3) 30-d mortality 3 (0.8) 0 1 (0.7) 2 (3.9) Values in parentheses are percentages unless mentioned otherwise. CCI indicates Comprehensive Complication Index; mL, milliliter. Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. www.annalsofsurgery.com e1273 Görgec et al Annals of Surgery Volume 277, Number 6, June 2023 TABLE 2. Baseline Characteristics of the First and Second 200 Consecutive Patients After Robotic Liver Surgery The First 200 RLS N= 200 The Second 200 RLS N= 200 P Patient characteristics Age, years, median (IQR) 64.0 (50.3–71.0) 64.0 (55.3–73.0) 0.293 Sex, male (%) 105 (52.5) 116 (58.0) 0.269 BMI, kg/m , median (IQR) 26.5 (23.0–30.6) 25.4 (23.0–29.0) 0.067 ASA grade 0.133 ASA 1 (%) 13 (6.6) 20 (10.1) ASA 2 (%) 137 (69.5) 116 (58.3) ASA 3 (%) 46 (23.4) 62 (31.2) ASA 4 (%) 1 (0.5) 1 (0.5) Charlston Comorbidity Index, median (IQR) 3 (2–5) 3 (1–6) 0.970 Neoadjuvant chemotherapy (%) 25 (12.5) 41 (20.5) 0.031 Cirrhosis (%) 17 (8.5) 12 (6.0) 0.335 Previous extrahepatic abdominal surgery (%) 84 (44.9) 111 (55.5) 0.038 Previous liver surgery (%) 16 (8.0) 28 (14.0) 0.057 Tumor characteristics Histologic diagnosis 0.100 CRLM (%) 98 (49.0) 120 (61.5) HCC (%) 36 (18.0) 18 (9.2) Cholangiocarcinoma (%) 6 (3.0) 7 (3.6) Gallbladder carcinoma (%) 1 (0.5) 2 (1.0) Non-CRLM (%) 11 (5.5) 12 (6.2) Other malignancy (%) 12 (6.0) 11 (5.6) Benign (%) 36 (18.0) 25 (12.5) Number of lesions, median (IQR) 1 (1–2) 1 (1–2) 0.877 Size of largest lesion, mm, median (IQR) 31.0 (20.0–50.0) 25.0 (15.0–40.0) 0.006 Distribution of lesions 0.105 Unilobar (%) 151 (89.3) 167 (83.5) Bilobar (%) 18 (10.7) 33 (16.5) Procedure characteristics Type of resection 0.005 Minor (%) 109 (54.4) 98 (49.0) Technically major (%) 57 (28.5) 84 (42.0) Anatomically major (%) 34 (17.0) 18 (9.0) Extent of resection 0.004 Wedge (%) 69 (34.5) 108 (54.0) Segmentectomy (%) 43 (21.5) 34 (17.0) Bisegmentectomy (%) 54 (27.5) 40 (20.0) Trisegmentectomy 4 (2.0) 1 (0.5) Left hemihepatectomy (%) 13 (6.5) 5 (2.5) Right hemihepactectomy (%) 15 (7.5) 11 (5.5) Extended right hemihepatectomy (%) 2 (1.0) — Other anatomically major (%) — 1 (0.5) Values in parentheses are percentages unless mentioned otherwise. Percentages may not add up due to rounding and missing data. Bold values indicate statistically significant P < 0.005. BMI indicates body mass index; CRLM, colorectal liver metastasis; HCC, hepatocellular carcinoma. 21 procedures for minor resections and 42 procedures for major centers with previous laparoscopic liver surgery experience and a resections (8% conversion vs 4% conversion, P = 0.074). learning curve of 35 major robotic liver resections in centers without There was no significant correlation between consecutive any previous laparoscopic liver surgery experience. procedures and severe complication (Rho =−0.071, P = 0.404) or mortality (Rho = 0.077, P = 0.063) and the CUSUM analysis DISCUSSION was inconclusive. There was a significant decrease in-hospital stay (Rho = This first nationwide retrospective study on RLS found a −0.091, P = 0.036), which remained consistent (P = 0.014) after remarkable increase in the use of RLS in the Netherlands from multivariate analysis adjusting for center and type of resection, 0.2% to 11.9% among all liver resections. Within the group of P = 0.318, P = 0.468, respectively). CUSUM analysis revealed an minimally invasive liver resections, there has been a substantial turning point after 19 procedures for minor resections and after movement toward robotics, with RLS accounting for over one- 47 for major resection. Hospital stay was significantly reduced fourth of all minimally invasive liver resections in 2020. Intra- after the turning points P = 0.043, median 4 [3–6] versus 3 [2–5] and postoperative outcome in these selected patients seem days (see supplement 11, Supplemental Digital Content 1, http:// promising with a conversion rate of 6.3%, severe complication links.lww.com/SLA/E50). rate of 7.0%, 30-day/in-hospital mortality of 0.8%, and a radical Center specific CUSUM analyses of blood loss in major oncological resection status of 83.2%. With CUSUM analysis, a resections stratified for previous laparoscopic liver surgery experi- learning curve of at least 33 procedures for major RLS was ence showed a learning curve of 33 major robotic liver resections in demonstrated. e1274 www.annalsofsurgery.com Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. Annals of Surgery Volume 277, Number 6, June 2023 Nationwide Implementation of Robotic Liver Surgery Nationwide retrospective studies on the implementation and outcomes of RLS are currently lacking and only a few ret- rospective multicenter studies have been performed. A recent multicenter study (2016–2018) investigated the outcomes of RLS as compared with open and laparoscopic liver surgery in 28 states in the United States including 351 RLS procedures with a 3.1% use of RLS. The authors reported an overall complica- tion rate of 7.2%, a mean hospital stay of 2 days and a mortality of 0.9%. Although in the current study the overall morbidity was higher and hospital stay longer, the authors did not stratify postoperative outcomes for type of resection. In addition, data regarding blood loss and conversion rates were missing. A sec- ond multicenter retrospective study from 5 Italian centers com- pared the outcome of RLS with laparoscopic liver surgery using propensity score matching and included 403 RLS procedures. With 12.9% major resections, a conversion rate of 4.8% to 7.9%, severe morbidity of 3.0% to 8.5% and mortality of 0.5% depending on the Kawaguchi difficulty level of the RLS proce- dure, their results are comparable with the outcomes of the current study. Of note, this study did not focus on nationwide implementation and only included high-volume expert centers. The 7% rate of severe postoperative complications in the current study was lower than the 11.1% reported in the overall DHBA cohort. Of note, outcomes in that study were not stratified for surgical approach. The R0 resection margin in the current study was lower than reported in previous studies, especially for ana- 30,31 tomically major liver resections. Also, the rate of R0 resec- tion rate in the current robotic cohort was lower than the overall R0 resection rate in the DHBA, regardless of surgical approach. It has been suggested that in certain patients with colorectal liver metastases and hepatocellular carcinoma, R1 resection may be inevitable and should not be considered directly as a technically error, especially in case of R1 vascular 34–36 resection. Nevertheless, the R0 rate in the current study clearly shows that anatomically major robotic liver resections are still technically demanding, and a thorough case selection should be followed until the learning curve of major RLS is reached. Although outcomes of the current study imply a safe and efficient application of RLS, it is important to interpret these results with respect to outcomes of laparoscopic liver surgery, which is currently considered the standard of care and the most used minimally invasive technique in liver surgery. A previous Dutch population-based study reported an increased use of laparoscopic liver surgery from 6% to 23% in 885 patients from 20 centers between 2011 and 2016. In contrast, building on the laparoscopic liver surgery experience, the nationwide use of RLS in the current study increased in an even faster pace. The con- version rate of 13%, median hospital stay of 5 days, severe morbidity of 8.6% and overall mortality of 1% with a R0 resection status of 89.4% in that previous study was rather comparable with the current outcomes of RLS. In addition, although selection criteria for laparoscopic liver resection in the Netherlands during implementation were limited to no need for vascular or biliary reconstruction and no need for a simultaneous anatomically major liver and colorectal resection, selection cri- teria for RLS as reported by the leading surgeons included the absence of centrally located lesions or indication for technically or anatomically major resection. We speculate that this careful inclusion may reflect the initial phase of adopting a new tech- nique. Another analysis of 1131 patients after laparoscopic liver resection in 272 US centers reported a postoperative complica- tion rate of 38.1%, mortality rate of 2.8%, and length of hospital stay of 5 days. Of note, outcomes from both laparoscopic liver surgery studies may seem to suggest the noninferiority of RLS as Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. www.annalsofsurgery.com e1275 TABLE 3. Surgical Outcome After Robotic Liver Resections Stratified for Type of Surgery and for the First and Second 200 Consecutive Patients Minor RLS Technically Major RLS Anatomically Major RLS First 200 N= 109 Second 200 N= 98 P First 200 N= 57 Second 200 N= 84 P First 200 N= 34 Second 200 N= 18 P Blood loss (mL), median[IQR] 100 (20–165) 100 (30–200) 0.333 300 (100–525) 200 (95–513) 0.620 300 (150–1250) 400 (225–850) 0.829 Conversion to laparotomy 5 (4.6) 3 (3.1) 0.570 5 (8.8) 7 (8.3) 0.927 4 (11.8) 1 (5.6) 0.470 Postoperative complications 24(22.0) 13 (13.3) 0.101 14 (24.6) 9 ((11.0) 0.034 11 (32.4) 5 (29.4) 0.830 Severe postoperative complications 7 (6.7) 4 (4.1) 0.407 6 (11.1) 4 (4.9) 0.173 4 (12.9) 2 (11.8) 0.909 CCI, mean (SD) 4.5 (11.1) 3.2 (9.1) 0.351 5.1 (11.7) 3.5 (13.1) 0.474 6.9 (13.2) 11.5 (25.2) 0.407 Postoperative hospital stay (d), median[IQR] 4 (3–5) 3 (2–4) < 0.001 4(3–6) 3 (1.8–4.3) < 0.001 7 (5.8–12) 4 (4–6) 0.002 Reoperation within 30 d 4 (3.8) 1 (1.0) 0.196 2 (3.7) 2 (2.4) 0.669 1 (3.2) 0 0.454 Readmission within 30 d 3 (2.8) 1 (1.0) 0.366 2 (3.5) 3 (3.7) 0.963 2 (6.1) 1 (5.9) 0.980 R0 resection in case of malignancy 65 (83.3) 71 (89.9) 0.089 43 (84.3) 59 (81.9) 0.731 18 (72.0) 8 (66.7) 0.740 30-d mortality 0 0 1 0 1 (1.2) 0.403 1 (2.9) 1 (5.9) 0.610 Values in parentheses are percentages unless mentioned otherwise. Values in bold are considered statistically significant (P < 0.05). CCI indicates Comprehensive Complication Index; mL, milliliter. Görgec et al Annals of Surgery Volume 277, Number 6, June 2023 compared with the laparoscopic approach, yet differences in peak at 33 RLS procedures followed by a second, final peak at patient populations might still be present. Although several 44 procedures. Furthermore, conversion diminished to 4%, 31,38 comparative studies have been published, future large series although this difference was not significant. Most likely this can and randomized trials comparing RLS and laparoscopic liver be attributed to the low event rate that comes with the low rate surgery are needed to determine the added value of RLS in the of conversion to begin with compared with laparoscopic liver current minimally invasive liver surgery practice. resections. The proficiency learning phase was reached after 19 Despite comparable outcomes of RLS and laparoscopic and 47 procedures for minor and major resections, respectively. liver surgery in the current literature, the robotic technique might Our results were within range of published data where 60 have several potential advantages as compared with the lapa- procedures were required for major RLS. Apart from the roscopic technique. Robotic instruments have an increased interpretation of the CUSUM analysis for blood loss, we dexterity as compared with the conventional rigid laparoscopic acknowledge that a proficiency learning curve could change instruments facilitating posterosuperior and major resections. when the entire cohort completes more than 100 inclusions or Another major benefit of the robotic system is the ability to build when we could acquire data on operative times. an interactive visual interface, rather than a basic operating field, The current study has several limitations. First, the retro- using customized software in which surgeons may be assisted by spective design might be accompanied with an inevitable risk of preoperative and/or intraoperative imaging such as intra- selection bias. Eligible patients have been selected specifically for operative ultrasound and Indocyanine green fluorescence imag- RLS. Second, the annual volume of RLS per center was rather low ing during parenchymal resection. The robotic system may also over the study period, especially between 2014 and 2017. This ensure a fine and safe dissection of the hepatic pedicle through its shows that the adoption of RLS in the Netherlands is probably still delicate movements and endowrist instruments, allowing it to in the early phase with acceptable outcomes and further extension reach the hilum and the portal bifurcation easier, especially might be observed in the next years. However, comparative data during anatomically major liver resection. Furthermore, from other countries are lacking as this is the first nationwide series robotic surgery is associated with less physical discomfort on RLS. Third, the current study used DHBA data with a fixed including fatigue as compared with open and laparoscopic sur- selection of collected variables. Specific information concerning gery, given postural differences during laparoscopic, open and operative time, intraoperative incidents, reason for conversion, robotic surgery. The main drawbacks of the robotic approach cause of mortality were not registered in the DHBA and therefore include the higher cost and suboptimal availability of robotic could not be reported in the current study. Furthermore, data on systems as compared with laparoscopy. Results of the current postoperative complications and mortality in the DHBA are reg- study suggest that outcomes of RLS including postoperative istered during hospital stay and (in case of earlier discharge) up complications, length of hospital stay, reoperations, and read- until 30 days after surgery instead of 90 days. Our results should be missions were favorable. These results imply that postoperative interpreted carefully as assessing postoperative outcomes just at costs of liver resection could be lowered with the robotic 30 days might miss a high number of major complications and approach. Also, with multiple new surgical robot devices cur- deaths as compared with 90 days. Fourth, no data were available rently developed, the hurdle to access and the costs of acquiring on costs of RLS, whereas previous studies highlighted its higher and maintaining a robotic system are expected to be lower. costs as compared with laparoscopic liver surgery. These data are These benefits are likely to enhance the use of RLS globally, not needed although to determine cost-effectiveness of RLS. Fifth, the just in high-volume centers but also in low-volume centers, but learning curves were assessed per center instead of per surgeon as require confirmation in randomized trials of robotic versus lap- specific data on which surgeon from each center performed the aroscopic liver surgery. procedures was not available. Sixth, there may be some variation in Interestingly, the survey results in the current study pre-, intra-, and postoperative management strategies at the indi- demonstrated that among robotic liver surgeons in the Neth- vidual centers. Such data may be valuable to gain more insights in erlands, there is a considerable variation in training they RLS outcome. However, the survey results demonstrated that completed before the start of their RLS program. This varia- surgical technique use for RLS was largely comparable. tion underlines the lack of a tailored and structured training program for RLS on national and international scale and, CONCLUSIONS subsequently, the lack of standardized minimal requirements for a surgeon to initiate a RLS program in a center, as pre- The present study showed that the nationwide imple- viously described. In the Netherlands, several training pro- mentation of RLS in the Netherlands has increased rapidly with grams for minimally invasive laparoscopic and robotic pan- currently one-tenth of all liver resections and one-fourth of all creatic surgery (LAELAPS-1, 2, 3) were shown to be feasible minimally invasive liver resections being performed robotically. 42–44 and effective. Similarly, in the Netherlands, a training RLS seems to be safe with promising outcomes for minor, program for both technically and anatomically major laparo- technically major, and anatomically major RLS. Nevertheless, scopic liver surgery (LAELIVE) was initiated including RLS remains complex and technically demanding and may detailed technique description and proctoring on-site. The benefit from the initiation of a structured and tailored (inter) national training program. Future large series or randomized aforementioned training programs support the feasibility and trials comparing RLS and laparoscopic liver surgery are needed effectiveness of uniform and structured training programs in to determine the added value of RLS in the field of minimally the field of minimally invasive HPB surgery and support the invasive liver surgery. design of an international training program in RLS. To our current understanding, this report is the only ACKNOWLEDGMENTS multicenter study investigating learning curves stratified for minor and major RLS. In trend analysis, the mean blood loss The authors thank the Dutch Institute for Clinical Auditing went down significantly for major resections with a range below (DICA) and the Dutch Hepato Biliary Audit (DHBA) for reg- 1000 mL after 40 to 50 consecutive procedures. On visual istering and providing the data on liver resections in the inspection of the blood loss CUSUM analysis, there is a clear Netherlands. e1276 www.annalsofsurgery.com Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. Annals of Surgery Volume 277, Number 6, June 2023 Nationwide Implementation of Robotic Liver Surgery outcomes and costs between the robotic and open technique in a single REFERENCES institution. J Laparoendosc Adv Surg Tech. 2017;27:375–382. 1. Nguyen KT, Gamblin TC, Geller DA. World review of laparoscopic liver 24. Wong DJ, Wong MJ, Choi GH, et al. Systematic review and meta- resection—2,804 patients. Ann Surg. 2009;250:831–841. analysis of robotic versus open hepatectomy. ANZ J Surg. 2019;89: 2. Müller PC, Kuemmerli C, Cizmic A, et al. Learning curves in open, 165–170. laparoscopic, and robotic pancreatic surgery. Ann Surg Open. 2022;3: 25. Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the e111. Reporting of Observational Studies in Epidemiology (STROBE): 3. Mirnezami R, Mirnezami AH, Chandrakumaran K, et al. Short- and Explanation and elaboration. Int J Surg. 2014;12:1500–1524. long-term outcomes after laparoscopic and open hepatic resection: 26. Strasberg SM. Nomenclature of hepatic anatomy and resections: a review of systematic review and meta-analysis. HPB. 2011;13:295–308. the Brisbane 2000 system. J Hepatobiliary Pancreat Surg. 2005;12:351–355. 4. Tzanis D, Shivathirthan N, Laurent A, et al. European experience of 27. Kawaguchi Y, Fuks D, Kokudo N, et al. Difficulty of laparoscopic liver laparoscopic major hepatectomy. J Hepatobiliary Pancreat Sci. resection: proposal for a new classification. Ann Surg. 2018;267:13–17. 2013;20:120–124. 28. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo 5. Martínez-Cecilia D, Cipriani F, Vishal S, et al. Laparoscopic versus open classification of surgical complications. Ann Surg. 2009;250:187–196. liver resection for colorectal metastases in elderly and octogenarian 29. Slankamenac K, Graf R, Barkun J, et al. The comprehensive patients. Ann Surg. 2017;265:1192–1200. complication index: a novel continuous scale to measure surgical 6. Ding G, Cai W, Qin M. Pure laparoscopic versus open liver resection in morbidity. Ann Surg. 2013;258:1–7. treatment of hepatolithiasis within the left lobes. Surg Laparosc Endosc Percutan Tech. 2015;25:392–394. 30. Aziz H, Wang JC, Genyk Y, et al. Comprehensive analysis of laparoscopic, robotic, and open hepatectomy outcomes using the 7. Ciria R, Cherqui D, Geller DA, et al. Comparative short-term benefits of nationwide readmissions database. J Robot Surg. 2022;16:401–407. laparoscopic liver resection: 9000 cases and climbing. Ann Surg. 31. Cipriani F, Fiorentini G, Magistri P, et al. Pure laparoscopic versus 2016;263:761–777. robotic liver resections: Multicentric propensity score-based analysis with 8. Kasai M, Cipriani F, Gayet B, et al. Laparoscopic versus open major stratification according to difficulty scores. J Hepatobiliary Pancreat Sci. hepatectomy: a systematic review and meta-analysis of individual patient 2021:1–16. data. Surgery. 2018;163:985–995. 32. Elfrink AKE, Kok NFM, den Dulk M, et al. Short-term postoperative 9. Fretland ÅA, Dagenborg VJ, Bjørnelv GMW, et al. Laparoscopic versus outcomes after liver resection in the elderly patient: a nationwide open resection for colorectal liver metastases. Ann Surg. population-based study. HPB. 2021;23:1506–1517. 2018;267:199–207. 33. van der Werf LR, Kok NFM, Buis CI, et al. Implementation and first results 10. Ratti F, Fiorentini G, Cipriani F, et al. Laparoscopic vs open surgery for of a mandatory, nationwide audit on liver surgery. HPB. 2019;21:1400–1410. colorectal liver metastases. JAMA Surg. 2018;153:1028–1035. 34. Viganò L, Procopio F, Cimino MM, et al. Is tumor detachment from 11. Abu Hilal M, Aldrighetti L, Dagher I, et al. The Southampton Consensus vascular structures equivalent to R0 resection in surgery for colorectal Guidelines for Laparoscopic Liver Surgery: from indication to imple- liver metastases? an observational cohort. Ann Surg Oncol. mentation. Ann Surg. 2018;268:11–18. 2016;23:1352–1360. 12. Takahara T, Wakabayashi G, Konno H, et al. Comparison of 35. Viganò L, Costa G, Cimino MM, et al. R1 resection for laparoscopic major hepatectomy with propensity score matched open colorectal liver metastases: a survey questioning surgeons about its cases from the National Clinical Database in Japan. J Hepatobiliary incidence, clinical impact, and management. J Gastrointest Surg. Pancreat Sci. 2016;23:721–734. 2018;22:1752–1763. 13. Abu Hilal M, Di Fabio F, Teng MJ, et al. Single-centre comparative 36. Donadon M, Terrone A, Procopio F, et al. Is R1 vascular hepatectomy study of laparoscopic versus open right hepatectomy. J Gastrointest Surg. for hepatocellular carcinoma oncologically adequate? Analysis of 327 2011;15:818–823. consecutive patients. Surgery. 2019;165:897–904. 14. Cipriani F, Ratti F, Cardella A, et al. Laparoscopic versus open major 37. He J, Amini N, Spolverato G, et al. National trends with a laparoscopic hepatectomy: analysis of clinical outcomes and cost effectiveness in a liver resection: results from a population based analysis. HPB. high-volume center. J Gastrointest Surg. 2019;23:2163–2173. 2015;17:919–926. 15. van der Poel MJ, Fichtinger RS, Bemelmans M, et al. Implementation 38. Ciria R, Berardi G, Alconchel F, et al. The impact of robotics in and outcome of minor and major minimally invasive liver surgery in the liver surgery: a worldwide systematic review and short-term outcomes meta- Netherlands. HPB. 2019;21:1734–1743. analysis on 2,728 cases. J Hepatobiliary Pancreat Sci. 2022;29:181–197. 16. Troisi RI, Pegoraro F, Giglio MC, et al. Robotic approach to the liver: 39. Giannone F, Felli E, Cherkaoui Z, et al. Augmented reality and image- open surgery in a closed abdomen or laparoscopic surgery with technical guided robotic liver surgery. Cancers (Basel). 2021;13:6268. constraints? Surg Oncol. 2020;33:239–248. 40. Wee IJY, Kuo LJ, Ngu JCY. A systematic review of the true benefit of 17. Sucandy I, Schlosser S, Bourdeau T, et al. Robotic hepatectomy for robotic surgery: Ergonomics. Int J Med Robot Comput Assist Surg. benign and malignant liver tumors. J Robot Surg. 2020;14:75–80. 2020;16:e2113. 18. Beard RE, Khan S, Troisi RI, et al. Long-term and oncologic outcomes 41. Zwart MJW, Görgec B, Arabiyat A, et al. Pan-European survey on the of robotic versus laparoscopic liver resection for metastatic colorectal implementation of robotic and laparoscopic minimally invasive liver cancer: a multicenter, propensity score matching analysis. World J Surg. surgery. Hpb. 2022;24:322–331. 2020;44:887–895. 42. De Rooij T, van Hilst J, Topal B, et al. Outcomes of a Multicenter Training 19. Wang J-M, Li J-F, Yuan G-D, et al. Robot-assisted versus laparoscopic Program in Laparoscopic Pancreatoduodenectomy (LAELAPS-2). Ann minor hepatectomy: a systematic review and meta-analysis. Medicine Surg. 2019;269:344–350. (Baltimore). 2021;100:e25648. 43. De Rooij T, Van Hilst J, Boerma D, et al. Impact of a nationwide 20. Nota CL, Woo Y, Raoof M, et al. Robotic versus open minor liver training program in minimally invasive distal pancreatectomy (LAE- resections of the posterosuperior segments: a multinational, propensity LAPS). Ann Surg. 2016;264:754–762. score-matched study. Ann Surg Oncol. 2019;26:583–590. 44. Zwart MJW, Nota CLM, de Rooij T, et al. Outcomes of a Multicenter 21. Croner RS, Perrakis A, Hohenberger W, et al. Robotic liver surgery for Training Program in Robotic Pancreatoduodenectomy (LAELAPS-3). minor hepatic resections: a comparison with laparoscopic and open Ann Surg. 2022;276:e886–e895. standard procedures. Langenbeck’s Arch Surg. 2016;401:707–714. 22. Chen PD, Wu CY, Hu RH, et al. Robotic versus open hepatectomy for 45. Chua D, Syn N, Koh Y-X, et al. Learning curves in minimally invasive hepatocellular carcinoma: a matched comparison. Ann Surg Oncol. hepatectomy: systematic review and meta-regression analysis. Br J Surg. 2017;24:1021–1028. 2021;108:351–358. 23. Daskalaki D, Gonzalez-Heredia R, Brown M, et al. Financial impact of 46. Chen PD, Wu CY, Hu RH, et al. Robotic major hepatectomy: Is there a the robotic approach in liver surgery: a comparative study of clinical learning curve? Surg (United States). 2017;161:642–649. Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. www.annalsofsurgery.com e1277
Journal
Annals of Surgery – Wolters Kluwer Health
Published: Jun 18, 2023