Robotic Liver Resections: Application of Difficulty Score Systems to an Initial Experience. Is a Specific Robotic Difficulty Score Necessary?

Abstract

Background:

Recently, the minimally invasive surgical approach has been available for performing liver resections (LRs) with laparoscopic and robotic techniques. The robotic approach for LRs seems to overcome several laparoscopic limitations, which is a valid alternative when performed in high volume and specialized centers. Laparoscopic difficulty score systems (DSSs) should serve to guide the surgeon's choice in the best surgical approach to adopt for every single patient, giving the possibility to switch to the open approach when needed. To this day, no specific robotic difficulty scores exist. The aim of our study was to verify the feasibility of applying these scores and related updates on robotic LRs performed in our Institute.

Materials and Methods:

Out of a total of 683 LRs performed from June 2010 to July 2019, 60 were performed through using a mini invasive approach and of these 18 were performed robotically. The Ban DSS and subsequently the modified Iwate DSS were applied to our cases.

Results:

Based on our findings, applying the DSS we divided our series into two groups: a low difficulty level group (1–3) made up of 5 patients, and an intermediate difficulty level group (4–6) consisting of 13 patients. Average Ban DSS and subsequently updated score system results were 4.6 ± 1.5 points (range 2–6) for both scores.

Conclusions:

Difficulties were encountered in applying the score when simultaneous multiple wedge resections were performed. The laparoscopic DSS is applicable to robotic LRs with some limitations due to the peculiarity of the two different minimally invasive approaches. A specific robotic difficulty rating score could be necessary to include these elements.

Introduction

Hepatobiliary procedures remain to be highly complex surgical operations with consistent morbidity and mortality rates. Minimally invasive liver resection (LR) began to be available from 1991, when Reich et al. performed and described the first laparoscopic-assisted surgery.¹ The new robotic approach seems to exceed laparoscopic limitations, representing a valid alternative when performed in selected patients and in high-volume centers,² since it combines the minimal invasiveness of laparoscopy with reduced hand tremor and improved imaging resolution.³ Last but not least, robotic surgery could be considered less strenuous for surgeons as well as less physically demanding.⁴ Recently, Ban et al.⁵ suggested to adopt difficulty score systems (DSSs) to employ the best surgical approach (open or laparoscopic) and maximize curative chances in LRs. The score was subsequently modified in 2016 into the IWATE DSS.⁶ The DSS was assessed to preoperatively predict possible difficulties of planned laparoscopic LRs.

In this study, we reviewed our initial experience with hepatic robotic surgery and verified the possible application of these scores and their relative updates in our robotic activity.

Materials and Methods

A retrospective analysis from a prospectively collected database of LRs carried out from June 2010 to July 2019 was performed.

Patients were evaluated for surgery after an extensive diagnostic work up to define the precise staging of their own disease. Indication for robotic surgery was placed after the evaluation of the following patient characteristics: American Society of Anesthesiology (ASA) score, comorbidity, pneumoperitoneum toleration, prior abdominal surgery, size, position, and number of nodules to be removed. Furthermore, a significant factor was the complete availability of the robot equipment that at our institute shares with gynecologists, urologists, thoracic surgeons, and otorhinolaryngologist surgeons on a monthly basis. Each patient was informed on the planned procedure, who then eventually signed an informed consent.

Patients' preoperative liver function was evaluated using Child-Pugh score,^7,8 the model for end-stage liver disease (MELD) score,⁹ and the indocyanine green retention rate at 15 minutes (ICG-R15) and clearance (ICG-CL).¹⁰

Intraoperative ultrasound (IOUS) was performed with the laparoscopic probe (Esaote^®, Italy LP323 laparoscopic probe).

Laparoscopic IOUS was performed in all cases before robot docking. All patients underwent LR using the da Vinci^® Si surgical system (Intuitive Surgical, Inc., Sunnyvale, CA). The camera port was usually placed through a small umbilical incision. The remaining three robotic trocars were placed appropriately on the type of LR to be performed. An additional 10 mm port was placed for laparoscopic access during the resection. A further skin incision was made for the extracorporeal Pringle maneuver with inserting the pleural tube. Transection of the parenchyma was usually performed with the use of monopolar scissors that were always placed in the right arm of the robot at the start of surgery. Hemostasis was reached with the use of bipolar forceps, usually placed in the left robotic arm. The Harmonic scalpel was not used as routine. Hepatic hilum was isolated and a tourniquet was predisposed to eventually perform extracorporeal Pringle maneuver. When performed, clamping of the pedicle was carried out for 15 minutes, and was delayed 5 minutes for warm reperfusion. Surgical specimens were extracted through a Pfannenstiel incision or a small subcostal incision, based on the characteristics of the patient.

The definition of LRs was based on the Brisbane classification.¹¹ For each surgical procedure the DSS,⁵ called BanDSS (BDSS), was calculated retrospectively. The subsequently modified IWATE score could not be different, since the hand-assisted variation could not be applied in robotic surgery. In case of multiple simultaneous resections, we applied the DSS to the nodule located in the more difficult segment or to the greater one in case of nodules positioned in the same segment.

The following patient data were collected: age, sex, body mass index (BMI), ASA score, liver function, diagnosis, nodule number, size, location, surgical procedure, Pringle maneuver; operation time, conversion to open surgery, presence of drain, postoperative complications, and hospital stay; and reintervention rate. Postoperative complications were ranked according to the Clavien–Dindo classification.¹²

Statistical analysis

Data are presented in the descriptive statistics reporting median with standard deviation and range percentages. Due to the small sample size, Fisher-test and unpaired t-test were used to assess statistical significance with a cutoff rate set at the .05 P-value probability level. The SPSS v20.00 (SPSS, Inc., Chicago, IL) was used when needed.

Results

Patient demographics and preoperative findings

Out of a total of 683 LRs performed from June 2010 to July 2019, 60 (8.8%) were performed through a miniinvasive approach and of these 18 (30%) were performed robotically. They are the subject of the present report. The first five cases were performed in 2011 and the others started from 2018 up to July 2019 after an interval time during which robotic surgery was not available for hepatobiliary surgery in our institute (Figure 1). Patient characteristics and indications for surgery are reported in Table 1. As a total, 18 patients were submitted to robotic surgery, 10 (55.6%) male and 8 (44.4%) female, the average age was 55.3 ± 17.4 years, and average BMI was 25.2 ± 4.6 (range 19.1–38.8). The preoperative assessment identified 2 patients (11.1%) as ASA I, 11 (64.1%) patients as ASA II, and 5 (27.8) patients as ASA III. All patients were Child A, 5 (27.8%) cases were reported HCV-related cirrhosis, and 1 (5.6%) case has HCV/HBV-related disease. Median MELD score was 7.6 ± 1.6 (range 6–11). Three (16.7%) patients underwent preoperative therapy for the neoplastic primitive condition (1 for seminoma, 1 for lung adenocarcinoma, and 1 for sinonasal undifferentiated carcinoma). Preoperative ICG test showed an average value of clearance %/min 20.3 ± 2.4 (range 16–23) and R15% rate of 5 ± 1.5 (range 3.4–7).

FIG. 1.

Temporal distribution of difficulty class level. IDL, intermediate difficulty level; LDL, low difficulty level; TOT, total.

Table 1.

Patient Characteristics

M/F (n, %)	10 (55.6)/8 (44.4)
Age (mean ± SD), years	55.3 ± 17.4
BMI (mean ± SD), kg/m²	25.2 ± 4.6
ASA n (%)
I	2 (11.1)
II	11 (64.1)
III	5 (27.8)
IV	0
Cirrhosis n (%)	6
HBV related	0
HCV related	5 (27.8)
HBV-HCV related	1 (5.6)
Child-Pugh
A	18
B	0
C	0
MELD score (mean ± SD)	7.6 ± 1.6
ICG clearance %/min (n ± SD)	20.3 ± 2.4
ICG R15 (%)	5 ± 1.5
Indications for surgery n (%)
HCC	6 (33.3)
FNH	3 (16.7)
Angioma	2 (11.1)
NET	1 (5.6)
Adenoma	1 (5.6)
CRLM	1 (5.6)
BCLM	1 (5.6)
Testicular seminoma LM	1 (5.6)
Pulmonary adenocarcinoma LM	1 (5.6)
Sinonasal undifferentiated carcinoma LM	1 (5.6)
Site of lesions n (%)
S1	0
S2	6 (33.3)
S3	3 (16.7)
S4	2 (11.1)
S5	3 (16.7)
S6	3 (16.7)
S7	5 (27.8)
S8	0

ASA, American Society of Anesthesiology; BCLM, breast cancer liver metastasis; BMI, body mass index; CRLM, colorectal liver metastasis; FHN, focal nodular hyperplasia; HBV, human B virus; HCC, hepatocellular carcinoma; HCV, human C virus; ICG, indocyanine green; MELD, model for end-stage liver disease; NET, neuroendocrine tumor; SD, standard deviation.

Operative findings

Operative findings are summarized in Table 2. All patients underwent robotic LR, in particular 4 (22.2%) left lateral sectionectomy, 12 (66.7%) wedge resections (WRs) 2 of which were multiple, and 2 (11.1%) segmentectomy (SEG) were performed. One (5.6%) case required conversion to open surgery for bleeding. One case of a patient carrying an intraparenchymal liver adenoma of the first series needed reintervention due to the partial resection of the nodule, in spite of the performance of the laparoscopic IOUS. Neither intraoperative blood transfusion nor other complications are noted. The overall median operative time was 252.1 ± 88.2 (range 101–385) minutes (low difficulty level [LDL] 239 ± 96.14 minutes, intermediate difficulty level [IDL] 252.3 ± 86.88 minutes); no statistical significance was noted between the LDL and IDL group (P = .78). No statistical differences were observed between the first and the second period cases (P = .89). In all cases, the hepatic hilum was dissected and a tourniquet was predisposed to eventually perform extracorporeal Pringle maneuver that was carried out in half cases. The median clamp duration was of 43.3 ± 21.8 (range 15–60) minutes.

Table 2.

Operative Findings

Surgical procedure n (%)	Overall	First period (2011)	Second period (2018/2019)
Left lateral sectionectomy	4 (22.2)	1 (5.6)	3 (16.7)
Wedge resection	12 (66.7)	4 (22.2)	8 (44.4)
Segmentectomy	2 (11.1)	0	2 (11.1)
Conversion rate, n (%)	1 (5.6)	1 (5.6)	0
Operative time (minutes), mean ± SD (range)	252.1 ± 88.2 (101–385)	253 ± 70.2 (150–325)	247 ± 89.2 (101–385)
Nodule size (mm), mean ± SD (range)	39.1 ± 29.5 (35–120)	25.4 ± 12.2 (10–40)	44.3 ± 3 (7–120)
Pringle time, mean ± SD (range)	43.3 ± 21.8 (15–60)	—	—

SD, standard deviation.

At the end of surgery, a single drainage tube was placed in 9 (50%) cases, in 1 (5.6%) case, instead two drains were placed. The average diameter of removed lesions was 39.1 ± 29.5 mm, (range 35–120) and 10 (55.6%) nodules measured over 30 mm. We have noted significant differences between two periods in terms of nodule size (25.4 ± 12.2 mm vs. 44.3 ± 3 mm, P = .0001).

Difficulty scoring

Based on our findings, applying the DSS, we divided our series into two groups: an LDL group (LDL 1–3) made up of 5 patients, and an IDL group (IDL 4–6) consisting of 13 patients.

DSS results are reported in Table 3. Average Ban DSS (BDSS) was 4.6 ± 1.5 points (range 2–6), in particular 2 (11.1%) cases have BDSS of 2, 3 (16.7%) cases BDSS of 3, 2 (11.1%) cases BDSS of 4, 4 (22.2%) cases BDSS of 5, and 7 (38.9%) cases BDSS of 6. As expected, the median Iwate DSS (IDSS) results were the same; particularly 2 (11.1%) cases have IDSS of 2, 3 (16.7%) cases have IDSS of 3, 2 (11.1%) cases have IDSS of 4, 4 (22.2%) cases have IDSS of 5, and 7 (38.9%) cases have IDSS of 6. No cases with BDSS or IDSS ≥7 were operated.

Table 3.

Difficulty Scoring and Postoperative Outcomes

DSS	Overall mean ± SD (range)	LDL (1–3) n (%)	IDL (4–6) n (%)
BanDSS	4.6 ± 1.5 (2–6)	5 (27.8)	13 (72.2)
Iwate DSS	4.6 ± 1.5 (2–6)	5 (27.8)	13 (72.2)
Morbidity n (%)	5 (27.8)	2 (11.1)	3 (16.7)
Clavien–Dindo
I		1 (5.6)	—
II		—	3 (16.7)
III		1 (5.6)	—
IV		—	—
V		—	—
LOS mean ± SD (range)	5.5 ± 3.3 (2–15)	4.8 ± 1.93 (3–8)	5.8 ± 3.6 (2–15)

DSS, difficulty score system; IDL, intermediate difficulty level; LDL, low difficulty level; LOS, length of stay; SD, standard deviation.

Postoperative outcomes

Overall postoperative morbidities were observed in 5 (27.8%) cases and are reported in Table 3; only 1 case, grade III Clavien–Dindo needed percutaneous drainage for perihepatic fluid collection. Three patients with grade II, two of which had cirrhosis, were treated with diuretics, antibiotics, and blood transfusions, in the postoperative period; one patient with grade I, required diuretics administration in the postoperative period. No mortalities occurred in our series. No statistical significance in the complication rate was noted between the LDL group and the IDL group (P = .58). The overall average postoperative length of stay was 5.5 ± 3.3 days (range 2–15). The LDL (4.8 ± 1.93 days range 3–8) and IDL groups (5.8 ± 3.6 days range 2–15) did not differ significantly in the length of stay (P = .56).

Discussion

This is the initial report reflecting the beginning of robotic surgical activity in performing LRs in recently established liver surgery. A total of 18 LRs were performed in two different periods.

Minimally invasive surgery was proposed for the LR recently¹; the first experiences reserved the laparoscopic approach to small resection, and now represents a valid alternative to open procedures. Liver laparoscopic resection in selected cases had a remarkable low-grade pain, a more rapid postoperative recovery, and return to daily activities. However, the difficulties in liver exposition, parenchymal frailty, and the risk of bleeding limited laparoscopy application in hepatic surgery to selected patients. Robotic technology has made it possible to overcome many of the limitations of laparoscopic surgery, which include: a reduction in tremors, magnifying images with three-dimensional technology, an increase in the effectiveness of surgical gestures through tools with seven degrees of freedom, and an improvement in fatigue in relation to the continuation of surgery.^4,13,14

One of our patients needed reintervention for the partial removal of a liver adenoma, in spite of the performance of the laparoscopic IOUS. This patient was in the initial group of our series. This was a common finding in the first experience of robotic surgery¹⁵ and it was mainly due to the complete absence of tactile perception during robotic surgery, making this event different from the pure laparoscopic approach. Such evidence can be overcome with the systematic and continuous use of the IOUS during surgery. If the institute does not own a robotic US probe, a close collaboration between the operator and their first assistant on the patient bedside is mandatory to avoid failure resulting from incorrect transection planes. On the other hand, the robotic equipment will soon allow to implement fusion image-guided techniques in the surgeon's surgical field, which could solve this technique's particular problem.¹⁶

It is our impression that recovery from surgery after robotic LR is much more rapid even in comparison to the pure laparoscopic technique, even though it is difficult to demonstrate considering the small number of patients. Nevertheless, we believe that a minimal-stay approach could be possible with implementing the robotic technique with the recent Enhanced Recovery After Surgery (ERAS) concepts. It seems possible that future surgery will bring minor procedures toward revolutionizing the management of liver patients, according to the complexity of the resection performed.^17–19

Some authors suggest to apply a DSS to predict the complexity of LRs applicable in laparoscopic surgical procedures.⁵ This score was modified in 2016 into the IWATE score, which mainly incorporates the laparoscopic hand-assisted technique, that is not employed during the use of the robotic technique. However, none of the specific score systems for robotic LRs is available nowadays. The aim of our study was to verify the feasibility of applying this score and relative updates on robotic LRs performed in our institute. In our series, during the calculation of DSS we did not observe any statistically significant differences in operative time, complication rate, and postoperative length of stay between the LDL and IDL groups. During minimally invasive laparoscopic or robotic interventions, the inability to have an immediate and direct control of bleeding represents a strong limitation, and hemorrhage represents the most frequent cause of conversion to the open technique.²⁰ We report one case of conversion to open surgery due to uncontrollable bleeding.

Hepatic WRs represent the most performed minimally invasive surgical procedures,^2,14,21 also in our experience. We had some difficulty during the process of applying the score in case of simultaneous multiple WRs. if we had calculated the difficulty score adding every singular nodule, we would have overvalued the results. In IDSS, there was a modification from difficulty level 10 to 12 and an added expert section of the difficulty level. We applied the original BDSS and updated it without modifying the score assigned. Furthermore, it was first thought to apply the score in laparoscopic LR but then used it in the robotic field. Like IDSS, the hand-assisted/Hybrid section had been added to the score, however, in our opinion a section needed to be added considering the specificity of the robotic/laparoscopic approach and the number of nodules treated. Perhaps the difficulty in applying the laparoscopic DSS to robotic resections derive from the intrinsic difference between the two minimally invasive approaches. From these considerations, we deduced the necessity of developing a specific robotic LR difficulty score.

There are some limitations in the present study. The first limitation is represented by the small sample size analyzed that may affect the results. Second, the first case and others were performed in a 7-year time frame due to the scarce availability of the robotic instrument for general and hepatobiliary surgery in our center, which could affect the learning curve of the surgical team and the technical aspects of the procedure and finally the outcomes. Finally, the presence of heterogeneity of liver diseases was reported in our series and the various types of LRs involved.

Conclusions

In this preliminary experience with robotic LRs, laparoscopic DSS is somehow applicable to the new technique, with some limitations due to the peculiarity of the two different minimally invasive approaches. The necessity to perform simultaneous multiple resections could create some trouble in estimating the level of difficulty. It is highly likely that we will need to develop a new specific robotic difficulty score that includes these considerations.

Footnotes

Acknowledgment

The authors thank TaniaMerlino from the Scientific Direction of IRCCS- Regina Elena National Cancer Institute for the English revision of the article.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Reich

, McGlynn

, DeCaprio

, Budin

. Laparoscopic excision of benign liver lesions. Obstet Gynecol, 1991; 78,(5 Pt 2):956–958.

, Wakabayashi

, Nitta

. Systematic review of robotic liver resection. Surg Endosc, 2013; 27:732–739.

Montalti

, Berardi

, Patriti

, Vivarelli

, et al. Outcomes of robotic vs laparoscopic hepatectomy: A systematic review and meta-analysis. World J Gastroenterol, 2015; 21:8441–8451.

Dalsgaard

, Jensen

, Hartwell

, Mosgaard

, Jørgensen

, Jensen

. Robotic surgery is less physically demanding than laparoscopic surgery: Paired cross sectional study. Ann Surg, 2020; 271:106–113.

Ban

, Tanabe

, Ito

, et al. A novel difficulty scoring system for laparoscopic liver resection. J Hepatobiliary Pancreat Sci, 2014; 21:745–753.

Wakabayashi

. What has changed after the Morioka consensus conference 2014 on laparoscopic liver resection?. Hepatobiliary Surg Nutr, 2016; 5:281–289.

Child

, TurcotteJG. Surgery and portal hypertension. Major Probl Clin Surg, 1964; 1:1–85.

Pugh

, Murray-Lyon

, Dawson

, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg, 1973; 60:646.

Kamath

, Wiesner

, Malinchoc

, Kremers

, Therneau

, Kosberg

, et al. A model to predict survival in patients with end-stage liver disease. Hepatology, 2001; 33:464–470.

10.

Haegele

, Reiter

, Wanek

, Offensperger

, Pereyra

, StremitzerS, et al. Perioperative non-invasive indocyanine green-clearance testing to predict postoperative outcome after liver resection. PLoS One, 2016; 11:e0165481.

11.

Strasberg

, Phillips

. Use and dissemination of the brisbane 2000 nomenclature of liver anatomy and resections. Ann Surg, 2013; 257:377–382.

12.

Dindo

, Demartines

, Clavien

. Classification of surgical complications. Ann Surg, 2004; 240:205–213.

13.

Choi

, Choi

, Kim

, Hwang

, Kang

, Choi

, et al. Robotic liver resection: Technique and results of 30 consecutive procedures. Surg Endosc, 2012; 26:2247–2258.

14.

Lee

, Cheung

, Chong

, et al. Laparoscopic and robotic hepatectomy: Experience from a single centre. ANZ J Surg, 2016; 86:122–126.

15.

Giulianotti

, Coratti

, Sbrana

, Addeo

, Bianco

, Buchs

, et al. Robotic liver surgery: Results for 70 resections. Surgery, 2011; 149:29–39.

16.

Buchs

, Volonte

, Pugin

, Toso

, Fusaglia

, Gavaghan

, et al. Augmented environments for the targeting of hepatic lesions during image-guided robotic liver surgery. J Surg Res, 2013; 184:825–831.

17.

Warner

, Jutric

, Nisimova

, Fong

. Early recovery pathway for hepatectomy: Data-driven liver resection care and recovery. Hepatobiliary Surg Nutr, 2017; 6:297–311.

18.

Melstrom

, Warner

, Woo

, Sun

, Lee

, Singh

, et al. Selecting incision-dominant cases for robotic liver resection: Toward outpatient hepatectomy with rapid recovery. Hepatobiliary Surg Nutr, 2018; 7:77–84.

19.

Viganò

, Torzilli

, Troisi

, et al. Minor hepatectomies: Focusing a blurred picture: Analysis of the outcome of 4471 open resections in patients without cirrhosis. Ann Surg, 2019; 270:842–851.

20.

Viganò

, Tayar

, Laurent

, Cherqui

. Laparoscopic liver resection: A systematic review. J Hepatobiliary Pancreat Surg, 2009; 16:410–421.

21.

Guerra

, Di Marino

, Coratti

. Robotic surgery of the liver and biliary tract. J Laparoendosc Adv Surg Tech A, 2019; 29141–146.