Background

The present study aimed to assess performance of the three-level complexity classification which stratified liver resection procedures into three complexity grades (grade I, low; grade II, intermediate; grade III, high complexity) and to evaluate whether the enhanced recovery after surgery (ERAS) protocol improves postoperative outcomes for each complexity grade.

Study Design

Consecutive patients undergoing open liver resection (OLR) and laparoscopic liver resection (LLR) at Lausanne University Hospital during 2010–2020 were assessed.

Results

A total of 437 patients were included. Operative time estimated blood loss, and length of hospital stay increased significantly with a stepwise increase of the grades from I to III in OLR and LLR (all, P<0.05). The same trend for comprehensive complication index (CCI) was found in OLR (P<0.005). Age (P=0.004), the three-level complexity classification (grade II vs. I, P=0.001; grade III vs. I, P<0.001), no use of the ERAS protocol (P=0.016), and biliary reconstruction (P<0.001) were significant predictors for postoperative complication defined as CCI ≥ 26.2 in a multivariable logistic regression analysis. The prediction model incorporating the 4 factors had a calculated concordance index of 0.735 and 0.742 based on the bootstrapping method. The use of ERAS protocol was associated with lower probability of postoperative complication for each complexity grade and age.

Conclusion

The use of ERAS protocol may decrease the probability of postoperative complication for each surgical complexity of liver resection and patient age. This finding emphasized the importance for tailoring perioperative managements by surgical complexity and patient age to improve outcomes after liver resection.

IntroductionRecent advances on surgical technique and perioperative management have improved postoperative outcomes in patients undergoing liver resection (1

Dokmak S, Fteriche FS, Borscheid R, et al. 2012 Liver resections in the 21st century: we are far from zero mortality. HPB (Oxford). 2013;15(11):908-915.

, 2Kenjo A. Miyata H. Gotoh M. et al.Risk stratification of 7,732 hepatectomy cases in 2011 from the National Clinical Database for Japan.). The incidence of postoperative complication after liver resection remains high compared to other oncological surgeries (3Cescon M. Vetrone G. Grazi G.L. et al.Trends in perioperative outcome after hepatic resection: analysis of 1500 consecutive unselected cases over 20 years.). To predict postoperative outcomes, the minor/major classification has been traditionally used to categorize liver resection procedures into two complex levels. The nomenclature of minor and major liver resections was originally introduced by Tung in 1979 (4

Tung TT. Les Résections Majeures et Mineures du Foie. Masson: Paris. 1979.

). In the study, major liver resection was defined as resection of two or more sections which were equivalent to Couinaud segment(s) II+III, IV, V+VIII or VI+VII. Recently, most studies have defined the term major liver resection as the resection of three or more contiguous segments (5Kawaguchi Y. Hasegawa K. Tzeng C.D. et al.Performance of a modified three-level classification in stratifying open liver resection procedures in terms of complexity and postoperative morbidity.). However, studies reported that the minor/major classification did not effectively stratify liver resection procedures in terms of surgical complexity and postoperative outcomes (6Kawaguchi Y. Fuks D. Kokudo N. Gayet B. Difficulty of Laparoscopic Liver Resection: Proposal for a New Classification., 7Jang J.S. Cho J.Y. Ahn S. et al.Comparative Performance of the Complexity Classification and the Conventional Major/Minor Classification for Predicting the Difficulty of Liver Resection for Hepatocellular Carcinoma., 8Lee MKt Gao F. Strasberg S.M. Completion of a Liver Surgery Complexity Score and Classification Based on an International Survey of Experts.). The new three-level complexity classification for laparoscopic liver resection (LLR), which categorized 11 LLR procedures as being of low, intermediate or high grades, was proposed on the basis of a study using a French cohort in 2018 (6Kawaguchi Y. Fuks D. Kokudo N. Gayet B. Difficulty of Laparoscopic Liver Resection: Proposal for a New Classification.). Subsequently, it was validated for LLR in a Japanese multi-institution cohort and an Italian multi-institution cohort (9

Kawaguchi Y, Tanaka S, Fuks D, et al. Validation and performance of three-level procedure-based classification for laparoscopic liver resection. Surgical endoscopy. 2019.

, 10Russolillo N. Aldrighetti L. Cillo U. et al.Risk-adjusted benchmarks in laparoscopic liver surgery in a national cohort.) and for open liver resection (OLR) in a North American cohort and a Japanese cohort (5Kawaguchi Y. Hasegawa K. Tzeng C.D. et al.Performance of a modified three-level classification in stratifying open liver resection procedures in terms of complexity and postoperative morbidity., 9

Kawaguchi Y, Tanaka S, Fuks D, et al. Validation and performance of three-level procedure-based classification for laparoscopic liver resection. Surgical endoscopy. 2019.

).The enhanced recovery after surgery (ERAS) programs was advocated to improve postoperative outcomes and allow cost reduction, and has been used for abdominal, orthopedic, urological and gynecological surgeries (11Adamina M. Kehlet H. Tomlinson G.A. et al.Enhanced recovery pathways optimize health outcomes and resource utilization: a meta-analysis of randomized controlled trials in colorectal surgery., 12Varadhan K.K. Neal K.R. Dejong C.H. et al.The enhanced recovery after surgery (ERAS) pathway for patients undergoing major elective open colorectal surgery: a meta-analysis of randomized controlled trials., 13Gustafsson U.O. Scott M.J. Schwenk W. et al.Guidelines for perioperative care in elective colonic surgery: Enhanced Recovery After Surgery (ERAS((R))) Society recommendations., 14Yamada T. Hayashi T. Cho H. et al.Usefulness of enhanced recovery after surgery protocol as compared with conventional perioperative care in gastric surgery., 15Jones C. Kelliher L. Dickinson M. et al.Randomized clinical trial on enhanced recovery versus standard care following open liver resection., 16Lassen K. Coolsen M.M. Slim K. et al.Guidelines for perioperative care for pancreaticoduodenectomy: Enhanced Recovery After Surgery (ERAS(R)) Society recommendations., 17Roulin D. Melloul E. Wellg B.E. et al.Feasibility of an Enhanced Recovery Protocol for Elective Pancreatoduodenectomy: A Multicenter International Cohort Study.). Our institution had been using the ERAS protocol for liver resection patients since July 2013, and showed benefits of the liver ERAS protocol (18Joliat G.R. Labgaa I. Hubner M. et al.Cost-Benefit Analysis of the Implementation of an Enhanced Recovery Program in Liver Surgery., 19Melloul E. Hubner M. Scott M. et al.Guidelines for Perioperative Care for Liver Surgery: Enhanced Recovery After Surgery (ERAS) Society Recommendations.). However, the effect of the ERAS protocol is unclear in the context of various surgical complexity and other factors associated with postoperative complication.

The present study aimed to assess the performance of the three-level complexity classification for LLR and OLR in a Swiss cohort and to evaluate whether the ERAS protocol was effective for improving postoperative outcomes after each complexity level of liver resection.

DefinitionsAnterolateral (AL) segments are defined as Couinaud segments II, III, IVb, V and VI, and posterosuperior (PS) segments are defined as Couinaud segments I, IVa, VII and VIII (6Kawaguchi Y. Fuks D. Kokudo N. Gayet B. Difficulty of Laparoscopic Liver Resection: Proposal for a New Classification., 20Wakabayashi G. Cherqui D. Geller D.A. et al.Recommendations for laparoscopic liver resection: a report from the second international consensus conference held in Morioka., 22Buell J.F. Cherqui D. Geller D.A. et al.The International Position on Laparoscopic Liver Surgery.). Surgical complexity of liver resection procedures was stratified using the three-level complexity classification, which classifies 11 different liver resection procedures as grade I (low complexity), grade II (intermediate complexity) or grade III (high complexity) (Figure 1) (5Kawaguchi Y. Hasegawa K. Tzeng C.D. et al.Performance of a modified three-level classification in stratifying open liver resection procedures in terms of complexity and postoperative morbidity., 6Kawaguchi Y. Fuks D. Kokudo N. Gayet B. Difficulty of Laparoscopic Liver Resection: Proposal for a New Classification., 9

Kawaguchi Y, Tanaka S, Fuks D, et al. Validation and performance of three-level procedure-based classification for laparoscopic liver resection. Surgical endoscopy. 2019.

, 23Kawaguchi Y. Lillemoe H.A. Vauthey J.N. Surgical Resection: Old Dog, Any New Tricks?.). Wedge resection was defined as “resection of less than one Couinaud segment for removal of a tumor less than 3 cm in diameter,” and segmentectomy was defined as “resection of less than one Couinaud segment for removal of a tumor at least 3 cm in diameter or anatomical removal of one Couinaud segment” (5Kawaguchi Y. Hasegawa K. Tzeng C.D. et al.Performance of a modified three-level classification in stratifying open liver resection procedures in terms of complexity and postoperative morbidity., 8Lee MKt Gao F. Strasberg S.M. Completion of a Liver Surgery Complexity Score and Classification Based on an International Survey of Experts.). Major liver resection was defined as the resection of more than three contiguous Couinaud segments according to the minor/major classification (5Kawaguchi Y. Hasegawa K. Tzeng C.D. et al.Performance of a modified three-level classification in stratifying open liver resection procedures in terms of complexity and postoperative morbidity., 8Lee MKt Gao F. Strasberg S.M. Completion of a Liver Surgery Complexity Score and Classification Based on an International Survey of Experts., 24Jarnagin W.R. Gonen M. Fong Y. et al.Improvement in perioperative outcome after hepatic resection: analysis of 1,803 consecutive cases over the past decade.). When two or more areas of the liver were resected, the higher grade was applied to this patient: e.g., patients undergoing right hepatectomy (grade III and major resection) and wedge resection of segment III (grade I and minor resection) was categorized as grade III and major resection. We defined morbidity and death which occurred within 90 days after surgery as postoperative complications and mortality, respectively. Postoperative complications were graded using the Clavien classification (25Dindo D. Demartines N. Clavien P.A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey.), and cumulative morbidity was measured using the Comprehensive Complication Index (CCI) (26Slankamenac K. Nederlof N. Pessaux P. et al.The comprehensive complication index: a novel and more sensitive endpoint for assessing outcome and reducing sample size in randomized controlled trials.).

Figure 1Three-level complexity classification. ∗Anterolateral (AL) segments are defined as Couinaud segments 2, 3, 4b, 5, and 6, and †posterosuperior (PS) segments are defined as Couinaud segments 1, 4a, 7, and 8. (Clin Liver Dis 24(4): 637-655, with permission.)

 Comparison of two classifications

The performance of the three-level classification was compared with that of the minor/major classification in stratifying liver resection procedures with respect to surgical and postoperative outcomes. To test whether the three-level complexity classification performed better than the minor/major classification, the areas under the receiver operating characteristic (ROC) curves (AUCs) were compared.

 Statistical analysisCategorical variables were expressed as n (%) and were compared between groups using Fisher’s exact test or the chi-square test, as appropriate. Continuous variables were expressed as median (interquartile range [IQR]) and were compared using the Kruskal–Wallis test for three groups. CCI scores was expressed as mean (standard deviation [SD]), and were compared among groups using the ANOVA test for three groups. The three-level complexity classification and the minor/major classification were evaluated using the ROC curve analysis, and AUCs were compared to evaluate the performance of each classification using the method described by DeLong and colleagues (27Investigation of the freely available easy-to-use software 'EZR' for medical statistics.). Median value was used as the threshold for estimated blood loss and duration of operation (estimated blood loss of 600 ml, duration of operation of 283 min). A CCI score of 26.2, which corresponds to one postoperative complication of Clavien classification grade IIIa, was used as the threshold between high (CCI score 26.2 or higher) and low (CCI score less than 26.2) complication severity (5Kawaguchi Y. Hasegawa K. Tzeng C.D. et al.Performance of a modified three-level classification in stratifying open liver resection procedures in terms of complexity and postoperative morbidity.). A logistic regression model analysis was used to predict the incidence of complication (CCI score 26.2 or higher). Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each factor. The multivariable prediction model was developed and validated based on the TRIPOD statement (28Moons K.G. Altman D.G. Reitsma J.B. et al.Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration.). The predictive performance of the logistic regression model was internally validated using the bootstrapping method (29Harrell Jr., F.E. Lee K.L. Mark D.B. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors.). Harrell’s C-statistic of the identified model was calculated using 100 bootstrap samples. P Discussion

The findings of the present study suggest that the recently-reported three-level complexity classification of LLR and OLR performed better than the minor/major classification with respect to surgical outcomes in patients at an European institution. Our regression model for predicting postoperative complications (CCI ≥ 26.2) incorporated the 4 factors (age, the three-level complexity classification, use of ERAS protocol, and biliary reconstruction), and showed a good discrimination for predicting patients with a high probability of postoperative complication. Importantly, the present study highlighted that the three-level complexity classification stratified the incidence of postoperative complications and use of ERAS protocol decreased the incidence of postoperative complication, regardless of patient age.

The presented regression model incorporated the recently-reported three-level complexity classification. The three-level complexity classification (6Kawaguchi Y. Fuks D. Kokudo N. Gayet B. Difficulty of Laparoscopic Liver Resection: Proposal for a New Classification.) was designed for LLR on the basis of a cohort in a single French institution to guide surgical complexity of LLR, because the traditional minor/major classification had limitations for stratifying surgical complexity of liver resections (9

Kawaguchi Y, Tanaka S, Fuks D, et al. Validation and performance of three-level procedure-based classification for laparoscopic liver resection. Surgical endoscopy. 2019.

). The minor/major classification was just useful for indicating the risk of liver failure. Recent advancement on liver volume calculation enabled to estimate accurate preoperative liver volume and postoperative future liver remnant volume, contributing to decrease the risk of postoperative liver failure (30Mise Y. Hasegawa K. Satou S. et al.How Has Virtual Hepatectomy Changed the Practice of Liver Surgery?: Experience of 1194 Virtual Hepatectomy Before Liver Resection and Living Donor Liver Transplantation.). As such, the new liver resection classification associated with surgical complexity may be useful for training and planning. It allows planning liver resection procedures, tailoring management after liver resection, and comparing intergroup differences of surgical complexity in cohort studies. The three-level complexity classification was recently validated using multi-institution data including patients who underwent LLR in 43 Japanese institutions (9

Kawaguchi Y, Tanaka S, Fuks D, et al. Validation and performance of three-level procedure-based classification for laparoscopic liver resection. Surgical endoscopy. 2019.

) and using 1752 patients of Italian Group of Minimally invasive Liver Surgery registry (10Russolillo N. Aldrighetti L. Cillo U. et al.Risk-adjusted benchmarks in laparoscopic liver surgery in a national cohort.). Although the three-level classification was originally designed for LLR, a recent bi-institution study in USA and Japan showed that the three-level classification also apply for OLR. It was correlated with surgical and postoperative outcomes and performed better than other classifications for liver resection. The current study based on a Swiss cohort supported that the three-level complexity classification did correlate with surgical and postoperative outcomes in patients undergoing both LLR or OLR (Figure 1). In the subgroup analysis including patients who underwent minor liver resection only, the three-level complexity classification showed a stepwise increase of surgical and postoperative outcomes. This clearly suggests the limitation of the minor/major classification because the category of minor liver resection includes heterogenous liver resection procedures in terms of surgical complexity and postoperative complication.ERAS protocol was proposed to temper patient’s response to surgical stress and so decrease postoperative complication (31Lassen K. Soop M. Nygren J. et al.Consensus review of optimal perioperative care in colorectal surgery: Enhanced Recovery After Surgery (ERAS) Group recommendations.). Our group had demonstrated the clinical and economic benefit of ERAS in patients undergoing colorectal surgery (32Roulin D. Donadini A. Gander S. et al.Cost-effectiveness of the implementation of an enhanced recovery protocol for colorectal surgery.). Other groups reported that the use of ERAS protocol improved postoperative outcomes after liver resection (19Melloul E. Hubner M. Scott M. et al.Guidelines for Perioperative Care for Liver Surgery: Enhanced Recovery After Surgery (ERAS) Society Recommendations., 33Hall T.C. Dennison A.R. Bilku D.K. et al.Enhanced recovery programmes in hepatobiliary and pancreatic surgery: a systematic review., 34Hughes M.J. McNally S. Wigmore S.J. Enhanced recovery following liver surgery: a systematic review and meta-analysis.). the present study confirmed that the use of ERAS protocol decreased the incidence of postoperative complication for all complexity levels of liver resection procedures. Specifically, the presented model suggests that the use of ERAS protocol effectively decreases complication probability in older patients with complex liver resection procedures. Use of ERAS protocol is only a modifiable factor in this regression model. As such, the present study may be useful in identifying patients with a high probability of developing postoperative complications and for selecting patients who should follow ERAS protocol with a high compliance.Studies reported that postoperative complication rate was lower in patients undergoing LLR than in patients undergoing OLR (35Kawaguchi Y. Hasegawa K. Wakabayashi G. et al.Survey results on daily practice in open and laparoscopic liver resections from 27 centers participating in the second International Consensus Conference., 36Nguyen K.T. Gamblin T.C. Geller D.A. World review of laparoscopic liver resection-2,804 patients., 37Kawaguchi Y. Otsuka Y. Kaneko H. et al.Comparisons of financial and short-term outcomes between laparoscopic and open hepatectomy: benefits for patients and hospitals.). A univariable logistic regression analysis confirmed this finding, because the probability of postoperative complication was significantly higher in OLR than in LLR patients. Interestingly, OLR (vs. LLR) was no longer a predictive factor for postoperative complication using a multivariable logistic regression analysis after adjusting for other predictive factors including the three-level complexity classification. This implies that the incidence of postoperative complication may not be different between LLR and OLR after adjustment for surgical complexity. In fact, the incidence of postoperative complication did not differ significantly between patients undergoing LLR with grade I and III procedures and patients undergoing OLR with grade I and III procedures (Figure 2). In contrast, the incidence of postoperative complication after grade II procedures was lower in patients undergoing LLR than in patients undergoing OLR. This finding emphasizes the importance of liver resection classification associated with surgical complexity and postoperative outcomes for comparing intergroup difference in cohort studies.Some limitations of the present study have to be addressed. First, this is a retrospective study in a single high-volume institution in Switzerland. The proportion of LLR was low, because laparoscopic approach had been frequently used since 2017 only. Finally, some data about the compliance of ERAS protocol were missing. Of note, high compliance to ERAS protocol is associated with the success of ERAS implementation (38Gustafsson U.O. Hausel J. Thorell A. et al.Adherence to the enhanced recovery after surgery protocol and outcomes after colorectal cancer surgery.), the compliance to ERAS protocol may have influenced the effect of ERAS on postoperative outcomes.In conclusion, this three-level complexity classification was useful for stratifying surgical complexity and postoperative outcomes in OLR and LLR patients. The use of ERAS protocol may decrease the probability of postoperative complication for each surgical complexity of liver resection and patient age. These finding emphasizes the importance for tailoring perioperative managements by surgical complexity and patient age to improve outcomes after liver resection.

eFigure 1. Comprehensive complication index (CCI) by 3-level complexity classification only in patients who had enhanced recovery after surgery (ERAS) protocol. CCI scores were significantly different in both patients undergoing laparoscopic liver resection and open liver resection (OLR) using the ANOVA test for 3 groups. In patients undergoing OLR procedures, CCI increased significantly with a stepwise increase of the grades from I to III (p < 0.05)

eFigure 2. Calibration plots of the logistic model using (A) the original cohort and (B) the bootstrap method.