Published online Nov 28, 2026. doi: 10.3748/wjg.120736
Revised: April 8, 2026
Accepted: May 20, 2026
Published online: November 28, 2026
Processing time: 207 Days and 12.9 Hours
Conventional cranial-caudal classification of hepatic segment 4 (S4) based on the falciform ligament does not fully reflect individual variation in portal inflow and may limit anatomical precision during parenchyma-sparing liver resection.
To characterize portal inflow variation in S4, develop a portal vein-based subsegmental classification, and assess its surgical applicability in laparoscopic paren
This retrospective single-center study included 207 patients who underwent laparoscopic liver resection (LLR) between June 2021 and April 2024. Preoperative contrast-enhanced computed tomography, magnetic resonance imaging, and three-dimensional reconstruction were performed to define the portal inflow patterns of S4. Classification-guided laparoscopic parenchyma-sparing resection was assessed using intraoperative indocyanine green fluorescence. Detailed perioperative biochemical and pathological analyses focused on the type II sub
Intermediate branches crossing the falciform ligament were identified in 192 of 207 patients (92.8%). S4 was classified as type I (cranial-caudal, 120/207, 58.0%), type II (left-right, 78/207, 37.7%), or type III (scattered, 9/207, 4.3%). In the type II subgroup, postoperative liver enzymes increased transiently and recovered by postoperative day 5, and all analyzed patients achieved negative pathological margins. In the comparative follow-up dataset, the pathological distribution included 113 hepatocellular carcinomas, 38 colorectal liver metastases, 36 intra
A portal vein-based classification of S4 may improve anatomical understanding and support individualized parenchyma-sparing LLR. The current data support anatomical plausibility and technical feasibility, while exploratory comparative analysis showed no significant difference in OS or PFS. Formal validation and broader long-term assessments remain necessary.
Core Tip: This study proposes a portal vein-based subsegmental classification of liver segment 4 (S4) derived from preoperative imaging and three-dimensional recons
- Citation: Huang K, Hu HY, Chen K, Chen XY, Chen Z, Sun Y, Du CY, Chang GX, Fu H, Hu C, Xiao H, Lan X. Portal vein-based subsegmental classification of liver segment four for parenchyma-sparing laparoscopic resection: A retrospective study. World J Gastroenterol 2026; 32(44): 120736
- URL: https://www.wjgnet.com/1007-9327/full/v32/i44/120736.htm
- DOI: https://dx.doi.org/10.3748/wjg.120736
Traditionally, segment 4 (S4) is defined as the region between the Cantlie line and the falciform ligament and is further subdivided into S4a (cranial portion) and S4b (caudal portion) based on the distribution of Glissonean pedicles[1,2]. Although widely accepted, this classification often fails to account for individual anatomical variations. As a result, it may lead to unnecessary resection of functional liver parenchyma or to an increased risk of positive surgical margins, thereby limiting its applicability in pre
However, conventional descriptions of S4 mainly emphasize the cranial-caudal subdivision into S4a and S4b and do not adequately address portal inflow patterns that extend across the falciform ligament. In particular, intermediate branches (IBs) arising between the Glissonean pedicles of segment 3 (G3) and Glissonean pedicles of segment 4 (G4) have not been sufficiently incorporated into the practical anatomical framework of S4. From a surgical perspective, these branches are clinically important because they may delineate perfused parenchyma beyond the traditional falciform ligament-based boundary, thereby influencing the extent of resection required to achieve adequate margins while preserving uninvolved liver tissue.
In this study, we sought to achieve three objectives. First, we aimed to characterize the portal inflow anatomy of S4, with particular attention given to IBs crossing the falciform ligament. Second, we aimed to develop a portal vein-based subsegmental classification of S4 that better reflects individual perfusion territories than the conventional falciform ligament-based model. Third, we aimed to explore the surgical applicability of this classification in laparoscopic paren
The primary aim of this retrospective study was to characterize the portal inflow anatomy of S4 and determine the anatomical significance of IBs crossing the falciform ligament. The secondary aim was to establish a portal vein-based subsegmental classification of S4. The third aim was to explore the clinical applicability of this classification in laparoscopic parenchyma-sparing liver resection.
This was a retrospective, single-center study conducted at the First Affiliated Hospital of Chongqing Medical Uni
All patients underwent multiphasic contrast-enhanced CT or gadolinium-enhanced MRI with thin-slice acquisition (≤ 1 mm). CT scans included arterial, portal venous, and delayed phases and were obtained with a multidetector scanner (Revolution CT, GE Healthcare, United States). MRI was performed on a 3.0 Tesla system (MAGNETOM Skyra, Siemens Healthineers, Germany) with hepatobiliary phase imaging after the administration of Gd-EOB-DTPA. DICOM data from the portal venous phase were imported into either the Hisense CAS system (Hisense Medical Technology, Qingdao, China) or Amira 3D software (Thermo Fisher Scientific, United States). Imaging evaluation and S4 classification were independently performed by two experienced researchers (Huang K and Xiao H). In cases of disagreement regarding the classification of S4, a third researcher (Lan X) was consulted. Final decisions were made through group discussion, using a majority-rule approach. Formal interobserver agreement statistics, such as, Cohen’s kappa were not prospectively planned or systematically recorded in this retrospective study and therefore were not reported.
All procedures were performed laparoscopically by experienced hepatobiliary surgeons. After induction of general anesthesia, patients were placed in a dorsal-elevated supine position with the right side slightly elevated. A five-trocar approach was generally used, with the trocar placement adjusted according to the tumor location and the planned transection plane. Intraoperative ultrasonography was routinely performed to confirm the tumor location, vascular relationships, and intended resection territory. Hepatic inflow occlusion was applied selectively using the intermittent Pringle manoeuvre when necessary. Liver parenchymal transection was performed using ultrasonic energy devices, with clips or sutures used for vascular and biliary control as appropriate. For classification-guided resection, the target Glissonean pedicle(s) of S4 were identified according to preoperative 3D reconstruction and intraoperative ultrasonography. Depending on the anatomical subtype, selective control of the relevant pedicle(s) was achieved to define the intended resection territory. Intraoperative indocyanine green (ICG) fluorescence imaging was then used to visualize portal territory demarcation and guide anatomical parenchymal transection. In the extended S4 resection, the umbilical fissure vein served as an important posterior landmark. This strategy enables individualized parenchyma-sparing resection while preserving uninvolved liver tissue whenever oncologically appropriate.
Clinical, pathological, and follow-up data were retrospectively extracted from electronic medical records and the comparator follow-up dataset. Continuous variables are reported as the mean ± SD or median with interquartile range, as appropriate. Categorical variables are presented as n (%). Repeated-measures analysis of variance was used for perioperative biochemical trends in the type II subgroup. Exploratory survival outcomes were estimated using the Kaplan-Meier method and compared using the log-rank test. All analyses were performed using R software (version 4.3.3).
Perioperative liver function parameters-including aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin, prothrombin time [international normalized ratio (INR)], albumin, and albumin-bilirubin (ALBI) score-were measured preoperatively and on postoperative days (PODs) 1, 3, and 5. All values were expressed as the mean ± SD.
Repeated-measures analysis of variance (ANOVA) was applied to evaluate overall temporal changes in these para
Pathological resection margins were defined as the shortest distance from the tumor edge to the transection plane. Two independent pathologists reviewed all the specimens, and discrepancies were resolved by consensus. R0 resection was defined as a microscopically tumor-free margin ≥ 1 mm, in accordance with the international consensus for hepatobiliary malignancies.
The fluorescence-pathology concordance rate, defined as the proportion of cases in which intraoperative ICG fluorescence demarcation coincided with the histopathological tumor margin, was calculated for all patients. The mean resection margin distance and concordance rate were recorded to evaluate anatomical accuracy and oncologic safety.
These parameters were analyzed to assess short-term biochemical recovery and pathological margin status in type II resections guided by the proposed S4 classification.
All 207 patients were included in the anatomical mapping and subtype classification analyses. Detailed perioperative biochemical and pathological analyses focused on the type II subgroup because this pattern most clearly reflected the incremental surgical relevance of the proposed classification beyond the conventional cranial-caudal S4a/S4b model. Specifically, the left-right portal inflow pattern of type II directly supports lateralized parenchyma-sparing resection. In contrast, type I largely overlaps with the traditional cranial-caudal concept, whereas type III was infrequent and anatomically heterogeneous, limiting meaningful subgroup outcome analysis. Pathology composition and follow-up outcomes were additionally assessed using the available comparator dataset. In this dataset, patients were categorized into novel and conventional groups according to the recorded surgical strategy. Overall survival (OS) was defined as the interval from surgery to death from any cause or the last follow-up. Progression-free survival (PFS) was defined as the interval from surgery to recurrence, progression, death, or last follow-up, whichever occurred first. Because the comparator grouping was closely associated with the anatomical subtype, these analyses were considered exploratory.
A total of 207 patients were enrolled in this study. The mean age was 58.2 ± 11.0 (range: 27.0-91.0) years, and 140 (67.6%) were male. All patients underwent LLR and had complete preoperative imaging data, including reconstruction of the Glissonean pedicles.
Redefining S4 boundaries and clinical applications: IBs arising from the bifurcation of the G3 and G4, which supplied a narrow area on both sides of the falciform ligament, were identified in 192 patients (92.8%). If these branches are considered part of G4, then the anatomical left boundary of S4 should be redefined as extending to the left of the falciform ligament (Figure 1). Specifically, 102 patients (49.3%) had one IB, 81 (39.1%) had two, and 9 (4.3%) had 3. Only 15 patients (7.2%) had no identifiable IBs. In a representative patient, the tumor was located near the falciform ligament within S4 and the anterior-ventral segment. In accordance with the conventional anatomical classification, left hemihepatectomy combined with resection of the anterior-ventral segment would have been needed to achieve negative surgical margins. However, when the new concept of S4 incorporating IBs was applied, laparoscopic sparing liver resection of the S4 and anterior-ventral segments was sufficient to achieve safe margins (Figure 2 and Video 1).
S4 subtypes and surgical applications: While classical segmentation divides S4 into cranial (S4a) and caudal (S4b) subsegments, our findings demonstrate that this cephalocaudal pattern is not consistently observed among patients. A substantial proportion of patients exhibited a left-right portal inflow pattern across the falciform ligament, indicating the need for a more comprehensive system. On the basis of portal vein anatomy, we developed a three-type classification scheme (Table 1 and Figures 3, 4, 5, 6, 7, and 8).
| Overall (n = 207) | |
| New S4 classification subtypes | |
| Type I (cranial-caudal subdivision) | 120 (58.0) |
| Ia (bilateral trunk type, Figure 3) | 55 (26.6) |
| Ib (unilateral trunk type, Figure 4) | 20 (9.7) |
| Ic (dispersed branching, Figure 5) | 45 (21.7) |
| Type II (left-right subdivision) | 78 (37.7) |
| IIa (bilateral trunk type, Figure 6) | 27 (13.0) |
| IIb (unilateral trunk type, Figure 7) | 51 (24.6) |
| Type III (scattered pattern, Figure 8) | 9 (4.3) |
From a surgical perspective, each subtype carries distinct surgical implications. Type I resection generally remained compatible with cranial-caudal subsegmental resection along the conventional S4a/S4b orientation, although branching variability still required individualized planning. Type II represented the most clinically distinctive pattern because its left-right portal inflow configuration directly supported lateralized parenchyma-sparing resection of the left and right subsegments of S4 (S4 L and S4R). In contrast, type III resection was uncommon and anatomically heterogeneous, and surgical planning in this subgroup relied mainly on individualized ICG fluorescence mapping rather than on a fixed pedicle-based resection plane. Because type II most clearly demonstrated the incremental surgical relevance of the proposed classification beyond the conventional cranial-caudal framework, detailed perioperative biochemical and pathological analyses were subsequently focused on this subgroup. A representative case (Figure 9 and Video 2) underscores the clinical utility of the new classification. The tumor was located at the junction of S4 and the anterior portion of segment 5, overlapping the right side of S4 and approaching the falciform ligament. Under traditional anatomical segmentation, resection would require the removal of the entire S4 and part of the right anterior section to ensure negative margins. However, preoperative 3D imaging revealed this case as type IIb, with a dominant right S4 pedicle (G4R) and scattered left S4 branches (G4 L). This anatomy enabled precise resection of segment 5 and S4R alone, preserving the uninvolved S4 L parenchyma while achieving adequate oncological margins. The resection line was guided by Glissonean control and confirmed by ICG fluorescence, demonstrating the value of portal-based segmentation in optimizing parenchyma-sparing liver surgery.
The perioperative changes in liver function and resection margin parameters among type II patients are summarized in Table 2. Repeated-measures ANOVA demonstrated significant time-dependent changes in AST, ALT, total bilirubin, INR, albumin, and ALBI scores (all P value < 0.001).
| Variable | Pre-operative | POD1 | POD3 | POD5 | P value |
| AST (U/L) | 31.5 ± 10.6 | 239.1 ± 81.5 | 76.3 ± 37.7 | 43.6 ± 15.8 | < 0.001 |
| ALT (U/L) | 28.3 ± 8.4 | 192.8 ± 77.6 | 70.3 ± 29.8 | 39.8 ± 12.2 | < 0.001 |
| Total bilirubin (µmol/L) | 12.9 ± 4.1 | 20.8 ± 7.2 | 25.6 ± 7.2 | 15.8 ± 4.8 | < 0.001 |
| INR | 1.00 ± 0.10 | 1.21 ± 0.10 | 1.18 ± 0.09 | 1.05 ± 0.05 | 0.002 |
| Albumin (g/L) | 42.0 ± 4.1 | 35.6 ± 4.2 | 36.9 ± 3.9 | 39.3 ± 3.4 | 0.004 |
| ALBI score | -2.67 ± 0.33 | -1.98 ± 0.41 | -2.12 ± 0.37 | -2.45 ± 0.32 | < 0.001 |
| Resection margin (mm) | - | - | - | 9.8 ± 3.2 | - |
| R0 resection rate (%) | - | - | - | 100 (78/78) | - |
AST and ALT levels peaked on POD1 (239.1 ± 81.5 U/L and 192.8 ± 77.6 U/L, respectively) and returned near baseline by POD5 (43.6 ± 15.8 U/L and 39.8 ± 12.2 U/L, respectively; all P > 0.05 vs preoperative), indicating transient hepatocellular injury with rapid recovery. The total bilirubin concentration peaked on POD3 (25.6 ± 7.2 μmol/L) and declined thereafter, whereas the prothrombin time (INR) showed mild, reversible prolongation on POD1-3. Albumin levels dec
The mean pathological resection margin was 9.8 ± 3.2 mm, and R0 resection was achieved in all patients (100%, 78/78). The fluorescence-pathology concordance rate was 97.4% (76/78), confirming precise anatomical alignment between portal-based ICG demarcation and histologic margins.
Collectively, these findings indicate mild, reversible postoperative biochemical disturbance together with negative pathological margins and precise anatomical targeting achieved through classification-guided type II resection.
The comparator follow-up dataset included 207 patients, of whom 78 were assigned to the Novel group and 129 to the Conventional group. Pathology revealed 113 hepatocellular carcinomas, 38 colorectal liver metastases, 36 intrahepatic cholangiocarcinomas, and 20 benign lesions. During follow-up, 32 patients developed recurrence, and 5 died. The median follow-up duration was 22.0 (interquartile range: 17.0-27.0) months.
Exploratory Kaplan-Meier analysis showed no significant difference in OS or PFS between the novel and conventional groups (Figure 10; log-rank P = 0.22 for OS and P = 0.21 for PFS). These findings suggest that the classification-guided strategy did not show an apparent adverse effect on medium-term oncologic outcomes in this retrospective comparator dataset.
A central finding of the present study is that IBs crossing the falciform ligament were identified in the vast majority of patients. This observation suggests that IBs are not rare incidental variants but rather a frequently overlooked component of S4 portal inflow anatomy. Classical descriptions of the medial segment have long emphasized the cranial-caudal subdivision into S4a and S4b and have treated the falciform ligament as a practical boundary. However, previous anatomical and imaging studies have also shown that the vascularization of S4 is substantially more variable than this traditional framework suggests[2,4,6]. In this context, our findings support portal territory-based reinterpretation of S4 anatomy and help explain why a falciform ligament-based definition may be insufficient for precision parenchyma-sparing resection in selected patients.
Building on these anatomical observations, we proposed a portal vein-based subsegmental classification of S4 with three major patterns: Type I (cranial-caudal), type II (left-right), and type III (scattered). From a surgical perspective, the greatest incremental value lies in type II anatomy because its left-right inflow configuration directly supports lateralized resection of S4 L or S4R beyond the conventional S4a/S4b concept. This point is clinically relevant because prior laparoscopic S4 resection techniques have largely been developed within the traditional cranial-caudal framework[7], whereas fluorescence-guided portal territory identification has increasingly shown value in defining functional segmental boundaries when the pedicle anatomy is fine, variable, or difficult to reproduce intraoperatively[8,9]. Accordingly, the proposed classification system should be understood not merely as an anatomical description but also as a practical fra
From a perioperative perspective, the analyzed type II cases showed transient postoperative biochemical disturbance with recovery by POD5, together with negative pathological margins in all analyzed cases. These findings support the technical feasibility and favorable short-term biochemical recovery of classification-guided type II resection. Moreover, they should not be overinterpreted. Hepatectomy is a complex metabolic and regenerative event, and the trajectories of AST or ALT alone cannot fully represent postoperative liver function. Therefore, although the current results are clini
From an oncologic perspective, the present findings also require cautious interpretation. In the comparator dataset, exploratory Kaplan-Meier analysis showed no significant difference in OS or PFS between the Novel and Conventional groups. This direction is consistent with findings reported in certain disease-specific settings, particularly those involving colorectal liver metastases, in which parenchyma-sparing and anatomic strategies may achieve comparable long-term oncologic outcomes in appropriately selected patients[10,11]. However, the disease-specific literature is not uniform. In hepatocellular carcinoma, some meta-analyses have favored anatomic resection for long-term survival or recurrence outcomes[12,13]. For this reason, the present exploratory comparative analysis should not be interpreted as proof of oncologic equivalence. Rather, it suggests that, within this mixed-pathology retrospective dataset, the proposed classification-guided strategy did not show an obvious adverse signal for medium-term survival outcomes.
Several limitations should be acknowledged. First, this was a retrospective single-center study subject to inherent selection bias. Second, although all 207 patients contributed to anatomical mapping and classification, detailed perioperative biochemical and pathological analyses focused on the type II subgroup, which limits direct extrapolation to type I and type III patterns. Third, formal reproducibility testing of the proposed classification was not performed because interobserver agreement statistics such as Cohen’s kappa were not prospectively recorded. Fourth, the comparative analysis was exploratory: The novel group corresponded to type II anatomy, whereas the conventional group consisted predominantly of type I and type III cases; therefore, the anatomical subtype and operative grouping were not in
A portal vein-based subsegmental classification of S4 improves the anatomical understanding of portal inflow variation and may assist individualized parenchyma-sparing liver resection. The present findings support anatomical plausibility and technical feasibility, while exploratory comparative analysis revealed no significant difference in OS or PFS. Formal reproducibility testing, broader perioperative assessments, and longer-term validation remain necessary.
We would like to thank the radiology and surgical teams at the First Affiliated Hospital of Chongqing Medical University for their invaluable support in data collection and imaging analysis.
| 1. | Machado MA, Makdissi FF, Surjan RC, Herman P, Teixeira AR, C Machado MC. Laparoscopic resection of left liver segments using the intrahepatic Glissonian approach. Surg Endosc. 2009;23:2615-2619. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 44] [Cited by in RCA: 46] [Article Influence: 2.7] [Reference Citation Analysis (0)] |
| 2. | Onishi H, Kawarada Y, Das BC, Nakano K, Gadzijev EM, Ravnik D, Isaji S. Surgical anatomy of the medial segment (S4) of the liver with special reference to bile ducts and vessels. Hepatogastroenterology. 2000;47:143-150. [PubMed] |
| 3. | Machado MA, Surjan RC, Basseres T, Schadde E, Costa FP, Makdissi FF. The laparoscopic Glissonian approach is safe and efficient when compared with standard laparoscopic liver resection: Results of an observational study over 7 years. Surgery. 2016;160:643-651. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 68] [Cited by in RCA: 55] [Article Influence: 5.5] [Reference Citation Analysis (0)] |
| 4. | Fischer L, Cardenas C, Thorn M, Benner A, Grenacher L, Vetter M, Lehnert T, Klar E, Meinzer HP, Lamadé W. Limits of Couinaud's liver segment classification: a quantitative computer-based three-dimensional analysis. J Comput Assist Tomogr. 2002;26:962-967. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 53] [Cited by in RCA: 42] [Article Influence: 1.8] [Reference Citation Analysis (0)] |
| 5. | Radulova-Mauersberger O, Weitz J, Riediger C. Vascular surgery in liver resection. Langenbecks Arch Surg. 2021;406:2217-2248. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 9] [Cited by in RCA: 33] [Article Influence: 6.6] [Reference Citation Analysis (0)] |
| 6. | Maurer R, Rivoire M, Basso V, Meeus P, Peyrat P, Dupré A. Portal supply of segment IV of the liver based on CT-scan. Surg Radiol Anat. 2017;39:471-476. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 11] [Cited by in RCA: 12] [Article Influence: 1.3] [Reference Citation Analysis (0)] |
| 7. | Kim YK, Han HS, Yoon YS, Cho JY, Lee W. Total anatomical laparoscopic liver resection of segment 4 (S4), extended S4, and subsegments S4a and S4b for hepatocellular carcinoma. J Laparoendosc Adv Surg Tech A. 2015;25:375-379. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 20] [Cited by in RCA: 18] [Article Influence: 1.6] [Reference Citation Analysis (2)] |
| 8. | Kobayashi Y, Kawaguchi Y, Kobayashi K, Mori K, Arita J, Sakamoto Y, Hasegawa K, Kokudo N. Portal vein territory identification using indocyanine green fluorescence imaging: Technical details and short-term outcomes. J Surg Oncol. 2017;116:921-931. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 56] [Cited by in RCA: 57] [Article Influence: 6.3] [Reference Citation Analysis (4)] |
| 9. | Xu Y, Chen M, Meng X, Lu P, Wang X, Zhang W, Luo Y, Duan W, Lu S, Wang H. Laparoscopic anatomical liver resection guided by real-time indocyanine green fluorescence imaging: experience and lessons learned from the initial series in a single center. Surg Endosc. 2020;34:4683-4691. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 27] [Cited by in RCA: 64] [Article Influence: 10.7] [Reference Citation Analysis (0)] |
| 10. | Moris D, Ronnekleiv-Kelly S, Rahnemai-Azar AA, Felekouras E, Dillhoff M, Schmidt C, Pawlik TM. Parenchymal-Sparing Versus Anatomic Liver Resection for Colorectal Liver Metastases: a Systematic Review. J Gastrointest Surg. 2017;21:1076-1085. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 120] [Cited by in RCA: 106] [Article Influence: 11.8] [Reference Citation Analysis (2)] |
| 11. | Wang K, Liu Y, Hao M, Li H, Liang X, Yuan D, Ding L. Clinical outcomes of parenchymal-sparing versus anatomic resection for colorectal liver metastases: a systematic review and meta-analysis. World J Surg Oncol. 2023;21:241. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 12] [Cited by in RCA: 13] [Article Influence: 4.3] [Reference Citation Analysis (0)] |
| 12. | Moris D, Tsilimigras DI, Kostakis ID, Ntanasis-Stathopoulos I, Shah KN, Felekouras E, Pawlik TM. Anatomic versus non-anatomic resection for hepatocellular carcinoma: A systematic review and meta-analysis. Eur J Surg Oncol. 2018;44:927-938. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 117] [Cited by in RCA: 120] [Article Influence: 15.0] [Reference Citation Analysis (3)] |
| 13. | Jiao S, Li G, Zhang D, Xu Y, Liu J, Li G. Anatomic versus non-anatomic resection for hepatocellular carcinoma, do we have an answer? A meta-analysis. Int J Surg. 2020;80:243-255. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 26] [Cited by in RCA: 67] [Article Influence: 11.2] [Reference Citation Analysis (2)] |