Imaging prediction of lymph node metastasis and early recurrence in patients with distal cholangiocarcinoma

Introduction

Biliary tract malignancies are classified based on tumor location as intrahepatic cholangiocarcinoma (ICC), peri-hilar cholangiocarcinoma, and distal cholangiocarcinoma (DCC) (1), among which DCC refers to the tumors that occur below the confluence of the cystic duct and above the ampulla of vater (2). The location of biliary tract malignancies determines their different characteristics. DCC has a higher lymph node (LN) metastasis rate and a poorer prognosis compared with the other types of biliary tract malignancies (3,4). Despite advances in perioperative care and surgical techniques, recent studies reported a concerning 0–50% 5-year survival rate following pancreaticoduodenectomy (PD) (3-5). The tendency of DCC to recur following surgical resection provides the necessity for preoperative accurate prediction of early recurrence (ER).

Some clinicopathologic factors have been investigated to predict the prognosis of resected DCC. The factors that are reported to be associated with ER, include tumor size, surgical margins, perineural invasion (PNI), microvascular invasion (MVI), pancreatic invasion, pathological differentiation, LN metastasis, molecular biomarkers, etc. (6,7). Among these factors, LN metastasis is regarded as one of the most potent predictors of ER after PD for DCC (8,9). Therefore, accurate preoperative diagnosis of LN status is crucial to determining a proper treatment strategy for DCC.

Upper abdominal enhanced computed tomography (CT) is a routine imaging method for preoperative assessment of DCC. Imaging features based on CT are used to evaluate the LN status of DCC before surgery (10,11). However, preoperative identification of LN metastasis on cross-sectional imaging in patients with DCC remains challenging for radiologists, and there is no international consensus yet. In clinical work, benign and malignant LNs are generally judged by measuring the diameter of the short axis in the LN. However, several reports described unsatisfactory sensitivity rates for CT in the detecting LN metastasis from cholangiocarcinoma when LNs ≥10 mm were considered to be positive for metastasis (12-15). Noji et al. conducted a retrospective analysis of clinical data and histopathological findings of 146 patients who had undergone regional lymphadenectomy with biliary cancer. The results showed that metastatic LNs were significantly larger than non-metastatic nodes (9.6 vs. 6.6 mm in short-axis diameter), and LN size >10 mm had a positive predictive value of only 28% (10). The diagnosis for metastatic LNs based on short-axis diameter alone was insufficient to provide a reliable pre-therapeutic evaluation of cholangiocarcinoma. Similar results were reported for esophageal cancer, lung cancer, colon cancer, and pancreatic cancer (15-18). Therefore, more diagnostic indicators, such as the short-to-long axis ratio, CT-value changes between arterial and venous phases, and central necrosis, have been found to improve CT diagnostic value for metastatic LNs (2,15). Besides, it was reported that the number of metastatic LNs was also associated with the ER of malignant tumor patients. According to the findings of Zhang et al., patients with 0, 1–2, or ≥3 LN metastasis had incrementally worse disease-specific and recurrence-free survival following curative-intent resection of ICC (19). Additionally, the American Joint Committee on Cancer (AJCC) 8^th edition guidelines (20) recommend that at least 6 LNs be assessed to stage the DCC accurately, again highlighting the importance of the number of LNs. Therefore, combining the CT imaging features of both size and number of LNs would be more reliable in determining the LN status. However, most previous studies on imaging features for evaluating the LN status focused on size and internal characteristics, but few examining the predictive value of the number of LNs. Moreover, few studies have reported the LN metastasis and prognostic relevance of CT for patients with DCC. More accurate standards in CT imaging to predict LN metastasis need to be explored in DCC patients.

This research aims to determine the value of imaging features of regional LNs in predicting LN metastasis, and thereby providing clues for predicting ER in patients with DCC. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2024-1017/rc).

Methods

Patient selection and clinical data

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Beijing Tsinghua Changgung Hospital (No. 23393-0-01) and written informed consent was waived by the Institutional Review Board.

Sixty-six patients who underwent PD for DCC in Beijing Tsinghua Changgung Hospital between January 2015 and August 2022 were retrospectively identified. The exclusion criteria were: (I) patients did not have upper abdominal enhanced multi-slice CT (MSCT) imaging within 1 month before the surgery; (II) the patient’s clinical records and follow-up information were incomplete; (III) patient underwent invasive treatment that was performed between the upper abdominal enhanced MSCT scan and the surgery; and (IV) the follow-up period was less than 12 months. In this study, 5 patients were excluded due to either a lack of upper abdominal enhanced MSCT imaging within 1 month before the surgery (n=4) or a follow-up period of less than 12 months (n=1). Therefore, the final sample comprised 61 patients, of whom 40 patients (65.6%) were male, with a mean age of 65.8±10.5 years (range, 27–85 years). The institutional internal review board approved the study.

Clinicopathologic variables of patients with DCC extracted included age, sex, preoperative serum levels of carbohydrate antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA), tumor size, PNI, MVI, pancreatic invasion, pathological differentiation (good, moderate, and poor), and LN metastasis (peri-pancreas, peri-bile duct, and peri-duodenum). Physical examinations for serum CA19-9 and CEA were performed within 1 week prior to surgery. Tumor size was determined and defined as the maximum diameter of the tumor at pathological examination. MVI and PNI were defined as microscopic tumor invasion of the microvascular or neural element of the surrounding normal tissue, respectively. The LNs removed were grouped as regional in line with the nodal classification of the eighth edition of the tumor-node-metastasis (TNM) staging manual by the AJCC (20). Two experienced pathologists evaluated all the histopathological slides in a blinded manner.

Image acquisition and evaluation

Scans were completed with the participants in the supine position. All the participants were subjected to upper abdominal enhanced MSCT imaging, which was performed using either a Discovery CT 750 high-definition (HD) scanner (GE Healthcare, Waukesha, WI, USA). All participants received the median cubital vein administration of 90 mL contrast media (Omnipaque 350, GE Healthcare, Shanghai, China) at a rate of 3 mL/s. After the start injection of the contrast medium, an abdominal CT scan was conducted in the arterial phase at 32 s, in the portal vein phase at 70 s, and in the delayed phase at 180 s, respectively. The following parameters were used: collection thickness, 0.625 mm; reconstruction thickness, 1.25 mm; pixel size, 0.684 mm × 0.684 mm; tube voltage, 120 kVp; tube current, 300 mA; and matrix size, 512×512.

Location, number, short axis, long axis, and central necrosis of LNs were evaluated on the axial images of the portal vein phase. Peri-pancreatic LN was defined as the LN in 13 and 17 stations (21) (Figure 1). The number of LN referred to the total number of peri-pancreatic LNs in the above stations. The size of LN was measured by the short axis and long axis of the largest LN, and then the ratio of short axis to long axis was calculated. Central necrosis was defined as the non-enhanced area appearing within the mass. CT-value changes between arterial and portal vein phases equaled to the CT value of portal vein phase minus the CT value of arterial phase. All the imaging features were evaluated independently by two radiologists in abdominal imaging with 5- and 10-year experience, respectively. The inconsistencies in the evaluation results between the two radiologists were resolved by consensus.

Figure 1 DCC with peri-pancreatic LN metastasis was confirmed by pathology in a 79-year-old man who experienced recurrence (distant metastasis) 7 months after PD. (A) Transverse enhanced CT image showing a mass in distal common bile duct with a size of 18.0 mm × 21.0 mm (white arrowhead). (B) Histopathologic image showing peripancreatic LN metastasis (hematoxylin-eosin; magnification, ×17). (C) Histopathologic image showing peripancreatic LN metastasis (hematoxylin-eosin; magnification, ×72). (D,E) Transverse enhanced CT images showing LNs with short diameter ≥6 mm in peripancreatic region by white arrows. (F) Enlarged view of the region LN marked by the dashed rectangle in (E) (white arrows). (G) Transverse enhanced CT image of portal vein phase showing the diameters of the largest LN was measured (white lines). (H,I) Transverse enhanced CT images of arterial phase and portal vein phase showing CT values of the largest LN were measured by region of interest (white circles) to calculate the CT value change. CT, computed tomography; DCC, distal cholangiocarcinoma; LN, lymph node; PD, pancreaticoduodenectomy.

Follow-up after resection

Follow-up was conducted at 1, 3, 6, and 12 months after PD. ER was defined as local recurrence or distant metastasis in DCC patients ≤12 months after PD.

Statistical analysis

Data were expressed as mean ± standard deviation (SD) or number and frequency [n (%)]. The Chi-squared test or Mann-Whitney U test was used to compare the clinical characteristics of the ER and non-ER groups. The Mann-Whitney U test was used to compare the imaging features between the LN metastasis and the non-LN metastasis groups. Then, the odds ratios (ORs) were calculated, and independent imaging features for predicting LN metastasis were identified through binary logistic regression analysis. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were calculated to determine the value of imaging features in predicting LN metastasis and ER in patients with DCC. True positive (TP) denoted the number of samples that were actually positive examples and were predicted to be positive examples, false negative (FN) was the number of samples that were actually positive but predicted to be negative, true negative (TN) was the number of samples that were actually negative cases and predicted to be negative cases, and false positive (FP) was the number of samples that were actually negative but predicted to be positive.

The formulas were used as follows: sensitivity = TP/(TP + FN) × 100%; specificity = TN/(FP + TN) × 100%; positive predictive value = TP/(TP + FP) × 100%; negative predictive value = TN/(FN + TN) × 100%; and accuracy = (TP + TN)/(TP + FP + TN + FN) × 100%. P values were two-sided for all tests, and P<0.05 indicated statistical significance. All statistical analyses were performed by using IBM SPSS Statistics (SPSS) 25.0 (SPSS, Inc., Chicago, IL, USA).

Results

Clinical characteristics of patients with DCC

Of 61 DCC patients, 19 (31.1%) experienced ER who were classified the as ER group, and the other 42 (68.9%) did not experience ER who were classified as the non-ER group. By comparing the clinical characteristics between these two groups, patients in the ER group showed significantly higher percentages in pancreatic invasion (68.4% vs. 40.5%), LN metastasis (57.9% vs. 28.6%), and peri-pancreatic LN metastasis (52.6% vs. 23.8%) than those in the non-ER group (P=0.045, 0.03, and 0.03, respectively). However, there were no significant differences observed between groups in the percentages of age, sex, CA19-9, CEA, pathological R0 resection, tumor size, PNI, MVI, pathological differentiation, and LN metastasis in the regions of peri-bile duct and peri-duodenum (all P values >0.05). The details of comparison results were listed in Tables 1,2.

Imaging features of LNs surrounding pancreas

Of these 61 DCC patients, 20 (32.8%) were confirmed to have peri-pancreatic LN metastasis by postoperative pathology, and included in the LN metastasis group. The other 41 (67.2%) DCC patients without peri-pancreatic LN metastasis were included in the non-LN metastasis group. Significantly more patients in LN metastasis group had at least 1 LN with short-axis diameter ≥10 mm, at least 2 LNs with short-axis diameter ≥8 mm, and at least 3 LNs with short-axis diameter ≥6 mm than in the non-LN metastasis group (P=0.04, 0.02, and 0.003, respectively). According to the results of multivariate analyses, at least 3 LNs with short-axis diameter ≥6 mm was an independent risk factor for peri-pancreatic LN metastasis (P=0.03) with an OR value of 4.831 [95% confidence interval (CI): 1.189–19.608].

Patients in LN metastasis group also showed higher numbers in at least 1 LN with short-axis diameter ≥8 mm, at least 1 LN with short-axis diameter ≥6 mm, at least 2 LNs with short-axis diameter ≥6 mm, ratio of short to long axis >0.8, CT-value changes between arterial and venous phases >15 Hounsfield unit (HU); however, there were no significant differences between two groups (all P values >0.05; Table 3).

Predictive value of imaging features in LN metastasis and ER

When compared with the commonly used diagnosing criteria for LN metastasis of at least 1 LN with short-axis diameter ≥10 mm or central necrosis in LN, the imaging feature of at least 3 LNs with short-axis diameter ≥6 mm showed better performance on predicting pancreatic LN metastasis in sensitivity (50.0%), specificity (85.4%), positive predictive value (62.5%), negative predictive value (77.8%), and accuracy (73.8%). The commonly used diagnosing criteria for LN metastasis had better performance on predicting ER in sensitivity (52.6%), specificity (78.6%), positive predictive value (52.6%), negative predictive value (78.6%), and accuracy (70.5%) (Table 4).

Discussion

LN metastasis is one of the most impacting prognostic factors for patient with DCC after surgical resection. It was reported that regional LN metastasis often portended a poor prognosis (9,22-24). In this study, patients experiencing ER after surgery showed a significantly higher percentage of peri-pancreatic LN metastasis than those not experiencing ER after surgery. To accurately predict the LN metastasis in this region preoperatively, we analyzed the imaging features of preoperative enhanced upper abdominal MSCT, and found that the imaging feature of at least 3 LNs with short-axis diameter ≥6 mm was an independent risk factor to predict peri-pancreatic LN metastasis with a high specificity.

In this study, patients with DCC after surgery had a significantly higher percentage of peri-pancreatic LN metastasis in preoperative CT imaging compared to those not experiencing ER (52.6% vs. 23.8%, P=0.03), while there were no significant differences in the regions of peri-bile duct and peri-duodenum. This finding may be contributed by the specific factors related to the anatomical and biological characteristics of DCC. One possible explanation is the anatomical proximity of the peri-pancreatic LNs to the distal bile duct. The peri-pancreatic LNs are located in close proximity to the distal bile duct, which is the primary site of the cancer (21,25). This close proximity may increase the likelihood of cancer cells spreading to the peri-pancreatic LNs in patients with DCC. Additionally, the lymphatic drainage pattern in the distal pattern in the distal bile duct region may play a role in the observed higher percentage of peri-pancreatic LN metastasis. The lymphatic vessels in this area may have a higher propensity to drain cancer cells from the distal bile duct to the peri-pancreatic LNs, leading to a higher rate of LN metastasis in this region (9,26). The preoperative CT imaging provided valuable information about the extent and location of LN involvement. The presence of metastatic LNs in the peri-pancreatic region on preoperative imaging may indicate a more aggressive tumor biology and a higher risk of ER after surgery in patients with DCC.

Among all the imaging features evaluated in this study, the feature of at least 3 LNs with short-axis diameter ≥6 mm were found to be the independent factor to predict peri-pancreatic LN metastasis (P=0.03). Although there were significant differences in the imaging features of at least 1 LN with short-axis diameter ≥10 mm and at least 2 LNs with short-axis diameter ≥8 mm between the LN metastasis group and the non-LN metastasis group, they were not found to be independent factors. In this study, we combined the size and the number of regional LN to predict LN metastasis, which was seldomly reported in previous studies. Most of previous reports studied the associations between the size of LN and metastasis, and there has been no consensus on the cutoff size of LN. Tseng et al. studied 198 patients with pancreatic head or periampullary carcinoma, and found that the imaging feature of size >10 mm of LN had the OR value of 2.1 (P=0.13) to predict LN metastasis (27). Kolck et al. enrolled 102 patients with ICC, and their results showed that the sensitivity and specificity of LN short axis diameter ≥10 mm to predict LN metastasis were 76.19% and 48.84%, of LN short axis diameter ≥8 mm were 92.86% and 20.93%, and of LN short axis diameter ≥6 mm were 100.00% and 0.00% (13). In this study, the combination of LN short-axis diameter ≥6 mm and 3 LNs showed higher specificity (85.4%) to predict LN metastasis than the findings in previous reports (11,13). In comparison with the commonly used diagnostic criteria of at least 1 LN with short-axis diameter ≥10 mm or central necrosis, at least 3 LNs with short-axis diameter ≥6 mm recommended in this study showed better results in all the aspects of sensitivity, specificity, positive predictive value, negative predictive value, and accuracy to predict LN metastasis. While the predict power for ER was not as high as with at least 1 LN with short-axis diameter ≥10 mm or central necrosis, the predictive accuracy may be enhanced by incorporating other clinical and pathological factors, such as pancreatic invasion, PNI, and neutrophil-to-lymphocyte ratio, a proposition we aim to validate in the future study (28,29). In addition, although the imaging features of at least 1 LN with short-axis diameter ≥10 mm and at least 2 LNs with short-axis diameter ≥8 mm were not independent risk factors to predict peri-pancreatic LN metastasis in this study, the results reflected that when the short axis of LN was less than 10 mm, the total number of regional LN could provide a clue for assessing LN metastasis.

There were other imaging features evaluated in this study; however, no significant differences were observed between the LN metastasis group and the non-LN metastasis group, including ratio of short to long axis >0.8, CT-value changes between arterial and venous phases >15 HU, and central necrosis, which could not use as predictive factors. Noji et al. reviewed the size and shape of LN on CT images from 57 patients with biliary carcinoma (11). Their results showed the positive predictive value of axial ratio of LN ≥0.7 was not substantial, ranging from 22% to 30%. Noji et al. also found that there were no significant differences in axial ratio of LN in the regions of peribiliary and around common hepatic artery, which are similar to our results (10). To the best of our knowledge, there has been no reports on the enhancement pattern of LN in patients with DCC. There was no intergroup difference in the imaging feature of CT-value changes between arterial and venous phases >15 HU in this study. This may be due to the low number of cases with LN necrosis of our study sample: 7 of 61 (11.5%) patients showed necrosis in LNs on the MSCT images. In addition, Tseng et al. studied the CT imaging features of patients with pancreatic head or periampullary carcinoma (14). They also found that the proportion of central necrosis occurrence was low (4.5%), and there was no significant difference between LN metastasis and non-LN metastasis (P=0.12).

In this study, we also compared the clinical and pathological features between patients experiencing ER and not experiencing ER after surgery. Apart from pancreatic invasion, there were no significant differences in intergroup comparisons of other factors, including age, sex, CA19-9, CEA, tumor size, PNI, MVI, and pathological differentiation. Some of these findings were inconsistent with previous reports. Ito et al. enrolled 61 patients with DCC to observe the risk factors of ER after surgery. By univariate analysis, PNI was significantly different between patients with ER and without ER (P<0.001) (28). Sahara et al. included 245 patients in analysis, and found that lymphovascular invasion was associated with a higher ER risk (29). The hazard ratio was 2.51 (95% CI: 1.47–4.42; P<0.001). Umezawa et al. retrospectively examined 88 patients who had undergone PD for DCC, and their results showed that significant differences between the ER group and the non-ER group in pathological differentiation (P=0.02) (30). The proportion of poorly differentiated cases was significantly higher among ER patients than non-ER patients (24.3% vs. 5.9%). These inconsistencies could be attributed to several factors, such as sample size, patient characteristics, and heterogeneity of DCC.

There were some limitations in this study. First, this investigation constituted a single-center study encompassing a cohort of DCC patients, thus necessitating future validation through a multicenter approach with an expanded sample size for broader applicability of the findings. Second, this retrospective study exclusively involved operable patients, potentially introducing selection bias. Third, this study did not entail a one-to-one correlation between imaging and pathology; rather, it focused on the predictive value of imaging features for LN metastasis. Subsequent prospective research is warranted to establish a direct association between LN imaging and pathological outcomes.

Conclusions

In conclusion, the imaging feature of at least 3 LNs with short-axis ≥6 mm could predict peri-pancreatic LN metastasis with a high specificity, as well as providing a clue for predicting ER after PD in patients with DCC.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2024-1017/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2024-1017/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2024-1017/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2024-1017/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Beijing Tsinghua Changgung Hospital (No. 23393-0-01) and written informed consent was waived by the Institutional Review Board.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Oneda E, Abu Hilal M, Zaniboni A. Biliary Tract Cancer: Current Medical Treatment Strategies. Cancers (Basel) 2020;12:1237. [Crossref] [PubMed]
2. Engels JT, Balfe DM, Lee JK. Biliary carcinoma: CT evaluation of extrahepatic spread. Radiology 1989;172:35-40. [Crossref] [PubMed]
3. Hirose Y, Sakata J, Nomura T, et al. Prognostic relevance of lymph node metastasis in pancreaticoduodenectomy for distal cholangiocarcinoma: Rational extent and number-based nodal classification for regional lymphadenectomy. Surgery 2025;180:109099. [Crossref] [PubMed]
4. Kato T, Okada K, Baba Y, et al. Preoperative prognostic stratification and prediction of long-term outcomes after pancreatoduodenectomy for distal cholangiocarcinoma. Eur J Surg Oncol 2024;50:108691. [Crossref] [PubMed]
5. Banales JM, Marin JJG, Lamarca A, et al. Cholangiocarcinoma 2020: the next horizon in mechanisms and management. Nat Rev Gastroenterol Hepatol 2020;17:557-88. [Crossref] [PubMed]
6. Xu S, Zhang XP, Zhao GD, et al. Development and validation of an online calculator to predict early recurrence and long-term survival in patients with distal cholangiocarcinoma after pancreaticoduodenectomy. J Hepatobiliary Pancreat Sci 2022;29:1214-25. [Crossref] [PubMed]
7. Zhou Y, Liu S, Wu L, et al. Survival after surgical resection of distal cholangiocarcinoma: A systematic review and meta-analysis of prognostic factors. Asian J Surg 2017;40:129-38. [Crossref] [PubMed]
8. Di Martino M, Koh YX, Syn N, et al. It is the lymph node ratio that determines survival and recurrence patterns in resected distal cholangiocarcinoma. A multicenter international study. Eur J Surg Oncol 2022;48:1576-84. [Crossref] [PubMed]
9. Kurahara H, Mataki Y, Idichi T, et al. Spread of lymph node metastasis and adjuvant therapy for distal cholangiocarcinoma. Int J Clin Oncol 2022;27:1212-21. [Crossref] [PubMed]
10. Noji T, Kondo S, Hirano S, et al. Computed tomography evaluation of regional lymph node metastases in patients with biliary cancer. Br J Surg 2008;95:92-6. [Crossref] [PubMed]
11. Noji T, Kondo S, Hirano S, et al. CT evaluation of paraaortic lymph node metastasis in patients with biliary cancer. J Gastroenterol 2005;40:739-43. [Crossref] [PubMed]
12. Kim NH, Lee SR, Kim YH, et al. Diagnostic Performance and Prognostic Relevance of FDG Positron Emission Tomography/Computed Tomography for Patients with Extrahepatic Cholangiocarcinoma. Korean J Radiol 2020;21:1355-66. [Crossref] [PubMed]
13. Kolck J, Auer TA, Walter-Rittel T, et al. Prediction of regional lymph node metastasis in intrahepatic cholangiocarcinoma: it's not all about size. Abdom Radiol (NY) 2023;48:3063-71. [Crossref] [PubMed]
14. Tseng DS, van Santvoort HC, Fegrachi S, et al. Diagnostic accuracy of CT in assessing extra-regional lymphadenopathy in pancreatic and peri-ampullary cancer: a systematic review and meta-analysis. Surg Oncol 2014;23:229-35. [Crossref] [PubMed]
15. Liu J, Wang Z, Shao H, et al. Improving CT detection sensitivity for nodal metastases in oesophageal cancer with combination of smaller size and lymph node axial ratio. Eur Radiol 2018;28:188-95. [Crossref] [PubMed]
16. Mönig SP, Baldus SE, Zirbes TK, et al. Lymph node size and metastatic infiltration in colon cancer. Ann Surg Oncol 1999;6:579-81. [Crossref] [PubMed]
17. Prenzel KL, Hölscher AH, Vallböhmer D, et al. Lymph node size and metastatic infiltration in adenocarcinoma of the pancreatic head. Eur J Surg Oncol 2010;36:993-6. [Crossref] [PubMed]
18. Prenzel KL, Mönig SP, Sinning JM, et al. Lymph node size and metastatic infiltration in non-small cell lung cancer. Chest 2003;123:463-7. [Crossref] [PubMed]
19. Zhang XF, Lv Y, Weiss M, et al. Should Utilization of Lymphadenectomy Vary According to Morphologic Subtype of Intrahepatic Cholangiocarcinoma? Ann Surg Oncol 2019;26:2242-50. [Crossref] [PubMed]
20. Amin MB, Edge SB, Greene FL, et al. AJCC cancer staging manual. 8th ed. Cham: Springer; 2017.
21. Takagi K, Nagai Y, Umeda Y, et al. Prognostic Value of the Regional Lymph Node Station in Pancreatoduodenectomy for Ampullary Carcinoma. In Vivo 2022;36:973-8. [Crossref] [PubMed]
22. Kiriyama M, Ebata T, Aoba T, et al. Prognostic impact of lymph node metastasis in distal cholangiocarcinoma. Br J Surg 2015;102:399-406. [Crossref] [PubMed]
23. Lyu S, Li L, Zhao X, et al. Prognostic impact of lymph node parameters in distal cholangiocarcinoma after pancreaticoduodenectomy. World J Surg Oncol 2020;18:262. [Crossref] [PubMed]
24. Qian X, Ni X, Miao G, et al. Association Between MRI-Based Radiomics Features and Regional Lymph Node Metastasis in Intrahepatic Cholangiocarcinoma and Its Clinical Outcome. J Magn Reson Imaging 2025;61:997-1010. [Crossref] [PubMed]
25. Tashiro K, Kuroki N, Einama T, et al. Prognostic significance of regional lymph node metastasis according to station in ampullary carcinoma. J Hepatobiliary Pancreat Sci 2020;27:712-20. [Crossref] [PubMed]
26. Lurje G, Bednarsch J, Czigany Z, et al. The prognostic role of lymphovascular invasion and lymph node metastasis in perihilar and intrahepatic cholangiocarcinoma. Eur J Surg Oncol 2019;45:1468-78. [Crossref] [PubMed]
27. Tseng DSJ, Pranger BK, van Leeuwen MS, et al. The Role of CT in Assessment of Extraregional Lymph Node Involvement in Pancreatic and Periampullary Cancer: A Diagnostic Accuracy Study. Radiol Imaging Cancer 2021;3:e200014. [Crossref] [PubMed]
28. Ito Y, Abe Y, Egawa T, et al. Predictive Factors of Early Recurrence in Patients with Distal Cholangiocarcinoma after Pancreaticoduodenectomy. Gastroenterol Res Pract 2018;2018:6431254. [Crossref] [PubMed]
29. Sahara K, Tsilimigras DI, Toyoda J, et al. Defining the Risk of Early Recurrence Following Curative-Intent Resection for Distal Cholangiocarcinoma. Ann Surg Oncol 2021;28:4205-13. [Crossref] [PubMed]
30. Umezawa S, Kobayashi S, Otsubo T. Low preoperative psoas muscle mass index is a risk factor for distal cholangiocarcinoma recurrence after pancreatoduodenectomy: a retrospective analysis. World J Surg Oncol 2022;20:176. [Crossref] [PubMed]