Original  Article

The mesenteric and antimesenteric location of colorectal cancer: The  relationship with lymph nodes metastases   

A. Benevento1, L. Boni1 G. Dionigi1 G. Carcano1 C. Capella2 G. Capriata3 G. Casula4 G. Dettori5 R. Dionigi1  

¹Department of Surgery and ²Department of Pathology, University of Insubria -Varese, ³Department of Surgery “Ospedale Valduce” - Como, ⁴Department of Surgery, University of Cagliari,⁵Department of Surgery, University of Sassari, Italy 

Correspondence to: L Boni, Department of Surgery, University of Insubria, Ospedale di Circolo di Varese, Viale Borri n.57, 21100 Varese, Italy Email:bonil@tin.it

Introduction Methods Statistics Results

Discussion Conclusion Acknowledgements References

Keywords: Prognostic factors, colorectal cancer, mesenteric, antimesenteric, vascular anatomy, lymph node metastases Surg J R Coll Surg Edinb Irel., 2 August 2004, 214-220

Aim: Purpose of the study was to evaluate if the circumferential location of colorectal cancer may be identified as a possible prognostic factor. The hypothesis is that tumours located on the antimesenteric (AM) side could have a better prognosis than tumours located on the mesenteric (M) side. Methods: All patients undergoing curative resection for colorectal cancer were enrolled in the study. The specimens were sent to the pathologist to define the exact location of the tumour, the histological type, grading, T, N status as well as lymphatic, vascular and neural invasion, peritumoural lymphoid reaction, desmoplasia and microsatellite instability. Statistical analyses were performed using the test for proportions (with continuity correction), the Pearson Chi-square test and generalised linear models; p<0.05 were considered statistically significant. Results: From August 2000 to August 2002, 255 patients were enrolled in the study. There was a significantly higher incidence of tumours located on the M (101) compared with the AM (37) site (p<0.0001). M located tumours were associated with higher numbers of metastatic lymph nodes (N1 and N2; p-value=0.014), whereas AM tumours were associated with involved lymph nodes in only 5/37 (13.5%)of tumours. There was no statistically significant relation between AM versus M location and T status: the Pearson Chi-Square test showed that the lymph node involvement and the location (M versus AM) are not statistically independent variables (p-value=0.014). Conclusions: Our preliminary results show that when M or AM tumour identification is possible, tumour location can be regarded as a prognostic factor. Further  longer studies on recurrence rate and survival are required to validate these findings and the clinical usefulness of this putative prognostic factor 

INTRODUCTION
Colorectal cancer is the second most common cause of death from cancer. In the USA, 150 000 new cases/year have been reported, leading to 60,000 deaths per anum.^1-4

The pathological staging, determined by the degree of tumour penetration through the bowel wall and involvement of lymph nodes, distant organs or both, have been identified as independent factors for survival by multiple randomised clinical trials.^5-8 Several other factors, such as the presence of obstruction or perforation, vascular or lymphatic invasion (or both), perineural invasion, peritumoural lymphocytic infiltration, the character of invasive margin and tumour type, tumour grade, DNA content and allelic loss of chromosome 18q, have all been reported as having an impact on the overall survival of patients with colorectal cancer, although their clinical relevance has still to be confirmed.^6,9-17 

In the early 1930s, Dukes (1940) demonstrated the direct relationship between the extent of local tumour spread and the incidence of lymph node metastasis.¹⁸ The Duke’s staging system, modified by Astler and Coller (1954) is routinely used in clinical practice.¹⁹ Several studies have also demonstrated that 10% to 20% of tumours, limited to the submucosa and muscularis propria already present lymph node metastases.^20,21 

We investigated the possible prognostic value of the circumferential location, mesenteric (M) or antimesenteric (AM) of colorectal cancer. The hypothesis being that tumours on the M site, closer to blood and lymphatic vessels, could result in more rapid and enhanced likelihood of tumour spread. 

We also evaluated if the location on the AM site in the colon and upper rectum, predisposed to the peritoneal spread of the tumour, whereas situated on the M site was responsible for enhanced haematogenous and lymphatic spread. This observation would not be valid for middle and low rectum cancers, where the nature of spread is different. Furthermore, the incidence of different circumferential locations of these tumours was documented, as there are no published data on this aspect. 

METHODS
All patients undergoing curative resection for colorectal cancer were enrolled in the study. Three Italian departments of surgery contributed to the data collection*. The type of resection performed depended on the location of the tumours and surgeon preference, but in all cases vessel ligation was at their origin and, with rectal tumours, total mesorectal excision, as described by Heald et al. (1995), was performed.²² Once the tumour was resected, the specimen was conventionally opened on the anti-mesenteric site and sent for the histopathological assessment (Figure 1). 

Figure 1: The surgical specimen conventionally opened along the antimesenteric side of the colon

The pathologist performed at least three sections on different parts of the tumour in order to analyse the circumferential involvement of the bowel wall at the point of maximum infiltration. The location of the tumour was reported in a designated diagram (Figure 2), where the mesenteric or antimesenteric location could be identified. According to this diagram, the resected colon was divided into eight sectors (12.5% each); tumours mainly involving sectors one and eight were considered AM, those located in sector five and four were M, whilst tumours involving sectors seven and six or two and three were classified as lateral tumours. Substenosing and stenosing tumours were those with advanced circumferential invasion in which identification of the original site of growth was not possible. 

Figure 2: Identification of the tumour location on a specifically designated diagram

The histopathological examination was carried out in the standard manner, reporting the histological type of the tumour, tumour grade, T and N status, as well as the presence of lymphatic, vascular or neural invasion, peritumoural lymphoid reaction, desmoplasia and microsatellite instability. All the data were collected in a specifically designated computerised database and analysed retrospectively. 

STATISTICS
Pearsons Chi-Square Test on the relevant contingency table was used to test whether the two attributes, N and location (M and AM ) were independent variables and a p-value = 0.05 was considered significant. A generalised linear model (Poisson regression) was also used to confirm the Pearson Chi-Square Test, regressing both T and N on location (M and AM) and other known prognostic factors; p-value = 0.05 was considered significant. 

A decision tree was also constructed where the variable to be explained was the N level, in order to predict N status in relationship to tumour location. 

RESULTS
From August 2000 to August 2002, 255 patients were enrolled in the study. There was no statistically significant difference between male and female numbers (54.9% versus 45.1%) and the mean age was 69 years (44-91). Left side sigmoid colon was the most frequent tumour site (37.2%), followed by the right colon (28.3%), high (10.9%), low (10.9%), and middle rectum (8.2%), and transverse colon (4.3%). The mean number of lymph nodes harvested with the resected specimen was 20. 

Table 1 and 2 report tumour sites and UICC stages in relation to the classification we proposed using the mentioned diagram. 

Figure 3 describes the incidence of different tumour location: 75 (29.4%) were stenosing, 19 (7.4%) substenosing, 23 (9%) lateral and 138 (54.1%) were identified as M or AM. Indeed, there was a significantly higher incidence of tumours located on the M (101) versus AM (37) site  (p<0.0001). 

Figure 3: The incidence of tumour location using the proposed classification

There was no statistically significant relation between AM versus M location and T status, hence the two groups were comparable regarding the degree of infiltration of the bowel wall. 

The Pearson Chi-Square test showed that the lymph node involvement and the location (M versus AM) were not statistically independent (p-value=0.014): M located tumours were associated with higher number (43/101 = 42.5%) of metastatic lymph nodes (N1 and N2), whereas AM tumours are associated with involved lymph nodes in only 5/37 (13.5%) tumours (Figure 4). This is confirmed by a generalised Poisson linear model (p-value=0.043).

Figure 4: Mesenteric tumours have more positive lymph nodes than antimesenteric (including M1 patients) 

 Moreover, the combination of the prognostic factors N and T is independent from the circumferential location ( pvalue=0.137). 

Using the algorithm for tree presentation we can forecast that T3 tumours located on the mesenteric site of the colon are likely to have metastatic spread to the regional lymph nodes, while T3 antimesenteric are likely to be N0. 

The analysis does not change upon exclusion of low and medium rectal tumours (34 cases). The low incidence of lymphatic and vascular invasion (only 15 cases: 7 in M and 8 in AM) did not allow statistical evaluation. 

DISCUSSION  
The degree of penetration through the bowel wall (T stage), the involvement of lymph nodes (N stage) or distant organs (M stage), as well as other factors, have been correlated with the prognosis of patients with colorectal cancer.^1-8 

In 1999, a consensus statement from the College of the American Pathologists divided prognostic factors for colorectal cancer into four categories according to the strength of published evidences (Table 3).6 T, N stage, presence of blood or lymphatic vessel invasion, residual tumour after surgery and pre-operative elevation of carcinoembryonic antigen (CEA) are the most important prognostic factors based on the evidence from multiple well designed trials; they are considered category I factors. Category II and III parameters require further investigation to be routinely used in clinical practice, while category IV factors have been demonstrated as having no prognostic value.⁶ 

To date, the prognostic significance of the bowel  circumferential location of colorectal cancer has  not been documented. 

Our hypothesis is that tumours located on the mesenteric site of the colon, being closer to blood and lymphatic vessels, might have more ready access to regional lymph nodes, resulting in a worse prognosis when compared with tumours located on the AM side. 

There is no data on the incidence of the different location of colorectal cancer. Using a specifically designed diagram, we found that more than 50% of the tumours could be identified as M or AM, Thus, a possible prognostic factor related to circumferential tumour location, could be useful for more than half of the patients affected with colorectal cancer. 

Although, in our small study, M tumours were significantly (p<0.001) more frequent than AM cancers. The two groups, however, did not differ statistically (p=0.513) regarding the degree of penetration of the bowel wall (T stage). Also, tumours located on the M site of the colon have a significantly (p=0.042) higher incidence of tumour involved lymph nodes, when compared with AM located cancers. Thus our preliminary study suggests possible new prognostic factor related to the location of the tumour and thereby possibly providing further statistically significant information to the already established prognostic factors (T and N). 

The arteries, veins and lymphatic vessels, supplying and draining each segment of the colon, are located in the mesocolon which also contains lymph nodes and nerves. 

Invasive carcinomas usually spread through the bowel wall in the submucosa, where they gain access to the lymphatic channels that drain to the epicolic lymph nodes located adjacent to the large bowel. Another group of nodes, known as paracolic, lie along the mesenteric side of the colon and are frequently infiltrated by cancer cells.²²  

Mesenteric located cancers are closer to lymph node stations, as well as to larger blood and lymphatic vessels; a different vascular pattern is said to be present on the two side of the colonic wall.²² 

Together with the Department of Anatomy of our University, we are currently running an experimental study on rats, in order to investigate the microcirculation of the colonic wall using a corrosion casting technique and scanning electron microscopy.²⁴ 

Preliminary results of this experimental study (unpublished data) seem to support the postulate that the AM side of the colon might significantly differ from the M, concerning the microvascular arterial density and distribution. As shown in Figure 5, the microvascularisation on the M side of the muscularis mucosae presents a rich network of vessels that is less well developed on the antimesenteric site. 

Figure 5: The different microvascularisation of the M and AM site of the colon as seen at scanning electron microscopy using a casting technique in rat model. The marginal artery (MA), located on the M site of the colon (red arrow), gives vessels which penetrate the bowel wall. On entering the muscularis mucosae on the M side the vessel produces a rich network of vessels (yellow arrow), which appears to be lacking on the antimesenteric side (white arrow). 

These findings need to be confirmed in humans, where the harvesting of the specimens presents some technical difficulties. 

The American College of Pathologists stated that the presence of vascular or lymphatic invasion should be considered primary prognostic factors, but, in our study, these factors were present in only eight and four of the patients with M or AM cancers, respectively. Thus, no conclusions can be made at this stage of the study.⁶ 

In our study the mean number of harvested lymph nodes was 22. It has been demonstrated that there is an increase in the percentage of lymph node metastases when 12 to 20 lymph node are 
recovered.²⁵ 

The technique of lymph node examination and detection of occult metastatic disease, as well as the prognostic significance of lymph node micrometastases, remains an important issue in haematoxylin-eosin staining in node-negative colorectal cancer.^25,26 It has been demonstrated that up to 26% of patients with lymph nodes free of tumour, analysed by routine histological techniques may harbour micrometastases in nodes.²⁷ The significance of occult lymph nodes micrometastases is still a matter of continuing debate.²⁷ 

Adjuvant chemotherapy is usually given to patients with colorectal cancer with involved lymph nodes. In a systematic review of chemotherapy effects in colorectal cancer, Ragnhammar et al. (2001) found that several phase III trials of post-operative adjuvant chemotherapy with fluorouracil and leucovorin in patients with node-involved colon cancer have demonstrated a similar statistically significant improvement in disease-free and overall survival, when compared with control groups.²⁸ 

On the contrary, no convincing benefit from adjuvant chemotherapy, has been fully demonstrated in cases of tumour-free lymph node colon cancer.^{29, 30} 

The preliminary results of our study may have an effect on current practice: patients with negative lymph nodes but M located tumours, may be considered suitable for adjuvant chemotherapy. 

A larger number of patients are needed to confirm our findings and to correlate circumferential location of colorectal cancer to other proven prognostic factors such as the presence of vascular, lymphatic and neural invasion or the degree of host lymphocytic response. 

Our study will continue collecting data on survival and recurrence rate in patients with A and AM colorectal cancer, as long-term outcomes are necessary involving large numbers of patients, to delineate more precisely the clinical relevance of our findings to date.

CONCLUSION 
Our results show that more than 50% of colorectal cancer are located on the M or AM side of the colon. When M or AM tumour identification is possible, tumour location appears to be a putative prognostic factor, as M located cancers have a significantly higher incidence of lymph node metastases. Further studies on recurrence rate and survival will be necessary to validate the usefulness of this putative prognostic factor in clinical practice. 

ACKNOWLEDGEMENTS  
The authors would like to thanks Professor A Mira and Dr. P.Omtzigt from the Department of Statistic of the University of Insubria, for the statistical analysis. 

Footnotes* Department of Surgery, University of Insubria (Director: Prof. R. Dionigi); Department of Surgery, University of Cagliari (Director: Prof. G. Casula); Department of Surgery, University of Sassari (Director: Prof. G. Dettori) General Surgery, “Valduce” Hospital - Como (Head: Dr. G. Capriata)

REFERENCES
1. Parkin DM, Pisani P, Ferlay J. Estimates of worldwide incidence of eighteen major cancers in 1985. Int J Cancer 1993;54:594-606. 
2. Hannisdal E, Thorsen G. Regression analyses of prognostic factors in colorectal cancer. J Surg Oncol 1988; 37: 109-112. 
3. Merrill RM, Capocaccia R, Feuer EJ, Mariotto A. Cancer prevalence estimates based on tumour registry data in the Surveillance, Epidemiology, and End Results (SEER) Program. Int J Epidemiol 2000 Apr;29:197-207. 
4. Chu KC, Tarone RE, Chow WH, Hankey BF, Ries LA. Temporal patterns in colorectal cancer incidence, survival, and mortality from 1950 through 1990. J Natl Cancer Inst 1994;86:997-1006. 
5. Newland RC, Dent OF, Lyttle MN, Chapuis PH, Bokey EL.. Pathologic determinants of survival associated with colon or rectal cancer with lymph node metastases: a multivariate analysis of 579 patients. Cancer. 1994; 73:2076-82. 
6. Compton CC, Fielding LP, Burgart LJ, Conley B, Cooper HS, Hamilton SR et al. Prognostic factors in colorectal cancer. Arch Pathol Lab Med. 2000; 124: 979-94. 
7. Crucitti F, Sofo L, Doglietto GB, Bellantone R, Ratto C, Bossola M, Crucitti A. Prognostic factors in colorectal cancer: current status and new trends. J Surg Oncol suppl. 1991; 2: 76-82. 
8. Hannisdal E, Thorsen G. Regression analyses of prognostic factors in colorectal cancer. J Surg Oncol 1988; 37: 109-112. 
9. Griffin MR, Bergstralh EJ, Coffey RJ, Beart RW Jr, Melton LJ 3rd. Predictors of survival after curative resection of carcinoma of the colon and rectum. Cancer 1987;60:2318-24. 
10. Minsky B, Mies C. The clinical significance of vascular invasion in colorectal cancer. Dis Colon Rectum 1989;32:794-803. 
11. Knudsen JB, Nilsson T, Sprechler M, Johansen A, Christensen N. Venous and nerve invasion as prognostic factors in post-operative survival of patients with resectable cancer of the rectum. Dis Colon Rectum. 1983; 26:613-17. 
12. Fielding LP, Phillips RKS, Fry JS, Hittinger R. Prediction of outcome after curative resection for large bowel cancer. Lancet 1986;2:904-7.  
13. Svennevig JL, Lunde OC, Holter J, Bjorgsvik D. Lymphoid infiltration and  prognosis in colorectal carcinoma. Br J Cancer. 1984; 49: 375-77. 
14. Jass JR. The pathological grading and staging of rectal cancer. Scand J Gastroenterol Suppl 1988;149:21-38. 15. Saccani Jotti G, Fontanesi M, Orsi N, Sarli L, Pietra N, Peracchia A, et al. DNA content in human colon cancer and non-neoplastic adjacent mucosa. Int J Biol Markers 1995;10:11-16. 
16. Lanza G, Matteuzzi M, Gafa R, Orvieto E, Maestri I, Santini A, del Senno L. Chromosome 18q allelic loss and prognosis in stage II and III colon cancer. Int J Cancer 1998;79:390-95. 
17. Ogunbiyi OA, Goodfellow PJ, Herfarth K, Gagliardi G, Swanson PE, Birnbaum EH, et al. Confirmation that chromosome 18q allelic loss in colon cancer is a prognostic indicator. J Clin Oncol 1998;16:427-33. 
18. Dukes CE. The classification of the cancer of the rectum. J Pathol Bacteriol. 1940; 50: 527. 
19. Astler VB, Coller FA. Prognostic significance of direct extension of  carcinoma of the colon and rectum, Ann Surg 1954; 139: 846-52. 
20. Grinnell RS. The grading and prognosis of carcinoma of the colon and rectum. Ann Surg 1939; 109: 500. 
21. Gilchrist R, David D. Lymphatic spread of carcinoma of the rectum. Ann Surg 1948;38:621. 
22 Heald RJ. Total mesorectal excision is optimal surgery for rectal cancer: a Scandinavian consensus. Br J Surg 1995;82:1297-99. 
23. De Vita VT, Hellman S, Rosemberg SA. Cancer Principles & Practice of Oncology. 6th edn. Philadelphia: Lipincott Williams & Wilkins. 2001:1216- 71.  
24. Hossler FE, West RF. Venous valve anatomy and morphometry: studies on the duckling using vascular corrosion casting. Am J Anat 1988 Apr;181:425- 32. 
25. Goldstein N, Hart J. Histologic features associated with lymph node metastasis in stage T1 and superficial T2 rectal adenocarcinomas in abdominoperineal resection specimens. Identifiying a subset of patients for whom treatment with adjuvant therapy or completion abdominoperineal resection should be considered after local excision. Am J Clin Pathol 1999;111:51-58. 
26. Mainprize KS, Kulacoglu H, Hewavisinthe J, Savage A, Mortensen N, Warren BF. How many lymph nodes to stage colorectal carcinoma? J ClinPathol 1998;51:165-66. 
27. Oberg A, Stenling R, Tavelin B, Lindmark G. Are lymph node micrometastasis of any clinical significance in Dukes stages A and B colorectal cancer? Dis Colon Rectum 1998;41:1244-49. 
28. Ragnhammar P, Hafstrom L, Nygren P, Glimelius B; SBU-group. Swedish Council of Technology Assessment in Health Care. A systematic overview of chemotherapy effects in colorectal cancer. Acta Oncol 2001;40:282-308. 
29. Ooi BS, Tjandra JJ, Green MD. Morbidities of adjuvant chemotherapy and radiotherapy for resectable rectal cancer: an overview. Dis Colon Rectum 1999 Mar;42:403-18. 
30. Schmoll HJ. Adjuvant chemotherapy in rectal cancer. Chirurg 1994 Jul;65:576-84. 

Copyright: 1 July 2004