|Ahead of print
Assessing the histopathology reports of colorectal carcinoma surgery: An audit of three years with emphasis on lymph node yield
Sourav Sarkar1, Kedar K Deodhar2, Atul Budukh3, Munita M Bal2, Mukta Ramadwar2
1 Ex Senior Registrar, Department of Pathology, Tata Memorial Hospital, Parel, Mumbai, Maharashtra, India
2 Department of Pathology, Tata Memorial Hospital, Parel, Mumbai, Maharashtra, India
3 Centre for Cancer Epidemiology, Advanced Centre for treatment and Research in Cancer (ACTREC), Tata Memorial Centre, Kharghar, Navi Mumbai, Maharashtra, India
|Date of Submission||19-Dec-2019|
|Date of Decision||22-Jun-2020|
|Date of Acceptance||02-Sep-2020|
|Date of Web Publication||24-Jun-2021|
Kedar K Deodhar,
Department of Pathology, Tata Memorial Hospital, Parel, Mumbai, Maharashtra
Source of Support: None, Conflict of Interest: None
Background: A comprehensive histopathology report of colorectal carcinoma surgery is important in cancer staging and planning adjuvant treatment. Our aim was to review histopathology reports of operated specimens of colorectal carcinoma in our institution between 2013 and 2015 to assess different histological parameters, including lymph node yield, and to evaluate compliance to minimum data sets.
Methods: After approval by the institutional review board (IRB), we analyzed 1230 histopathology reports of colorectal carcinoma between 2013 and 2015. Various gross and microscopic findings (along with age, sex) were noted, for example, specimen type, tumor site, resection margins including circumferential resection margin (CRM), lymphovascular invasion, perineural invasion, pTNM stage, lymph node yield, etc.
Results: Out of 1230 patients, 826 (67.15%) were men and 404 (32.85%) were women. The overall mean age was 52 (range: 18 - 90) years. There were 787 surgeries for rectal cancers. All reports commented on the type of specimen, tumor size (mean = 4.38 cm), proximal, and distal margins. Lymphovascular invasion (LVI) and the pT stage were mentioned in 98.06% and 99.84%, respectively. The overall mean lymph node yield was 18.38 (median = 15, range = 0-130 lymph nodes). A statistically significant difference in lymph node yield was detected between rectal and colonic cancer patients (14.79 and 27.26); post neoadjuvant therapy (NACT) cases, and NACT naive cases (13.51 and 25.11); and high tumor stage and low tumor stage disease (20.60 and 15.22). Not commenting on extramural vascular emboli, tumor budding, and CRM in non-rectal cancer cases were the lacunae.
Conclusion: Our compliance with minimal data sets is satisfactory. The overall mean lymph node yield was 18.38 (median = 15). Extramural vascular emboli, tumor budding need to be captured.
Keywords: Colorectal neoplasms, lymph nodes, pathology
Key Message: Lymph node yield at time of surgery is an important prognostic factor in colorectal cancer. Along with other histopathology details, extramural vascular emboli, tumour budding, and CRM in nonrectal colon
cancer have prognostic significance.
|How to cite this URL:|
Sarkar S, Deodhar KK, Budukh A, Bal MM, Ramadwar M. Assessing the histopathology reports of colorectal carcinoma surgery: An audit of three years with emphasis on lymph node yield. Indian J Cancer [Epub ahead of print] [cited 2021 Jul 27]. Available from: https://www.indianjcancer.com/preprintarticle.asp?id=319332
| » Introduction|| |
A complete cancer histopathology report helps in confirming the diagnosis, gives optimum staging which helps plan the adjuvant therapy. It can also assess the effectiveness of neoadjuvant therapy. Furthermore, it is a useful tool to evaluate the quality of services, collect data for epidemiology.
The practice of using the templates and minimum datasets for cancer surgery reporting is in use for nearly two decades. Such data sets from various reputed organizations (such as the Royal College of Pathologists, UK; College of American Pathologists, USA, etc.) are widely and freely available.,
Aims and objectives
1) To review surgical pathology reports of surgical specimens of colorectal carcinoma operated at our institution between 2013 and 2015. 2) To assess different histological parameters that are mentioned in the histopathology report of colorectal carcinoma, list their frequencies and see the compliance to minimum data sets. 3) To assess the lymph node yield from colorectal carcinoma specimens with a review of the literature.
| » Materials and Methods|| |
This study was conducted after obtaining approval from the institutional review board on June 24, 2016. We searched archives for colorectal adenocarcinoma patients operated at our institution between 2013 and 2015. All the histopathology reports were reviewed from electronic medical records (EMR). The following gross and microscopic parameters were noted from the reports.
The history of neoadjuvant therapy was noted.
After enlisting the study data on the Excel sheet, the frequency of these parameters was calculated. The enlisted data on the Excel sheet were analyzed using SPSS/STATA software (version 23). The mean lymph node yield by different age groups, different treatment categories, and the site of the tumor was calculated. The comparison between the mean lymph node of different pathological and treatment categories was compared by applying the t-test.
In our laboratory, we receive all the specimens in fresh, unopened state in double-sealed plastic bags. After delivering them from the bag, they are grossly inspected. For rectal/rectosigmoid tumors, the non-peritonealised surface/circumferential resection margin (CRM) is identified [Figure 1] and inked. The rectum is usually opened from the anterior wall, allowed to fix for 24–48 hours, and sectioned/sliced as per the standard protocol [Figure 2]. At the perceived tumor location, transverse sections of the bowel ring/segments are taken. Usually, a complete cut through the tumor is avoided so as to allow good inspection of the tumor and CRM. Appropriate sections for histology are selected which are showing a maximum depth of tumor invasion. The lymph nodes are also dissected after fixation. They are dissected on the bowel segment away from the tumor by careful palpation and dissection. At the tumor location, after tumor sections are taken, lymph nodes are selected from the pericolic/perirectal fat, by naked eye examination and palpation. If these are located close to the CRM, then the lymph node section is submitted along with CRM. Fat-dissolving techniques are not used.
|Figure 1: Gross photograph of APR specimen. A- from anterior surface , B- from right lateral surface showing complete total mesorectal excision and non peritonealised surface (arrows show peritoneal reflections)|
Click here to view
|Figure 2: Transverse cuts/slices from the rectal carcinoma specimen. The black rectangles are the area to be examined for maximum depth of tumour invasion in the wall and assessing the CRM|
Click here to view
This study was conducted as an MD pathology dissertation topic. Hence, this is a relatively older data. However, such studies remain useful for quality control measures. For staging, the then prevailing tumor nodes metastases-7 classification is considered.
| » Results|| |
A total of 1230 patients of colorectal adenocarcinoma were operated at our institution between 2013 and 2015. These were operated at two different sites but were operated by the same team, experienced in colorectal cancer surgery. Out of 1230 patients, 826 (67.15%) patients were men, and 404 (32.85%) patients were women. The overall mean age was 52 (range = 18 - 90) years, the commonest age group being 50–59 years (24.63% patients).
The total number of surgeries for rectal carcinoma and colonic carcinoma were 1230. Surgeries were of 13 different types, and their distribution is as given in [Table 1].
|Table 1: Types of surgeries for colorectal carcinoma 810 (74.72%) reports|
Click here to view
The first six types of surgeries [anterior resection (AR), abdominoperineal resection (APR), low AR, ultra-low AR, inter sphincteric resection (ISR), and pelvic exenteration] were undertaken for rectal/rectosigmoid tumors. Hence, this group is broadly taken as surgeries for rectal cancers (n = 787) for lymph node calculation purposes, as the lymph node yield will depend on the type, completeness of surgery including total mesorectal excision (TME) and pathologist's handling of the specimen. All other surgeries (number 7–13) were taken as surgeries for colonic cancer [Table 1].
The type of surgical 1230 (100%) reports. The mean tumor size was 4.38 cm (range = 0.5 cm–22 cm).
The distance of the tumor from the proximal and distal cut margin was mentioned in all (100%) reports. The proximal cut margin was grossly uninvolved in all cases, whereas, in two (0.16%) cases distal cut margin was grossly involved.
Comment about the tumor site perforation was made in all (100%) reports. Out of these, grossly tumor site perforation was noted in 36 (2.95%) patients..
The total number of patients operated by APR, AR, low AR, ultra-low AR, ISR, and pelvic exenteration was 787. Out of these specimens, the quality of TME was mentioned in 741 (94.16%) reports and was not mentioned in 46 (5.84%) reports grossly. The mesorectal excision was complete in 723 (97.57%) patients, nearly complete in 14 (1.89%) patients and was incomplete in 4 (0.54%) patients.
There were 750 patients of rectal/rectosigmoid carcinomas. The distance of the tumor (grossly) from the CRM was mentioned in 721 (96.13%) reports and was not mentioned in 29 (3.87%) reports.
Tumor differentiation was stated as follows and is given in [Table 2]. The majority of the tumors were moderately differentiated.
|Table 2: Tumor differentiation in all cases irrespective of neoadjuvant therapy status|
Click here to view
Neoadjuvant therapy response
Out of all colorectal carcinoma patients (1230), neoadjuvant therapy was given in 619 (50.33%) patients. Out of these, no residual tumor was detected in 146 (23.59%) patients, and the residual viable tumor was detected in 473 (76.41%) patients. The rest 611 (49.67%) patients did not receive any preoperative neoadjuvant therapy. Out of the 619 patients, who received preoperative neoadjuvant therapy, tumor regression grade (TRG) was mentioned in 616 (99.51%) reports and not mentioned in 3 (0.49%) reports. We used the five-point scoring system for TRG scoring. TRG 1 (no residual tumor) was seen in 146 (23.70%) patients. The total number of patients having a viable tumor was 1084.
Tumor invasion in the bowel wall/pT stage -the pT stage was mentioned in 1228/1230 (99.84%) reports. Stage T0 was detected in 146 (11.89%) patients, T1 in 40 (3.26%) patients, T2 in 244 (19.87%) patients, T3 in 636 (51.79%) patients and T4 in 162 (13.19%) patients. T4a stage was mentioned in 52 reports and T4b in 36 reports. In the rest of the 74 reports of T4 stage tumors, subclassification was not mentioned.
Lymphovascular invasion and perineural invasion
Lymphovascular invasion (LVI) was mentioned in 1063/1084 (98.06%) reports. Perineural invasion (PNI) was mentioned in 274 (25.28%) reports but was not mentioned in 810 (74.72%) reports.
Proximal and distal margins
Proximal and distal resection margins were commented upon in all 1084 (100%) reports. Microscopically proximal cut margin was involved in one out of 1084 cases. The distal cut margin was involved in 11/1084 (1.01%) reports, microscopically.
Circumferential resection margin
Out of all rectal/rectosigmoid carcinomas (750 patients), the tumor was located below the peritoneal reflection in 424 patients. In these, a microscopically viable tumor was detected in 331 patients. Microscopic CRM status was mentioned in 321 (96.98%) reports and was not mentioned in 10 (3.02%) reports. Out of these 321 patients, CRM was involved by a tumor in 30 (9.35%) patients.
All these 30 patients, where CRM was involved microscopically, had received preoperative neoadjuvant therapy.
Histological features suggestive of microsatellite instability
Histological features suggestive of microsatellite instability (MSI) were mentioned in 553 (51.01%) reports and were not mentioned in 531 (48.99%) reports. Out of these 553 cases, histological features suggestive of MSI were seen in 115 (9.35% of all 1230) patients.
Lymph node yield in colorectal cancer specimen
The overall mean lymph node yield was 18.38 (range 0–130, median = 15, Standard deviation (SD) = 12.94). The number of lymph nodes at the 25th and 75th percentile were 10 and 24, respectively.
Mean and median lymph node by surgery type and by age groups are listed in [Table 3] and [Table 4], respectively.
The mean lymph nodes dissected in the age group less than and equal to 39 years is 18.73, whereas the mean lymph node for the age group greater than 39 years is 18.28. There is no statistically significant difference in the lymph node yield (P = 0.615) between these age groups [Table 5].
|Table 5: Comparison between mean lymph node by age group, neoadjuvant therapy, rectal cancer surgery versus colonic cancer surgery, and depth of bowel wall invasion|
Click here to view
The mean lymph node yield in patients who received preoperative neoadjuvant therapy is 12.89 and for patients without neoadjuvant therapy is 23.95. This difference is statistically significant (P < 0.001) [Table 5].
There was also a vstatistically significant difference in the lymph node yield between surgery for rectal cancer cases (mean lymph node 14.20) and surgery for colonic cancer cases (mean lymph node 25.81) (P < 0.001) [Table 5].
To calculate the statistical correlation between the depth of tumor invasion (T stage) and lymph node yield, we divided the cases into two groups. One group included patients of T0, T1, and T2 stage patients (n = 430) and the other group includes T3 and T4 stage patients (n = 798). A statistically significant difference has been detected between lymph node yield in low tumor stage (14.24 in stage T0, T1, T2) and high tumor stage (20.64 in stage T3, T4) (P < 0.001) [Table 5].
Lymph node stage- (N Stage)
Lymph nodes (N status) were mentioned in all 1230 (100%) reports.. Stage N0 was detected in 689 (56.02%) patients,, N1a in 134 (10.89%) patients,, N1b in 159 (12.93%) patients, N1c in 28 (2.28%) patients,, N2a in 123 (10.00%) patients and N2b in 95 (7.72%) patients. In two (0.16%) patients no lymph nodes were identified; hence, lymph node status was not known (NX).
| » Discussion|| |
Histopathology reports of colorectal adenocarcinoma surgery provide important information both for clinical management of the patient and for evaluation of the delivery of healthcare services. Inadequate histopathology reports can result in suboptimal therapy. In this retrospective study, we assessed the compliance to minimum data set parameters and calculated their frequencies.
All the specimens have been grossed by surgical pathology residents, and histopathology of these specimens has been reported by consultants of our pathology department. we have not investigated the specimen handling as regards to individuals based on their experience in years, nor reviewed any slides for this study.
Longitudinal margins: Both longitudinal margins (proximal and distal) in a resection specimen should be examined routinely. If the margin is more than 30 mm away from the primary tumor grossly, then it can be regarded as free margin and histological examination is not required, except for unfavorable histology such as signet ring cell carcinoma or tumors having extensive lymphovascular emboli, infiltrative growth pattern. It is also not necessary to examine doughnuts histologically if the distance of the resection margin is >30 mm from the tumor. However, as a routine practice, we perform histological examination of both the longitudinal margins and doughnuts in all cases.
CRM: It is one of the most important predictive markers for local recurrence and survival. The involvement of CRM by tumor is an indicator for adjuvant therapy for those who have not received prior therapy. It is also important in the colonic cancers mainly in the cecum and ascending and descending colon.
The minimum distance between the CRM and the tumor is recorded. If the distance is 1 mm or less, then it is regarded as involved by tumor. This involvement may be through direct extension from the main tumor, by a tumor in lymphovascular space, or by a metastatic node at the margin.,,
In our study, we observed that in 10 cases of rectal cancers located below the peritoneal reflection, the CRM was not mentioned. The CRM was involved in 9.35% of cases of rectal cancers located below the peritoneal reflection. Adam et al. found CRM involvement in 36% cases of their study and those are associated with poor outcomes and a higher rate of local recurrences. This audit, however, showed a drawback that the comment about CRM in non-rectal tumors has not been captured uniformly and is required.
Quality of the TME should be mentioned in all rectal cancer cases, which has bearing on prognosis and also serves as a quality indicator. In our study, comment on TME was given in 94.16% of rectal surgery cases. In the study conducted by Garcia-Granero et al., complete TME was achieved in 82.3% cases. Maslekar et al. and Nagtegaal et al. have also stressed the importance of incomplete TME and adverse outcomes., We observed complete TME in 97.57% of cases in this study.
LVI and PNI should be mentioned in all surgical pathology reports of colorectal cancer specimens., We observed that in 98.06% reports, lymphovascular emboli have been mentioned. In this comparison, the perineural invasion was mentioned in 25.28% cases only. We later realized that the column of perineural invasion was not present in our template/form. It has now been corrected and is recorded in the current reporting system.
Although, lymphatic emboli (sometimes reported as lymphovascular emboli) have been uniformly mentioned in our reports and have been captured; extramural vascular emboli, separately has been infrequently mentioned and has not been tabulated. This remains a lacuna in the reports. Comment about "tumor budding" is a relatively recent parameter. Although currently, some experts consider it as a noncore data item,, it is an important parameter of adverse prognosis and has not been mentioned in our reports. This is a lacuna in our reports.
During the signing outs of these cases, although, we used a template, the reports were written in a free text form. We have changed that practice and have now adopted a computerized synoptic system that records these parameters individually.
Another observation we made is that we are using a 5-point tumor regression grade system (for post neoadjuvant chemoradiotherapy (NACT RT) rectal carcinoma assessment), whereas, the international datasets advise using a 3-grade system., Although, our system has been working well in the local hospital multidisciplinary team, adopting a 3-grade system would be useful for inter-institutional comparison and reducing interobserver variability.
Lymph node yield in surgical specimens in colorectal cancer
For microscopy, all nodes are to be examined. Detection of lymph node metastasis in colorectal cancer places it in stage-III disease and makes the subject a candidate for adjuvant therapy. In a colorectal cancer resection specimen, it is recommended to retrieve at least 12 lymph nodes., Studies have shown that, as more numbers of lymph nodes are dissected, there are more chances to detect lymph node metastasis. They suggest that for accurate staging more lymph nodes should be examined.,
In literature, it has been documented that several factors influence the number of lymph nodes dissected from a surgical specimen. These include age of the patient, obesity, location of the tumor, neoadjuvant therapy, surgical technique, and pathologist's handling of the specimen. Out of these, surgical technique and pathologist's handling are two actionable points. If fewer than 12 lymph nodes are dissected, it is recommended to reexamine the specimen for additional lymph nodes and such a comment is added to the report.,
In our study, of the 387 patients with lymph node yield less than 12, 311 (80.36%) patients had received preoperative neoadjuvant therapy. In all the cases where lymph node harvest was less than 12, repeat grossing was performed and that was mentioned on the histopathology report.
In our previous, similar study (2012), where we investigated 170 colorectal cancer cases, the mean lymph node yield calculated were 10.30 and 16.71 for rectal and colon cancer cases, respectively. The overall mean lymph node yield was 12.68. In that study, when the lymph node yield was less than 12, only 21.3% of reports commented about repeat grossing. Since then (2012 onwards), we have changed our practice and comment on repeat grossing is added. Additional sections were taken for more lymph nodes and was documented in histopathology reports.
In a large population-based study (n = 153,483), which included colonic as well as rectal cancer cases, Chou et al. noted mean (+SD) lymph node harvest was 12 (+ 9.3). For colonic cancer, mean (+SD) was 12.4 (+9.4) and for rectal cancer mean (+SD) was 10.2 (+8.8). In our present study, we found mean (+SD) lymph node yield to be 18.38 (+12.94). For rectal surgery cases mean (+SD) lymph node was 14.20 (+8.70) and for colonic surgery cases mean (+SD) was 25.81 (+15.64) (P value < 0.001).
In the same study, age was consistently an important determinant of lymph node yield. They found, for every 10 years increase in age, there was an average reduction of 9% in lymph node harvest. These authors hypothesized that in older patients with associated comorbidities, the surgeons perform the less extensive dissection. However, Tekkis et al. hypothesized that with increasing age, lymph nodes undergo involution. In our present study, we didn't find any significant correlation between the age of the patient and the mean lymph node harvest (P = 0.615).
Another factor that definitely influences the lymph node yield is preoperative neoadjuvant therapy. The reason for a low lymph node count in these cases could be due to inflammation, tissue fibrosis, or shrinkage of lymph nodes. Chou et al. observed a 25% reduction in lymph node harvest in post-radiotherapy cases. In our current study, a statistically significant difference was seen in lymph node yield for colorectal cancer with neoadjuvant therapy (mean lymph node yield 12.89) and neoadjuvant therapy-naive cases (mean lymph node yield 23.95).
The number of lymph nodes retrieved also depends upon the depth of tumor invasion (T stage). It has been hypothesized that larger tumors induce an immunologic reaction in the regional lymph nodes, which results in lymph node enlargement and dissection of more lymph nodes in higher stage disease. Chou et al. found an increase in average lymph node dissection for individual T3 and T4 tumors versus T1 tumors. In our study, we observed similar/statistically significant findings (mean lymph node in T3 and T4 cases was 20.64, and in T0, T1, and T2 cases was 14.24). However, Leung et al. did not find any significant correlation between the depth of tumor invasion and lymph node yield.
| » Conclusion|| |
After reviewing the contents of 1230 histopathology reports of the colorectal carcinoma surgery, we found that overall compliance was satisfactory (most of the parameters were mentioned in more than 95% reports). The lacunae in colorectal histopathology reports were, not capturing extramural vascular emboli and not uniformly commenting about non-rectal CRM. Tumor budding needs to be recorded. The mean lymph node yield in surgeries for rectal cancers was 14 and for surgeries for colonic cancers was 26. The overall mean lymph node yield was 18.38. A statistically significant difference in lymph node yield was detected between rectal and colonic surgery cases; post-NACT and neoadjuvant naive cases; and high tumor stage and low tumor stage diseases. No significant association was detected with the age of the patient and mean lymph node yield.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Ng IO, Luk IS, Yuen ST, Lau PW, Pritchett CJ, Ng M, et al.
Surgical lateral clearance in resected rectal carcinomas. A multivariate analysis of clinicopathological features. Cancer 1993;71:1972-6.
Bateman AC, Carr NJ, Warren BF. The retroperitoneal surface in distal caecal and proximal ascending colon carcinoma: The Cinderella surgical margin? J Clin Pathol 2005;58:426-8.
Wittekind C, Compton C, Quirke P, Nagtegaal I, Merkel S, Hermanek P, et al.
A uniform residual tumor (R) classification: Integration of the R classification and the circumferential margin status. Cancer 2009;115:3483-8.
Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet 1986;2:996-9.
Hermanek P, Wittekind C. The pathologist and the residual tumor (R) classification. Pathol Res Pract 1994;190:115-23.
Wittekind C, Compton CC, Greene FL, Sobin LH. TNM residual tumor classification revisited. Cancer 2002;94:2511-6.
Adam IJ, Mohamdee MO, Martin IG, Scott N, Finan PJ, Johnston D, et al.
Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet 1994;344:707-11.
García-Granero E, Faiz O, Muñoz E, Flor B, Navarro S, Faus C, et al.
Macroscopic assessment of mesorectal excision in rectal cancer: A useful tool for improving quality control in a multidisciplinary team. Cancer 2009;115:3400-11.
Maslekar S, Sharma A, Macdonald A, Gunn J, Monson JR, Hartley JE. Mesorectal grades predict recurrences after curative resection for rectal cancer. Dis Colon Rectum 2007;50:168-75.
Nagtegaal ID, van de Velde CJH, van der Worp E, Kapiteijn E, Quirke P, van Krieken JHJM, et al
. Macroscopic evaluation of rectal cancer resection specimen: Clinical significance of the pathologist in quality control. J Clin Oncol 2002;20:1729-34.
Compton CC, Fielding LP, Burgart LJ, Conley B, Cooper HS, Hamilton SR, et al.
Prognostic factors in colorectal cancer. College of American Pathologists Consensus Statement 1999. Arch Path Lab Med 2000;124:979-94.
Goldstein NS. Lymph node recoveries from 2427 pT3 colorectal resection specimens spanning 45 years: Recommendations for a minimum number of recovered lymph nodes based on predictive probabilities. Am J Surg Pathol 2002;26:179-89.
Cserni G, Vinh-Hung V, Burzykowski T. Is there a minimum number of lymph nodes that should be histologically assessed for a reliable nodal staging of T3N0M0 colorectal carcinomas? J Surg Oncol 2002;81:63-9.
Rajput A, Romanus D, Weiser MR, Ter Veer A, Niland J, Wilson J, et al
. Meeting the 12 lymph node benchmark in colon cancer. J Surg Oncol 2010;102:3-9.
Deodhar KK, Budukh A, Ramadwar M, Bal MM, Shrikhande SV. Are we achieving the benchmark of retrieving 12 lymph nodes in colorectal carcinoma specimens? Experience from a tertiary referral center in India and review of literature. Indian J Pathol Microbiol 2012;55:38-42. [Full text]
Chou JF, Row D, Gonen M, Liu YH, Schrag D, Weiser MR. Clinical and pathologic factors that predict lymph node yield from surgical specimens in colorectal cancer: A population-based study. Cancer 2010;116:2560-70.
Tekkis PP, Smith JJ, Heriot AG, Darzi AW, Thompson MR, Stamatakis JD, et al.
A national study on lymph node retrieval in resectional surgery for colorectal cancer. Dis Colon Rectum 2006;49:1673-83.
Shia J, McManus M, Guillem JG, Leibold T, Zhou Q, Tang LH, et al.
Significance of acellular mucin pools in rectal carcinoma after neoadjuvant chemoradiotherapy. Am J Surg Pathol 2011;35:127-34.
Leung AM, Scharf AW, Vu HN. Factors affecting number of lymph nodes harvested in colorectal cancer. J Surg Res 2011;168:224-30.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]