Ann Surg Oncol (2015) 22:203–208 DOI 10.1245/s10434-014-3972-3
ORIGINAL ARTICLE – COLORECTAL CANCER
Diagnostic Performance of Multidetector Row Computed Tomography for Assessment of Lymph Node Metastasis in Patients with Distal Rectal Cancer Hirotoshi Kobayashi, MD, FACS1, Akifumi Kikuchi, MD2, Satoshi Okazaki, MD2, Megumi Ishiguro, MD3, Toshiaki Ishikawa, MD2, Satoru Iida, MD2, Hiroyuki Uetake, MD3, and Kenichi Sugihara, MD2 Center for Minimally Invasive Surgery, Tokyo Medical and Dental University, Tokyo, Japan; 2Department of Surgical Oncology, Graduate School, Tokyo Medical and Dental University, Tokyo, Japan; 3Department of Translational Oncology, Tokyo Medical and Dental University, Tokyo, Japan 1
ABSTRACT Background. The accurate preoperative diagnosis of depth of tumor invasion and nodal status in distal rectal cancer is important because neoadjuvant chemotherapy or lateral pelvic lymph node dissection is indicated for patients with T3–T4 tumor or nodal involvement. This study aimed to determine the optimal cutoff value for predicting lymph node metastasis in patients with distal rectal cancer using multidetector row computed tomography (MDCT). Methods. The study investigated 77 patients who had undergone surgery for distal rectal cancer at a single institution between 2008 and 2011. Diagnostic performance for depth of tumor invasion and mesorectal and lateral pelvic lymph node metastases was evaluated. The optimal cutoff value was determined by receiver operating characteristic curve analysis. Results. For predicting mesorectal and lateral pelvic lymph node metastasis, MDCT had a sensitivity of 0.36 and 0.89 and a specificity of 0.78 and 0.97, respectively. The optimal cutoff values of major and minor axes lengths for predicting mesorectal lymph node metastasis were 6.5 mm and 5.7 mm, respectively. The areas under the curve (AUCs) were 0.82 and 0.88, respectively. For
Ó Society of Surgical Oncology 2014 First Received: 2 March 2014; Published Online: 15 August 2014 H. Kobayashi, MD, FACS e-mail: [email protected]
predicting lateral lymph node metastasis, the optimal cutoff values were 9 mm for the major axis and 6 mm for the minor axis. Both AUCs were 1. Conclusions. Using MDCT, the optimal cutoff value of minor axis length for predicting mesorectal and lateral pelvic lymph node metastases in patients with distal rectal cancer was 6 mm. The accuracy of MDCT was satisfactory for predicting lateral pelvic lymph node metastasis.
Among the causes of cancer mortality, colorectal cancer ranks second in the United States and third in Japan.1,2 Furthermore, the incidence of colorectal cancer is increasing rapidly in Japan.2,3 The prognosis of patients with rectal cancer is known to be worse than that of patients with colon cancer. One of the reasons for this is the high local recurrence rate in rectal cancer patients.4 Whereas the standard therapy for T3–T4 rectal cancer is preoperative chemoradiotherapy plus total mesorectal excision (TME) in Western countries, TME plus lateral pelvic lymph node dissection is preferred in Japan. The Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines list the indications for lateral pelvic lymph node dissection as T3–T4 cancer with a distal edge located below the peritoneal reflection and distal rectal cancer with mesorectal lymph node metastasis.5 Therefore, accurate diagnosis of the depth of tumor invasion and lymph node metastasis enables identification of patients who do not need lateral pelvic lymph node dissection. Although the reported sensitivity of computed tomography (CT) for detecting nodal involvement of colorectal cancer is 45–73 %,6–10 the quality of recent multidetector row CT scanners is better than those used in previous studies.
204
This study aimed to clarify the diagnostic accuracy of multidetector row CT for T3–T4 distal rectal cancer and lymph node metastasis. Furthermore, the optimal cutoff values of major and minor axes lengths for predicting lymph node metastasis in patients with distal rectal cancer were evaluated. METHODS Patients The medical charts of 77 patients who underwent curative surgery for distal rectal carcinoma in the Department of Surgical Oncology at the Tokyo Medical and Dental University in Tokyo, Japan between January 2008 and December 2011 were reviewed. In this study, distal rectal cancer was defined as a tumor whose distal edge was located below the peritoneal reflection. The CT scanner used in the current study had 64 detectors. The lengths of the major and minor axes were measured. All the patients underwent TME or tumor-specific mesorectal excision (TSME). In addition, the patients with T3–T4 distal rectal cancer underwent lateral pelvic lymph node dissection, as well as TME or TSME. The resected rectal specimens were stretched and pinned to a cork board. The surgeon identified and isolated the lymph nodes and recorded their number and distribution. The surgeons in this study harvested the lymph nodes from the specimen to evaluate the precise location of nodal metastases. This technique is not routinely performed in North America or Europe. After formalin fixation, the specimens and lymph nodes were examined by the pathologist. The associations between the lengths of the major and minor axes of the lymph nodes shown on preoperative CT and the status of pathologic lymph node metastases were evaluated retrospectively. The depth of tumor invasion was diagnosed preoperatively by colonoscopy, barium enema, magnetic resonance imaging, and endoscopic ultrasonography.
H. Kobayashi et al. TABLE 1 Clinicopathologic characteristics of 77 patients with curative surgical treatment for distal rectal cancer Clinicopathologic features
n
Age: years (range)
66 (35–93)
%
Gender Male
48
62
Female
29
38
Low anterior resection
40
52
Intersphincteric resection Abdominoperineal resection
3 28
4 36
Hartmann’s operation
4
5
Total pelvic exenteration
2
3
Present
40
52
Absent
37
48
Present
17
22
Absent
60
78
T0
2
3
T1
6
8
T2
18
23
T3
45
58
6
8
Well differentiated
22
29
Moderately differentiated
46
60
Poorly differentiated
5
6
Mucinous
3
4
Others
1
1
Present
40
52
Absent
37
48
Surgical procedure
Lateral pelvic lymph node dissection
Laparoscopic surgery
Pathologic T-category
T4b Pathology
Lymph node metastasis
RESULTS Patients’ Characteristics
Statistical Analysis All data are expressed as means ± standard deviations. Differences in continuous variables were compared using the Mann–Whitney U test, whereas differences in categorical variables were analyzed using the v2 test. The optimal cutoff values for predicting lymph node metastasis were analyzed using receiver operating characteristic (ROC) curves. Areas under the curve (AUCs) were also calculated. Data were analyzed using JMP 9 (SAS Institute Inc., Cary, NC, USA). Statistical significance was established at a p value lower than 0.05.
The characteristics of all the patients are listed in Table 1. No patient received preoperative chemoradiotherapy in the current study. The majority of the patients underwent sphincter-sparing surgery. Of the 77 patients, 40 also underwent lateral pelvic lymph node dissection. Laparoscopic surgery was performed in 17 patients. Diagnostic Performance for T-Category Five patients were excluded from the evaluation of diagnostic performance for T-category because of additional
Diagnostic Accuracy of MDCT for Lower Rectal Cancer
205
TABLE 2 Diagnostic performance for T-category pT0
pT1
pT2
pT3
cT0
0
cT1
1
cT2
pT4b
Total
0
0
1
0
1
0
5
0
0
6
0
3
4
3
0
10
cT3
0
0
9
38
4
51
cT4b
0
0
0
2
2
4
Total
1
3
18
44
6
72
TABLE 3 Diagnostic performance for depth of tumor invasion of T3–T4 pT1–pT2
pT3–pT4
Total
cT1–cT2 cT3–cT4
13 9
4 46
17 55
Total
22
50
77
TABLE 4 Diagnostic performance for mesorectal lymph node metastasis Pathologically positive
Pathologically negative
Total
Preoperatively positive
14
4
18
Preoperatively negative
25
34
59
Total
39
38
77
resection after endoscopic or transanal resection. Thus, 72 patients were included in the current study. The accuracy of diagnostic performance for T-category was 61 % (Table 2). From the viewpoint of clinical use, the diagnostic performance for the depth of tumor invasion of T3–T4 also was investigated (Table 3). The accuracy was 77 %. Diagnostic Performance for Mesorectal Lymph Node Metastasis The median number of mesorectal lymph nodes retrieved was 15 (range, 3–41). Among the 77 patients with a curative resection for distal rectal cancer, 39 had mesorectal lymph node metastases (Table 4). The median number of involved mesorectal nodes was 2 (range, 1–18). The sensitivity was 36 %, and the specificity was 89 % (P = 0.0071). The false-positive and false-negative rates were respectively 11 and 64 %, and the positive and negative predictive values were respectively 78 and 58 %.
Diagnostic Performance for Lateral Pelvic Lymph Node Metastasis The median number of lateral lymph nodes retrieved was 14 (range, 2–34). Among the 40 patients with lateral pelvic lymph node dissection, 8 (20 %) had lateral pelvic lymph node metastases. The median number of involved lateral nodes was 3 (range, 1–29). The sensitivity was 78 %, and the specificity was 97 % (P \ 0.0001). The false-positive and false-negative rates were respectively 3 and 22 %, whereas the positive and negative predictive values were respectively 88 and 94 % (Table 5). Optimal Cutoff Values for Predicting Lymph Node Metastasis The mean lengths of the major axis of positive and negative mesorectal lymph nodes shown on MDCT were respectively 9.4 and 5.3 mm (P \ 0.0001), whereas those of the minor axis were respectively 8.2 and 3.7 mm (P \ 0.0001). The ROC curves for predicting mesorectal lymph node metastasis using the major and minor axis lengths are shown in Fig. 1. The AUCs using the major and minor axis lengths were respectively 0.82 and 0.88. The optimal cutoff values of the major and minor axis lengths for predicting mesorectal lymph node metastasis were respectively 6.5 and 5.7 mm. As for lateral lymph node metastasis, the mean lengths of the major and minor axes in cases with positive lateral nodes were 14.4 and 11.4 mm, respectively. The optimal cutoff values of the major and minor axis lengths were 9 and 6 mm, respectively. The AUCs of the major and minor axis lengths were both 1. After the cutoff value of 6 mm for the minor axis was adopted, the sensitivity and specificity for predicting mesorectal lymph node metastasis were respectively 49 and 92 % (P \ 0.0001). The positive and negative predictive values were 86 and 64 %, respectively. The sensitivity and specificity for predicting lateral lymph node metastasis were improved to be 78 and 100 %, respectively (P \ 0.0001). The positive and negative predictive values were 100 and 94 %, respectively. DISCUSSION The current study demonstrated that the best cutoff value of minor axis length for predicting mesorectal and lateral pelvic lymph node metastases in patients with distal rectal cancer was 6 mm. The lengths of both the major and minor axes of the lymph nodes shown on MDCT were useful for predicting lymph node metastasis.
206
TABLE 5 Diagnostic performance for lateral pelvic lymph node metastasis Pathologically positive
Pathologically negative
Total
Preoperatively positive
7
1
8
Preoperatively negative
2
30
32
Total
9
31
40
a
1.00 0.90 0.80
Sensitivity
0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.00
0.20
0.40
0.60
0.80
1.00
0.80
1.00
1-specificity
b
1.00 0.90 0.80 0.70
Sensitivity
Using the length of the major axis on MDCT, 6.5 mm in the mesorectum and 9 mm in the lateral pelvic area were the cutoff values that provided the greatest predictive value. On the other hand, using the length of the minor axis, 5.7 mm in the mesorectum and 6 mm in the lateral pelvic area provided the greatest predictive value. These differences might reflect limitations of a small sample size. The current study also demonstrated that the AUC of the minor axis was greater than that of the major axis for predicting mesorectal lymph node metastasis. The AUC of the major axis was equal to that of the minor axis for predicting lateral pelvic lymph node metastasis. Therefore, a minor axis length of 6 mm is the most appropriate cutoff value for predicting lymph node metastases in patients with distal rectal cancer. The aforementioned results are consistent with those previously reported. Kanemoto et al. 11 reported that the minor axis, not the major axis, was appropriate for predicting colorectal lymph node metastasis. After the cutoff value of 6 mm was adopted for the minor axis, the sensitivity and specificity for predicting both mesorectal and lateral pelvic lymph node metastasis were improved. Especially, the specificity for predicting mesorectal and lateral pelvic lymph node was respectively 92 and 100 % if this cutoff value was used to evaluate the nodal status in lower rectal cancer. The use of MDCT will help in selecting patients for lateral node dissection. The most common method for the diagnosis of hematogenous and nodal metastases from colorectal cancer is CT, although a distinct criterion for diagnosing lymph node metastasis from colorectal cancer does not exist. The most widely used criterion is any node larger than 1 cm or a cluster of three or more nodes smaller than 1 cm.6, 12–17 However, we often see lymph node metastases smaller than 1 cm. The most appropriate cutoff value for predicting lymph node metastasis from colorectal cancer has not been established. Dighe et al. 18 reported the diagnostic precision of CT in local staging of colon cancers. In their meta-analysis, the sensitivity and specificity for detecting nodal involvement using CT ranged from 0.66 to 0.87 and from 0.7 to 0.8, respectively. These were unsatisfactory compared with the rate for detecting tumor invasion using CT. The sensitivity and specificity for detecting tumor invasion were 0.84 to 0.95 and 0.74 to 0.86, respectively. However, in their investigation, the best accuracy was reported by the studies that had used MDCT. Thus, a precise criterion for detecting lymph node metastasis from colorectal cancer using MDCT needs to be developed. The current study investigated both lateral pelvic lymph node metastasis and mesorectal lymph node metastasis from distal rectal cancer. The lateral pelvic lymph node is an important location as a recurrence site after surgery for
H. Kobayashi et al.
0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.00
0.20
0.40
0.60
1-specificity FIG. 1 The receiver operating characteristics curves of the diagnostic performance for lymph node metastasis. The lengths of the major (a)and minor (b) axes of lymph nodes on preoperative multi-detector row CT were investigated. The area under the curve was 0.82 (a) and 0.88 (b), respectively
distal rectal cancer. This study demonstrated that the AUC for lateral pelvic lymph node metastasis was greater than that for mesorectal lymph node metastasis. This might mean that MDCT is more useful for predicting lateral pelvic lymph node metastasis than for predicting mesorectal lymph node metastasis. One of the reasons for the difference in the AUC was the size of the metastatic lymph nodes. No metastatic lymph node had a minor axis shorter than 6 mm in the lateral
Diagnostic Accuracy of MDCT for Lower Rectal Cancer
pelvic area, although such nodes were found in the mesorectum. At the same time, the positive lateral pelvic lymph nodes were larger than the positive mesorectal lymph nodes. The current results support those of Yano et al. who reported that the size of positive mesorectal lymph nodes (5.7 mm) was significantly smaller than the size of positive lateral pelvic lymph nodes (11.5 mm). In their study, the accuracy of detecting positive lymph nodes in the lateral pelvic area was superior to that in the mesorectum. Approximately 15 % of patients with distal rectal cancer have lateral pelvic lymph node metastasis.19 The standard therapeutic approach for controlling lateral pelvic lymph node metastasis is preoperative chemoradiotherapy in Western countries and lateral pelvic lymph node dissection in Japan. The accurate preoperative diagnosis of lateral pelvic lymph node metastasis is very important because both preoperative chemoradiotherapy and lateral lymph node dissection can lead to postoperative sexual and urinary dysfunction.20–23 There is merit in omitting such invasive therapies to prevent such postoperative complications. The sensitivity for the preoperative diagnosis of local invasion (i.e., depth of tumor invasion) of T3–T4 was 92 %. This was satisfactory, whereas the specificity was 59 %. It is important to decrease false-positive cases in detecting the preoperative depth of tumor invasion using MDCT. In all false-positive cases, pathologic T2 tumors were preoperatively diagnosed as T3 tumors. The accurate diagnosis of T3–T4 distal rectal cancer is very important because patients with those tumors usually receive preoperative chemoradiotherapy or lateral pelvic lymph node dissection.19, 24–30 In the current study, transrectal ultrasonography (TRUS) was not used. The use of TRUS will be necessary to improve the accuracy of the preoperative diagnosis of tumor invasion depth. A previous pooled analysis demonstrated that the sensitivity and specificity for the preoperative diagnosis of tumor invasion depth using TRUS were 0.97 and 0.87, respectively.31 In conclusion, the optimal cutoff value of minor axis length for predicting lymph node metastasis in patients with distal rectal cancer using MDCT was 6 mm. The accuracy of MDCT was satisfactory for predicting lateral pelvic lymph node metastasis in distal rectal cancer. REFERENCES 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012;62:10–29. 2. Kotake K, Honjo S, Sugihara K, et al. Changes in colorectal cancer during a 20-year period: an extended report from the multi-institutional registry of large bowel cancer, Japan. 2003;46:S32–43. 3. Muto T, Kotake K, Koyama Y. Colorectal cancer statistics in Japan: data from JSCCR registration, Int J Clin Oncol. 1974–1993. 2001;6:171–6.
207 4. Kobayashi H, Mochizuki H, Sugihara K, et al. Characteristics of recurrence and surveillance tools after curative resection for colorectal cancer: a multicenter study. Surgery. 2007;141:67–75. 5. Watanabe T, Itabashi M, Shimada Y, et al. Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol. 2012;17:1–29. 6. Balthazar EJ, Megibow AJ, Hulnick D, Naidich DP. Carcinoma of the colon: detection and preoperative staging by CT. AJR Am J Roentgenol. 1988;150:301–6. 7. Hundt W, Braunschweig R, Reiser M. Evaluation of spiral CT in staging of colon and rectum carcinoma. Eur Radiol. 1999;9: 78–84. 8. Isbister WH, al-Sanea O. The utility of preoperative abdominal computerized tomography scanning in colorectal surgery. J R Coll Surg Edinb. 1996;41:232–4. 9. McAndrew MR, Saba AK. Efficacy of routine preoperative computed tomography scans in colon cancer. Am Surg. 1999;65:205–8. 10. Taylor AJ, Youker JE. Imaging in colorectal carcinoma. Semin Oncol. 1991;18:99–110. 11. Kanamoto T, Matsuki M, Okuda J, et al. Preoperative evaluation of local invasion and metastatic lymph nodes of colorectal cancer and mesenteric vascular variations using multidetector-row computed tomography before laparoscopic surgery. J Comput Assist Tomogr. 2007;31:831–9. 12. Acunas B, Rozanes I, Acunas G, Celik L, Sayi I, Gokmen E. Preoperative CT staging of colon carcinoma (excluding the rectosigmoid region). Eur J Radiol. 1990;11:150–3. 13. Ashraf K, Ashraf O, Haider Z, Rafique Z. Colorectal carcinoma, preoperative evaluation by spiral computed tomography. J Pak Med Assoc. 2006;56:149–53. 14. Cademartiri F, Luccichenti G, Rossi A, Pavone P. Spiral hydroCT in the evaluation of colo-sigmoideal cancer. Radiol Med. 2002;104:295–306. 15. Freeny PC, Marks WM, Ryan JA, Bolen JW. Colorectal carcinoma evaluation with CT: preoperative staging and detection of postoperative recurrence. Radiology. 1986;158:347–53. 16. Gazelle GS, Gaa J, Saini S, Shellito P. Staging of colon carcinoma using water enema CT. J Comput Assist Tomogr. 1995;19: 87–91. 17. Harvey CJ, Amin Z, Hare CM, et al. Helical CT pneumocolon to assess colonic tumors: radiologic-pathologic correlation. AJR Am J Roentgenol. 1998;170:1439–43. 18. Dighe S, Purkayastha S, Swift I, et al. Diagnostic precision of CT in local staging of colon cancers: a meta-analysis. Clin Radiol. 2010;65:708–19. 19. Kobayashi H, Mochizuki H, Kato T, et al. Outcomes of surgery alone for lower rectal cancer with and without pelvic sidewall dissection. Dis Colon Rectum. 2009;52:567–76. 20. Akasu T, Sugihara K, Moriya Y. Male urinary and sexual functions after mesorectal excision alone or in combination with extended lateral pelvic lymph node dissection for rectal cancer. Ann Surg Oncol. 2009;16:2779–86. 21. Lange MM, Maas CP, Marijnen CA, et al. Urinary dysfunction after rectal cancer treatment is mainly caused by surgery. Br J Surg. 2008;95:1020–8. 22. Lange MM, Marijnen CA, Maas CP, et al. Risk factors for sexual dysfunction after rectal cancer treatment. Eur J Cancer. 2009;45:1578–88. 23. Marijnen CA, van de Velde CJ, Putter H, et al. Impact of shortterm preoperative radiotherapy on health-related quality of life and sexual functioning in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol. 2005;23:1847–58. 24. Bosset JF, Collette L, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med. 2006;355: 1114–23.
208 25. Gerard JP, Azria D, Gourgou-Bourgade S, et al. Clinical outcome of the ACCORD 12/0405 PRODIGE 2 Randomized Trial in Rectal Cancer. J Clin Oncol. 2012;30:4558–65. 26. Gerard JP, Conroy T, Bonnetain F, et al. Preoperative radiotherapy with or without concurrent fluorouracil and leucovorin in T3–4 rectal cancers: results of FFCD 9203. J Clin Oncol. 2006;24:4620–5. 27. Kobayashi H, Mochizuki H, Kato T, et al. Is total mesorectal excision always necessary for T1–T2 lower rectal cancer? Ann Surg Oncol. 2010;17:973–80. 28. Kobayashi H, Mochizuki H, Kato T, et al. Lymph node ratio is a powerful prognostic index in patients with stage III distal rectal
H. Kobayashi et al. cancer: a Japanese multicenter study. Int J Colorectal Dis. 2011;26:891–6. 29. Sugihara K, Kobayashi H, Kato T, et al. Indication and benefit of pelvic sidewall dissection for rectal cancer. Dis Colon Rectum. 2006;49:1663–72. 30. Sugihara K, Moriya Y, Akasu T, Fujita S. Pelvic autonomic nerve preservation for patients with rectal carcinoma: oncologic and functional outcome. Cancer. 1996;78:1871–80. 31. Solomon MJ, McLeod RS. Endoluminal transrectal ultrasonography: accuracy, reliability, and validity. Dis Colon Rectum. 1993;36:200–5.
ORIGINAL ARTICLE – COLORECTAL CANCER
Diagnostic Performance of Multidetector Row Computed Tomography for Assessment of Lymph Node Metastasis in Patients with Distal Rectal Cancer Hirotoshi Kobayashi, MD, FACS1, Akifumi Kikuchi, MD2, Satoshi Okazaki, MD2, Megumi Ishiguro, MD3, Toshiaki Ishikawa, MD2, Satoru Iida, MD2, Hiroyuki Uetake, MD3, and Kenichi Sugihara, MD2 Center for Minimally Invasive Surgery, Tokyo Medical and Dental University, Tokyo, Japan; 2Department of Surgical Oncology, Graduate School, Tokyo Medical and Dental University, Tokyo, Japan; 3Department of Translational Oncology, Tokyo Medical and Dental University, Tokyo, Japan 1
ABSTRACT Background. The accurate preoperative diagnosis of depth of tumor invasion and nodal status in distal rectal cancer is important because neoadjuvant chemotherapy or lateral pelvic lymph node dissection is indicated for patients with T3–T4 tumor or nodal involvement. This study aimed to determine the optimal cutoff value for predicting lymph node metastasis in patients with distal rectal cancer using multidetector row computed tomography (MDCT). Methods. The study investigated 77 patients who had undergone surgery for distal rectal cancer at a single institution between 2008 and 2011. Diagnostic performance for depth of tumor invasion and mesorectal and lateral pelvic lymph node metastases was evaluated. The optimal cutoff value was determined by receiver operating characteristic curve analysis. Results. For predicting mesorectal and lateral pelvic lymph node metastasis, MDCT had a sensitivity of 0.36 and 0.89 and a specificity of 0.78 and 0.97, respectively. The optimal cutoff values of major and minor axes lengths for predicting mesorectal lymph node metastasis were 6.5 mm and 5.7 mm, respectively. The areas under the curve (AUCs) were 0.82 and 0.88, respectively. For
Ó Society of Surgical Oncology 2014 First Received: 2 March 2014; Published Online: 15 August 2014 H. Kobayashi, MD, FACS e-mail: [email protected]
predicting lateral lymph node metastasis, the optimal cutoff values were 9 mm for the major axis and 6 mm for the minor axis. Both AUCs were 1. Conclusions. Using MDCT, the optimal cutoff value of minor axis length for predicting mesorectal and lateral pelvic lymph node metastases in patients with distal rectal cancer was 6 mm. The accuracy of MDCT was satisfactory for predicting lateral pelvic lymph node metastasis.
Among the causes of cancer mortality, colorectal cancer ranks second in the United States and third in Japan.1,2 Furthermore, the incidence of colorectal cancer is increasing rapidly in Japan.2,3 The prognosis of patients with rectal cancer is known to be worse than that of patients with colon cancer. One of the reasons for this is the high local recurrence rate in rectal cancer patients.4 Whereas the standard therapy for T3–T4 rectal cancer is preoperative chemoradiotherapy plus total mesorectal excision (TME) in Western countries, TME plus lateral pelvic lymph node dissection is preferred in Japan. The Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines list the indications for lateral pelvic lymph node dissection as T3–T4 cancer with a distal edge located below the peritoneal reflection and distal rectal cancer with mesorectal lymph node metastasis.5 Therefore, accurate diagnosis of the depth of tumor invasion and lymph node metastasis enables identification of patients who do not need lateral pelvic lymph node dissection. Although the reported sensitivity of computed tomography (CT) for detecting nodal involvement of colorectal cancer is 45–73 %,6–10 the quality of recent multidetector row CT scanners is better than those used in previous studies.
204
This study aimed to clarify the diagnostic accuracy of multidetector row CT for T3–T4 distal rectal cancer and lymph node metastasis. Furthermore, the optimal cutoff values of major and minor axes lengths for predicting lymph node metastasis in patients with distal rectal cancer were evaluated. METHODS Patients The medical charts of 77 patients who underwent curative surgery for distal rectal carcinoma in the Department of Surgical Oncology at the Tokyo Medical and Dental University in Tokyo, Japan between January 2008 and December 2011 were reviewed. In this study, distal rectal cancer was defined as a tumor whose distal edge was located below the peritoneal reflection. The CT scanner used in the current study had 64 detectors. The lengths of the major and minor axes were measured. All the patients underwent TME or tumor-specific mesorectal excision (TSME). In addition, the patients with T3–T4 distal rectal cancer underwent lateral pelvic lymph node dissection, as well as TME or TSME. The resected rectal specimens were stretched and pinned to a cork board. The surgeon identified and isolated the lymph nodes and recorded their number and distribution. The surgeons in this study harvested the lymph nodes from the specimen to evaluate the precise location of nodal metastases. This technique is not routinely performed in North America or Europe. After formalin fixation, the specimens and lymph nodes were examined by the pathologist. The associations between the lengths of the major and minor axes of the lymph nodes shown on preoperative CT and the status of pathologic lymph node metastases were evaluated retrospectively. The depth of tumor invasion was diagnosed preoperatively by colonoscopy, barium enema, magnetic resonance imaging, and endoscopic ultrasonography.
H. Kobayashi et al. TABLE 1 Clinicopathologic characteristics of 77 patients with curative surgical treatment for distal rectal cancer Clinicopathologic features
n
Age: years (range)
66 (35–93)
%
Gender Male
48
62
Female
29
38
Low anterior resection
40
52
Intersphincteric resection Abdominoperineal resection
3 28
4 36
Hartmann’s operation
4
5
Total pelvic exenteration
2
3
Present
40
52
Absent
37
48
Present
17
22
Absent
60
78
T0
2
3
T1
6
8
T2
18
23
T3
45
58
6
8
Well differentiated
22
29
Moderately differentiated
46
60
Poorly differentiated
5
6
Mucinous
3
4
Others
1
1
Present
40
52
Absent
37
48
Surgical procedure
Lateral pelvic lymph node dissection
Laparoscopic surgery
Pathologic T-category
T4b Pathology
Lymph node metastasis
RESULTS Patients’ Characteristics
Statistical Analysis All data are expressed as means ± standard deviations. Differences in continuous variables were compared using the Mann–Whitney U test, whereas differences in categorical variables were analyzed using the v2 test. The optimal cutoff values for predicting lymph node metastasis were analyzed using receiver operating characteristic (ROC) curves. Areas under the curve (AUCs) were also calculated. Data were analyzed using JMP 9 (SAS Institute Inc., Cary, NC, USA). Statistical significance was established at a p value lower than 0.05.
The characteristics of all the patients are listed in Table 1. No patient received preoperative chemoradiotherapy in the current study. The majority of the patients underwent sphincter-sparing surgery. Of the 77 patients, 40 also underwent lateral pelvic lymph node dissection. Laparoscopic surgery was performed in 17 patients. Diagnostic Performance for T-Category Five patients were excluded from the evaluation of diagnostic performance for T-category because of additional
Diagnostic Accuracy of MDCT for Lower Rectal Cancer
205
TABLE 2 Diagnostic performance for T-category pT0
pT1
pT2
pT3
cT0
0
cT1
1
cT2
pT4b
Total
0
0
1
0
1
0
5
0
0
6
0
3
4
3
0
10
cT3
0
0
9
38
4
51
cT4b
0
0
0
2
2
4
Total
1
3
18
44
6
72
TABLE 3 Diagnostic performance for depth of tumor invasion of T3–T4 pT1–pT2
pT3–pT4
Total
cT1–cT2 cT3–cT4
13 9
4 46
17 55
Total
22
50
77
TABLE 4 Diagnostic performance for mesorectal lymph node metastasis Pathologically positive
Pathologically negative
Total
Preoperatively positive
14
4
18
Preoperatively negative
25
34
59
Total
39
38
77
resection after endoscopic or transanal resection. Thus, 72 patients were included in the current study. The accuracy of diagnostic performance for T-category was 61 % (Table 2). From the viewpoint of clinical use, the diagnostic performance for the depth of tumor invasion of T3–T4 also was investigated (Table 3). The accuracy was 77 %. Diagnostic Performance for Mesorectal Lymph Node Metastasis The median number of mesorectal lymph nodes retrieved was 15 (range, 3–41). Among the 77 patients with a curative resection for distal rectal cancer, 39 had mesorectal lymph node metastases (Table 4). The median number of involved mesorectal nodes was 2 (range, 1–18). The sensitivity was 36 %, and the specificity was 89 % (P = 0.0071). The false-positive and false-negative rates were respectively 11 and 64 %, and the positive and negative predictive values were respectively 78 and 58 %.
Diagnostic Performance for Lateral Pelvic Lymph Node Metastasis The median number of lateral lymph nodes retrieved was 14 (range, 2–34). Among the 40 patients with lateral pelvic lymph node dissection, 8 (20 %) had lateral pelvic lymph node metastases. The median number of involved lateral nodes was 3 (range, 1–29). The sensitivity was 78 %, and the specificity was 97 % (P \ 0.0001). The false-positive and false-negative rates were respectively 3 and 22 %, whereas the positive and negative predictive values were respectively 88 and 94 % (Table 5). Optimal Cutoff Values for Predicting Lymph Node Metastasis The mean lengths of the major axis of positive and negative mesorectal lymph nodes shown on MDCT were respectively 9.4 and 5.3 mm (P \ 0.0001), whereas those of the minor axis were respectively 8.2 and 3.7 mm (P \ 0.0001). The ROC curves for predicting mesorectal lymph node metastasis using the major and minor axis lengths are shown in Fig. 1. The AUCs using the major and minor axis lengths were respectively 0.82 and 0.88. The optimal cutoff values of the major and minor axis lengths for predicting mesorectal lymph node metastasis were respectively 6.5 and 5.7 mm. As for lateral lymph node metastasis, the mean lengths of the major and minor axes in cases with positive lateral nodes were 14.4 and 11.4 mm, respectively. The optimal cutoff values of the major and minor axis lengths were 9 and 6 mm, respectively. The AUCs of the major and minor axis lengths were both 1. After the cutoff value of 6 mm for the minor axis was adopted, the sensitivity and specificity for predicting mesorectal lymph node metastasis were respectively 49 and 92 % (P \ 0.0001). The positive and negative predictive values were 86 and 64 %, respectively. The sensitivity and specificity for predicting lateral lymph node metastasis were improved to be 78 and 100 %, respectively (P \ 0.0001). The positive and negative predictive values were 100 and 94 %, respectively. DISCUSSION The current study demonstrated that the best cutoff value of minor axis length for predicting mesorectal and lateral pelvic lymph node metastases in patients with distal rectal cancer was 6 mm. The lengths of both the major and minor axes of the lymph nodes shown on MDCT were useful for predicting lymph node metastasis.
206
TABLE 5 Diagnostic performance for lateral pelvic lymph node metastasis Pathologically positive
Pathologically negative
Total
Preoperatively positive
7
1
8
Preoperatively negative
2
30
32
Total
9
31
40
a
1.00 0.90 0.80
Sensitivity
0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.00
0.20
0.40
0.60
0.80
1.00
0.80
1.00
1-specificity
b
1.00 0.90 0.80 0.70
Sensitivity
Using the length of the major axis on MDCT, 6.5 mm in the mesorectum and 9 mm in the lateral pelvic area were the cutoff values that provided the greatest predictive value. On the other hand, using the length of the minor axis, 5.7 mm in the mesorectum and 6 mm in the lateral pelvic area provided the greatest predictive value. These differences might reflect limitations of a small sample size. The current study also demonstrated that the AUC of the minor axis was greater than that of the major axis for predicting mesorectal lymph node metastasis. The AUC of the major axis was equal to that of the minor axis for predicting lateral pelvic lymph node metastasis. Therefore, a minor axis length of 6 mm is the most appropriate cutoff value for predicting lymph node metastases in patients with distal rectal cancer. The aforementioned results are consistent with those previously reported. Kanemoto et al. 11 reported that the minor axis, not the major axis, was appropriate for predicting colorectal lymph node metastasis. After the cutoff value of 6 mm was adopted for the minor axis, the sensitivity and specificity for predicting both mesorectal and lateral pelvic lymph node metastasis were improved. Especially, the specificity for predicting mesorectal and lateral pelvic lymph node was respectively 92 and 100 % if this cutoff value was used to evaluate the nodal status in lower rectal cancer. The use of MDCT will help in selecting patients for lateral node dissection. The most common method for the diagnosis of hematogenous and nodal metastases from colorectal cancer is CT, although a distinct criterion for diagnosing lymph node metastasis from colorectal cancer does not exist. The most widely used criterion is any node larger than 1 cm or a cluster of three or more nodes smaller than 1 cm.6, 12–17 However, we often see lymph node metastases smaller than 1 cm. The most appropriate cutoff value for predicting lymph node metastasis from colorectal cancer has not been established. Dighe et al. 18 reported the diagnostic precision of CT in local staging of colon cancers. In their meta-analysis, the sensitivity and specificity for detecting nodal involvement using CT ranged from 0.66 to 0.87 and from 0.7 to 0.8, respectively. These were unsatisfactory compared with the rate for detecting tumor invasion using CT. The sensitivity and specificity for detecting tumor invasion were 0.84 to 0.95 and 0.74 to 0.86, respectively. However, in their investigation, the best accuracy was reported by the studies that had used MDCT. Thus, a precise criterion for detecting lymph node metastasis from colorectal cancer using MDCT needs to be developed. The current study investigated both lateral pelvic lymph node metastasis and mesorectal lymph node metastasis from distal rectal cancer. The lateral pelvic lymph node is an important location as a recurrence site after surgery for
H. Kobayashi et al.
0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.00
0.20
0.40
0.60
1-specificity FIG. 1 The receiver operating characteristics curves of the diagnostic performance for lymph node metastasis. The lengths of the major (a)and minor (b) axes of lymph nodes on preoperative multi-detector row CT were investigated. The area under the curve was 0.82 (a) and 0.88 (b), respectively
distal rectal cancer. This study demonstrated that the AUC for lateral pelvic lymph node metastasis was greater than that for mesorectal lymph node metastasis. This might mean that MDCT is more useful for predicting lateral pelvic lymph node metastasis than for predicting mesorectal lymph node metastasis. One of the reasons for the difference in the AUC was the size of the metastatic lymph nodes. No metastatic lymph node had a minor axis shorter than 6 mm in the lateral
Diagnostic Accuracy of MDCT for Lower Rectal Cancer
pelvic area, although such nodes were found in the mesorectum. At the same time, the positive lateral pelvic lymph nodes were larger than the positive mesorectal lymph nodes. The current results support those of Yano et al. who reported that the size of positive mesorectal lymph nodes (5.7 mm) was significantly smaller than the size of positive lateral pelvic lymph nodes (11.5 mm). In their study, the accuracy of detecting positive lymph nodes in the lateral pelvic area was superior to that in the mesorectum. Approximately 15 % of patients with distal rectal cancer have lateral pelvic lymph node metastasis.19 The standard therapeutic approach for controlling lateral pelvic lymph node metastasis is preoperative chemoradiotherapy in Western countries and lateral pelvic lymph node dissection in Japan. The accurate preoperative diagnosis of lateral pelvic lymph node metastasis is very important because both preoperative chemoradiotherapy and lateral lymph node dissection can lead to postoperative sexual and urinary dysfunction.20–23 There is merit in omitting such invasive therapies to prevent such postoperative complications. The sensitivity for the preoperative diagnosis of local invasion (i.e., depth of tumor invasion) of T3–T4 was 92 %. This was satisfactory, whereas the specificity was 59 %. It is important to decrease false-positive cases in detecting the preoperative depth of tumor invasion using MDCT. In all false-positive cases, pathologic T2 tumors were preoperatively diagnosed as T3 tumors. The accurate diagnosis of T3–T4 distal rectal cancer is very important because patients with those tumors usually receive preoperative chemoradiotherapy or lateral pelvic lymph node dissection.19, 24–30 In the current study, transrectal ultrasonography (TRUS) was not used. The use of TRUS will be necessary to improve the accuracy of the preoperative diagnosis of tumor invasion depth. A previous pooled analysis demonstrated that the sensitivity and specificity for the preoperative diagnosis of tumor invasion depth using TRUS were 0.97 and 0.87, respectively.31 In conclusion, the optimal cutoff value of minor axis length for predicting lymph node metastasis in patients with distal rectal cancer using MDCT was 6 mm. The accuracy of MDCT was satisfactory for predicting lateral pelvic lymph node metastasis in distal rectal cancer. REFERENCES 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012;62:10–29. 2. Kotake K, Honjo S, Sugihara K, et al. Changes in colorectal cancer during a 20-year period: an extended report from the multi-institutional registry of large bowel cancer, Japan. 2003;46:S32–43. 3. Muto T, Kotake K, Koyama Y. Colorectal cancer statistics in Japan: data from JSCCR registration, Int J Clin Oncol. 1974–1993. 2001;6:171–6.
207 4. Kobayashi H, Mochizuki H, Sugihara K, et al. Characteristics of recurrence and surveillance tools after curative resection for colorectal cancer: a multicenter study. Surgery. 2007;141:67–75. 5. Watanabe T, Itabashi M, Shimada Y, et al. Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol. 2012;17:1–29. 6. Balthazar EJ, Megibow AJ, Hulnick D, Naidich DP. Carcinoma of the colon: detection and preoperative staging by CT. AJR Am J Roentgenol. 1988;150:301–6. 7. Hundt W, Braunschweig R, Reiser M. Evaluation of spiral CT in staging of colon and rectum carcinoma. Eur Radiol. 1999;9: 78–84. 8. Isbister WH, al-Sanea O. The utility of preoperative abdominal computerized tomography scanning in colorectal surgery. J R Coll Surg Edinb. 1996;41:232–4. 9. McAndrew MR, Saba AK. Efficacy of routine preoperative computed tomography scans in colon cancer. Am Surg. 1999;65:205–8. 10. Taylor AJ, Youker JE. Imaging in colorectal carcinoma. Semin Oncol. 1991;18:99–110. 11. Kanamoto T, Matsuki M, Okuda J, et al. Preoperative evaluation of local invasion and metastatic lymph nodes of colorectal cancer and mesenteric vascular variations using multidetector-row computed tomography before laparoscopic surgery. J Comput Assist Tomogr. 2007;31:831–9. 12. Acunas B, Rozanes I, Acunas G, Celik L, Sayi I, Gokmen E. Preoperative CT staging of colon carcinoma (excluding the rectosigmoid region). Eur J Radiol. 1990;11:150–3. 13. Ashraf K, Ashraf O, Haider Z, Rafique Z. Colorectal carcinoma, preoperative evaluation by spiral computed tomography. J Pak Med Assoc. 2006;56:149–53. 14. Cademartiri F, Luccichenti G, Rossi A, Pavone P. Spiral hydroCT in the evaluation of colo-sigmoideal cancer. Radiol Med. 2002;104:295–306. 15. Freeny PC, Marks WM, Ryan JA, Bolen JW. Colorectal carcinoma evaluation with CT: preoperative staging and detection of postoperative recurrence. Radiology. 1986;158:347–53. 16. Gazelle GS, Gaa J, Saini S, Shellito P. Staging of colon carcinoma using water enema CT. J Comput Assist Tomogr. 1995;19: 87–91. 17. Harvey CJ, Amin Z, Hare CM, et al. Helical CT pneumocolon to assess colonic tumors: radiologic-pathologic correlation. AJR Am J Roentgenol. 1998;170:1439–43. 18. Dighe S, Purkayastha S, Swift I, et al. Diagnostic precision of CT in local staging of colon cancers: a meta-analysis. Clin Radiol. 2010;65:708–19. 19. Kobayashi H, Mochizuki H, Kato T, et al. Outcomes of surgery alone for lower rectal cancer with and without pelvic sidewall dissection. Dis Colon Rectum. 2009;52:567–76. 20. Akasu T, Sugihara K, Moriya Y. Male urinary and sexual functions after mesorectal excision alone or in combination with extended lateral pelvic lymph node dissection for rectal cancer. Ann Surg Oncol. 2009;16:2779–86. 21. Lange MM, Maas CP, Marijnen CA, et al. Urinary dysfunction after rectal cancer treatment is mainly caused by surgery. Br J Surg. 2008;95:1020–8. 22. Lange MM, Marijnen CA, Maas CP, et al. Risk factors for sexual dysfunction after rectal cancer treatment. Eur J Cancer. 2009;45:1578–88. 23. Marijnen CA, van de Velde CJ, Putter H, et al. Impact of shortterm preoperative radiotherapy on health-related quality of life and sexual functioning in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol. 2005;23:1847–58. 24. Bosset JF, Collette L, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med. 2006;355: 1114–23.
208 25. Gerard JP, Azria D, Gourgou-Bourgade S, et al. Clinical outcome of the ACCORD 12/0405 PRODIGE 2 Randomized Trial in Rectal Cancer. J Clin Oncol. 2012;30:4558–65. 26. Gerard JP, Conroy T, Bonnetain F, et al. Preoperative radiotherapy with or without concurrent fluorouracil and leucovorin in T3–4 rectal cancers: results of FFCD 9203. J Clin Oncol. 2006;24:4620–5. 27. Kobayashi H, Mochizuki H, Kato T, et al. Is total mesorectal excision always necessary for T1–T2 lower rectal cancer? Ann Surg Oncol. 2010;17:973–80. 28. Kobayashi H, Mochizuki H, Kato T, et al. Lymph node ratio is a powerful prognostic index in patients with stage III distal rectal
H. Kobayashi et al. cancer: a Japanese multicenter study. Int J Colorectal Dis. 2011;26:891–6. 29. Sugihara K, Kobayashi H, Kato T, et al. Indication and benefit of pelvic sidewall dissection for rectal cancer. Dis Colon Rectum. 2006;49:1663–72. 30. Sugihara K, Moriya Y, Akasu T, Fujita S. Pelvic autonomic nerve preservation for patients with rectal carcinoma: oncologic and functional outcome. Cancer. 1996;78:1871–80. 31. Solomon MJ, McLeod RS. Endoluminal transrectal ultrasonography: accuracy, reliability, and validity. Dis Colon Rectum. 1993;36:200–5.
