| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the First Department of Surgery, Nagasaki University School of Medicine, Nagasaki, Japan.
Correspondence: Address correspondence and reprint requests to: Tohru Nakagoe, MD, First Department of Surgery, Nagasaki University School of Medicine, 1-7-1 Sakamoto, Nagasaki 852-8501, Japan; Fax: 81-95-849-7306; E-mail: nakagoe{at}net.nagasaki-u.ac.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods: Microscopic distal intramural spread was examined in 134 consecutive specimens of resected rectal cancer. Correlations among distal intramural spread, established clinicopathologic factors, and patients’ prognoses were examined by univariate and multivariate analyses. American Joint Committee on Cancer classification and stage groupings were used for tumor assessment.
Results: Thirty-three patients (24.6%) had distal intramural spread. Multivariate logistical regression analysis revealed that T3/T4 and M1 were independent predictive variables for the presence of distal intramural spread. Patients with distal intramural spread had a shorter disease-specific or disease-free survival time after curative surgery than those without distal intramural spread (P = .0003 and P = .0006, respectively). Most patients with distal intramural spread developed distant recurrence. Cox’s regression with multiple covariates showed that distal intramural spread is an independent factor in predicting distant recurrence and worse outcomes after curative surgery in patients with rectal cancer.
Conclusions: Distal intramural spread is an independent risk factor for distant metastasis and poor prognosis in patients with rectal cancer.
Key Words: Rectal cancer • Distal intramural spread • Distal resection margin • Distant metastasis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Thus, a correlation between distal margin clearance (macroscopic distal resection margin minus microscopic distal spread equals microscopic distal margin clearance) and patient prognosis has been fully discussed. However, the clinical significance of microscopic distal intramural spread as a prognostic factor for survival and recurrence in patients with rectal cancer has not been fully addressed.6–9 It has been reported that most patients with distal intramural spread had a poor prognosis and developed distant metastasis rather than local recurrence, even after curative surgery.6–9 These data were evaluated only by univariate analysis. We believe that the univariate-analyzed data should be verified by means of multivariate analysis.
In this study, we first examined the extent of microscopic distal intramural spread of rectal cancer. Second, we determined correlations among distal intramural spread and well-established clinicopathologic variables, survival, and the incidence of distant and local recurrence to clarify whether these clinical end points bore any relationship to the presence of distal intramural spread. These relationships were examined by univariate and multivariate analyses.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Definitions
Tumor location was measured in distance (cm) from the dentate line, which was determined by preoperative rigid or flexible colonoscopy. Macroscopic types of tumor (localized type or infiltrative type) were determined by colonoscopy and/or by the resected specimen according to the Japanese Classification of Colorectal Carcinoma.10 Tumor size (cm) was documented as the longest diameter of the tumor on gross examination of the fresh specimen. The distal resection margin (cm) was determined on gross examination of the fresh specimen that was spread on a wooden specimen board by means of fine needles.
Curability of surgical resection was determined both macroscopically and microscopically according to the Japanese Classification of Colorectal Carcinoma.10 Namely, noncurative resection indicated that there was definite residual tumor, whereas curative resection indicated no evidence of residual tumor. Curative resection also indicated that no residual tumor was estimated when distant metastases to liver or lung were excised completely.
Operative Technique
We principally removed lymph nodes at the root of the inferior mesenteric artery. Total mesorectal excision was performed for patients with cancer in the middle or lower third of the rectum, but not for patients with cancer in the upper third of the rectum. Total mesorectal excision involved sharp dissection in the avascular plane between the mesorectum and surrounding parietal tissue under direct vision. The choice of operation (anterior resection or abdominoperineal resection) depended on the location of the tumor relative to the dentate line and stage. The policy has been to perform the anastomosis in the abdomen by anterior resection whenever possible. In anterior resections, the distal rectum was divided with a 1-cm margin for stage 0 or I or at least a 2-cm distal resection margin for stage II, III, and IV tumors. Lateral lymph node dissection was indicated for patients with N1 or N2 cancer of the lower third of the rectum.
Histopathologic Evaluation
Two of the authors (E.Y. and K.T.) examined all rectal cancer specimens included in this study. After the operation, mesenteric and pararectal lymph nodes were removed from the resected specimens. Each specimen was pinned under tension onto a wooden board and then fixed in 10% formalin. After fixation in formalin for 1 week, the whole tumor was sliced into longitudinal sections approximately 5 mm thick. The distal normal mucosa of the tumor was also entirely sliced from the anal edge of the tumor down to the distal resection margin. All specimens were embedded in paraffin and were available for hematoxylin and eosin and Elastica van Gieson staining. Depth of tumor invasion, lymph node metastasis, histological differentiation, venous invasion, and lymphatic invasion were determined. American Joint Committee on Cancer classification and stage groupings were used for tumor assessment.11
Distal intramural spread of rectal carcinoma (direct extension, lymphatic invasion, or venous invasion) was determined according to Shirouzu et al.6 The maximum length of distal intramural spread was measured by using a micrometer from the microscopically defined distal mucosal edge of the tumor. Distal donuts included in the circular stapler were examined histologically. All were histologically free of carcinoma.
Follow-Up and Study End Points
Follow-up was completed through office records or by telephone or written contact with the patient, primary care physician, or both. The patients were examined at 3-month intervals for the first three postoperative years and every 6 months thereafter. Complete clinical examinations were performed and included digital examinations and rectal palpation, as well as a total colonoscopy. In addition, patients underwent periodic metastatic work-ups with chest and abdominal computed tomography.
End points of the study were disease-specific and disease-free survivals and tumor recurrence after curative resection. The criteria for establishing recurrent disease were histological confirmation; palpable disease or evidence of disease on radiographical studies, with subsequent clinical progression; and supportive biochemical data (e.g., increasing levels of carcinoembryonic antigen). Disease-specific survival was defined as the time from the date of the operation to the date of rectal cancer death. Data from patients who died of causes other than rectal cancer were censored in the survival analysis. Disease-free survival was defined as the time from the date of the operation to the date of first recurrence.
Follow-up was completed with a median of 87.7 months (range, 49.8–142.5 months) for patients alive at the conclusion of the study. A minimum of 6 years of follow-up is available for all patients. One patient who died on the fifth postoperative day secondary to a pulmonary embolism was excluded from the survival analysis.
Patients with recurrence were defined in terms of local or distant recurrence.12,13 Local recurrence was defined as tumor limited to the rectum, pelvis, or adjacent organs. Distant recurrence was defined as any tumor recurrence outside the pelvis and included metastases to the abdominal cavity, liver, lung, brain, or bone. The location of recurrence was determined at the time of the first clinical relapse.
Statistical Analysis
Continuous data were expressed as mean ± SD. Categorical data were analyzed with Fisher’s exact test. The influence of each variable on the survival time was calculated according to the Kaplan-Meier method,14 and differences between survival times were tested for significance with the log-rank test.15 The prognostic relevance of a single factor was determined by univariate Cox’s regression analysis, whereas clinicopathologic factors were analyzed by multivariate Cox’s regression analysis.16 We used the logistical regression analysis to study the predictive value of risk factors because the outcome variable was binary, such as rectal cancer without versus with distal intramural spread.17,18 By means of univariate analysis, potentially predictive variables were identified at a significance level of P < .25, and these variables were then used for multivariate analysis. A 25% significance level was selected on the basis of Hosmer and Lemeshow’s recommendations for building multivariate models.17 All tests were two tailed, and a P value of <.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.0 and <2.0 cm in length; only 1 patient had distal intramural spread 2.0 cm in length (2.1 cm). The mode of distal spread—direct invasion, lymphatic invasion, or venous invasion—was found in 5, 18, and 10 patients, respectively.
Logistical Regression Analysis to Determine Predictive Factors for Distal Intramural Spread
Univariate analysis indicated that eight variables (tumor location, tumor size, histological grade, lymphatic invasion, venous invasion, T stage, N stage, and M stage) were associated with distal intramural spread at a significance level of P < .25. Therefore, these variables were included in a multivariate analysis. Distal intramural spread was found to be independently related to two variables: T stage and M stage (Table 2).
|
1.0 cm in length. In patients with T4 tumors, distal intramural spread occurred in 11 (36.7%) of 30 patients and was <2.0 cm. None of the patients with curative resection had distal intramural spread 2.0 cm in length.
|
|
As shown in Table 4, recurrence developed in 22 (23.4%) of 94 patients without distal intramural spread and 13 (56.5%) of 23 patients with distal intramural spread. The difference in frequency of recurrence was statistically significant (P = .0042). Of patients with distal intramural spread, 11 patients developed distant recurrence and 2 patients developed both distant and local recurrences, whereas none of the patients developed only local recurrence.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The multivariate logistical analysis was used in this study to determine predictive factors for distal intramural spread. T stage and M stage were significant factors (P = .040 and P = .0031, respectively). In addition, tumor location and lymphatic invasion also approached significant values (P = .056 and P = .072, respectively). A larger patient sample size would likely show that tumor location and lymphatic invasion were significant predictive factors. Grinell7 reported that distal intramural spread was usually found in advanced cancers of the rectum with a much higher incidence of node metastasis, vein invasion, and annular involvement of the bowel than those without distal intramural spread. In addition, it occurred most often in tumors so advanced that only a palliative resection could be performed. Even in curative resections, the prognosis for the patient was very poor. Penfold8 reported that patients with intramural extension of >1 cm did not survive >5 years. Recently, Lazorthes et al.9 reported that the extent of distal intramural spread of rectal cancer was directly related to lymph node involvement and increased incidence of hepatic metastasis. Shirouzu et al.6 concluded that distal spread (including both intramural and extramural spread) seemed to be an important risk factor for distant metastasis after surgery. These results, analyzed by means of univariate analysis,6–9 are supported and strengthened by the multivariate-analyzed data reported in this study.
Lymphatic blockage develops when lymph node metastasis occurs. When a large number of lymph node metastases occur along the inferior mesenteric artery, lymph flow may change to a retrograde direction. This may be the cause for the common occurrence of abnormal retrograde distal spread.5,6 In patients with distal mural spread, the cancer seems to be extremely advanced, with occult cancer cells potentially remaining all over the body. That is why the prognosis of patients with distal spread may be dismal.
Distal extramural spread, as well as distal intramural spread, can occur.6,19 Microscopic spread to the distal mesorectum (extramural) may exceed the intramural spread of rectal cancer.20,21 Patients with tumors in the distal mesorectum have worse outcomes, a great risk of local recurrence, and an increased frequency of distant metastasis.21 Therefore, the extent of both distal intramural and extramural spread is extremely important in deciding the optimal distal resection margin from the tumor when performing sphincter-preserving operations. Shirouzu et al.6 studied the correlation between distal spread (both intramural and extramural) and patients’ prognosis and reported that a distal resection margin of 1 cm may be appropriate clearance for most rectal cancers. The authors stressed that a longer resection margin does not necessarily improve the prognosis of patients if the extent of distal spread is confirmed to be longer than 1 cm. In fact, the disease is incurable in such patients, and they usually die as a result of distant metastasis.
In this study, distal extramural spread was not evaluated. However, our results suggest that none of the patients with curative resection for Tis/T3 tumors had distal intramural spread 1.0 cm in length, whereas none of the patients with curative resection for T4 tumors had distal intramural spread 2.0 cm in length. In addition, our study, using multivariate analysis, also revealed that the distal intramural spread was independently related to T stage. In general, carcinoma in the middle or lower third of the rectum should be excised concurrently with total mesorectum excision.1,22 Therefore, we believe that a distal resection margin of 1 or 2 cm may be adequate distal clearance for carcinoma with Tis/T3 or T4 stage in the middle or lower third of the rectum, respectively.
In conclusion, distal intramural spread is an independent risk factor, separate from stage, for distant metastasis and poor prognosis in patients with rectal cancer. Recently, preoperative evaluation of distal intramural spread, as well as the depth of tumor invasion, of rectal cancer has included the practice of using transrectal ultrasonography.23 Preoperative detection for distal intramural spread by using such ultrasonography might be helpful in planning appropriate adjuvant therapies.
|
Footnotes |
|---|
Deceased. 
Thirty-three (24.6%) of 134 consecutive specimens of resected rectal cancer had microscopic distal intramural spread of tumor. Multivariate analysis revealed that distal intramural spread is an independent risk factor for distant metastasis and poor prognosis in patients with rectal cancer.
Received for publication May 23, 2002. Accepted for publication September 23, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
