(Radiology. 2000;215:108-114.)
© RSNA, 2000
Rectal Carcinoma: Double-Contrast MR Imaging for Preoperative Staging1
Nils-Olof Wallengren, MD,
Stig Holtås, MD,
Åke Andrén-Sandberg, MD,
Eva Jonsson, MSc,
Doris T. Kristoffersson, MSc and
Stephen McGill, PhD
1 From the Departments of Radiology (N.O.W., S.H.) and Surgery (A.A.S.), Center for Medical Imaging and Physiology, University Hospital, S-221 85 Lund, Sweden; and Nycomed-Amersham, Stockholm and Oslo, Sweden (E.J., D.T.K., S.M.). From the 1995 RSNA scientific assembly. Received January 27, 1998; revision requested April 7; final revision received June 14, 1999; accepted July 20. Address reprint requests to N.O.W.
 |
Abstract
|
|---|
PURPOSE: To evaluate and compare the imaging findings and staging of rectal carcinoma by using conventional magnetic resonance (MR) imaging, MR imaging with an enema of superparamagnetic ferristene-based contrast material, and MR imaging with an enema of ferristene solution plus intravenous injection of gadodiamide.
MATERIALS AND METHODS: Twenty-nine patients (17 women, 12 men; age range, 39–91 years) referred with a diagnosis of rectal carcinoma were examined. Analysis of the rectal wall and staging of the tumor were performed. In all patients, the MR imaging findings were correlated with the histopathologic findings.
RESULTS: The contrast material enema caused distention of the rectum and an intraluminal signal void, whereas the gadodiamide injection caused enhancement of the mucosa on T1-weighted images. This enhancement enabled evaluation of the normal rectal wall and differentiation of the mucosa, tunica muscularis, and perirectal space, which was not possible on the nonenhanced images. Double-contrast (ferristene solution plus gadodiamide) MR imaging was superior to imaging with only ferristene-based contrast material and had a sensitivity of 100%, specificity of 70%, and accuracy of 90% in distinguishing tumor stages worse than Dukes A.
CONCLUSION: Double contrast material–enhanced MR imaging enables accurate rectal carcinoma staging, which is not possible at nonenhanced imaging.
Index terms: Magnetic resonance (MR), contrast media, 757.121411, 757.12143 • Neoplasms, staging, 757.32 • Rectum, MR, 757.121411, 757.12143 • Rectum, neoplasms, 757.32
|
Introduction
|
|---|
Rectal cancer is one of the most common malignant tumors. At present, staging is based on findings at surgery and histopathologic analysis of the specimens. In an early pilot study (1), our results showed that it is possible to stage rectal tumors preoperatively with double contrast material–enhanced magnetic resonance (MR) imaging—that is, MR imaging with an intravenous injection of gadodiamide combined with an enema of superparamagnetic contrast material—and obtain better results than those obtained with MR imaging performed with specially designed endorectal surface coils, conventional MR imaging, ultrasonography (US), or computed tomography (CT) (2–12). It is of practical importance to be able to determine the stage of rectal carcinomas preoperatively, because patients with Dukes A carcinoma can be operated on without delay and, possibly, by using a transrectal route (13–15). In Sweden, Dukes B, C, and D carcinomas are routinely irradiated preoperatively to improve survival rates (16); however, this is not considered to be necessary for patients with Dukes A carcinoma.
In patients with widespread perirectal growth, in whom radical surgery offers no hope for cure because of spread to the adjacent organs, MR imaging can reveal this situation, and unnecessary surgery can be avoided (14). Tissue characterization is excellent with MR imaging and can be improved further for staging of rectal carcinoma by enhancement of the mucosa with an intravenously administered gadolinium-based contrast material combined with an enema that contains superparamagnetic iron oxide particles. As shown in our pilot study (1), this contrast material enema causes distention of the bowel and a signal void in the bowel lumen, which facilitates analysis of tumor growth in the rectal wall. The aim of the present study was to compare conventional MR imaging, MR imaging with ferristene-based contrast material, and MR imaging with a ferristene solution enema plus gadodiamide injection to investigate whether reproducible and good results, similar to those in the pilot study, could be obtained in a larger patient series.
|
MATERIALS AND METHODS
|
|---|
Patients
The original study population consisted of 31 consecutive patients with a diagnosis of rectal carcinoma, based on manual palpation and proctoscopic and/or barium examination results. The study was approved by the local ethics committee, and written informed consent was obtained from each patient. The study lasted for 20 months. One patient was excluded from the study because the tumor found at barium examination was caused by an artifact. A second patient was excluded when histopathologic examination revealed that staging was not possible because the tumor was not radically extirpated, and, thus, precise staging based on the results of definite histopathologic examination of the whole specimen could not be performed. The final analyses were based on findings in the remaining 29 patients (17 women, 12 men; age range, 39–91 years). All patients with a final diagnosis of rectal carcinoma underwent surgery; in one patient, who had a final diagnosis of leiomyoma, only biopsy was performed. In all cases, the final diagnosis was made by means of histopathologic evaluation.
Contrast Material
The contrast materials used in this study were the paramagnetic contrast medium gadodiamide (Omniscan; Nycomed Imaging, Oslo, Norway), which was administered by means of intravenous injection at a dose of 1 mmol/per kilogram of body weight, and a superparamagnetic ferristene-based contrast material (Abdoscan; Nycomed Imaging), which was administered through the rectum. Injected gadodiamide, in the doses described, acts as a positive contrast material on T1-weighted images.
The superparamagnetic ferristene-based contrast material consisted of a nonbiodegradable, sulphonated styrene-divinylbenzene copolymer coated with superparamagnetic iron oxide (a mixture of Fe2O3 and Fe3O4). It is a negative contrast material compound—that is, it reduces the signal intensity by shortening the T2*. Although this effect is most pronounced on T2-weighted images, this contrast material also causes reduced signal intensity on T1-weighted images (17,18). The optimal concentration of superparamagnetic contrast material, that is, the concentration that causes a signal void without causing artifacts, was determined previously in a phantom study (19) and found to be 0.1 g/L of particles, which is equal to 20 mg/L (3,581 µmol/L) iron. In this study, we used a concentration of 23.4 mg/L (419 µmol/L) iron. To get the contrast material to fill and stay in the rectum, it was mixed with viscosity-increasing microcrystalline cellulose– and xantham gum–based granules. These granules were mixed with 600 mL of water to produce a suspension with a viscosity of approximately 16 P, which was measured at a shear rate of 1.46 s-1 at room temperature by using a WOR rheometer (Bohlin Reologi AB, Lund, Sweden). The high viscosity was necessary to achieve an even distribution of the suspension in the rectum and sigmoid colon.
The patients were not allowed to drink or eat during the 4 hours immediately before the examination and were prepared as they would be for proctoscopy—that is, with a bowel enema to clean the distal part of the colon and rectum. The contrast material mixture was administered by means of a rectal tube, and the rectum was filled until the patient indicated discomfort. The normal volume used was between 200 and 400 mL, and the maximum volume used was 450 mL. To reduce bowel motion and keep the contrast material in the rectum, the patients were given 40 mg of butyl-scopolamine hydrobromide (Buscopan; Boehringer Ingelheim Pharma KG, Ingelheim, Germany) intravenously prior to the acquisition of the postcontrast images. In three patients, 0.5 mg of glucagon was given intravenously instead of butyl-scopolamine hydrobromide.
MR Imaging Examination
MR imaging was performed with a 0.3-T beta-3000 imaging unit (Fonar, Melville, NY) with a vertical magnetic field by using either a body coil or solenoidal surface coil wrapped around the area of interest. For T1-weighted imaging, parameters of 600/30 (repetition time msec/echo time msec) and two signals acquired were used. For T2-weighted imaging, a spin-echo sequence was used with the following parameters: 1,500/85, one signal acquired, 256 x 256 matrix, 1-mm pixel size, 7-mm section thickness, and 3-mm intersection gap. Before contrast material administration, sagittal and transverse T1-weighted images were obtained. Immediately after the rectal administration of ferristene-based contrast material, sagittal T1-weighted images were obtained. In addition to the previously administered ferristene-based contrast material, gadodiamide was injected intravenously, and then sagittal and transverse T1-weighted images were obtained. In addition, a T2-weighted image was acquired in the best projection, as judged from the T1-weighted images.
Analyses
The three different image series were evaluated by using consensus of two experienced radiologists (N.O.W., S.H.), who stated which image series gave the most diagnostic information. The following findings were recorded: number of detectable tumors; quality of tumor delineation, which was graded as not possible, poor, adequate, or excellent; and presence of enlarged (>10-mm) regional lymph nodes. Radiologic staging was performed according to the Dukes stage system (20), and the findings were divided into two groups: Dukes A carcinoma or benign tumor and Dukes B carcinoma or worse. The MR imaging–based diagnosis was given with five levels of certainty, as is appropriate for receiver operating characteristic analyses, and compared with the histopathologic diagnosis. The histopathologic definition of tumor stage worse than Dukes A is a tumor that has broken through the tunica muscularis propria of the rectal wall (20). The quality of filling with ferristene-based contrast material and the presence of MR imaging artifacts also were evaluated. Any adverse manifestations associated with the examinations were recorded.
Statistical Evaluation
The estimate of sensitivity was calculated as the proportion of patients considered by the radiologists to have tumors of a stage possibly, probably, or definitely worse than Dukes A compared with the proportion of patients with Dukes A tumors, as staged at surgery and/or histopathologic analysis.
The estimate of specificity was calculated as the proportion of patients considered by the radiologists to have tumors of a stage probably or definitely not worse than Dukes A out of all the patients with benign or Dukes A tumors at surgery and/or histopathologic analysis.
The precision of these estimates was expressed as a binomial, exact 95% CI (21). In addition, a receiver operating characteristic curve (Fig 1) for radiologic evaluation was constructed. The coordinates for the curve were established by using the ROCFIT program (C. Metz, University of Chicago, Ill). The area under the curve was calculated by using the trapezoid method.
|
RESULTS
|
|---|
Staging
Twenty-eight of the 29 consecutive patients were operated on, and postoperative histopathologic analysis with staging of carcinoma was performed. In one patient referred with rectal carcinoma, the tumor growth emanated from the tunica muscularis, and biopsy results showed leiomyoma (Fig 2) without signs of tumor infiltration. Because the tumor was an incidental finding at routine medical examination and the patient had no symptoms, no treatment was required. In the 28 patients who underwent surgery, the double-contrast MR imaging classification of carcinoma correlated well with the classification obtained by using postoperative histopathologic analysis (Table 1, Fig 1). The results of distinguishing tumor stages worse than Dukes A yielded the following results: 100% sensitivity, 70% specificity, and 90% accuracy.
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. (a) Sagittal and (b) transverse double-contrast T1-weighted spin-echo MR images (600/30, two signals acquired) obtained in a 68-year-old woman. At routine medical examination, a tumor was identified dorsally in the rectum by means of rectal palpation. Proctoscopy and barium enema examination did not enable identification of a tumor. a and b show the tumor (arrow) to be situated in the rectal wall and emanating from the tunica muscularis, and it appears as a local thickening of the muscular layer. The histopathologic specimen obtained at biopsy revealed a leiomyoma. The patient had undergone hysterectomy, and a metal clip (arrowhead in a) at the vaginal cuff caused a typical artifact.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. (a) Sagittal and (b) transverse double-contrast T1-weighted spin-echo MR images (600/30, two signals acquired) obtained in a 68-year-old woman. At routine medical examination, a tumor was identified dorsally in the rectum by means of rectal palpation. Proctoscopy and barium enema examination did not enable identification of a tumor. a and b show the tumor (arrow) to be situated in the rectal wall and emanating from the tunica muscularis, and it appears as a local thickening of the muscular layer. The histopathologic specimen obtained at biopsy revealed a leiomyoma. The patient had undergone hysterectomy, and a metal clip (arrowhead in a) at the vaginal cuff caused a typical artifact.
|
|
Tumor Delineation
The tumor delineation on sagittal MR images is presented in Table 2. In all except four patients, tumor delineation was not possible on the precontrast images. On the sagittal postcontrast images obtained with ferristene-based contrast material only, the delineation improved considerably after the rectal administration of the contrast material and yielded adequate or excellent results in the majority of cases. In three cases, the delineation was not possible. Of these three patients, one had inadequate filling of contrast material, and in two, biopsy performed before the examination caused local bleeding that made differentiation between the bowel wall layers difficult.
On the sagittal postcontrast (ferristene solution plus gadodiamide) T1-weighted images, the results improved further after the rectal administration of ferristene-based contrast material and intravenous injection of gadodiamide. In two cases, delineation was not possible due to incomplete filling of contrast material and local bleeding caused by biopsy. The results were similar on the T1-weighted transverse images: It was possible to delineate the entire tumor in 24 cases, and 16 of these studies were of excellent quality and eight were of adequate quality. In five cases there was poor delineation because of artifacts, inadequate filling, and biopsy.
On the sagittal postcontrast (ferristene solution plus gadodiamide) T2-weighted images obtained in 13 patients, the findings were inferior compared with those on the T1-weighted images. In 15 cases, T2-weighted MR imaging was performed in the transverse plane. The delineation quality was considered to be adequate in 11 cases and excellent in one. In two cases there was poor tumor delineation, and in one case, delineation was not possible. In one of the 29 patients, T2-weighted imaging could not be performed due to patient fatigue.
When a comparison between sequences was performed, in all 29 cases, the T1-weighted images obtained after the administration of ferristene-based contrast material plus gadodiamide were considered to have provided the most diagnostic information.
Number of Tumors
In 28 of the 29 patients, the primary tumor was identified on the postcontrast images. One patient was incontinent, and, therefore, no adequate filling of rectal contrast material could be achieved and tumor identification was difficult. In one patient, histopathologic analysis revealed that there were two tumors present, and the postcontrast images obtained in this patient showed two separate tumors (Fig 3). In one patient, two of the four postcontrast images obtained indicated that there were two separate tumors present, but in this case, the results of histopathologic analysis confirmed the presence of only one continuous tumor.
View larger version (175K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Sagittal double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in a 51-year-old woman shows two tumors in the rectum. Note the malignant stricture (arrowhead) caused by a Dukes B tumor in the upper part of the rectum. A polypoid tumor (arrow) that does not penetrate the tunica muscularis is seen ventrally in the lower part of the rectum. The results of histopathologic analysis showed Dukes A carcinoma.
|
|
Presence of Enlarged Lymph Nodes
At histopathologic analysis, two patients had enlarged (>1-cm) lymph nodes, and these were depicted on the postcontrast MR images (Fig 4d). In two other patients, the postcontrast MR imaging findings suggested enlarged lymph nodes. In one of these patients, the tumor growth was so extensive that the tumor could not be removed and only biopsy specimens were obtained; thus, the presence of lymph nodes could not be verified. In the other patient, histopathologic analysis results verified the presence of malignant lymph nodes, but these had a diameter of less than 1 cm. In four patients, histopathologic analysis results showed malignant infiltration into lymph nodes that had a diameter of less than 1 cm, and, retrospectively, these small lymph nodes could be identified on the postcontrast MR images.
View larger version (180K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. (a-c) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained (a) before and (b-c) after the administration of contrast material in a 65-year-old man. (a) Before contrast material administration, the tumor cannot be identified, but the presence of multiple lymph nodes (arrow) indicates the possibility of tumor. (b) After rectal contrast material administration, a tumor (short arrow) causes narrowing of the lumen. The superior and inferior boundaries of the tumor are indicated by the long arrows. Although this is not a large tumor, the narrowed lumen makes tumor delineation by means of transrectal US or MR imaging with an endorectal coil impossible. (c) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrowhead) can be seen in the rectal wall, and tumor growth (short arrow) through the tunica muscularis into the mesorectal tissue can be clearly seen. The superior and inferior boundaries of the tumor are indicated by the long arrows. (d) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, enlarged lymph nodes (arrows) are clearly seen in the perirectal tissue adjacent to the tumor (arrowhead).
|
|
View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. (a-c) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained (a) before and (b-c) after the administration of contrast material in a 65-year-old man. (a) Before contrast material administration, the tumor cannot be identified, but the presence of multiple lymph nodes (arrow) indicates the possibility of tumor. (b) After rectal contrast material administration, a tumor (short arrow) causes narrowing of the lumen. The superior and inferior boundaries of the tumor are indicated by the long arrows. Although this is not a large tumor, the narrowed lumen makes tumor delineation by means of transrectal US or MR imaging with an endorectal coil impossible. (c) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrowhead) can be seen in the rectal wall, and tumor growth (short arrow) through the tunica muscularis into the mesorectal tissue can be clearly seen. The superior and inferior boundaries of the tumor are indicated by the long arrows. (d) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, enlarged lymph nodes (arrows) are clearly seen in the perirectal tissue adjacent to the tumor (arrowhead).
|
|
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. (a-c) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained (a) before and (b-c) after the administration of contrast material in a 65-year-old man. (a) Before contrast material administration, the tumor cannot be identified, but the presence of multiple lymph nodes (arrow) indicates the possibility of tumor. (b) After rectal contrast material administration, a tumor (short arrow) causes narrowing of the lumen. The superior and inferior boundaries of the tumor are indicated by the long arrows. Although this is not a large tumor, the narrowed lumen makes tumor delineation by means of transrectal US or MR imaging with an endorectal coil impossible. (c) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrowhead) can be seen in the rectal wall, and tumor growth (short arrow) through the tunica muscularis into the mesorectal tissue can be clearly seen. The superior and inferior boundaries of the tumor are indicated by the long arrows. (d) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, enlarged lymph nodes (arrows) are clearly seen in the perirectal tissue adjacent to the tumor (arrowhead).
|
|
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4d. (a-c) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained (a) before and (b-c) after the administration of contrast material in a 65-year-old man. (a) Before contrast material administration, the tumor cannot be identified, but the presence of multiple lymph nodes (arrow) indicates the possibility of tumor. (b) After rectal contrast material administration, a tumor (short arrow) causes narrowing of the lumen. The superior and inferior boundaries of the tumor are indicated by the long arrows. Although this is not a large tumor, the narrowed lumen makes tumor delineation by means of transrectal US or MR imaging with an endorectal coil impossible. (c) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrowhead) can be seen in the rectal wall, and tumor growth (short arrow) through the tunica muscularis into the mesorectal tissue can be clearly seen. The superior and inferior boundaries of the tumor are indicated by the long arrows. (d) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, enlarged lymph nodes (arrows) are clearly seen in the perirectal tissue adjacent to the tumor (arrowhead).
|
|
Adverse Events and Artifacts
No adverse events were observed in this study. Substantial artifacts were not a problem with any of the sequences used. Metallic artifacts were observed in five patients. In two of these patients, metallic artifacts outside the colon were observed before contrast material administration, and they were not caused by the contrast material after its administration. Three patients had minor artifacts caused by the contrast material, but these did not disturb image interpretation to a major degree. One patient had peristaltic motion artifacts caused by the bowel, and another patient's movement caused the artifacts.
|
DISCUSSION
|
|---|
In a previous pilot study (1), our results showed that it is possible to stage rectal carcinomas correctly by using double contrast–enhanced MR imaging. In this larger study, this was confirmed, and the sensitivity for the detection of tumor stages worse than Dukes A was 100%. Single-contrast studies (ie, with ferristene-based contrast material only) were compared with precontrast studies and with double-contrast studies (ie, with ferristene-based contrast material plus gadodiamide) (Fig 4). The difference between precontrast and postcontrast images was immense, and the examinations performed without the administration of intraluminal contrast material were found to be of almost no value (Table 2). In this study, most (25 of 29) of the tumors found by using manual palpation, proctoscopy, and/or barium examinations were invisible on the MR images obtained prior to ferristene-based contrast material administration.
The reason for this improved diagnostic information is that the ferristene formulation expands the bowel lumen, stretches the rectal wall, and thus causes a signal void in the luminal space, which makes the tissue layers visible. Injected gadodiamide perfuses the highly vascular mucosal layers and thus causes increased signal intensity on T1-weighted images. The mucosal layers of the rectal wall appear as a sharp white line. The tunica muscularis appears as a dark band between the mucosa and the perirectal fat. This enables the extent of tumor growth to be evaluated, because the tumor tends to appear hypointense compared with the gadolinium-based contrast material–enhanced mucosa, and tumor extension to the mucosal layer appears as an interruption of this bright structure. Therefore, images obtained with injected gadodiamide have further improved tumor delineation compared with those obtained with only ferristene-based contrast material administration (Table 2).
In our study, the T2-weighted images obtained after the administration of ferristene-based contrast material gave less information than did the T1-weighted images obtained after the administration of ferristene-based contrast material plus gadodiamide. The MR imaging unit used in this study could not provide fast spin-echo images; thus, the T2-weighted images had a high signal-to-noise ratio and low resolution. Therefore, the resolution on T2-weighted images possibly could be improved with this technique; however, this needs to be evaluated in a new study. Furthermore, fat saturation could, theoretically, improve the conspicuity of the tumor following contrast material injection, but owing to the technical limitations of the imaging unit that we used, fat saturation could not be performed. In addition, fat saturation works less efficiently at the low field strength used in this study.
No adverse events or discomfort was observed in this study; this indicates that the two contrast materials used were well tolerated. No acute reactions to the rectal administration of ferristene-based contrast material or problems with the procedure were encountered, and the administration of the ferristene solution enema was straightforward and rapid: It took 3 minutes or less. A radiographer could conduct this procedure in a routine clinical setting. Taking into consideration that only the postcontrast images would be required for the clinical routine, the whole imaging procedure could be completed in less than 20 minutes.
Tumor staging is extremely important for the surgical planning and prognosis of individual patients (22–27). In Sweden, where preoperative irradiation of rectal carcinomas is performed, the stage is very important (16). Because the 5-year survival rate of patients with Dukes A carcinoma is so good—more than 95% (25)—these individuals would not benefit from preoperative irradiation, which might also cause side effects, whereas patients with the worst stages of cancer do benefit from irradiation and thus have improved survival (16). Because the 5-year survival rate of patients with Dukes B or Dukes C carcinoma is so much poorer than that of patients with Dukes A carcinoma and these patients need a totally different treatment, no distinction between Dukes B and C tumors was made in this study. Because the examinations were confined to the pelvis, Dukes D tumors—that is, those with distant metastases—could not be staged.
The patients in this study were consecutive, but the prevalence of Dukes A tumor was relatively high (34%) compared with the expected prevalence of 15% (20). The high prevalence of Dukes A tumor in our series underlined the diagnostic superiority of the double-contrast MR imaging technique, because it is much more difficult to correctly stage a small tumor than a big one.
Furthermore, Dukes A carcinomas often can be operated on by using the transrectal route (Fig 5), which makes it possible to avoid open surgery (13–15). At the other end of the scale are those patients who have very advanced tumor disease with overgrowth in adjacent organs, in whom radical surgery offers no hope for cure (23). In elderly and weak patients especially, unsuccessful surgery that could do more harm than good can be avoided, and the patient will perhaps benefit more from irradiation only. Because the tumor stage cannot be determined prior to surgical examination, expensive resources must be made available during surgery, and possibly presurgical radiation treatment, even though frequently these may turn out to be unnecessary.
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. (a, b) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained after the administration of (a) transrectal contrast material and (b) both transrectal and intravenous contrast material in a 38-year-old man. (a) After transrectal contrast material administration, a small tumor (arrowhead) can be seen dorsally in the rectum. The tunica muscularis (arrow) is not clearly seen. (b) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrow) can be clearly identified; the tumor does not extend into the tunica muscularis. The results of histopathologic analysis verified a Dukes A tumor. (c) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, the same tumor (arrow) is seen, without any signs of infiltration into the tunica muscularis. The patient was successfully operated on by using the transrectal route.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. (a, b) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained after the administration of (a) transrectal contrast material and (b) both transrectal and intravenous contrast material in a 38-year-old man. (a) After transrectal contrast material administration, a small tumor (arrowhead) can be seen dorsally in the rectum. The tunica muscularis (arrow) is not clearly seen. (b) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrow) can be clearly identified; the tumor does not extend into the tunica muscularis. The results of histopathologic analysis verified a Dukes A tumor. (c) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, the same tumor (arrow) is seen, without any signs of infiltration into the tunica muscularis. The patient was successfully operated on by using the transrectal route.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. (a, b) Sagittal T1-weighted spin-echo MR images (600/30, two signals acquired) obtained after the administration of (a) transrectal contrast material and (b) both transrectal and intravenous contrast material in a 38-year-old man. (a) After transrectal contrast material administration, a small tumor (arrowhead) can be seen dorsally in the rectum. The tunica muscularis (arrow) is not clearly seen. (b) After the administration of both transrectal and intravenous contrast material, the tunica muscularis (arrow) can be clearly identified; the tumor does not extend into the tunica muscularis. The results of histopathologic analysis verified a Dukes A tumor. (c) On the transverse double-contrast T1-weighted spin-echo MR image (600/30, two signals acquired) obtained in the same patient, the same tumor (arrow) is seen, without any signs of infiltration into the tunica muscularis. The patient was successfully operated on by using the transrectal route.
|
|
The staging of Dukes A versus Dukes B carcinoma with high sensitivity and specificity is not the only medically useful information offered by double-contrast MR imaging of rectal carcinoma. Multiplanar imaging can contribute further to surgical planning by giving a complete view of the pelvic regions surrounding the tumor. The contrast-enhanced MR imaging technique used in this study represents a possible one-step procedure for both regional detection of a suspected tumor and surgical planning.
Conventional colon examination, proctoscopy, and, more recently, flexible sigmoidoscopy are highly sensitive methods for the detection of rectal cancer. However, none of these methods can enable the staging of tumors preoperatively. Both transrectal US and MR imaging with an intrarectal coil can enable the staging of rectal carcinoma, but they have limitations (3,10). When the tumors are causing stricture of the lumen, it is not possible to use this method if the probe and the coil, respectively, cannot be advanced through the stricture and thus properly placed; this makes tumor delineation impossible. Furthermore, these methods, with their limited examination fields, fail to enable visualization of the whole pelvis and therefore often fail in showing the pathologic lymph nodes and overgrowth in adjacent organs. MR imaging with a transrectal coil can be improved by using a multicoil array, but at present, this is possible with only a few MR imaging units.
In conclusion, double-contrast (ferristene solution enema plus gadodiamide injection) MR imaging is efficient for staging of rectal carcinoma. In the present study, the sensitivity was 100%; specificity, 70%; and accuracy, 90% in distinguishing tumor stages worse than Dukes A. This examination technique also helps in the identification of inoperable tumors. The double-contrast MR imaging technique was superior to MR imaging with only the ferristene solution enema.
|
Footnotes
|
|---|
Author contributions: Guarantor of integrity of entire study, N.O.W.; study concepts and design, N.O.W., S.H.; definition of intellectual content, N.O.W., S.H.; literature research, N.O.W.; clinical studies, A.A.S., N.O.W., E.J.; data acquisition, N.O.W.; data analysis, N.O.W., S.H., S.M.; statistical analysis, D.T.K.; manuscript preparation and editing, N.O.W., S.H.; manuscript review, E.J., S.M., N.O.W., S.H.
|
References
|
|---|
-
Wallengren NO, Holtås S, Andrén-Sandberg Å. Preoperative staging of rectal carcinoma using double-contrast MR imaging: technical aspects and early clinical experiences. Acta Radiol 1996; 36:791-798.
-
Schnall MT, Connick T, Hayes CE, Lemkinski RY, Kressel HY. MR imaging of the pelvis with an endorectal: external multicoil array. J Magn Reson Imaging 1992; 2:229-232.[Medline]
-
Okizuka H, Sugimura K, Yshida T. Preoperative local staging of rectal carcinoma with MR imaging and a rectal balloon. J Magn Reson Imaging 1993; 3:329-335.[Medline]
-
Joosten FP, Jansen IP, Joosten HJ, Rosenbusch G. Staging of rectal carcinoma using MR double surface coil, MR and rectal coil and intrarectal ultrasound: correlation with histopathological findings. J Comput Assist Tomogr 1995; 19:752-758.[Medline]
-
Guinet C, Buy JN, Ghossain MA, et al. Comparison of magnetic resonance imaging and computed tomography in the preoperative staging of rectal cancer. Arch Surg 1990; 125:385-388.[Abstract/Free Full Text]
-
Kusunoki M, Yanagi H, Kamikonya N, Hishikawa Y, Shoji Y, Yamamura T. Preoperative detection of local extension of carcinoma of the rectum using magnetic resonance imaging. J Am Coll Surg 1994; 179:653-656.[Medline]
-
de Lange EE, Fechner RE, Edge SP, Spaulding CA. Preoperative staging of rectal carcinoma with MR imaging: surgical and histopathological correlation. Radiology 1990; 176:623-628.[Abstract/Free Full Text]
-
McNicholas MN, Joyce WP, Dolan J, Gibney RG, MacErlaine TP, Hyland J. Magnetic resonance imaging of rectal carcinoma: a prospective study. Br J Surg 1994; 81:911-914.[Medline]
-
Thaler V, Watzka S, Martin F, et al. Preoperative staging of rectal cancer by endoluminal ultrasound versus magnetic resonance imaging: preliminary results of a prospective comparative study. Dis Colon Rectum 1994; 37:1089-1093.
-
Rafaelsen SR, Kronborg O, Fenger C. Digital rectal examination transrectal ultrasonography in staging of rectal cancer: a prospective, blind study. Acta Radiol 1994; 35:300-304.[Medline]
-
Fedyaev EB, Volkova EA, Kuznetsova EE. Transrectal and transvaginal ultrasonography in the preoperative staging of rectal carcinoma. Eur J Radiol 1995; 20:35-38.[Medline]
-
Thoeni RF. Colorectal cancer: cross-sectional imaging for staging of primary tumor in detection of local recurrence. AJR Am J Roentgenol 1991; 156:909.[Abstract/Free Full Text]
-
Lock FR, Ritchie JK, Hawley BR. Reappraisal of radical local excision for carcinoma of the rectum. Br J Surg 1993; 80:928-929.[Medline]
-
Rouanet P, Saint-Aubert B, Fabre JM, et al. Conservative treatment for low rectal carcinoma by local excision with or without radiotherapy. Br J Surg 1993; 80:1752-1756.
-
Tanaka S, Yokota T, Saito T, Okamoto S, Oguro Y, Yoshida S. Clinicopathologic features of early rectal carcinoma and indications for endoscopic treatment. Dis Colon Rectum 1995; 38:959-963.[Medline]
-
Cedermark B, Johansson H, Rutqvist ME, Wilking N. The Stockholm I trial of preoperative short term radiotherapy in operable rectal carcinoma: prospective randomized trial—Stockholm Colorectal Cancer Study Group. Cancer 1995; 75:2269-2275.[Medline]
-
Bach-Gansmo T, Gundersen HG, Klaveness J, et al. Superparamagnetic oral contrast agents for abdominal MRI. In: Rinck PA, eds. Contrast and contrast agents in magnetic resonance imaging. Mons, Belgium: European Workshop of Magnetic Resonance in Medicine, 1989; 94.
-
Bach-Gansmo T, Hemmingsson A, Lönnemark M, et al. Safety and tolerance of oral magnetic particles for abdominal magnetic resonance imaging: phase I—clinical study. Clin Trials J 1989; 26:216.
-
Holtås S, Wallengren NO, Ericsson A, Bach-Gansmo T. Signal alterations, artifacts and image distortion induced by a superparamagnetic contrast medium: a phantom study in a 0.3 tesla MR system. Acta Radiol 1990; 31:213.[Medline]
-
Goligher JC, ed. Surgery of the anus, rectum and colon London, England: Bailliere Tindall, 1977; 512-513.
-
Lentner C, ed. Geigy scientific tables West Caldwell, NJ: Ciba-Geigy, 1982; 89-102, 221.
-
Fabre JM, Rouanet P, Ele M, et al. Colorectal carcinoma in patients aged 25 years and more: factors influencing short and long-term operative mortality. Int Surg 1993; 78:200-203.[Medline]
-
Jessup JM, Bleday R, Busse P, Steele G. Conservative management of rectal carcinoma: efficacy of a multimodality approach. Semin Surg Oncol 1993; 9:39-45.[Medline]
-
Hermanek P, Wiebelt H, Staimmer D, Ryedl S. Prognostic factors of rectal carcinoma: experience of the German multicenter study SGCRC—German studies group colo-rectal carcinoma. Tumori 1995; 81(suppl 3):60-64.[Medline]
-
Hricak H, Carrington BM. MRI of the pelvis Norwalk, Conn: Appleton & Lange, 1991; 470.
-
Cawthorn SJ, Parums DW, Gibbs NM, et al. Extent of mesorectal spread and involvement of lateral resection margin as prognostic factors after surgery for rectal cancer. Lancet 1990; 335:1055-1059.[Medline]
-
MacFarlane JK, Ryall RD, Heald RJ. Mesorectal excision for rectal cancer. Lancet 1993; 341:457-460.[Medline]
This article has been cited by other articles:
|
|
|
|
 
S. H. Kim, J. M. Lee, M. W. Lee, G. H. Kim, J. K. Han, and B. I. Choi
Sonography Transmission Gel as Endorectal Contrast Agent for Tumor Visualization in Rectal Cancer
Am. J. Roentgenol.,
July 1, 2008;
191(1):
186 - 189.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
X. M. Zhang, H. L. Zhang, D. Yu, Y. Dai, D. Bi, M. R. Prince, and C. Li
3-T MRI of Rectal Carcinoma: Preoperative Diagnosis, Staging, and Planning of Sphincter-Sparing Surgery
Am. J. Roentgenol.,
May 1, 2008;
190(5):
1271 - 1278.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Nakai, K. Togashi, K. Kosaka, A. Kido, M. Kataoka, T. Koyama, and S. Fujii
Do Anticholinergic Agents Suppress Uterine Peristalsis and Sporadic Myometrial Contractions at Cine MR Imaging?
Radiology,
December 19, 2007;
(2007)
2461062091.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
K. P. Dhamanaskar, W. Thurston, and S. R. Wilson
Transvaginal sonography as an adjunct to endorectal sonography in the staging of rectal cancer in women.
Am. J. Roentgenol.,
July 1, 2006;
187(1):
90 - 98.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. F. A. Vliegen, G. L. Beets, M. F. von Meyenfeldt, A. G. H. Kessels, E. E. M. T. Lemaire, J. M. A. van Engelshoven, and R. G. H. Beets-Tan
Rectal Cancer: MR Imaging in Local Staging--Is Gadolinium-based Contrast Material Helpful?
Radiology,
January 1, 2005;
234(1):
179 - 188.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M.-J. Kim, J. S. Lim, Y. T. Oh, J. H. Kim, J.-J. Chung, S. H. Joo, N. K. Kim, K. Y. Lee, W. H. Kim, and K. W. Kim
Preoperative MRI of Rectal Cancer With and Without Rectal Water Filling: An Intraindividual Comparison
Am. J. Roentgenol.,
June 1, 2004;
182(6):
1469 - 1476.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. H. Fuchsjager, A. G. Maier, W. Schima, E. Zebedin, F. Herbst, M. Mittlbock, F. Wrba, and G. L. Lechner
Comparison of Transrectal Sonography and Double-Contrast MR Imaging When Staging Rectal Cancer
Am. J. Roentgenol.,
August 1, 2003;
181(2):
421 - 427.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T H Saunders, H K Mendes Ribeiro, and F V Gleeson
New techniques for imaging colorectal cancer: the use of MRI, PET and radioimmunoscintigraphy for primary staging and follow-up
Br. Med. Bull.,
December 1, 2002;
64(1):
81 - 99.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. S. K. Goh, J. P. N. Goh, and G. K. L. Wansaicheong
Methylcellulose as a Rectal Contrast Agent for MR Imaging of Rectal Carcinoma
Am. J. Roentgenol.,
May 1, 2002;
178(5):
1145 - 1146.
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
