|
|
||||||||
Original report |
1
Department of Diagnostic Radiology, Asan Medical Center, University of Ulsan
College of Medicine, 388-1 Poongnap Dong Songpa Ku, Seoul, 138-040,
Korea.
2
Department of Pathology, Asan Medical Center, Seoul, 138-040 Korea.
3
Department of Surgery, Asan Medical Center, Seoul, 138-040 Korea.
Received May 14, 1999;
accepted after revision July 2, 1999.
Address correspondence to H. K. Ha.
Abstract
|
|
|---|
CONCLUSION. Metastatic linitis plastica to the rectum should be considered when CT shows a long segment of circumferential rectal wall thickening, especially in patients with peritoneal carcinomatosis from gastric cancer. In such patients, CT helps avoid unnecessary extensive surgery.
|
|
|---|
When the rectal wall is circumferentially involved with metastatic linitis plastica, the rectal lesion might be mistaken for primary rectal cancer, other inflammatory or ischemic processes, or changes after radiation therapy. Misdiagnosis may occur even with a knowledge of the patient's history of primary tumor or prior surgery for tumor.
According to researchers [1, 2, 6], metastatic linitis plastica to the rectum is frequently associated with peritoneal carcinomatosis. However, the incidence and routes for this type of rectal involvement are not well understood. Moreover, most reports of such patients describe radiologic findings on double-contrast barium enema results [7, 8] but, to our knowledge, no analysis of the CT features of this condition. We examined the CT features of metastatic linitis plastica to the rectum in 22 patients.
|
|
|---|
CT was performed using a GE 9800 Quick System (General Electric Medical Systems, Milwaukee, WI) and Somatom Plus-S, Plus-4, and Plus-40 scanners (Siemens, Erlangen, Germany) with 8- or 10-mm slice thickness at 8- or 10-mm intervals from the diaphragm to the pubis. Approximately 600-900 ml of oral contrast material (2% barium sulfate suspension; E-Z-CAT, E-Z-EM, Westbury, NY) was given 1 hr before scanning. Approximately 100-120 ml of IV iopamidol (Iopamiro 300; Bracco, Milan, Italy) or iopromide (Ultravist; Schering, Berlin, Germany) was given as a bolus (rate, 3.0 ml/sec) to 15 patients. Scanning was started 40-60 sec after IV infusion with a scanning time of 0.8-2.0 sec and an interscan delay of 1.8-3.5 sec. A traditional bolus rapid-drip infusion technique was used for the remaining seven patients. Rectal contrast material was not administered to all patients.
CT images were analyzed for the length and thickness of the rectum, patterns of rectal wall thickening (even or uneven), contrast enhancement (homogeneous, heterogeneous, or target sign), degree of infiltration in the perirectal fat plane (not visible; grade I, confined to the perirectal space but not reaching the perirectal fascia; grade II, beyond the perirectal fascia but not reaching the pelvic side wall; and grade III, reaching the pelvic side wall), and changes in other abdominal organs and sites. CT images were independently reviewed by two radiologists. If interpretations differed, consensus findings were used for a final decision.
|
|
|---|
Thirteen of the 22 patients underwent proctosigmoidoscopic examination to confirm rectal abnormality. Biopsy specimens yielded tumor cells in seven patients. In six patients, biopsy indicated chronic proctitis without tumor cells. Surgery for rectal lesions was attempted in 14 of 22 patients: low anterior resection in one, palliative colostomy in eight, palliative ileostomy in one, right hemicolectomy with ileostomy in one, and exploratory laparotomy in three. In 22 patients, the tumor cells obtained were identical to those found during histopathologic examination.
Table 1 summarizes the CT features of 22 patients with linitis plastica to the rectum. On CT, rectal wall thickening (range, 1.0-2.2 cm; mean, 1.6 cm) appeared to extend downward to the lower rectum nearly to the level of the anal verge in 19 patients (86%) (Fig. 1A, 1B). In 12 patients (55%), rectal wall thickening appeared to extend upward to the sigmoid colon. In six patients (27%), the thickened rectal wall showed three zones (target sign): a hyperattenuated inner zone, a hyperattenuated outer zone, and a hypoattenuated middle zone (Figs. 1A, 1B and 2). Of the 16 patients without the target sign, three showed marked contrast enhancement in the rectum.
|
|
|
|
Table 2 summarizes changes in other abdominal organs and sites. Masses were present in the pelvic cavity of 11 patients (50%) (Figs. 3A, 3B, 3C and 4). Except for two patients who had a recurrent ovarian or bladder tumor (10.0 and 10.5 cm, respectively), the masses in the remaining nine patients ranged from 1.8 to 5.0 cm in diameter (mean, 3.1 cm). In these patients, the masses were directly contiguous with the involved rectum. The bladder wall was thickened in five patients (23%) (Fig. 1A), and cystoscopic confirmation of tumor infiltration was made in two patients.
|
|
|
|
|
In nine (41%) of the 22 patients, bowel wall thickening similar to that of the rectum was seen in the gastrointestinal tract; a total of 19 intestinal segments were involved including transverse colon in seven patients, ascending colon in four, terminal ileum in three, descending colon in two, proximal ileum in two, and jejunum in one. Intestinal obstruction developed in 11 patients (50%) and included the rectum in five patients, sigmoid colon in one, transverse colon in one, ascending colon in one, ileum in one, and jejunum in two.
Resected surgical specimens of the rectum were available for five patients. On histopathologic examination, the mucosa was focally involved and eroded in most instances and the submucosa or subserosa was thickened by reactive fibrosis and tumor cell infiltration (Fig. 3C). The muscularis propria layer was thickened to a lesser degree, probably because of tumor infiltration and mild hypertrophic change of the muscle fiber; in this layer, reactive fibrosis was unidentified. Histopathologic examination showed similar findings for all surgical specimens. For patients with the target sign on rectal images, CT-histopathologic correlation was impossible because these patients did not undergo surgical resection.
|
|
|---|
In our study, the most common CT feature in patients with metastatic linitis plastica to the rectum was concentric bowel wall thickening in a long segment (>10 cm). However, because we could not obtain biopsy specimens, we might have overestimated the extent of segment involvement in which ischemic colitis coexisted proximal to rectal cancer [10]. Furthermore, other conditions such as fluid overload, heart failure, or postoperative state may contribute to rectal wall thickening. Additionally, the same finding can be seen in neoplastic, inflammatory, and vascular disorders; therefore, a pattern of bowel wall involvement may be nonspecific for diagnosis. The perirectal changes in patients with linitis plastica did not differ from those of other diseases. Rectal wall thickening extended downward to the lower rectum nearly to the level of the anal verge in 19 (86%) of our 22 patients. As Fernet et al. [1] speculated, this occurrence probably results from the milking force associated with peristalsis and antiperistalsis.
An important pathologic characteristic of linitis plastica is the exuberant desmoplastic response that the tumor cells or their products elicit in the stroma, especially in the submucosa and subserosa. In reviewing the literature [1, 2, 11], we found similar histopathologic findings. The target sign seen on CT reflected these pathologic findings. Although the target sign is an important indicator of benign gastrointestinal tract diseases [12], our study shows that this sign can be seen in malignant conditions as well. Moreover, the hypoattenuated middle zone of the target sign represents the muscularis propria. This result differs from the usual understanding of target signs in that hypoattenuated zone (submucosa).
Many conditions appear on CT with a long length of rectal wall thickening; these include primary linitis plastica without primary tumors at other sites, primary rectal cancer with proximal obstructing colitis [10], ulcerative colitis, Crohn's disease, pseudomembranous and ischemic colitis, and endometriosis. Use of only the rectal wall thickening pattern in diagnosis might cause difficulties in differentiating metastatic linitis plastica from other conditions. Therefore, knowledge of a patient's history of primary cancer at other sites and the ancillary findings in abdominal organs may help to suggest the possibility of metastatic linitis plastica to the rectum. In conclusion, metastatic linitis plastica to the rectum should be considered when CT shows a long segment of circumferential rectal wall thickening, especially for patients with peritoneal carcinomatosis from gastric cancer. In such patients, CT helps avoid unnecessary extensive surgery.
Acknowledgments
We thank Bonnie Hami for her editorial assistance in preparing this
manuscript.
|
|
|---|
This article has been cited by other articles:
![]() |
M. J. Gollub, M. B. Schwartz, and J. Shia Scirrhous Metastases to the Gastrointestinal Tract at CT: The Malignant Target Sign Am. J. Roentgenol., April 1, 2009; 192(4): 936 - 940. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Rudralingam, M. J. Dobson, M. Pitt, D. J. Stewart, A. Hearn, and S. Susnerwala MR Imaging of Linitis Plastica of the Rectum Am. J. Roentgenol., August 1, 2003; 181(2): 428 - 430. [Full Text] [PDF] |
||||
![]() |
K. W. Kim, B. I. Choi, J. K. Han, T. K. Kim, A. Y. Kim, H. J. Lee, Y. H. Kim, J.-I. Choi, K.-H. Do, H. C. Kim, et al. Postoperative Anatomic and Pathologic Findings at CT Following Gastrectomy RadioGraphics, March 1, 2002; 22(2): 323 - 336. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. F. Keogh, J. A. Brown, and P. T. Phang Linitis Plastica of the Rectum: Utility of Transrectal Ultrasonography J. Ultrasound Med., January 1, 2002; 21(1): 103 - 106. [Full Text] [PDF] |
||||
![]() |
A. Y. Kim, H. K. Ha, B. K. Seo, E.-S. You, K. S. Cho, P. N. Kim, M.-G. Lee, H. Y. Jeong, S. K. Yang, and Y. I. Min CT of Patients with Right-Sided Colon Cancer and Distal Ileal Thickening Am. J. Roentgenol., November 1, 2000; 175(5): 1439 - 1444. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |