Previous studies have found a widely variable prevalence of dysplasia and cancer in colonic strictures in patients with ulcerative colitis. Consequently, therapeutic recommendations are conflicting. To better assess the prevalence, we reviewed the clinical and pathological findings in all 27 patients with ulcerative colitis complicated by stricture who were entered into our Inflammatory Bowel Disease Registry. A true stricture was defined as a persistent localized narrowing of the colon found on air-contrast barium enema or on colonoscopy. Upon careful review, 12 of 27 patients were found to have transient colonic spasm, not a stricture, and were excluded. The remaining 15 patients with true strictures represented 3.2% of all ulcerative colitis patients in the registry. Strictures were identified at 13.3 +/- 9.9 years following the diagnosis of ulcerative colitis. Eleven patients had multiple strictures that were principally located in the left colon. Of the 15 patients, 11 had dysplasia and two had cancer found on colonoscopic biopsy. Ultimately, six patients had carcinoma found at colonoscopy or colectomy (three modified Dukes' stage A, one stage B, and two stage D). All cancers were at the site of a stricture. These findings indicate that a true colonic stricture in ulcerative colitis is frequently associated with dysplasia and cancer, which can be diagnosed with colonoscopic biopsy. A stricture should be considered a strong risk factor for cancer, requiring intensive colonoscopic surveillance. If dysplasia is discovered, or if the stricture cannot be adequately biopsied, consideration should be given to total colectomy.

Barium examinations of the large and small bowel were analyzed in six of seven patients who had adenocarcinoma in areas of the intestine affected with Crohn disease; radiographic changes were correlated with clinical, surgical, and pathologic findings. Radiographic examinations were available in five of these patients at the time of diagnosis of tumor. Two of the five patients demonstrated classic radiographic changes associated with carcinoma. In the other three cases, the radiographic changes were atypical for carcinoma and demonstrated progression of disease over time to include more portions of the bowel and presence of fistulas, strictures, and obstruction. The most frequent clinical presentation of adenocarcinoma in these patients was a recrudescence of symptoms after a long quiescent period. In patients with long-standing Crohn disease plus these clinical features and the above radiographic findings, the diagnosis of a coexisting carcinoma should be considered.

We conducted a comprehensive study on the application of ultra-high-strip-density (UHSD) grids to mammography for the improvement of image contrast. These UHSD grids have strip densities of 70 to 100 lines/cm. After investigating the performance of mammographic grids with various design parameters through Monte Carlo simulation studies, we made prototypes of UHSD grids having lead strip thicknesses of 20 to 37 micron, aluminum interspacer thicknesses of 80 to 120 micron, and grid ratios of 2 to 6. The UHSD grids can be placed inside the cassette without a Bucky tray and thus will not increase geometric unsharpness. The measured physical characteristics and the phantom images obtained confirmed our findings in the simulation studies. A 40 to 90% increase in contrast was achieved at Bucky factors of 2 to 3.5; these results are comparable with those from conventional antiscatter grid techniques in mammography. The potential usefulness of the UHSD grids was demonstrated in a preliminary clinical comparison of mammograms.

The imaging performance of new high-strip-density (HSD) grids having 57 lines/cm was compared with that of conventional low-strip-density (LSD) grids having 33 or 40 lines/cm. The unique advantage of HSD grids is that, under most standard radiographic conditions, the grid lines are not noticeable on the final image, even if the grid is stationary. This is due to the combined effect of the high fundamental spatial frequency of HSD grids, the modulation transfer function of screen-film systems and of the human visual system, and scattered radiation. Monte Carlo simulation studies, phantom images, and clinical evaluation indicate that HSD grids can provide contrast improvement factors and Bucky factors that are comparable to or slightly better than those obtained with LSD grids. Therefore, it may now be possible to eliminate moving Bucky trays from radiographic tables and fluoroscopic devices.

No abstract