Technetium-99m scintigraphy: More accurate assessment of ulcerative colitis with exametazime-labelled leucocytes than with antigranulocyte antibodies

To compare two technetium-99m scintigraphic techniques - Tc-99m-labelled antibodies against granulocyte non-specific cross-reacting antigen-95 and Tc-99m-exametazime labelled leucocytes in ulcerative colitis - 23 consecutive patients undergoing colonoscopy were investigated in a prospective and randomized study, In each patient the two scans and colonoscopy and biopsy were performed within 10 days. Scans, endoscopy and histology were independently graded for degree of inflammation in eight different colorectal segments for each patient. Active inflammation in one or more segments was present on endoscopy in 22 patients and on histology in 17 patients. Twenty-two patients had increased uptake on the antibody scan and 21 patients on the exametazime scan. Twelve patients showed the same disease extent with both scan methods (total colitis, n=10; distal colitis, n=2), Compared with endoscopy, sensitivity for inflammation in individual segments was 0.51 for antibody scan and 0.87 for exametazime scan; specificity was 0.67 and 0.55, respectively. The predictive value for presence of inflammation was 0.66 for antibody scan and 0.72 for exametazime; the predictive value for absence of inflammation was 0.52 and 0.77, respectively. Segmental scan uptake of endoscopically or histologically visualized inflammation was consistently lower for antigranulocyte antibodies than for exametazime. It is concluded that in patients with active ulcerative colitis, inflammation is better visualized with Tc-99m-exametazime labelled leucocytes than with Tc-99m-labelled antigranulocyte antibodies. The antibody technique offers the advantage of in vivo labelling, but is less reliable than the exametazime method for imaging of colonic inflammation.