Specificity and sensitivity of distinctive chest radiographic and/or 67Ga images in the noninvasive diagnosis of sarcoidosis.
Academic Article
Overview
abstract
An analysis of gallium-67 (67Ga) uptake in the lacrimal and salivary glands and intrathoracic lymph nodes was made in 162 patients with sarcoidosis, consisting of a large number with chronic "fibrotic" disease, and 167 HIV-positive patients (most of whom have/had AIDS). This study was designed to further assess the diagnostic sensitivity and to fully evaluate the diagnostic specificity of chest radiographic and/or 67Ga uptake findings found to be characteristic of sarcoidosis. A lambda 67Ga uptake image or a panda 67Ga uptake image with associated bilateral, symmetrical hilar lymphadenopathy (BSHL) or bilateral, symmetrical parenchymal infiltration indicative of pulmonary fibrosis on chest radiograph (BSIF) was commonly present in sarcoidosis. These distinctive 67Ga uptake images were frequently observed in patients with normal chest radiographs (stage 0 [33 percent]), as well as in patients with BSHL on chest radiograph whether the disease was in an "early" stage, ie, stages I (74 percent) and II (90 percent) or a chronic "fibrotic" stage, ie, IVa (71 percent). A panda 67Ga uptake image was observed in 8 percent of HIV-positive patients; however, a lambda 67Ga uptake image alone or any of the other chest radiographic and/or 67Ga uptake patterns distinctive for sarcoidosis were not observed in any of 167 HIV-positive patients. We conclude that (1) a lambda 67Ga thoracic image (usually associated with a panda 67Ga uptake image) or a panda 67Ga uptake image together with BSHL or BSIF on chest radiograph represent distinctive patterns that are highly specific and sensitive for the noninvasive diagnosis of the majority of sarcoidosis patients, and (2) the finding of a panda 67Ga uptake image, not associated with BSHL or BSIF on chest radiograph, should suggest, in addition to a limited number of readily diagnosable disorders, the presence of a positive HIV status.