Pulmonary high-resolution computed tomography versus gallium scintigraphy: diagnostic utility in the diagnosis of patients with AIDS who have chest symptoms and normal or equivocal chest radiographs.

Fifty-six consecutive symptomatic patients with AIDS referred for gallium scintigraphy were prospectively studied with chest high-resolution computed tomography (HRCT). Results of gallium and HRCT were correlated with findings of bronchoscopy or clinical follow-up for 1 month from time of discharge. Twenty-two patients were eventually diagnosed with at least one of the following: Pneumocystis carinii, cytomegalovirus, Mycobacterium avium complex, bacteria, Kaposi's sarcoma, or lymphocytic interstitial pneumonitis. HRCT was more sensitive (82%) and more specific (91%) than gallium (59% and 75%, respectively). HRCT yielded higher positive predictive values (86%) and negative predictive values (88%) than did gallium (62% and 73%, respectively). HRCT was more helpful in guiding the method of biopsy and directing the brochoscopist to the diseased lung segment that would maximize diagnostic yield.

Pleurocutaneous fistula as a complication of oleothorax. CT findings in three patients.

The purpose of this study was to highlight pleurocutaneous fistula, documented on computed tomography (CT) scans, as a complication of oleothorax. CT images of the chest were evaluated retrospectively in three patients with pleurocutaneous fistula complicating oleothorax. The images documented oleothorax rupture in all three, with only one patient (case 2) showing concomitant surrounding chest wall and axillary inflammatory reaction. Surgery performed on one patient (case 1) confirmed the CT findings. While CT is invaluable in the demonstration of complications of oleothorax, to our knowledge CT depiction of oleothorax has not been reported in the literature. This complication may be more common than described in earlier literature, with the advent of cross-sectional imaging. In the right clinical setting, CT findings of inflammation or edema surrounding the oleothorax rupture may warrant surgical intervention.

Thoracic aortic trauma: role of dynamic CT.

Two protocols were developed to study the efficacy of both chest radiography and dynamic computed tomography (CT) of the thorax in screening for traumatic aortic injury. Both protocols demonstrated a definite role for dynamic CT in excluding aortic injury in patients with normal initial chest radiographs. The authors studied 164 patients, and preliminary results indicated a high reliability for this method. Their algorithm, with 6-month follow-up, yielded both 100% sensitivity and a 100% negative predictive value. Unnecessary aortography, with its associated morbidity and cost, was decreased by 73%.

Benign lymphoepithelial parotid tumors in AIDS patients: CT and MR findings in nine cases.

Benign lymphoepithelial parotid neoplasm with associated cervical adenopathy is a newly described head and neck manifestation of HIV infection, characterized previously as purely cystic masses on both CT and MR evaluations. The CT and MR findings in nine cases of surgically proved benign lymphoepithelial parotid lesions are presented. Five (56%) of the nine cases demonstrated masses that appeared other than cystic, reflecting a greater variety of radiologic appearances than previously reported. The majority of these cystic masses contained solid components and therefore could not be labeled as cysts; instead, they were considered to be benign lymphoepithelial parotid lesions. For patients with parotid gland enlargement, the lack of a characteristic CT/MR appearance of these lesions necessitates an evaluation of clinical history and a determination of whether cervical adenopathy is present in order to determine their underlying origin.

Contrast media adversely affect oxyhemoglobin dissociation.

Effects of ionic (Hypaque-76) and nonionic (Isovue-370 and Omnipaque-350) contrast media on oxyhemoglobin dissociation of normal human red blood cells were evaluated. In series 1, 4-mL venous blood samples were obtained from 15 normal human volunteers. One blood sample served as control, and 1 mL of either of the three contrast media was added in vitro to the other 4-mL blood samples. P50 values were estimated from the linear portion of the oxyhemoglobin dissociation curve obtained by tonometry. Determinations of P50 were performed at either pH 7.4 or 7.2. At pH 7.4, P50 in the absence of contrast media was 26.3 +/- 0.4 mm Hg (mean +/- SEM). The contrast media caused comparable decreases in P50 from this value (Hypaque-76, 20.0 +/- 0.5 mm Hg; Omnipaque-350, 21.6 +/- 0.4 mm Hg; Isovue-370, 20.7 +/- 0.4 mm Hg). Reducing pH to 7.2 in the absence of contrast media increased P50 to 33.3 +/- 1.0 mm Hg, evidence of the Bohr effect. The presence of contrast media either completely abolished (Hypaque-76 and Omnipaque-350) or markedly attenuated (Isovue-370) this effect. In series 2 (five patients), blood samples were withdrawn from the external iliac artery during injection of Isovue-370 (60-78 mL) into the proximal abdominal aorta to evaluate peripheral vascular disease. Measurement of P50 of these samples yielded findings consistent with those of series 1. The present findings demonstrate that both ionic and nonionic contrast media increase the affinity of hemoglobin for oxygen and, therefore, that they may inhibit oxygen delivery to body tissues.

Unsuspected upper cervical spine fractures associated with significant head trauma: role of CT.

Several recent studies report the sensitivity of computed tomography (CT) to be far greater than that of traditional plain film radiographic studies for evaluation of cervical spine fractures and spinal cord pathology. Nevertheless, plain films continue to be the standard screening examination. CT is used only if fractures are demonstrated or suspected on plain film survey. Recently, three patients with significant head and neck trauma (all three patients had intracranial hemorrhage) had cervical spine evaluation by computed tomography and standard plain film views. CT demonstrated significant C1-C2 fractures, while plain films were completely normal in all three cases. Prospectively studying the next 50 patients with significant head trauma, we added a few more slices to the routine head scan protocol to cover the first three cervical vertebrae. This added very little time or cost to the procedure. The additional CT images demonstrated four upper cervical fractures that could not be seen on plain films, even in retrospect. Our findings suggest that routine inclusion of the upper cervical spine with head CT is appropriate in the evaluation of patients with significant head trauma as defined by intracranial hemorrhage or skull fracture.