The clinical and high-resolution computed tomographic (CT) findings in 71 patients (142 orbits) with Graves orbitopathy and 20 healthy patients (40 orbits) were retrospectively reviewed. The orbits with orbitopathy were subgrouped at clinical examination into those with (n = 18) and those without (n = 124) optic neuropathy. Mean extraocular muscle diameters and the calculated muscle diameter index were significantly increased in all orbits with ophthalmopathy, particularly in those with optic neuropathy. Graves orbitopathy affected the superior muscle group (63.4%) more than the medial (61.3%) or inferior (57%) recti. The most common pattern of muscle involvement involved all five measured extraocular muscles. Solitary muscle involvement most frequently involved the superior muscle group (6.3%). Significant enlargements of the retrobulbar optic nerve sheath and superior ophthalmic vein were noted only in orbits with optic neuropathy. Anterior displacement of the lacrimal gland at CT correlated with clinical palpability and occurred more frequently in patients with optic neuropathy. Severe apical crowding was the most sensitive indication of optic neuropathy at CT.

No abstract

A patient had massive hematemesis due to a fistula between an unsuspected aberrant right subclavian artery and the esophagus, with the fistula caused by prolonged esophageal intubation. Lack of awareness of this complication delayed the correct diagnosis, which was eventually made at a second angiographic study. With knowledge of the fistula, this potentially fatal situation can be promptly diagnosed by endoscopy and emergency angiography. Transcatheter embolization is suggested as a potential alternative to surgery in the treatment of the aberrant right subclavian artery--esophageal fistula.

The evaluation of loose bodies in the elbow is usually done by means of clinical examination, radiography, and postarthrographic computed tomography (CT). The authors review their experience with magnetic resonance (MR) imaging in place of post-arthrographic CT for the evaluation of loose bodies in the elbow. The prospective interpretation of MR studies of the elbow in 20 patients was compared with arthroscopic findings. All elbows were imaged in multiple planes with thin sections, surface coils, and combinations of T1, T2, and proton-density weighting. The sensitivity for showing loose bodies with MR imaging was 100%, and the specificity was 67%. Because this was a nonblinded study, the results are biased and caution must be used when extrapolating these results to the general population. In this limited experience, MR imaging has reliably shown loose bodies in the elbow, and in the authors' institutions has replaced postarthrographic CT for that purpose.