Magnetic resonance imaging of the functional anatomy of the inferior oblique muscle in superior oblique palsy.

PURPOSE To study size and contractility of the normal inferior oblique (IO) muscle using high-resolution magnetic resonance imaging (MRI) and to evaluate abnormalities of the superior oblique (SO) and IO muscles in chronic SO palsy. DESIGN Prospective, case control study. PARTICIPANTS Thirteen patients with SO palsy and 17 orthotropic subjects. METHODS High-resolution, surface coil MRI was used to obtain sets of contiguous, 2-mm thick coronal and sagittal images repeated in multiple gaze directions. Digital image analysis was used to measure IO and SO muscle cross-sectional areas for evaluation of size and contractility. Diagnosis of SO palsy in one bilateral and 12 unilateral cases was based on subnormal contractility and SO size less than the normal 95% confidence limit. Ipsilesional and contralesional oblique muscles were compared with controls and correlated with clinical characteristics. RESULTS In all subjects, anterior movement and contractile thickening of the IO were observed in supraduction, with posterior movement and relaxational thinning in infraduction. The mean (+/- standard deviation) cross-sectional area of 15 normal control IO muscles was 13.4 +/- 3.9 mm(2), with mean contractile increase from infraduction to supraduction of 5.7 +/- 2.6 mm(2). Subjects with SO palsy had incomitant hypertropia with a wide range of overelevation and underelevation in adduction (i.e., upshoot, downshoot). SO atrophy correlated with underdepression in adduction (P < 0.0001). Contralesional SO cross-section was slightly greater than normal (P = 0.004). The IO cross-section ipsilesional and contralesional to SO palsy did not, however, differ significantly from normal and did not correlate with elevation in adduction (P > 0.2). CONCLUSIONS Quantitative morphometry by MRI can demonstrate IO size and contractility. Even in cases of unequivocal SO palsy associated with ipsilesional SO atrophy and deficient contractility, the degree of elevation in adduction was not correlated with IO size. This finding suggests that the associated overelevation in adduction, commonly termed "inferior oblique overaction," actually arises from some other mechanism than IO hypertrophy or excess contractility. Revision of clinical terminology seems warranted.