How sensitive is the three-step test in the diagnosis of superior oblique palsy?

Abstract

The Parks-Bielschowsky three-step test is the classical cornerstone of cyclovertical strabismus. We evaluated sensitivity of the 3-step test in clinical diagnosis of SO palsy in patients with unequivocal magnetic resonance imaging (MRI) evidence of SO atrophy. 51 patients were selected from a prospective MRI study of strabismus because they exhibited significant SO atrophy. Detailed ocular motility data, including three-step testing, were evaluated to determine sensitivity of single and combined clinical findings in diagnosis of SO palsy. Maximum mean ± SD ipsilesional SO cross-section was reduced to 9.7 ± 3.9 mm2 in SO palsy, representing 53% of the 18.5 ± 4.5 mm2 contralesional SO cross-section, and 53% of the 18.4 ± 3.6 mm2 normal SO cross-section (P < 0.0001). Only 35 patients (69%) with SO atrophy fulfilled the entire three-step test. Two steps were fulfilled in 16 (29%) patients, and only one step was fulfilled in 1 patient (2%). Affected SO cross-section was similar in orbits that fulfilled the 3-step test (10.0 ± 4.1 mm2) versus those that did not (9.0 ± 3.4 mm2; P = 0.51). Since maximum SO cross section correlates with contractility, it seems reasonable to regard the MRI finding of SO atrophy as a sufficient objective confirmation of SO palsy. The complete 3-step test misses 31% of cases of SO atrophy. While acceptance of only two steps would increase sensitivity to 98%, relaxation of diagnostic rigor would probably make the test non-specific. The three-step test fails to detect SO palsy in one-third of cases proven by MRI. Often, only two of three steps are positive in SO palsy.