It is my pleasure to have discussion with Dr Demer regarding our article. What we revealed in the study was (1) that paretic side/normal side ratios of the cross-sectional areas of the superior oblique (SO) muscle in patients with idiopathic SO palsy showed a wide range of variation; (2) that left side/right side ratios of the cross-sectional areas of the SO in control subjects did not show any variation; and (3) that left side/right side ratios of the cross-sectional areas of the 4 rectus muscles in both SO palsy patients and control subjects did not show any variation. We did not state in our article that the 95% confidence interval of the paretic side-to-normal side ratios of the cross-sectional areas of the SO in the SO palsy patients should be used for diagnostic purpose of the SO muscle atrophy. Rather, we stated in the Conclusions of the Abstract that “the muscle hypoplasia could be defined as such when the paretic side/contralateral side ratios of cross sectional areas of the SO muscle on magnetic resonance images fall outside the 95% confidence interval of the ratios in normal controls .” We also suggested in the Discussion of the article that “another way of defining the muscle hypoplasia is to use the paretic side/contralateral side ratios which are located outside the range of the left side/right side ratios of cross sectional areas of the SO muscle in normal controls .”

Our study suggests that patients diagnosed with idiopathic SO palsy without the muscle atrophy, based on a preceding standard, indeed might have the atrophy at some levels. I believe that patients with the simulated or masquerading condition of idiopathic SO palsy would exist, but that the number of such patients would be small. Loosening of the connective tissue that supports the eye globe and the extraocular muscles, designated as the pulley by Dr Demer, is a good candidate to explain the simulated or masquerading SO palsy. We did not use magnetic resonance images in quasicoronal sections, taken specially to be aligned perpendicular to the long axis of the orbit; rather, we used magnetic resonance images in the usual coronal sections, taken as a radiologic standard in every hospital. The digital images are available on electronic medical records systems, and the cross-sectional areas of the SO muscle can be measured in a short duration by image programs. Such an easily available method in a daily clinical setting will enhance the understanding of idiopathic SO palsy.