Abstract
Purpose:
Testing of torsional diplopia is commonly performed in the ophthalmology clinic using the double maddox rod. It is not uncommon during testing for the patient to have a vertical component to the diplopia, either pathologic or artificially introduced by the tester. This study aims to determine to what degree, if any, vertical diplopia interferes with an individual's ability to properly judge torsional status.
Methods:
This study was approved by the Institutional Review Board at the University of South Carolina. Exclusion criteria included previous strabismus surgery or diplopia, or a history of amblyopia. Medical students who did not meet these exclusion criteria were allowed to enroll in the study. For each individual, a double maddox rod was placed in the vertical orientation. Variable amounts of prism of 3, 10, 16, 20, and 30 prism diopters (PD) were placed in front of the eye in sequential fashion to create vertical diplopia. For each strength of prism, the left maddox rod was rotated 45 degrees off of axis. Individuals were then asked to align the double maddox rod to parallel as accurately as possible. Measurements were taken for each strength of prism.
Results:
A total of thirty seven individuals were recruited. The average measured torsion at 3, 10, 16, 20, and 30 PD was 0.95, 1.62, 1.89, 2.08, and 2.16, respectively. Normalizing the data to values obtained at 3 PD yields values of 0.68, 0.95, 1.06, and 1.22 for 10, 16, 20, and 30 PD, respectively.
Conclusions:
The average measured torsion at baseline was 0.95 degrees and there was a trend toward increasing torsion as vertical diplopia was increased. While it is possible subclinical torsion existed in these patients and progressed with increasing vertical diplopia, it is more likely the torsion at baseline is explained by difficulty ensuring perfect ninety degree vertical orientation of the maddox rod prior to testing. The progression of torsion with increasing vertical diplopia is most likely due to increased difficulty for the examinee to properly align the maddox rod in a parallel fashion as the diplopia became more widely spaced. These results indicate that while the double maddox rod may qualitatively measure torsion, it may not be ideal for quantitatively measuring one to two degree changes in torsion, at least not in its typical design. A larger lens radius with more graduations would help alleviate this problem.