Cyclotorsion Measured in a Patient Population Using Three Different Methods: A Comparative Study

To establish if cyclotorsion measurements obtained in a patient population using different methods of testing give equivalent results, and if the results from each test are repeatable over time.

Twenty referrals, from the orthoptic waiting list at Sahlgrenska University Hospital, Mölndal, Sweden, with vertical deviation stated as the primary reason for referral, were examined. Subjective cyclotorsion measurements using three different methods were performed. The methods included two standard tests: a synoptophore, with slides A17a/18a, the single Maddox rod test (SMR), and a new test method: the KMScreen method. Measurements were repeated 3 times for fixing either eye, and on two separate test occasions, for repeatability within each of the methods.

During the first test measurements, 11 (55%) showed excyclotorsion measured with the synoptophore method, 18 (90%) with the SMR test, and 15 (75%) with the KMScreen method. During the second set of measurements, 11 (55%) showed excyclotorsion with the synoptophore method, 17 (85%) with the SMR test, and 10 (50%) with the KMScreen test. Median perceived torsional measurements at the first visit for the synoptophore, SMR, and KMScreen tests were -1°, -5°, and -1°, respectively. There were significant differences between the synoptophore and SMR (P=0.025), and the SMR and KMScreen (P=0.025), but not between the synoptophore and KMScreen (P=0.90) methods on measuring the affected eye. All three tests showed repeatable values. The 95% limits of agreement (LoAs) between visits, defined as mean torsional difference ± 1.96 SD of difference, for the synoptophore, SMR, and KMScreen methods were -0.5° ± 4.5°, -0.3° ± 5°, and -0.7° ± 3.5°, respectively. P values for differences between visits were P=0.45, P=0.75, and P=0.90, respectively, using an exact permutation test.

In this small study group, the three different methods used for testing cyclotorsion showed significantly different results. However, all methods showed significant repeatability (test-retest reliability) over time. This finding is of clinical importance when investigating cyclotorsion in patients. Clinicians need to be aware of the differences between methods when monitoring recovery or changes in patients, and in particular when planning surgical correction for cyclotorsion.