Accuracy of the positioning of virtual toric intraocular lenses by unassisted axis evaluation

Abstract

Purpose: To assess the unassisted accuracy of axis evaluation in an experimental model simulating toric intraocular lens (IOL) placement.

Methods: This experimental study enrolled 45 volunteers, 15 senior ophthalmologists, 15 residents in ophthalmology and 15 control subjects from professions out of the medical field. All subjects were given drawings of 10 cm circles with markings similar to those showing the axis of toric IOLs. The drawings were at scale with the design of the optic of the SN6ATn Acrysof® IOL, with 2 x 3 dots indicating the IOL’s axis. Two dots at 0° and 180° were marked outside the representation of the IOL optic, similar to the pre-operative markings used for IOL placement. Twenty-four drawings representing virtual IOLs were prepared for 24 angles between 0 and 180°. Four remarkable angles (0°, 45°, 90° and 135°) were selected and the 20 other angles were randomly chosen. In face-to-face sessions with a single examiner, subjects were asked to estimate, without assistance of any kind, the angle of virtual toric IOLs. The subjects were masked to the list of angles tested and were not told that 4 remarkable angles were included among the drawings. The drawings were shown to all subjects in the same order, which corresponded to a random distribution of the angles tested. The subjects’ responses were compared to the exact angular values.

Results: The mean age of participants (26 males and 19 females) was 44.6 ± 10.4 years for senior ophthalmologists, 26.9 ± 1.5 years for residents and 26.9 ± 10.2 years for controls. Among the 45 participants, the mean axial error was 4.8 ± 4.7°. The mean error of angle assessment by senior ophthalmologists was 3.9 ± 3.8° and was statistically smaller (p=0.012) than the mean error in the ophthalmology residents group (5.3 ± 4.9°) and in the controls group (5.1 ± 5.2°). The assessment of the 4 remarkable angles was better in all groups (p<0.001) and the assessment of acute angles was better than that of obtuse angles (p=0.0036).

Conclusions: Senior surgeons were able to assess the angles of IOL placement with an average error of less than 5° with a naked eye. Current sophisticated guiding systems are a necessary requirement to assist in the placement of toric IOLs, yet they often increase experienced surgeons’ self-assessments by only a few degrees’ precision.

Keywords: 567 intraocular lens • 428 astigmatism • 445 cataract