Operator learning effect and interoperator reproducibility of the scanning laser polarimeter with variable corneal compensation

Purpose: To ascertain operator learning effect, and to quantify the interoperator reproducibility, using the newly introduced GDx VCC (variable corneal compensator) scanning laser polarimeter. Design: Prospective instrument validation study. Participants: Three operators with no prior experience in operating the GDx VCC examined one randomly chosen eye of each of 30 randomly ordered subjects (15 glaucoma patients and 15 normal subjects). Methods: Each study eye was scanned by the 3 operators in a random sequence during a single session. Five GDx parameters were analyzed: TSNIT (temporal, superior, nasal, inferior, temporal) Average, TSNIT Standard Deviation (SID), Superior Average, Inferior Average, and Nerve Fiber Indicator (NFI). Main Outcome Measures: Retinal nerve fiber layer (RNFL) thickness GDx parameters across operators. Results: A learning effect was not found for any of the operators for any of the 5 GDx parameters studied. Reproducibility for each GDx parameter was measured as the mean SID value of the measurements taken by the 3 operators. Interoperator reproducibilities for normal and glaucomatous eyes, respectively, were 0.8 and 1.6 mum (TSNIT Average), 1.7 and 2.2 mum (Superior Average), 1.6 and 2.7 mum (Inferior Average), 1.3 and 1.7 mum (TSNIT SID), and 1.4 and 4.6 mum (NFI). The data obtained by each operator were highly correlated with those of the other 2 operators. Conclusions: With the commercially available GDx VCC, a learning effect was not found for 3 novice operators. In addition, RNFL measurements seem highly reproducible across operators. (C) 2005 by the American Academy of Ophthalmology.