Interinstrument variability in Hertel-type exophthalmometers.

Abstract

PURPOSE: To investigate variability in measurements of axial globe position taken with different exophthalmometers of the same style and manufacturer by the same observer. METHODS: In this small clinical study, 10 Hertel-type mirror exophthalmometers were used to measure the degree of exophthalmos of a plastic head. The deepest points on the lateral orbital margins were marked to ensure accurate placement of the instrument. A single observer, masked to previous measurements, recorded multiple measurements with each instrument over several days. RESULTS: Variation was seen between instruments from different manufacturers and more interestingly, between the same design of instruments from the same company. The measurement of axial globe position varied by up to 2.9 mm between instruments of the same make. There was also variation in the measurement of the relative axial globe position of the model. This measurement varied from 0.6 to 2.4 mm. The mean variation of readings taken with the same instrument was 0.5 mm. CONCLUSIONS: It is already considered good practice to minimize interobserver error by recording the make of exophthalmometer used. This small study also suggests that the same instrument should be used for each examination to minimize error caused by variation between instruments of the same make. Variation of 1.5 to 2.0 mm could affect patient treatment when a difference in measurement of 2 mm is regarded as significant during serial or relative exophthalmometry.