Autokeratometer, corneal topographer provide similar measurements except in low astigmatism

This study in the April issue of the Journal of Cataract & Refractive Surgery prospectively assessed the repeatability and agreement of corneal biometry measurements obtained with an autokeratometer and a corneal topographer in healthy subjects. The authors found that both devices provided excellent repeatability and comparability of corneal powers and corneal astigmatism, suggesting they can be used interchangeably for measurement of these variables in healthy eyes. However, disagreement in axis location between the two devices was not negligible in some eyes, especially in those with low astigmatism.

The meridian is actually the hard part to nail down, especially with small amounts of astigmatism. It is critical to get this right with multiple means of measurement, since there is no point in having a perfect reproducible way to fix astigmatism with a femtosecond laser if we don't know where to fix it.

The authors say that few studies have assessed the comparability of corneal astigmatism and axis location and believe this to be the first study to compare these astigmatic parameters between autokeratometry and corneal topography.

Fifty-seven healthy volunteers (57 eyes) underwent keratometric readings with an ARK-700A autokeratometer (Nidek Co. Ltd.) and an Atlas corneal topographer (Carl Zeiss Meditec AG).

The authors found that the repeatability of both devices in the flattest and steepest meridians, mean corneal power, corneal astigmatism, axis location, J0 and J45 was high. The 95 percent limits of agreement between the two devices were from -0.51 to 0.48 D for the flattest meridian, -0.74 to 0.71 D for the steepest meridian, -0.56 to 0.53 D for the mean corneal power, -0.58 to 0.58 D for corneal astigmatism, -15.3 to 17.5 degrees for axis location, -0.32 to 0.30 D for J0 and -0.22 to 0.20 D for J45.

All eyes were within ±0.50 D of the differences in astigmatic vector components between the two devices, suggesting that they can be used interchangeably for measuring these parameters in a clinical setting.

However, in eyes with a small amount of astigmatism, there was wider disagreement in axis location between the two devices. A subgroup analysis showed that in eyes with corneal astigmatism greater than 1.00 D, the 95 percent limits of agreement width of the disagreement in axis location was narrower than in eyes with corneal astigmatism of 1.00 D or less. This is presumably because it is more difficult to identify small differences in axis location in eyes with low astigmatism than in eyes with high astigmatism.