Study finds considerable variability in specific optical path variables when measuring the same eyes with SS-OCT and OLCR biometry devices

"The most significant finding being that lens thickness demonstrated the greatest difference with the swept source OCT consistently measuring this value greater than the OLCR. We hypothesize there is a reason for this, which will be discussed in another presentation."

This transcript has been lightly edited for clarity.

Hi, my name is Deanna Dang, I am a fourth year medical student at the University of Oklahoma. This project is a collaboration with Dr. Riaz and Dean McKee Eye Institute and Dr. Cooke, at Great Lakes Eye Care, along with other students and researchers at the University of Oklahoma.

Our project is called "Variability between Two Bioimeters." We have no financial disclosures. And ocular biometric measurements are crucial in the calculation of IOL power before cataract surgery. And this paper compares the swept source OCT (IOL Master 700), with the optical low coherence reflectometry biometer (Lenstar LS 900), and seeing if these devices were similar, or different in any meaningful way.

Looking at previous studies, there appeared to be better agreement between both biometers. Some notable differences were that the swept source OCT (IOL Master 700) was able to measure axial length more often than the Lenstar, take three-fourths less time than the Lenstar and also had higher success in measuring eyes with dense media. Of note studies have noticed statistical differences in several variables, but none were clinically significant.

Our paper's different because of the large volume of eyes able to be studied. In the previous studies, the highest number of study eyes was 164. We are very fortunate that at Dean McGee Eye Institute we have both biometers to compare. Patients considering cataract surgery at our institution, and it does sequential calculations with both biometers and the research team here was able to pull precise data from both biometers allowing us to study this large volume of eyes.

Both biometers would give warnings for extreme standard deviations. Even after eliminating those eyes from the data, the volume of eyes studied was still much larger than previous studies in the several 1000s range.
For methods we started with about 24,000 eyes, and then removed eyes with the exclusion criteria listed on the slide. And we were left with 8000 to 13,000 eyes. We studied those 8000 to 13,000 eyes, which also included eyes that received warnings from either biometer and compare those values.

Afterwards, we remove those eyes that received warnings from either biometer which resulted in a significant reduction in the number of eyes, we were left with about 3000 to 5000 eyes depending on which variable was being studied. But these numbers are still much higher than previous studies.

The following charts in this presentation show values which were obtained by taking the absolute value of the difference between the swept source OCT value and the OLCR value for each measured variable. These are the results for mean K and axial length, comparing all eyes versus the eyes that did not receive any warnings or alerts on either biometer.

Interestingly, after removing the eyes with alerts, there was a larger difference in mean K between biometers, than before alerts were excluded. Differences in axial length between both biometers after warnings were excluded showed a mild improvement, especially considering the number of eyes excluded.

Differences in central corneal thickness between both biometers after warnings were excluded also showed a mild improvement. Notably the differences in interior chamber depth between both biometers showed a drastic decrease in difference at greater or equal to 0.1 millimeters.

The OLCR (Lenstar) on average was estimating a 0.18 millimeter increase for anterior chamber depth compared to the IOL Master 700 in all eyes. It also measured 0.13 millimeter increase when eyes that received alerts were excluded. This is a very important finding that we will discuss in another presentation.

AQD and lens thickness also showed significant decrease in differences once eyes with warnings were excluded. The variable with the greatest inner biometer difference is lens thickness with the OLCR averaging 0.39 millimeter decrease in lens thickness compared to the swept source OCT and all eyes and a 0.30 millimeter decreased once eyes with alerts were excluded.

Especially at a difference of 0.3 millimeters or greater, we noted that nearly half of lens thicknesses measured by the OLCR are smaller than the lens thickness as measured by the swept source OCT.

Differences in white to white between both biometers, after warnings for excluded, showed a mild improvement, especially compared to the number of eyes excluded.

For conclusions there is considerable variability in variables when measuring the eyes with both biometers, which may affect formula performance and postoperative refractive predictions when using formulas that utilize values like interior chamber depth and lens thickness.

Here are the conclusions restated again, but the most significant finding being that lens thickness demonstrated the greatest difference with the swept source OCT consistently measuring this value greater than the OLCR. We hypothesize there is a reason for this, which will be discussed in another presentation.