New IOLMaster algorithm improves measurement success, especially in severe cataract

The authors of this study examined the measurement capabilities of the new version 5 software of the IOLMaster partial coherence interferometry device. The authors found that the composite algorithm improves the signal-to-noise ratio, resulting in fewer failures in eyes with severe cataract. However, they say that some cases, in particular white cataracts, could not be measured successfully and remain the domain of ultrasound biometry.

The algorithm used in version 4 calculates the average of the axial length value in millimeters of the single scans, whereas version 5 creates a composite of all measurements from multiple scans by digital processing of signals so that the signal-to-noise ratio improves.

Researchers evaluated the ability of the new composite software in 49 cataract patients unsuccessfully measured with the algorithm used in version 4. They found that the proportion of eyes with an insufficient signal-to-noise ratio decreased from 10.6 percent to 6.8 percent with version 5, allowing measurement in 30 percent of eyes that previously failed, including 42 percent of those with posterior subcapsular cataract.

The authors conducted a second, retrospective study of 1712 patients (2255 eyes) that had undergone biometry with the new composite software. Of these, 105 eyes (4.7 percent) were considered measurement failures due to a signal-to-noise ratio of less than 2.0. The reason for unsuccessful measurement was a white or very dense cataract in 70 eyes, a posterior subcapsular cataract in 14 eyes, a dense nuclear cataract combined with posterior subcapsular cataract in 13 eyes and other reasons in eight cases (traumatic cataract in two, congenital cataract in three and no discernible reason in three eyes). All 105 eyes were measured successfully with ultrasound.

The authors conclude that the new algorithm has the greatest potential in eyes with posterior subcapsular cataract, whereas only 7 percent of eyes with white or very dense nuclear cataract benefited from the improved algorithm.