The early detection and treatment of glaucoma are necessary to prevent progressive vision loss and blindness. The machine-to-machine (M2M) model is a new method of predicting which eye in a patient who is at high risk for glaucoma is more likely to develop vision loss.

The M2M model uses an algorithm to predict the thickness of an eye’s retinal nerve fiber layer (RNFL) from a series of optic disc photographs, known as fundus photographs. A thinning RNFL indicates high glaucoma risk. Until recently, M2M had not been tested on the eyes of “glaucoma suspects,” which are patients who don’t currently have glaucoma, but who present with issues that strongly suggest glaucoma will develop.

A new study led by medical student Terry Lee, BA, of the Vision, Imaging and Performance Laboratory at Duke Eye Center, investigated whether or not the M2M’s predictions could be used to accurately gauge glaucoma risk prior to vision loss. 

What They Did

The study involved 1,072 eyes of 827 patients from the Duke Glaucoma Registry. All the participants had a history of elevated intraocular pressure (IOP) and other risk factors, but had normal visual fields at the beginning of the study. The average follow-up was 4.4 years. Researchers took new photographs and administered visual field tests during each follow-up session. M2M was applied to all new fundus photographs, to provide updated RNFL measurements and make new predictions.

Individual eyes were classified as “glaucoma conversors” if they demonstrated vision defects during at least two consecutive visual field tests during follow-up. 

What They Found

By the end of the study, 196 eyes (18%) had ‘converted’ to glaucoma. Both the baseline measurements and rates of change over time were significant factors in predicting whether or not a given eye would develop glaucoma. 

“We found that the mean rate of change in the RNFL thickness slope was faster for conversors than non-conversors,” Lee said. “Each 1 μm (micron)/year faster decrease was associated with a 99% increase in the hazard. These made the eye twice as likely to develop glaucoma.” 

The predictive performance of the M2M model was similar to that of the optical coherence tomography (OCT) measurements from a previous study. OCT is the standard method for measuring RNFL and predicting glaucoma risk. 

What it Means

Though its performance is comparable to OCT, one advantage of M2M is its ability to update glaucoma risk predictions over time as it acquires new data, and the study showed that the rates of change over time were even more significant than the baseline measurements when it comes to predictive ability.

While it will not likely replace OCT in the near future, the M2M model is less expensive than the traditional method, and may be available in settings where OCT is not, whether due to cost or the need for specialized technicians, making it a viable alternative.

*Ophthalmology Times. (2021, June 23). Method of Predicting Glaucoma Conversion Proves to Be Faster. https://www.ophthalmologytimes.com/view/method-of-predicting-glaucoma-conversion-proves-faster