Analyses exploring relationships between structural and functional losses in glaucoma show that early structural damage as measured by spectral-domain optical coherence tomography (OCT) can occur in the absence of visual field (VF) changes. However, VF changes occur once when about 20% of retinal nerve fiber layer (RNFL) thickness is lost, said Dr Gadi Wollstein, at the 20th annual meeting of the American Glaucoma Society.

The analyses were based on data collected from 72 healthy participants and 40 patients with glaucoma representing the spectrum from early to advanced disease, said Dr Wollstein, associate professor of ophthalmology and director, Ophthalmic Imaging Research Laboratories, University of Pittsburgh School of Medicine.

Data on RNFL thickness were obtained from 200 × 200 cube scans using spectral domain OCT (Cirrus HD-OCT, Carl Zeiss Meditec), and VF testing was performed with automated perimetry (Humphrey Field Analyzer, Carl Zeiss Meditec). A 'broken-stick' method of analysis was used to find the inflection or 'tipping point' where there was a change in slope in the relationship between VF and RNFL thickness.

Using data for global RNFL thickness, the analysis showed VF threshold values remained stable at high RNFL thickness values but then began to decrease at an average RNFL thickness of about 75 µm, which was about 20% below the average RNFL thickness for healthy controls at the same age. The slopes for the relationships between VF threshold values and RNFL thickness were statistically significantly different for thicknesses higher and lower than the inflection point.

"Our findings are consistent with previous data based on histological analysis that shows substantial tissue loss occurs before glaucomatous VF changes are detectable," he said.

The message is clinicians can expect to find a substantial zone where patients can experience glaucomatous structural damage without functional change. However, they need to beware that VF loss can be expected to occur once the patient reaches the tipping point so that they can plan for closer follow-up and consider the need for alternative treatment options, he said.

The relationship between structure and function in glaucoma has been the subject for extensive research, and previous studies have shown there is not a simple linear relationship. Regardless of the technology used to measure structure, the results showed a curvilinear relationship where VFs remained relatively unchanged with early structural loss followed by a steep decline demonstrating accelerated functional loss.

"The statistical methods used in previous studies had a 'one-size-fits-all' approach, but the relationship between structure and function in glaucoma may not be the same across the entire spectrum of the disease," Dr Wollstein said. "The broken-stick method of analysis is a piecewise linear regression that addresses this possibility."

Sectoral analyses

In addition to determining the global RNFL thickness value at which VF damage became associated with structural loss, the data were analyzed using sectoral RNFL thickness data. This was done using previously published maps that match the threshold value for all 52 points in the VF to their corresponding optic nerve head locations.

Results of the sectoral analyses showed that tipping points could be defined for temporal, superior, and inferior quadrants, but not for the nasal quadrant. For the temporal, superior and inferior quadrants, the results were similar to those observed in the global RNFL thickness analysis, with the significant change in slope occurring after about 20% of the normative RNFL thickness was lost (range: 17% to 28%).

In the clock-hour analyses, an inflection point could be found for half of the clock hours, and for those locations, it again occurred after about 20% loss in the RNFL thickness.

"For some of the clock hours there was only one corresponding point in the VF," he said. "This limited the reliability of the data and may explain why we did not consistently identify inflection points."

Discussing the limitations of the study, Dr Wollstein mentioned that it had a cross-sectional, population-based design and was not a longitudinal investigation.

"Unfortunately, most studies in glaucoma suffer from this same drawback, since longitudinal studies are difficult to conduct when they involve a slowly progressing disease," he said. "The population-based nature means that clinicians may not directly extrapolate the findings of the study to predict the disease course for individual patients because they may have an RNFL that is thinner or thicker than average as well as exhibit different behaviour in terms of structural and functional loss."