Abstract
Purpose :
Retinal blood vessel density is lower in glaucomatous eyes. However the relative timing of changes in blood flow and vessel density remains unclear. We measured both on the same day, and compared them against current functional status and the rate of prior functional change.
Methods :
71 eyes of 36 participants performed visual field testing 6 times at 6-month intervals. Functional status was defined as 10Mean Deviation/10, to enable linear modelling of the structure-function relation, and its rate of loss was calculated by linear regression. On the last test date, eyes were imaged using optical coherence tomography angiography (OCTA) and laser speckle flowgraphy (LSFG). A 6° radius circumpapillary B-scan centered on the optic nerve head was derived from the OCT to segment and measure Retinal Nerve Fiber Layer Thickness (RNFLT). Cross-sectional Vessel Area at 6° radius was defined by summing the OCTA flow probability signal from all pixels within the RNFL. Blood Flow was defined as the mean blur rate reported by LSFG, minus the background choroidal flow signal, summed across the major vessels just outside the optic nerve head.
Results :
Flow per unit Vessel Area was higher in eyes with worse function (p<0.001, generalized estimating equation model). The rate of prior functional loss was non-significantly worse in eyes with lower RNFLT or Vessel Area (p=0.060 and 0.075), and unrelated to Flow (p=0.317); see figure. However, Flow per unit Vessel Area was higher in more rapidly progressing eyes (p=0.014). In a multivariable model, Flow per unit Vessel Area remained a significant predictor of the prior rate of functional change (p=0.029) even after adjusting for RNFLT and Vessel Area.
Conclusions :
Vessel density decreases in glaucoma, but blood flow per mm2 of remaining vasculature increases. This new biomarker correlates with the amount of functional loss, and also the recent rate of progression. This could reflect increased flow due to increased metabolic demand, disrupted neurovascular coupling, and/or altered blood flow autoregulation. It could also reflect proportionately greater loss of area for capillaries than higher-flow major vessels. Prospective data collection is needed to determine whether flow per unit vessel area predicts subsequent progression; potentially allowing us to resolve questions of cause-and-effect between changes in blood flow and ganglion cell loss.
This is a 2020 ARVO Annual Meeting abstract.