The purpose of this study was to determine the relationship among visual field loss, structural loss of neural tissue, and blood flow reduction in glaucoma. The authors' multivariate regression analysis revealed that blood flow reduction and structural loss were both significant and independent predictors of visual field loss in glaucoma patients. In addition, there was no correlation or paradoxical correlation between blood flow and structural changes. These findings suggest that structural evaluation alone can, at best, only provide a partial understanding of disease severity. Retinal blood flow, as measured by OCT, was more than twice as important as structural variables in explaining the variation in visual field MD. Therefore blood flow measurements may potentially be at least as important as structural measurements in the diagnostic and prognostic evaluation of glaucoma.
There are a number of possible causal relationships among blood flow, IOP, ONH structure, and visual field loss (). In clinically detectable glaucoma, the ONH has already undergone significant structural change with the loss of nearly 50% of retinal nerve fibers.32
It is possible that neural tissue loss drives the reduction in blood flow through decreased metabolic demand33
(A) or that neural structure loss is a consequence of ischemic damage due to perfusion deficits (B). Alternatively, visual field loss may arise from reduced blood flow that is independent of structural loss (C). The distinction is critical because the identification of a vascular dysfunction independent of structural loss raises the possibility of a new metric for glaucoma evaluation and the potential for a novel therapeutic target. The authors' analysis suggests that structural thinning of the RNFL or disc rim is not a critical intermediate link or confounding variable when evaluating the relationship between reduced blood flow and visual field loss. Therefore, reduced blood flow in glaucoma appears to be neither a major consequence nor a strong driver of neural tissue loss, and is largely independently associated with visual field loss in glaucoma.
Figure 2. Possible relationships among blood flow, intraocular pressure, structure, and visual field loss in glaucoma. (A) Reduced blood flow could be a consequence of neural tissue loss arising from elevated intraocular pressure. (B) Reduced blood flow and elevated (more ...)
While this study identifies blood flow as a predictor of visual loss, the precise pathophysiological link between blood flow and glaucoma remains unclear. Ocular perfusion pressure is the driving force for the blood circulation in the eye, and is defined as the difference between mean arterial blood pressure and venous pressure, which is approximately equal to IOP. Several studies have demonstrated significant associations between ocular perfusion pressure, blood flow, and visual function.34–38
Some studies have demonstrated a positive correlation between systemic blood pressure and glaucoma,39
while others identified a negative correlation40
or no correlation.41
These conflicting results suggest that the link between blood pressure and glaucoma is complex, and may depend on many other factors such as medications and vascular autoregulation.
Intraocular pressure has been evaluated extensively and is an important risk factor for glaucoma. However, IOP is an imperfect correlate of visual field loss. In some patients, disease progression can occur despite adequate IOP control,4–7
while in ocular hypertensive patients there may be no structural or functional consequence despite sustained elevation of IOP.6
Therefore, it seems that the level of glaucomatous damage depends on the susceptibility of an individual eye to a certain level of IOP. Similarly, it is possible that the amount of RNFL loss depends on the susceptibility of an individual eye to a certain level of retinal blood flow. A number of population-based studies identified a strong association between low perfusion pressure and glaucoma.40–43
This study did not demonstrate any significant correlation between visual field loss or total retinal blood flow, and either diastolic or systolic perfusion pressures. Further studies are warranted to elucidate the precise relationship between the various vascular factors and glaucoma. Direct measurement of retinal blood flow may provide insight on how other vascular factors affect ocular perfusion, which in turn could lead to glaucoma progression.
Vascular deficiencies in glaucoma may also be related to unstable perfusion and dysfunctional autoregulation. Nocturnal circadian dips in blood pressure with presumed ischemic damage have been implicated in disease progression.44–46
However, blood pressure is an imperfect surrogate of ocular perfusion, since brachial and ocular blood pressures likely vary in diseased vascular beds and significant differences can arise with positional change. Further studies assessing hour-to-hour variability in blood flow using Doppler OCT would allow a stronger understanding of the influence of circadian variation on ocular perfusion and glaucoma.
Autoregulation of ocular capillary beds causes shifts in regional blood flow in accordance with tissue demand and adds further complexity to assessing ocular perfusion. Autoregulation may mitigate shifts in blood pressure and IOP, and recent studies suggest that autoregulation may be dysfunctional or absent in glaucoma patients.47
In healthy eyes, retinal and ONH blood flow are autoregulated and maintained at a relatively constant level, despite diurnal variations of IOP and ocular perfusion pressure. Ocular blood flow in the eye is autoregulated via local mechanisms, and includes metabolic and mechanical autoregulation. Metabolic autoregulation is influenced by local conditions, such as nutritional needs, carbon dioxide or oxygen levels, and adenosine levels when ATP is not being produced because of hypoxia.48
Mechanical autoregulation is triggered by changes in systemic blood pressure, IOP, ocular perfusion pressure, or shear stress. It seems that some factors such as endothelin-1 and nitric oxide, released by the vascular endothelium, are responsible for the regulation of local perfusion in the retina and ONH.49
In glaucoma, ocular flow dysregulation is common and is closely associated with the systemic vascular dysregulation mediated by endothelial dysfunction,50
and systemic factors such as vasospasm or atherosclerosis may influence it.51
The severity of glaucoma may affect retinal function, and thereby the metabolic need and the level of retinal blood flow. The fact that each dB reduction in blood flow was associated with nearly 2 dB reduction in visual field MD suggests that glaucoma may affect the functioning of not only the ganglion cell layer and nerve fiber layer, but other inner retinal layers as well. Additional studies employing Doppler OCT may advance our understanding of vascular dysregulation in glaucoma.
The principal finding of this study, that retinal blood flow is a largely independent metric of visual function, raises new questions. Retinal blood flow may not only act as a useful complementary variable for the diagnosis and monitoring of disease progression, but may also serve as a potential therapeutic target. However, critical questions need to be addressed before these possibilities are realized. First, we need to gain a stronger understanding of ocular blood flow physiology and its diagnostic potential. When do changes first manifest in the glaucoma disease process? What is the sensitivity and specificity of reduced ocular blood flow? Are there racial differences in normative values? Is there a circadian influence on ocular perfusion? What are the implications of reduced retinal blood flow on retinal function in glaucomatous eyes? Second, we need to understand the mechanisms by which ocular blood flow is lowered or raised. Does lowering IOP increase blood flow? Are systemic factors at play? Can increasing blood pressure improve ocular perfusion? What are the systemic risks? Third, we need to assess the functional benefits of improving ocular perfusion. Could enhanced blood flow improve visual field function? Or slow visual field loss? Or reduce structural change?
There are a number of limitations to this study. First, the study employed a non-interventional case-control design, which can identify associations but cannot determine causation. Although the pattern of correlation is highly suggestive of causation, further assessments using alternative study designs will be needed to confirm a causal link between reduced blood flow and glaucoma. Second, the case-control design is also susceptible to confounding elements not accounted for in control selection. The study's glaucoma and normal groups were age-matched and there were no significant differences between the groups for history of hypertension and diabetes mellitus. There was also no significant difference in use of antihypertensive medications, diastolic ocular perfusion pressure, and systolic ocular perfusion pressure. However, there may be relevant confounding elements that remain unidentified and the potential confounding impact of both ocular and systemic medications warrants further investigation. Third, this study assessed total retinal blood flow, but not localized retinal flow. RNFL loss is more localized in early to moderate glaucoma, and there may be significant correlations between localized RNFL loss and localized retinal flow. This study was not able to evaluate the microcirculation of the neuroretinal rim, because the double-circular Doppler OCT used in the study was not designed for this purpose. Similarly, technological limitations precluded precise, non-invasive measurements of blood flow in the short posterior ciliary arteries. These may be relevant to the glaucoma disease process because they partly supply the ONH. Fourth, there is variability in repeat measurements of retinal blood flow using Doppler OCT, and previous studies reported the coefficient of variation to be approximately 10–14%.27,52
Future studies confirming the reliability of the authors' findings would be instructive.
In summary, this study demonstrated that retinal blood flow measurement may be useful as a novel variable in glaucoma assessment. The authors' analysis demonstrated that reduced blood flow is associated with visual field loss and largely independent of thinning of the disc rim or the RNFL. Therefore, blood flow assessment provides information on glaucoma disease severity that is not captured by structural evaluation alone. Further investigations in ocular perfusion are warranted, and may open the possibility of a paradigm shift in the diagnosis and treatment of glaucoma.