In their recent report entitled “Correlation of optical intensity on optical coherence tomography (OCT) and visual outcome in central retinal artery occlusion (CRAO),” Chen et al1 assert that the increased inner retinal reflectivity seen on OCT after CRAO is an important guide to ischemia severity, especially when set against the associated ghosting of the outer retina. However, the utility of the “optical density ratio” so derived is open to serious question given that the extent of panretinal hypoperfusion can be more readily stratified by reference to the oxygenation status of the superficial and middle retinal components of the macular inner retina.2 That is to say, using fundoscopy, high-resolution OCT, and pupil assessment, three broad grades of oxygenation-based hypoperfusion maculopathy (OHM) are distinguishable at lower volume rates of retinal blood flow. In essence, the superficial inner retina (SIR), containing multilayered ganglion cells and their processes, is “normoxic” in OHM Grade 1 (OHM1), is predominantly “hypoxic” in OHM2, and is predominantly “anoxic” in OHM3.2 These grades are likewise applicable to panretinal hypoperfusion from central retinal vein occlusion (CRVO), and intermediate grades can also be recognized (i.e., OHM1/2 or OHM2/3).
In all instances, the macular middle retina (MR), which comprises the inner nuclear layer and adjacent plexiform tissue, is more vulnerable to panretinal hypoperfusion than the SIR. This appears to be principally because, owing to the late embryological development of the deep capillary plexus and middle capillary plexus from the superficial plexus by capillary sprouting,3 the MR is effectively the central artery's most distal field of macular supply. Thus, although the tissue oxygen tension (pO2) within the MR is similar to that within the SIR under normal physiological conditions,4 the MR will receive blood that is increasingly deoxygenated during stepwise reductions in panretinal perfusion due to disproportionate oxygen extraction within the SIR.2,5–9 This preferential discharge of oxygen from the proximal portion of the vascular path means that, within the retinal arterial tree, the descending gradient of hemoglobin-oxygen saturation (sO2) will steepen, and the sO2 in venous blood draining from the macula will decrease from a normal value of ≈60% to, eventually, 0%. The reciprocal increase in the “oxygen extraction fraction” (i.e., from ≈40% to 100%) is one of the defining characteristics of the so-called “misery perfusion.”
Another corollary of preserved periarterial tissue oxygenation within the SIR, despite the reduced blood volume flow rate, is the development of perivenous infarction within the MR as tissue normoxia gives way to ischemic hypoxia and ischemic anoxia further downstream. Thus, in keeping with the Krogh oxygenation model wherein cylindrical volumes of cells are supplied by axial vessels (in this instance by macular arteries and paired veins with countercurrent blood flow), a “hypoxic funnel” is intercalated between proximal normoxic tissue and distal anoxic tissue.2,8 Of note, hypoxic neural parenchyma, otherwise known as the “ischemic penumbra,” is functionally impaired but structurally intact, the latter feature rendering it indistinguishable from normoxic tissue on fundoscopy and on OCT.2,9 That is to say, hypoxic retina is transparent and has unchanged reflectivity, so it is not susceptible to specific evaluation using measures such as the optical density ratio.1 The functional flaws identified within the hypooxygenated neural tissue comprise “electrical silence” (i.e., failure of impulse and synaptic transmission, including the afferent component of the pupillary light reflex) and impaired intraaxonal transportation of mitochondria (among other cargoes).2 Nevertheless, the hypometabolic neurons maintain their resting membrane potential, and the functional deficits seem to be largely reversible after tissue reperfusion during a time-window of at least a few days.