The history of medicine is characterized by discovery, subclassification, reorganization, and bifurcation of physiological concepts or pathologic entities, driven by advances in investigative technologies or by recognition of clinical details previously overlooked. The story of paracentral acute middle maculopathy (PAMM) is no exception. While it was initially described by Sarraf and associates as a variant of acute macular neuroretinopathy (AMN), it is becoming increasingly evident that PAMM should be considered a clinical finding related to ischemia of the deep retinal circulation, not a specific retinal disease.

In this issue of the Journal , Chen and associates describe the occurrence of PAMM in 9 patients with diverse risk factors for retinal ischemia, including diabetes mellitus, hypertension, hyperlipidemia, sickle cell disease, a “flu-like illness,” and transient orbital compression. Additionally, they report that PAMM lesions occurring in a patient with recent motor vehicle trauma are indistinguishable from polygonal Purtscher flecken, suggesting that certain lesions in Purtscher retinopathy might in fact be lesions of PAMM. Their findings significantly expand the known spectrum of etiologies that underlie PAMM and require a shift in consideration of PAMM from a disease entity to a newly recognized clinical finding with a wider differential diagnosis that warrants further medical investigations and optimization of systemic risk factors. Moreover, as PAMM lesions are being discovered with relative rapidity in an array of systemic diseases, it appears that PAMM is likely to be more common than the wider ophthalmic community currently appreciates.

In a previous report, Yu and associates described the occurrence of PAMM in 35 patients with retinal arterial occlusive disease and hypothesized that transient ischemia of the intermediate and deep capillary plexuses may account for PAMM. They described the PAMM lesion as a deeper form of the more familiar “cotton-wool spot.” In the acute phase, a cotton-wool spot presents as a fluffy white nonexudative focal loss of transparency in the nerve fiber layer due to ischemic axoplasmic stasis. After resolution, the cotton-wool spot leaves a focus of inner retinal thinning or depression. The relative subtlety of PAMM lesions when compared with cotton-wool spots may be due to differences in neurosensory architecture and optical characteristics between inner plexiform, outer plexiform, and nerve fiber layers, the latter of which contains the longest axons, which are perpendicularly oriented to the axis of visualization and least translucent to visible light.

The organization of the retinal microcirculation into superficial, intermediate, and deep capillary plexuses is well known from histologic and confocal imaging studies. The localization of PAMM lesions mostly within the inner nuclear layer bordered by the intermediate and deep capillary plexus etiologically implicates these vascular layers, but the precise hemodynamic factors leading to their occurrence remain incompletely understood.

The recent discovery of PAMM in patients with sickle cell retinopathy, not only by Chen and associates but also by Ilginis and associates, provides some insight into its underlying mechanism. The capillary-occlusive manifestations of sickle cell retinopathy are attributed to the change in aspect ratio and loss of compliance of erythrocytes as they enter a zone of reduced oxygen tension. It is therefore believed that retinal vascular occlusive and neovascular events in sickle cell retinopathy occur in the peripheral retina because these areas are perfused by the smallest capillaries, which are most distally located from the oxygen source, the so-called watershed zones. The observation that PAMM can occur parafoveally in sickle cell patients, and even in the papillomacular bundle, suggests that the middle retina, distal in circulatory terms relative to the inner retinal circulation and to the choriocapillaris, does indeed represent a functional watershed zone in the anteroposterior axis.

Hemoglobinopathies vary in their predilection for neovascular complications and severe, vision-threatening retinopathy, and the Hb-SC variant carries the highest risk. It would be interesting to find out whether the risk distribution of PAMM lesions across different hemoglobinopathies is similar to that of classical sickle cell retinopathy and whether PAMM lesions antecede and predict the development of classical sickle cell retinopathy.

Our view of the relationship between PAMM and AMN is evolving as we question whether they are truly distinct entities. In 1975, Bos and Deutman described a series of patients, mostly young female subjects, who presented with well-defined central scotomata that mapped to radially oriented perifoveal wedge-shaped lesions of AMN. They reported from biomicroscopic observations that AMN appeared to involve the superficial retinal layers, but imaging of AMN with time-domain optical coherence tomography (OCT) demonstrated that these lesions in fact reside in the outer retina. The superiority of near-infrared reflectance over red-free imaging in demonstrating AMN was not reported until 2007.

With the application of ultrahigh-resolution and spectral-domain OCT, further studies reported that AMN lesions specifically involve the outer plexiform and outer nuclear layers, where they manifest as hyperreflective bands with associated ellipsoid disruption, subsequent loss of hyperreflectivity, and a legacy of permanent thinning of the outer nuclear layer.

Subclassification of AMN was proposed in 2013 by Sarraf and associates, who designated the outer plexiform–outer nuclear layer lesions as “type 2” and described PAMM as a new “type 1” variant, with its characteristic parafoveal hyperreflective band on spectral-domain OCT, mostly at the level of the inner nuclear layer. The PAMM variant was also noted to occur more commonly in older patients, mostly male, with vasculopathic risk factors. By contrast, AMN continues to be relatively rare and to have a demographic and risk profile that is distinctly different from that of PAMM. Acute macular neuroretinopathy occurs more commonly in younger female subjects who often have had a recent history of a hypotensive or hypertensive episode induced by febrile illness, trauma, or medication.

The pathogenesis of AMN remains unknown, although our knowledge of the clinical circumstances surrounding its onset, and of the affected demographic, might inform our hypotheses. Andrews and associates have demonstrated immunohistochemically that the distribution of markers of mitochondrial enzyme activity in the axons of the Henle fiber layer is biased toward the outer retina/photoreceptors, implying that these axons may lie on the functional circulatory watershed between the choriocapillaris and the most distal reaches of the deep capillary plexus. It is possible that while PAMM results from occlusive events of the retinal arterial tree, AMN results from a simultaneous hypoperfusion more proximally at the level of the ophthalmic artery that leads to reduced perfusion of both the deep capillary plexus and choriocapillaris. The severity of ischemia in the deep capillary plexus in AMN may not be sufficient to produce PAMM lesions but, combined with choriocapillaris ischemia, may reach the threshold to cause AMN at the functional anteroposterior watershed. If this is the case, it would place AMN at the mild end of a perfusion spectrum defined in its worst manifestation by conditions that cause simultaneous infarcts of the retina and choroid, albeit with differing cellular pathology (eg, giant cell arteritis). Supporting the requisite for simultaneous effects on both retinal and choroidal circulations in the pathogenesis of AMN is the observation that even in complete central retinal artery occlusion, spectral-domain OCT typically shows minimal outer retinal involvement.

Optical coherence tomography angiography (OCTA) is a new modality that represents time-based changes in reflectivity of intravascular components as quantifiable flow maps with depth resolution. In the immediate future, OCTA promises to reveal new information about retinal capillary function in health and disease, with quantitative indices.

The application of new modalities, diagnostic or therapeutic, increases the collective knowledge of diseases and informs the constantly evolving systems of classification that we employ. At present we do not possess the tools for restorative intervention at the capillary level, and there is therefore no specific treatment for AMN or PAMM. It remains to be seen whether a deeper understanding of these conditions as “capillaropathies” further subclassifies them or reunites them as having a common mechanism and potential for intervention.

The relatively recent description of PAMM, together with rapidly growing interest in studying it over the last 2–3 years, suggests that it is more common than previously thought and that it accounts for visual debilitation in a significant number of patients for whom definitive diagnosis may previously have been elusive. Its association with an expanding spectrum of systemic risk factors makes it an important finding with which all ophthalmologists should attain familiarity.