Though Gass postulated that the average age of onset might be between the 3^rd and 5^th decades of life, subsequent studies have suggested a much later onset, with most patients not developing symptoms or being diagnosed until they are between 50 to 70 years of age (1, 3, 5, 13, 16). However, the age of onset is highly variable, with studies frequently reporting ranges of 30 years of age up to 80 years (3, 5). Rates of diagnosis between men and women are not regularly reported, but three studies reported a slightly greater proportion of their cohorts being female, between 57% and 66% (5, 16, 17). Other studies report an equal distribution between male and female participants (3).

Race and ethnicity trends have also not been well studied, with most studies failing to describe the race or ethnicity of their participants. However, studies on AOFVD are published from diverse geographic locations worldwide, including China, Japan, the U.S., South America, the Middle East, and across Europe (18–22). This suggests that the disease can be found in most populations.

AOFVD is known to be a rare disease, and its prevalence has seldom been reported. Dalvin et al. reported the prevalence of AOFVD in Olmsted County, Minnesota as being between 1 in 7400 to 8200 individuals (23). Prevalence otherwise remains undescribed. Misdiagnosis of AOFVD or miscoding of the disease as AMD in the electronic health record can make it difficult to assess its true prevalence.

Early articles exploring AOFVD suggested an AD inheritance pattern (1, 7, 12). However, it has since become clear that most cases of AOFVD are sporadic and do not follow a clear inheritance pattern (3, 5, 6). That said, several genes have been associated with AOFVD, including PRPH2, BEST1, IMPG1, and IMPG2.

Histopathologic examination of AOFVD remains sparse despite the important role it plays in revealing disease etiology and progression. No new histologic papers specific to AOFVD have been published since 2003. In total there are five papers that discuss AOFVD histopathology: Gass’s original paper, Patrinely et al., Jaffe and Schatz, Dubovy et al., and Arnold et al. (1, 8, 40–42)

Disruption of photoreceptors in the foveal region is common across all studies. Some show RPE hypertrophy in the macula, while others demonstrate peripheral hypertrophy (8, 40). Across all these papers, there is a general consensus that AOFVD is a clinical spectrum of disease that results from the disordered metabolism of RPE cells, resulting in the accumulation of material in the subretinal space. As the material accumulates, RPE atrophy may ensue, triggering changes in the overlying retina.

Both Gass and Patrinely et al. noted disruption and atrophy of photoreceptors in the foveal region. Gass demonstrated hypertrophy of the RPE in the macula (corresponding to hyperpigmentation on fundoscopy), while Patrinely et al. saw peripheral hypertrophy of RPE with atrophy of the RPE in the macula. Both studies also demonstrated pigment-laden macrophages that had migrated from the RPE into the overlying retina and periodic-acid Schiff staining material that had accumulated between the RPE and Bruch’s membrane. Both reported the presence of calcific bodies, with Gass reporting the bodies as being between the RPE and Bruch’s membrane and Patrinely et al. reporting them as extending into the retina.

Where Gass and Patrinely et al. diverge is in the reporting of lipofuscin. Lipofuscin is a pigment produced through oxidative damage and as a metabolic product. It is seen in other retinal diseases such as Best disease, fundus flavimaculatus, Stargardt disease, and some melanomas (40). Lipofuscin could contribute to the yellow appearance of the vitelliform lesion (40). Patrinely et al. noted a large amount of lipofuscin within abnormal RPE cells and macrophages whereas Gass reported minimal lipofuscin. Similarly, Jaffe, Schatz and Arnold et al. did not find evidence of excessive lipofuscin. However, Dubovy et al. did find a significant amount of lipofuscin in the RPE and macrophages in 3 of their cases. Some hypothesize that the amount of lipofuscin in the histopathologic samples correlates with the stage of disease, with earlier vitelliform lesions possessing greater amounts and later stage disease demonstrating reabsorption of the lipofuscin (42).

As mentioned, the classic AVL associated with AOFVD fits an “egg-yolk” appearance: a yellow, elevated lesion within the fovea with a central pigmented spot. At times, the central spot may not appear pigmented but, instead, appears as a separate elevated “figure” underlying the lesion (3). The vitelliform lesion may also be described as a pigmented clump with hypopigmented halos (3), though some groups choose to exclude these morphologies from AOFVD clinical studies. The AVL associated with AOFVD is rounded and regular in shape, mostly centered on the fovea and sometimes at the perifoveal area. Drusen can be observed surrounding the AVL, though they are less prominent than in AMD (45). The AVLs associated with AOFVD are generally less than 1 disc diameter in size (13).

Studies have demonstrated that most patients diagnosed with AOFVD (either with the yellow lesion type or pigmented type) will possess lesions bilaterally (3, 13, 45, 46). However, bilateral lesions are not required for diagnosis. Additionally, despite the presence of bilateral lesions, the progression of the lesions may not be symmetric; the two eyes may be observed at different stages (3).

The vitelliform and pseudohypopyon stages are indistinguishable on color fundus photography. During the vitelliruptive stage, the AVL may take on a “scrambled egg” appearance due to the patchy reabsorption of the lesion. This correlates on fundoscopy to a less homogeneous lesion with the dispersion of the raised, yellow material without corresponding atrophy or subretinal fibrosis (43).

The vitelliform lesion in stage I of AOFVD is hyperautofluorescent on fundus autofluorescence (FAF) ( Figures 1 , 2 ) (13). It is hypothesized that its hyperautofluorescence is derived from the autofluorescent precursors of lipofuscin (A2E, A2PE-H2, A2PE, and A2-rhodopsin), which supports histopathologic findings (46). In the pseudohypopyon stage, on FAF, the lesion will have a hyperautofluorescent inferior half with a hypoautofluorescent superior half (45). As the lesion progresses to the vitelliruptive and atrophic stages, the AVL disappears, and the lesion left behind is hypoautofluorescent due to underlying RPE atrophy ( Figure 2 ) (45, 46).

On fluorescein angiography (FA), in the vitelliform stage, the lesion appears as a non-fluorescent central spot with a hyperfluorescent “spoked” ring without leakage (3, 10, 13, 47). Late phase images may demonstrate central staining of the lesion ( Figure 1 ) (10). As the disease progresses towards the pseudohypopyon and vitelliruptive stages, the hyperfluorescence typically resolves, though a “stars-in-the-sky” appearance may be seen as focal nodules and cuticular drusen persist throughout these two stages (48). The stars-in-the-sky appearance and persistent hyperfluorescence in the vitelliruptive stage on FA and FAF may contribute to difficulty in distinguishing choroidal neovascularization (CNV). In these instances, other techniques, such as OCTA and indocyanine green angiography (ICGA), play an important role in establishing the diagnosis and planning treatment, especially for patients presenting at this stage for the first time in the clinic (14, 17).

Battaglia et al. and Lanzetta et al. both characterized in 1996 the ICGA findings in eyes with AOFVD and vitelliform lesions (14, 53). ICGA reveals an oval, central dark spot beginning in the early frames of ICGA and persisting throughout all later frames (14, 53). The dark spot masks the underlying choroidal vessels. In later frames (between 8 to 15 minutes), a hyperindocyanescent area appears in approximately 81% of eyes (14, 53). The hyperindocyanescent area is irregularly round and located in the central area of the hypoindocyanescent dark spot. It is hypothesized that the blockage seen in the early frames is due to lipofuscin and melanin accumulation within the lesion, whereas the late frame hyperindocyanescence is evidence of more extensive RPE damage (53).

On SD-OCT, the vitelliform lesion corresponds to dome-shaped subretinal hyperreflective material (SHRM) between the RPE and ellipsoid zones in eyes with AOFVD in the vitelliform stage ( Figures 1 , 2 and 4 ) (10, 16, 43). The ellipsoid zone is regularly disrupted in AOFVD, though this disruption may not impact VA (45). The height of the vitelliform lesion in AOFVD tends to grow more so than other AVLs, such as those due to AMD (45).

In all stages of AOFVD, there are frequently cuticular drusen at the RPE-Bruch’s membrane complex with associated RPE elevation, which may be intra- or extra-lesional ( Supplementary Figure 1 ) (16, 46, 54). Querques et al. hypothesized that these cuticular drusen might represent RPE hyperplasia or macrophages containing large amounts of melanolipofuscin (16). Additionally, Wilde et al. describe that many patients (40%) newly presenting with AOFVD have subretinal drusenoid deposits (reticular pseudodrusen), which form secondary to damage in Bruch’s membrane and the RPE (48).

In the pseudohypopyon stage, the vitelliform lesion will consist of two different “zones.” In the upper zone, the material is relatively hyporeflective, with a few limited clumps of SHRM. The lower zone is characterized by SHRM. As in the vitelliform stage, the ellipsoid zone overlying the lesion is often disrupted (43). All other retinal layers from the internal limiting membrane to the external limiting membrane remain normal appearing (43). Intraretinal pseudocysts may also be seen during this stage. In the vitelliruptive stage, the lesion may appear on SD-OCT as if it is fragmented, with a mix of hyperreflective and hyporeflective spaces overlying the RPE ( Figure 2 ). This process is frequently accompanied by overlying photoreceptor loss (16). As the vitelliruptive stage progresses, the lesion becomes progressively optically empty and hyporeflective with the hyperreflective material “clumping” and resolving along the RPE/photoreceptor interface (43). As the vitelliform lesion resolves, the overlying sensory retina may progressively attenuate, ultimately resulting in macular atrophy ( Figure 3 ). CNV itself may present at any stage of AOFVD and will present with findings of subretinal fluid on SD-OCT ( Figure 4 ).

Only a few studies have utilized enhanced depth imaging (EDI) in the analysis of AOFVD, but these studies have helped elucidate some of the findings associated with each stage of disease. EDI demonstrates that the pseudohypopyon stage resembles more of a “croissant” configuration, with the lower hyporeflective space consisting of fluid (55). Additionally, studies have found that eyes in the pseudohypopyon stage have significantly increased choroidal thickness in the temporal and subfoveal regions (55, 56). Once eyes reached the vitelliruptive stage, increased choroidal thickness was seen by Grenga et al. across all locations in the retina (subfoveal, nasal, temporal, superior, and inferior regions) (56). This is in contrast to findings which support that choroidal thickness during the vitelliform stage is not significantly different from controls without AOFVD (55, 56).

Optical coherence tomography angiography (OCTA) has emerged in the last 10 years as a new, non-invasive modality for assessing retinal blood vessel density and flow. Since 2016, many investigators have published reports which have characterized OCTA findings in patients with AOFVD.

Multiple studies have demonstrated a reduction in choriocapillaris vessel density in the foveal region corresponding to the AVL (19, 49, 51, 52). Cennamo et al. further elucidated that this reduction in choriocapillaris vessel density was seen only in the vitelliform and pseudohypopyon stages, whereas there was actually a significant increase in choriocapillaris vessel density during the vitelliruptive stage (19).

Multiple studies have demonstrated reduced blood flow in the superficial capillary plexus and deep capillary plexus in the foveal region corresponding to the vitelliform lesion (19, 49, 50). It is hypothesized that this is likely due to the shadowing effect from the vitelliform material, though it may be attributable to the mechanical compression of blood vessels by the vitelliform material, thus decreasing flow. Additionally, reduced flow in the choriocapillaris can be observed (19, 49, 50).

OCTA routinely demonstrates that patients with AOFVD possess an increased subfoveal choroidal thickness, which corroborates what has been reported on EDI (19, 49, 50). The choroidal thickness is increased compared to both healthy eyes as well as eyes with AMD (50). Both the increased choroidal thickness as well as reduced flow in the choriocapillaris supports that AOFVD possesses a pachychoroid, a feature which is defined by the attenuation of the choriocapillaris and dilation of choroidal veins (57). Further studies are needed in order to elucidate whether the decreased capillary plexus blood flow is an artifact of imaging.

Finally, OCTA may possess advantages over FA for the diagnosis of CNV in the context of AOFVD. As noted above, the vitelliform lesion may mask or be mistaken as neovascularization due to its late-phase staining in FA, with Joshi et al. identifying CNV in 1 eye out of 8 using OCTA, which had not been seen in FA (58). Additionally, Joshi et al. described a reduced hyporeflectivity in the vitelliform lesion area during the pseudohypopyon stage that is best appreciated by OCTA as opposed to FA (58). OCTA may thus be considered as an alternative to FA given its non-invasive approach and potentially more sensitive results.

The incidence of CNV in patients with AOFVD remains inadequately characterized with marked variance between the few estimates that are published, which may be secondary to differences in length of follow up between studies. Total incidence of CNV has been reported as being 2.1% (17), 7.7% (64), and 11.7% (65). Among papers which report rates of CNV, two describe the type of CNV. In both of these studies, all patients with CNV are described as having type 1 CNV (17, 65). For those patients with CNV, visual acuity is significantly affected. In Da Pozzo et al’s study of 51 eyes with AOFVD, VA was reduced from 20/80 to 20/250 among the 6 patients with CNV. Among the 3 eyes with diagnosed and treated CNV in Wilde et al’s study, there was an average change in BCVA -0.21 logMAR. Risk factors for CNV similarly remain poorly reported. Wilde et al. found a significant increase in risk for CNV with the presence of subretinal drusenoid deposits seen on OCT (64). Balaratnasingam et al. found that eyes with AVLs secondary to AMD had a greater risk of conversion to CNV than eyes with AVLs secondary to AOFVD (17). Other risk factors such as age, sex, and medical comorbidities are scarcely described, underscoring the need for research on this topic.

The incidence of macular atrophy in patients with AOFVD is perhaps less understood than in CNV. Balaratnasingam et al. reported an incidence of 21.3% in their cohort of 61 patients (17). Wilde et al. characterized the incidence of macular atrophy among a cohort of 26 patients with AVLs and found that 26.9% of individuals developed incident macular atrophy during the course of follow-up. Among those patients, 42% developed bilateral macular atrophy (64). Out of a larger prospective cohort of 237 eyes with AVLs, Chandra et al. found 21.9% exhibited macular atrophy at five years of follow-up (66).

Macular atrophy, unsurprisingly, has been associated with a significant decrease in vision, with studies reporting an associated decrease of up to 0.29 and 0.323 logMAR (17, 64). Increased risk of developing macular atrophy has been associated with patients who have lower baseline BCVA, greater maximum width of vitelliform lesions, and larger maximum height. Additionally, eyes that progress to the pseudohypopyon stage or vitelliruptive stage are more likely to progress towards macular atrophy (66). Similar to CNV, eyes with subretinal drusenoid deposits are also more likely to progress toward macular atrophy (64). Finally, similar to CNV, no studies have correlated patient or medical risk factors with conversion to macular atrophy.

Apart from the risk factors described above for progression to macular atrophy or CNV, very few studies have identified risk factors associated with progression to the later stages of AOFVD. It remains unclear whether medical factors such as hypertension or chronic disease play any role in AOFVD or its progression. The only patient factor which is suggested to increase risk of progression is initial BCVA; worse baseline BCVA portends increased risk for macular atrophy and CNV (66).

Adult onset vitelliform dystrophy shares many phenotypic features with Best disease. Both present with a vitelliform lesion that share the same features across fundus photography, FA, and SD-OCT (67). Best disease is also known as vitelliform macular dystrophy, which may lead to confusion when AOFVD is referred to as adult vitelliform macular dystrophy or adult vitelliform macular degeneration. The shared characteristics prompted Gass to propose in his original 1974 paper that AOFVD exists as a subtype of Best disease (1). Others have suggested that AOFVD is perhaps a more mild phenotype of Best disease (3). However, there are several distinct characteristics that are useful in distinguishing these two diseases.

AOFVD presents exclusively in adults, with the average age at diagnosis being in the 6^th decade of life (1, 3, 5, 13, 16). As described above, most patients diagnosed with AOFVD have a relatively benign disease course with preservation of VA. In contrast, Best disease is characterized by juvenile onset without loss of visual acuity until the sixth or seventh decades of life (25). Because many patients are not seen until they become symptomatic, it can be difficult to differentiate AOFVD from Best disease. In general, vitelliform lesions associated with Best disease are larger than those seen with AOFVD (68). That said, AOFVD may become quite large, so measuring lesion size is not a reliable outcome for differentiating the two diseases.

Best disease follows an AD inheritance pattern: most patients with Best disease will exhibit a missense mutation in BEST1 (67). As discussed above, BEST1 mutations have been associated with AOFVD, but they are not present in the majority of patients with AOFVD (31). Additionally, AOFVD has been shown to have a sporadic presentation. While there may be increased risk within families, it rarely presents in an AD fashion.

Prior studies have demonstrated that the most reliable way to differentiate between Best disease and AOFVD is by performing an electrooculogram (EOG). In Best disease, the EOG is abnormal with a reduced Arden’s ratio, whereas it is normal in the majority of patients with AOFVD (68).

AMD with vitelliform lesions is frequently difficult to differentiate from AOFVD. Patients present in the 6^th and 7^th decades of life for both diseases. Furthermore, because AOFVD patients are older, they frequently have comorbid drusen and subretinal drusenoid deposits (16, 48). Some studies define AMD with vitelliform lesions purely by the presence of an AVL with foveal drusen (46). However, this may be masking cases that might actually be AOFVD, which underscores the importance of strict diagnostic guidelines both for AMD with vitelliform lesions and for AOFVD.

Other studies differentiate AOFVD from AMD by the size of drusen (17). The consensus definition for AMD is drusen greater than 63 um with pigmentary abnormalities or large drusen (>125 um). The drusen must be present within 2 disc diameters of the fovea in persons older than 55 (69). Eyes with vitelliform lesions and drusen who do not meet these diagnostic standards should instead be diagnosed with AOFVD.

A recent study suggests that patients with AOFVD have greater macular choroidal thickness and subfoveal choroidal thickness than those with AMD, which may prove useful for clinical differentiation of the two diseases (70). Finally, AOFVD in the atrophic stage may be differentiated from AMD via the pattern of atrophy. In AOFVD, the atrophy will generally correspond to the area of the prior vitelliform lesion in the foveal region, whereas in AMD, atrophy may be diffuse and extrafoveal.

Pachychoroid disease, as briefly discussed above, is a subset of retinal disease characterized by the dilatation of choroidal vessels in Haller’s layer and, to a lesser degree, in Sattler’s layer of the choroid. These vessel changes are often accompanied by choroidal filling defects and decreased flow within the choriocapillaris layer (57). Pachyvitelliform maculopathy is a subset of Pachychoroid disease where an acquired vitelliform lesion is observed in an area of enlarged pachyvessels. RPE dysfunction secondary to reduced flow within the choriocapillaris likely drives the development of the AVL (71, 72).

In pachyvitelliform maculopathy, the AVL may regress or migrate. Additionally, the AVL may resolve, only to reform in the same or different location (73). These phenomena are not typical of AOFVD. Additionally helpful in differentiating the two disease is the development of choroidal vessel dilation relative to the AVL. In, AOFVD, there is a reduction in choriocapillaris vessels in the vitelliform and pseudohypopyon stages of the disease, with a dilation of choroid vessels occurring in the vitelleruptive stage, suggesting that the AVL causes the pachyvessels (19). Finally, the association of AOFVD with mutations in genes known to be associated with the function of RPE cells further supports that the AVLs in AOFVD are not exclusively due to choroidal disease but rather primary RPE dysfunction (28, 33).