Abstract
Objective
To characterize distinct precursor patterns associated with the development of patchy atrophy (PA) in highly myopic eyes and to describe their structural evolution and clinical outcomes over time.
Design
A retrospective multicenter cohort study.
Participants
Highly myopic patients with documented onset of PAs and at least 3 months of multimodal imaging (MMI) prior to the PA development.
Methods
Clinical, demographic, and MMI findings were retrospectively collected from 3 tertiary referral centers. Based on MMI findings preceding PA onset, eyes were categorized into 3 patterns: (1) inflammatory PAs in the context of punctate inner choroidopathy; (2) neovascular PAs, associated with extrafoveal macular neovascularization (MNV); and (3) idiopathic PAs, in the absence of identifiable inflammatory or neovascular precursor lesions.
Main Outcome Measures
Precursor patterns, clinical and MMI characteristics, clinical complication profile, and atrophy growth rate (AGR).
Results
The study included 103 eyes (84 patients; 28 [27%] men) with a mean (standard deviation) age of 62 (13.4) years and mean refractive error of –13.2 (5.8) diopters at PA onset. Median observation periods prior to PA onset and post-PA follow-up were 2.1 years (interquartile range 5.20) and 3.44 years (interquartile range 7.23), respectively. Inflammatory PAs were detected in 33 (32%) eyes, neovascular PAs in 9 (9%) eyes, and 61 eyes (59%) had idiopathic PAs. Number of pre-existing PAs at baseline was the highest in inflammatory PAs (6.1 ± 6.4), followed by neovascular PAs (1.6 ± 1.1) and idiopathic PAs (1.5 ± 1.1; P < 0.001). Late-stage complications, including MNV (P < 0.001), subretinal fibrosis (P < 0.001), and macular atrophy (P = 0.01), occurred more frequently in inflammatory and neovascular PA than in idiopathic PA. Exploratory tree-based analysis identified baseline lesion number as a key feature distinguishing inflammatory PAs from idiopathic PAs. Mean AGR in the overall cohort was 0.73 ± 1.06 mm2/year (95% confidence interval, 0.52–0.94), with progression fastest in inflammatory PAs (1.3 ± 1.4 mm2/year), followed by neovascular PAs (0.7 ± 0.7 mm2/year) and slowest in idiopathic PAs (0.4 ± 0.7 mm2/year; P < 0.001). Higher baseline lesion number was associated with faster AGR (P = 0.005).
Conclusions
Patchy atrophies in highly myopic eyes develop with 3 precursor patterns (inflammatory, neovascular, and idiopathic) and differ in structural characteristics, atrophy progression, and complication profile. Recognition of PA patterns may improve clinical assessment of highly myopic eyes.
Financial Disclosure(s)
The authors have no proprietary or commercial interest in any materials discussed in this article.
Keywords: Patchy atrophy, High myopia, Myopic maculopathy, Punctate inner choroidopathy, Myopic neovascularization
Myopia represents a growing global public health challenge, with projections indicating that nearly 1 billion individuals worldwide will be affected by high myopia by 2050.1 This expanding population is at substantially increased risk of sight-threatening ocular complications.2, 3, 4, 5 Among patients with high myopia, myopic maculopathy is one of the leading causes of irreversible visual loss.6,7 Longitudinal data have shown that patchy atrophy (PA) plays a crucial role within this spectrum, with nearly 80% of eyes with PA showing progression and approximately half ultimately developing macular atrophy (MA).7
Clinically, PAs appear as well-defined, yellow–gray lesions that typically spare the central fovea.8 Upon OCT, PAs correspond to localized defects of the retinal pigment epithelium (RPE) layer, without or with corresponding Bruch membrane (BM) defects, often accompanied by overlying outer retinal atrophy.9 Despite their prognostic relevance, the mechanisms underlying the development of PAs have remained incompletely understood so far, and current classification systems rely largely on morphologic descriptors rather than on underlying pathogenic processes.
Growing evidence suggests that PAs may arise through distinct mechanisms. One proposed pathway is neovascularization-related PAs, in which an extrafoveal myopic macular neovascularization (mMNV) precedes the PA formation and leads to a fibrotic remodeling and subsequent degeneration.10, 11, 12 An inflammatory-mediated pathway has additionally been proposed, with punctate inner choroidopathy (PIC) lesions implicated in the development of this type of PAs,13 particularly along lacquer cracks, which, however, may also independently result in patchy atrophy and were therefore not considered sufficient alone to define inflammatory lesions.14 In contrast, an idiopathic form of PAs characterized by BM defects has been linked to progressive myopic axial elongation and choroidal thinning, suggesting a predominantly mechanical process.9
Taken together, these observations indicated that PA is unlikely to represent a single disease entity but rather a shared morphologic endpoint of multiple pathophysiological processes. However, the relative contribution of these precursor mechanisms and their impact on the disease course has remained poorly defined, especially with respect to the further enlargement of PAs.
Our aim was to characterize retrospectively and longitudinally eyes that had developed PAs, in order to characterize the morphological changes preceding the onset of PAs and to delineate distinct phenotypic patterns according to the 3 proposed pathogenic pathways. Furthermore, we assessed the progression of PAs according to the precursor pattern and compared the complication rate between the groups. We also aimed to build a classification tree analysis integrating baseline clinical and imaging features associated with different PA patterns.
Methods
Study Design
The retrospective, longitudinal, multicenter study included highly myopic patients with PA, defined according to the Meta-Analysis of Pathologic Myopia (META-PM). They had been examined in 3 tertiary referral centers: the Department of Ophthalmology at Lariboisière Hospital, Paris, France (46 eyes; 45%); the Retina Department of the French Institute of Myopia, Paris, France (14 eyes; 14%); and the Department of Ophthalmology at IRCCS San Raffaele Hospital, Milan, Italy (42 eyes; 41%). The study adhered to the tenets of the Declaration of Helsinki and received approval by the institutional review boards of all participating centers. All patients signed a written informed consent, were notified of the potential use of their anonymized clinical data for research purposes, and were given the opportunity to refuse the use of their data.
Criteria to be included into the study were (1) diagnosis of high myopia, defined as an axial length of > 26.5 mm and/or a spherical equivalent refractive error of < –6.0 diopters (D);15 (2) ophthalmoscopic signs of fundus tessellatus or diffuse chorioretinal atrophy;8 (3) availability of multimodal imaging (MMI) documenting the retinal area that subsequently developed PA at least 3 months prior to the detection of PAs;8 and (4) an age of ≥18 years. Exclusion criteria included poor-quality or incomplete clinical or MMI data; absence of OCT scans covering the lesion site; foveal neovascularization, defined as mMNV within 200 μm of the foveal center;16 ocular diseases unrelated to high myopia; systemic inflammatory or infectious conditions that could mimic PIC (e.g., tuberculosis, syphilis, sarcoidosis, and histoplasmosis); and prior ocular procedures known to induce chorioretinal atrophy (e.g., focal laser photocoagulation, photodynamic therapy, or posterior segment surgery). Both eyes of the same patient were eligible for inclusion if all criteria were met.
Baseline Assessment
Patients were included at the time of the first documented development of PA, which was defined as the earliest visit at which PA was clinically and structurally detected. As required by the inclusion criteria, multimodal images were available that had been taken at least 3 months prior to the PA detection and which covered the retinal area that eventually developed the PA. It allowed a retrospective identification of precursor lesions at the same topographic location.
Baseline demographic and ocular variables, including sex, age, axial length (measured by optical biometry [IOL Master, Carl Zeiss Meditec]), and spherical equivalent refractive error were recorded at the time of PA onset. Best-corrected visual acuity (VA) was measured using decimal charts and converted to logarithm of the minimum angle of resolution units for analysis.
Multimodal imaging included, at minimum, wide-angle pseudocolor fundus photography (Silverstone; Optos), blue fundus autofluorescence, multicolor and infrared imaging, and spectral-domain OCT (Spectralis HRA + OCT; Heidelberg Engineering), with 6 × 6 radial and high-density volumetric cube scans covering the posterior pole (30° × 25°, 121 B-scans). Using the OCT images, PA onset was confirmed by the development of a focal disruption of the RPE layer, without or with an accompanying defect in the underlying BM.9 Fluorescein angiography, indocyanine green angiography, and OCT angiography were performed in selected situations at the discretion of the treating physician. All images obtained after PA onset were reviewed to complement the assessment of the PA evolution, development of late-stage complications, and calculation of atrophy growth rate (AGR). Since patients were included only if they had no PA, the AGR assessed the enlargement of PAs that developed during follow-up, measured from the time of their first detection.
Precursor Lesions and Complications Analysis
Precursor lesions were categorized as neovascular, inflammatory, or idiopathic at the eye level. A neovascular precursor was defined by the presence of extrafoveal mMNV (> 200 μm from the fovea) identified by a hyperreflective material above the RPE with intraretinal or subretinal fluid on OCT images and posterior hypo-transmission, and a corresponding flow signal on OCT angiography. When available, fluorescein angiography images revealed an early hyperfluorescence and progressive leakage.12,16
An inflammatory precursor was defined by imaging features consistent with PIC preceding PA onset, according to SUN II and MMI in uveitis guidelines.17,18 Punctate inner choroidopathy was defined by the presence of “punctate” choroidal spots < 250 μm in diameter, involving the posterior pole with or without mid-peripheral extension, and with absent to minimal anterior chamber and vitreous inflammation. Upon OCT, these lesions were characterized by a focal outer retina disruption with RPE–BM splitting caused by a homogeneous, ill-defined hyperreflective material, associated with defects in the interdigitation and ellipsoid zone, focal inner choroidal thickening, and hypertransmission of the corresponding RPE layer. Corresponding hypo-autofluorescent or hyper-autofluorescent changes were observed upon fundus autofluorescence, while a neovascular flow was not detectable on the OCT angiography images.17,19
Idiopathic precursor lesions were defined when PA development was not preceded by neovascular or inflammatory changes. Upon OCT, idiopathic lesions showed a disruption of the RPE layer without or with corresponding defects in BM, with a thinning of the outer retinal layers and choroidal disorganization, in the absence of neovascular or inflammatory features.17 Hemorrhagic BrM ruptures without any sign of macular neovascularization (MNV) were included into the category of idiopathic precursor lesions.
Macular atrophy was classified according to the META-PM system and defined on the OCT images as a complete RPE and outer retinal atrophy in accordance with the Classification of Atrophy Meetings criteria.20 Subretinal fibrosis was defined as a well-circumscribed, elevated yellowish-white lesion on fundus photography, with late-phase staining on fluorescein angiography images.21
Central macular thickness was automatically measured, and the readings were extracted using the Heidelberg Spectralis software. Subfoveal choroidal thickness was manually measured as the distance between the outer border of BM and the choroidoscleral interface in the foveal center.22 A posterior staphyloma was defined according to Ohno-Matsui.23 Lacquer cracks were assessed using infrared imaging and, when available, on late-phase indocyanine green angiography images.24 Perforating scleral vessels were identified upon OCT as retrobulbar vessels perforating the sclera. We additionally recorded the presence of a Fuchs´ spot.24 The presence of a V-shaped appearance was additionally evaluated on cross-sectional OCT images.25 This configuration was defined as a sharply angulated, symmetric inward deflection of the outer plexiform layer and inner retinal layers converging toward a focal point at the site of RPE–BM disruption.25
Assessment of Atrophic Areas and Treatment
The presence and extent of PA lesions were evaluated from the time of first detection through each subsequent follow-up. Atrophic areas were identified on 30° infrared images as well-demarcated regions of increased reflectivity and manually delineated using the built-in measurement tools, using a hypertransmission of the RPE band on the OCT scans as confirmation, as previously described by Moukadem et al.26 In eyes with multiple discrete atrophic lesions, each PA lesion was measured separately, and lesion counts were recorded.
Eyes with inflammatory precursors were treated with corticosteroids, when feasible, administered orally or intravitreally, according to the lesion activity and physician judgment. In selected situations with recurrent or refractory disease, immunomodulatory therapy was initiated.
In eyes with mMNV, intravitreal anti-VEGF injections were administered on a pro re nata regimen until the lesion had regressed. All treatments were performed as part of routine clinical care.
Statistical Analysis
Categorical variables were summarized as counts and percentages, and continuous variables as mean (standard deviation) or median (interquartile range), as appropriate. Analyses were structured around 3 objectives: first, precursor mechanisms leading to the PA were identified and their frequencies quantified. Second, baseline demographic, clinical, and MMI characteristics were compared across PA patterns using mixed-effects models to account for correlation between fellow eyes. Linear mixed-effects models were used for continuous variables, and mixed-effects logistic regression models were applied for categorical variables, with a random intercept for patient identity. When the number of observations in some categories was very small, resulting in unstable estimates or quasi-complete separation, effect sizes were not estimated, and results were reported descriptively. Third, atrophy progression was assessed by estimating AGR at the eye level. For each eye, AGR was calculated by fitting a linear regression model with atrophic area as the dependent variable and follow-up time as the independent variable, using all available measurements. Atrophy growth rate was then analyzed in linear mixed-effects models to identify predictors of faster progression, with results reported as β coefficients and 95% confidence intervals (CIs). A backward stepwise selection strategy, guided by clinical relevance and improvement in model fit, was used to construct the final multivariable model. An exploratory classification and regression tree analysis was additionally performed using baseline variables only to identify combinations of demographic, clinical, and imaging features that best discriminated the PA pathway. Tree complexity was controlled by cost-complexity pruning with cross-validation. Given the exploratory nature of this analysis and the limited sample size, classification and regression tree results were interpreted descriptively to support clinical pathway recognition rather than prediction.
All analyses were performed using R software (version 4.5.1; R Foundation for Statistical Computing). A 2-sided P value of <0.05 was considered statistically significant.
Results
Population Baseline Characteristics
Among 1953 eyes screened (Fig S1, available at www.ophthalmologyscience.org), 103 eyes from 84 patients were included. Twenty-eight patients were male (33%). At PA onset, mean age was 62 ± 13.4 years, and mean spherical equivalent refractive error was –13.2 ± 5.7 D. The median observation period before PA onset was 2.1 years (interquartile range 5.20), and median post-PA follow-up was 3.44 years (interquartile range 7.23). Overall, 1486 longitudinal observations were analyzed. At the time of the first fundus imaging, 3 eyes (3%) had a tessellated fundus, while 100 eyes (97%) showed a diffuse chorioretinal atrophy (Table 1).
Table 1.
Baseline Demographic and Clinical Characteristics of the Study Cohort Stratified by PA Pathway
| Patients’ Characteristics | Total (n = 84 Patients) | Inflammatory Pattern (n = 26) | Neovascular Pattern (n = 8) | Idiopathic Pattern (n = 50) |
|---|---|---|---|---|
| Age at PA onset (yrs) | 61.92 (13.44) | 59.42 (14.01) | 64.84 (12.33) | 62.84 (13.29) |
| Male gender | 28 (33%) | |||
| Bilateral disease | 19 (23%) | |||
| Eyes’ Characteristics | Total (n = 103 eyes) | Inflammatory Pattern (n = 33) | Neovascular Pattern (n = 9) | Idiopathic Pattern (n = 61) |
| SCT in μm | 63.32 (47.76) | 70.70 (48.43) | 88.11 (88.69) | 55.67 (37.36) |
| CMT in μm | 233.58 (64.16) | 226.61 (63.21) | 242.89 (66.65) | 235.98 (65.03) |
| Baseline VA in logMAR | 0.25 (0.28) | 0.21 (0.30) | 0.28 (0.21) | 0.27 (0.28) |
| Baseline PA lesion count (n) | 2.99 (4.30) | 6.12 (6.4) | 1.56 (1.13) | 1.51 (1.10) |
| PA lesions at the vascular arcades | 18 (17%) | (0%) | 0 (%) | 18 (30%) |
| Lacquer cracks | 75 (73%) | 31 (94%) | 6 (67%) | 38 (62%) |
| SER in D∗ | –13.21 (5.65) | –15.32 (5.32) | –9.29 (2.92) | –12.60 (5.76) |
| AL in mm∗ | 30.88 (2.10) | 30.81 (1.67) | 29.24 (1.89) | 31 (2.40) |
| Presence of staphyloma | 74 (72%) | 19 (58%) | 5 (56%) | 50 (82%) |
| PV along PA | 28 (27%) | 6 (18%) | 2 (22%) | 20 (33%) |
| Fuchs spot along PA | 18 (18%) | 12 (36%) | 2 (22%) | 4 (7%) |
| BM rupture at PA lesions | 58 (56%) | 28 (85%) | 4 (44%) | 26 (43%) |
| Pseudophakic | 26 (25%) | 9 (27%) | 2 (22%) | 15 (25%) |
AL = axial length; BM = Bruch membrane; CMT = central macular thickness; D = diopters; logMAR = logarithm of the minimum angle of resolution; PA = patchy atrophy; PV = perforating vessel; SCT = subfoveal choroidal thickness; SER = spherical equivalent; VA = visual acuity.
Quantitative data are expressed as their mean (standard deviation); qualitative data are expressed as their absolute count and percentage proportion.
Missing data.
Pathways to Patchy Atrophy
An inflammatory or neovascular precursor lesion preceding PA development was observed in 42 eyes (41%). Inflammatory precursors consistent with PIC were the most frequent and were identified in 33 out of 103 (32%) eyes included in the study (Figs 1 and 2). In most eyes with PIC (31/33, 94%), PA onset coincided with active inflammation, while in two (6%) eyes, PA developed 0.17 ± 0.11 years after lesion inactivation.
Figure 1.
Long-term follow-up of a highly myopic eye with an inflammatory pattern of PA. The figure illustrates the long-term multimodal imaging in the follow-up of a highly myopic eye with an inflammatory pattern of PAs (AL: 31.59 mm, age at PA onset: 66 years). At baseline, infrared imaging of the posterior pole (A) shows the presence of branched lacquer cracks. The corresponding spectral-domain OCT (B) confirms the absence of structural atrophy or inflammatory activity and shows the presence of large choroidal vessels. After 3 years of follow-up, infrared imaging (C) demonstrates the development of multiple punctate inner choroidopathy (PIC) lesions located along the lacquer cracks. The adjacent horizontal OCT B-scans (D–F) reveal active inflammatory foci, characterized by localized choroidal thickening (arrowhead), inflammatory pigment epithelial detachments (star) and disruption of Bruch membrane (circle). At 4 years after baseline, a marked extension of patchy atrophic areas involving the posterior pole has developed, as shown on infrared image and further highlighted by fundus auto-fluorescence (G and H). At this stage, OCT imaging (green boxed area in G) shows an inactive stage of the disease, characterized by outer nuclear layer herniation and Bruch membrane defects, in the absence of active inflammatory foci. After 8 years of follow-up, the posterior pole is extensively involved, with the development of macular atrophy leading to severe visual loss, as illustrated by fundus pseudocolor imaging (I) and fundus auto-fluorescence imaging (L). The disease course is further complicated by the presence of subretinal macular fibrosis (blue boxed area), representing the sequela of a previous neovascular event. AL = axial length; PA = patchy atrophy.
Figure 2.
Rapidly progressive inflammatory pattern leading to macular atrophy. This figure illustrates an inflammatory pattern of patchy atrophy development, characterized by a rapid and aggressive disease course (AL: 28.98 mm; age at PA onset: 65 years). Infrared imaging obtained at annual follow-up visits (A–E) documents the stepwise evolution of the disease. Early infrared images (A and B) show the appearance of focal inflammatory lesions, which progressively enlarge and coalesce over time. Corresponding OCT B-scans (upper panels) demonstrate active inflammatory changes, including outer retinal disruption, inflammatory pigment epithelial detachments, and choroidal thickening. As the disease progresses, OCT imaging (third OCT panel) reveals a neovascularization, with subsequent development of a subretinal fibrosis, which becomes evident in the following scans. This phase is followed by a rapid progression toward macular atrophy, as shown in the later infrared images (D and E) and corresponding OCT sections. At the end of the follow-up at 6 years after baseline, multimodal imaging (F–H) reveals a fully developed macular atrophy, with extensive loss of outer retinal layers and fibrotic changes. At this stage, the final atrophic phenotype is indistinguishable from other forms of macular atrophy, and no residual imaging features allow recognition of the initial inflammatory (PIC-related) origin of the disease. AL = axial length; PA = patchy atrophy; PIC = punctate inner choroidopathy.
Neovascular precursors were identified in 9 eyes (9%), in which PA developed secondary to mMNV-related atrophy (Fig 3). In these eyes, PA occurred at a mean of 1.13 ± 0.86 years after the first detection of mMNV.
Figure 3.
Neovascular pathway leading to PA in pathologic myopia. At baseline (A), a highly myopic eye with peripapillary staphyloma shows no evidence of neovascular or atrophic complications (AL: 29.24 mm; age at PA onset: 55 years). During follow-up (B), extrafoveal myopic macular neovascularization activity develops, with OCT revealing intraretinal cystoid spaces and a double-layer sign on magnified scans. After 6 months (C), a well-demarcated atrophic area becomes evident, corresponding upon OCT with an outer retinal atrophy and increased RPE band hypertransmission. At the end of follow-up, fundus autofluorescence and color fundus photography (D and E) demonstrate a single patchy chorioretinal atrophy lesion involving the posterior pole. AL = axial length; PA = patchy atrophy; RPE = retinal pigment epithelium.
The remaining 61 eyes (59%) were classified as idiopathic PA. Within this group, PA lesions were located most often along the temporal vascular arcades (22 eyes, 21%) or spatially corresponded with an underlying perforating vessel (17 eyes, 16%). Less frequent idiopathic PA presentations included lesions observed after regression of hyperreflective subretinal material (2 eyes, 2%) or arising at the margin of a posterior staphyloma (1 eye, 1%). In 20 eyes (19%), no specific spatial association could be identified (Figure 4, Figure 5, Figure 6, Figure 7, Figure 8).
Figure 4.
Long-term development of perivascular chorioretinal atrophy. This figure illustrates an eye with idiopathic PA developing along the temporal vascular arcades and evolving into perivascular chorioretinal atrophy over a 15-year follow-up (AL: 32.15 mm; age at PA onset: 61 years). At baseline, infrared imaging (A) shows no atrophic lesions. The corresponding OCT scan (orange boxed area) demonstrates preserved outer retinal architecture, without evidence of inflammatory activity or neovascularization. During follow-up, infrared imaging (B and C) reveals the progressive appearance of atrophic changes aligned with the vascular arcades, as highlighted by the dashed boxes. Corresponding OCT scans (orange and blue boxed area) show gradual thinning of the outer retinal layers and retinal pigment epithelium-Bruch membrane interruptions, consistent with a slowly progressive atrophic process, in the absence of inflammatory foci or signs of macular neovascularization. Fundus auto-fluorescence imaging (D) demonstrates well-demarcated, elongated perivascular atrophic areas distributed along the vascular arcades. After 15 years of follow-up, at the final examination (E and F), the patient exhibited an extensive degree of atrophy, characterized by near-complete loss of all retinal layers (green dashed box). The atrophic configuration assumed a “crab-shaped” pattern, extending from the optic disc along both the superior and inferior vascular arcades, while sparing the macular region. AL = axial length; PA = patchy atrophy.
Figure 5.
Idiopathic PA developing over an underlying perivascular vein with a stable longitudinal course. Pseudocolor fundus imaging demonstrates chorioretinal atrophy involving the posterior pole (AL: 28.97 mm; age at PA onset: 56 years). A type 1 posterior staphyloma is clearly evident, together with the presence of lacquer cracks and a small atrophic area in the temporal sector. The atrophic lesion is also clearly identifiable on the corresponding fundus auto-fluorescence image shown alongside. The lower OCT scans depict the structural appearance of the atrophic area. demonstrating the presence of a posterior scleral vessel located directly beneath the atrophic area (arrowheads). PA = patchy atrophy; AL = axial length.
Figure 6.
Retinal pigment epithelium thickening and subsequent reabsorption preceding the development of PA. This figure illustrates the evolution of retinal pigment epithelium (RPE) thickening followed by its progressive absorption, ultimately leading to the development of a PA (AL: 32.41 mm; age at PA onset: 76 years). At baseline, infrared imaging (A) shows no evidence of atrophy. The corresponding OCT B-scan (B) demonstrates localized RPE thickening with associated involvement of the outer retinal layers and choroid, in the absence of definite inflammatory foci. After 5 months of follow-up, OCT (C) shows partial regression of the RPE thickening, accompanied by early structural disorganization of the outer retinal layers. At 7 months after baseline, infrared enface image demonstrates changes become evident with the development of a small atrophic area in the temporal sector (D). The corresponding OCT (E) demonstrates a complete reabsorption of the previously thickened RPE material, associated with irreversible outer retinal loss, without disruption of Bruch membrane and without frank RPE loss. AL = axial length; FU = follow-up; PA = patchy atrophy.
Figure 7.
Idiopathic PA developing along the edge of a posterior staphyloma. Pseudocolor fundus photography shows a localized atrophic area at the edge of a staphyloma, highlighted in the magnified view (AL: 27.11 mm; age at PA onset: 68 years). The corresponding infrared image confirms the presence of two well-demarcated atrophic lesions located nasally to the macula along the staphyloma border. The upper OCT B-scan (green dashed box) demonstrates atrophy of the outer retinal layers with preservation of the Bruch membrane–retinal pigment epithelium (BM–RPE) complex. In contrast, the lower OCT B-scan shows disruption of the BM–RPE complex. In both cases, there is no evidence of inflammatory scars or neovascular changes. AL = axial length; PA = patchy atrophy.
Figure 8.
Idiopathic patchy atrophy in an asymptomatic young patient. This figure illustrates a case of idiopathic development of patchy atrophies in an asymptomatic 42-year-old woman (AL 29.46 mm). At baseline, infrared imaging (A) and the corresponding OCT B-scan (B) show no detectable abnormalities on en face imaging or cross-sectional analysis. In particular, no lacquer cracks or signs of outer retinal or choroidal pathology are evident. During follow-up, infrared imaging (C) reveals the appearance of a small focal atrophic area located temporally to the macula. The corresponding OCT B-scan (D) demonstrates structural alterations at the site of atrophy. The magnified OCT view (blue dashed circle) highlights a disruption of the Bruch membrane-retinal pigment epithelium complex, associated with herniation of the outer nuclear layer and outer retinal atrophy. No signs of inflammatory or neovascular activity were detected, and no posterior scleral vessels were identified in spatial correspondence with the atrophic area. AL = axial length.
Baseline Structural Differences between PA Pathways
Baseline structural characteristics differed between the PA pathways (Table S1, available at www.ophthalmologyscience.org). Inflammatory PAs, as compared with idiopathic PAs, showed a significantly higher number of pre-existing lesions at baseline (P < 0.001), and higher prevalences of pre-existing BM ruptures (P < 0.001) and pre-existing V-shaped configurations (P = 0.04).
Lacquer cracks were observed more frequently in inflammatory PAs than in idiopathic PAs (P = 0.03), whereas no difference in that parameter was observed between neovascular PAs and idiopathic PAs. Subfoveal choroidal thickness tended to be lower in eyes with idiopathic PAs, although pairwise differences were not statistically significant. No significant differences were observed between PA pathways with respect to the prevalences of perforating scleral vessels and posterior staphylomas, baseline VA, or sex distribution.
Late-Stage Complications
Late-stage complication profiles differed between PA pathways (Table 2). In the overall cohort, the incidence rate was 8.53 events per 100 eye-years (95% CI, 6.29–11.31) for MNV, 5.79 events per 100 eye-years (95% CI, 4.07–7.98) for fibrosis, and 1.56 events per 100 eye-years (95% CI, 0.83–2.66) for MA. Across all outcomes, incidence rates were lowest in idiopathic eyes and highest in neovascular eyes, with inflammatory eyes showing intermediate values (Table S2, available at www.ophthalmologyscience.org).
Table 2.
Late-Stage Clinical Manifestations and Comparison of Macular Complications According to PA Pathways
| Eyes’ Characteristics | Total (n = 103 Eyes) | Inflammatory Pathway (n = 33) | Neovascular Pathway (n = 9) | Idiopathic Pathway (n = 61) |
|---|---|---|---|---|
| Final VA in logMAR | 0.50 (0.63) | 0.61 (0.76) | 0.56 (0.47) | 0.43 (0.56) |
| Final PA lesion count (n) | 2.39 (2.44) | 4.48 (3.36) | 1.54 (1.01) | 1.92 (1.48) |
| MNV cases at the end of FU | 48 (47%) | 26 (79%) | 9 (100%) | 13 (21%) |
| Subretinal fibrosis cases at the end of FU | 37 (36%) | 17 (52%) | 9 (100%) | 11 (18%) |
| MA cases at the end of FU | 13 (13%) | 7 (21%) | 3 (33%) | 3 (5%) |
| Number of anti-VEGF injection | 2.89 (4.02) | 4.67 (4.85) | 7.88 (3.06) | 1.18 (2.28) |
| Outcome | Comparison | OR | 95% CI | P Value |
|---|---|---|---|---|
| MNV (ref: no) | Neovascular vs. Idiopathic | N/A | N/A | N/A |
| Inflammatory vs. Idiopathic | 13.2 | 4.7–36.9 | <0.001∗ | |
| Subretinal fibrosis (ref: no) | Neovascular vs. Idiopathic | 1.04 | 0.86–1.51 | 0.6 |
| Inflammatory vs. Idiopathic | 4.8 | 1.8–12.9 | 0.002∗ | |
| MA (ref: no) | Neovascular vs. Idiopathic | 9.7 | 1.7–54.2 | 0.014 |
| Inflammatory vs. Idiopathic | 5.4 | 1.3–22.1 | 0.021 |
CI = confidence interval; FU = follow-up; logMAR = logarithm of the minimum angle of resolution; MA = macular atrophy; MNV = macular neovascularization; N/A = not available; OR = odds ratio; PA = patchy atrophy; VA = visual acuity.
Quantitative data are expressed as their mean (standard deviation); qualitative data are expressed as their absolute count and percentage proportion. Odds ratio not reported due to quasi-complete separation and resulting model instability. Significant associations are presented in bold.
Significant associations (P < 0.05) are marked with an asterisk (∗).
Among idiopathic PA eyes, new MNV developed in 13 of 61 eyes (21%), subretinal fibrosis in 11 eyes (18%), and MA in 3 eyes (5%). In eyes with neovascular PA, MNV and subsequent subretinal fibrosis were observed in all eyes (9/9, 100%), while MA was documented in 3 (33%) eyes. In eyes with inflammatory PA, MNV was present in 25 of 32 eyes (78%), subretinal fibrosis developed in 16 eyes (50%), and MA developed in 7 eyes (22%). Compared with the eyes with idiopathic PA, eyes with inflammatory PA showed higher odds of MNV (odds ratio [OR] 13.2; 95% CI, 4.7–36.9; P < 0.001). Subretinal fibrosis developed more frequently also in eyes with inflammatory PAs than in eyes with idiopathic PAs (OR 4.8; 95% CI, 1.8–12.9; P = 0.002) (Table 2).
Progression to MA differed between the PA pathways. Both inflammatory PAs and neovascular PAs showed higher odds of developing MA compared with eyes with idiopathic PAs (inflammatory PA: OR = 5.4; 95% CI, 1.3–22.1; P = 0.02; neovascular PA: OR = 9.7; 95% CI, 1.7–54.2; P = 0.01), whereas no significant difference was observed between eyes with inflammatory PAs and eyes with neovascular PAs.
Classification Tree Analysis
Exploratory classification and regression tree analysis selected baseline lesion burden as the first splitting variable among PA patterns. Eyes with fewer than 2.75 lesions were more frequently classified as idiopathic PA, whereas eyes with higher lesion counts were more often classified as inflammatory PA (Fig S2, available at www.ophthalmologyscience.org).
Within the subgroup with a lower lesion count, age at PA onset and the presence of posterior staphyloma were selected as subsequent splits, with idiopathic PA more frequently observed in older eyes and in eyes with staphylomas. A smaller subset of eyes with a lower lesion count but younger age was classified as eyes with inflammatory PAs.
Among eyes with a higher lesion count, baseline VA was selected as an additional splitting variable, with inflammatory PAs more frequently observed in eyes with lower baseline VA.
Atrophy Progression According to PA Patterns
Mean AGR in the overall cohort was 0.73 ± 1.06 mm2/year (95% CI 0.52–0.94). The AGR values differed between PA patterns, with lower values observed in eyes with idiopathic PAs (0.41 ± 0.70 mm2/year), intermediate values in eyes with neovascular PA (0.68 ± 0.73 mm2/year), and the highest values in eyes with inflammatory PA (1.34 ± 1.39 mm2/year, P < 0.001; Fig 9). Detailed AGR values across the different idiopathic subtypes are reported in Table S3 (available at www.ophthalmologyscience.org). Within the same PA pathway, higher baseline lesion burden (β = 0.31; 95% CI = 0.11–0.50, P = 0.005) and inflammatory PAs (β = 0.71; 95% CI = 0.24–1.19, P = 0.004) were associated with a faster atrophy progression (Table 3).
Figure 9.
Progression of PA area over time according to etiologic subtype. Scatterplots illustrate the longitudinal change in atrophic area (mm2) as a function of time since the first visit (months). Each dot represents an individual eye measurement, while the solid line indicates the fitted linear regression model with shaded 95% confidence intervals. The upper panel shows the overall progression of PAs in the entire cohort, demonstrating a steady increase in atrophic area over time. Stratified analyses are presented in the lower panels. Eyes with inflammatory PAs exhibited the steepest slope, indicating the most rapid enlargement of atrophic lesions. Idiopathic PAs showed an intermediate rate of progression with a more gradual increase in atrophy size. In contrast, neovascular PAs demonstrated the slowest progression, with smaller increments in atrophic area over comparable follow-up periods. PA = patchy atrophy.
Table 3.
Predictors of Atrophy Growth Rate in Eyes with PA
| Predictor | Regression Estimate (mm2/yr) | Standard Error | 95% CI | P Value |
|---|---|---|---|---|
| Age at atrophy onset (for each year) | 0.01 | 0.008 | –0.003 to 0.032 | 0.12 |
| Male gender (ref: no) | –0.184 | 0.22 | –0.54 to 0.31 | 0.58 |
| Number of atrophy lesions at baseline (for each lesion) | 0.31 | 0.1 | 0.11–0.50 | 0.004∗ |
| Staphyloma (ref: no) | –0.11 | 0.22 | –0.60 to 0.38 | 0.65 |
| Inflammatory PA (ref: no) | 0.71 | 0.25 | 0.24–1.19 | 0.004∗ |
CI = confidence interval; PA = patchy atrophy.
Significant associations are presented in bold.
Significant associations (P < 0.05) are marked with an asterisk (∗).
Discussion
In this multicenter longitudinal study, we found that PAs in highly myopic eyes emerged as a heterogeneous phenotype arising through different precursor pathways rather than a single pathogenic process. Using longitudinal imaging, we identified 3 main trajectories leading to PAs: inflammatory, neovascular, and idiopathic. Each pathway was associated with distinct structural features, complication profiles, and rates of atrophy expansion.
Approximately one-third of eyes in our cohort showed imaging features consistent with PICs. This prevalence was higher than previously reported and was likely related to the availability of images taken prior to the PA onset. It allowed the detection of prior inflammatory activity that might no longer be detectable once atrophy has developed. As PAs progressed, different pathogenic routes converged toward a common atrophic appearance, making inflammatory and noninflammatory forms difficult to distinguish in advanced stages. It may be taken into account that a referral bias due to the tertiary nature of the centers participating on our study could not be excluded and might have led to an overproportion of eyes with inflammatory PAs in our study cohort.
The role of inflammation in the development of PAs has only recently been explored.25,27, 28, 29, 30 Hady et al13 reported the occurrence of active PIC lesions in highly myopic eyes with PAs, suggesting that inflammatory damage may evolve into atrophy eventually indistinguishable from idiopathic forms of PAs. Our data extended these observations by showing that inflammatory lesions were frequently identifiable before the development of PAs and were associated with a higher lesion burden, more frequent BM defects, and a higher rate of complications. These findings were consistent with the hypothesis that structural breaks in BM may expose retinal antigens and trigger localized inflammatory responses, as previously proposed.14 The high prevalence of lacquer cracks in eyes with inflammatory PAs supported this notion and suggests a potential broader role in the pathogenesis of myopic maculopathy.8
A neovascular pathway, as compared to the inflammatory pathway for the development of PAs, was less common but clearly identifiable. Extrafoveal mMNV preceded the development of PAs in a subset of eyes, and atrophy developed after a variable interval. Despite anti-VEGF therapy, the natural history of mMNV often included progressive chorioretinal atrophy12,31 likely related to repeated exudation, ischemic damage, and fibrotic remodeling.9,12 In this context, PAs may represent a late structural consequence of sustained neovascular disease. Of note, foveal mMNVs were excluded from this study, as the progression toward extensive MA has been known.6 Here, we focused on PAs developing outside of the foveal area.
The majority of eyes included in our study, however, did not show signs of inflammatory or neovascular precursors and were classified as eyes with idiopathic PAs. One may discuss that existing and potentially ongoing axial elongation might have played a pathogenetic role in these eyes. In these situations, PAs might have arisen out of preceding focal BM ruptures or lacquer cracks, without inflammation triggering the event. Interestingly, many idiopathic PAs were located along the temporal vascular arcades or were spatially correlated with perforating scleral vessels. It might suggest that local anatomic features may influence the development of PAs.9,32 Perivascular chorioretinal atrophy has recently been described and hypothesized to be secondary to focal anterior scleral protrusions.33 Similarly, perforating scleral vessels have been proposed to represent points of reduced structural resistance within the sclera, potentially predisposing adjacent tissues to focal damage during progressive globe elongation.34 The observations made in our study are descriptive and hypothesis-generating, supporting the concept that mechanical forces in highly myopic eyes may act in a spatially heterogeneous manner.
From a clinical perspective, although PA subtypes may appear similar once advanced, their earlier trajectories differed. Idiopathic PAs generally showed a lower lesion burden and fewer complications, whereas inflammatory PAs were associated with more frequent neovascularization, fibrosis, and faster atrophy enlargement.30 Neovascular PAs showed an intermediate profile. Our results align with the findings obtained in previous studies demonstrating that inflammatory phenotypes were associated with a more rapid and aggressive atrophy development and progression.35 In addition, MNV were more often associated with, and were more active in eyes with inflammatory PAs. Therefore, early recognition of inflammatory precursors is critical, as it carries important prognostic implications and may justify closer surveillance, earlier therapeutic intervention and adapted patient information, even when late-stage atrophy appears clinically similar.
Our study has limitations, including its retrospective design, relatively small sample size, and a potential referral bias, limiting the interpretation of the reported prevalences. However, it distinguished three PA precursor patterns, highlighting the inflammatory component as one of the potential causal factors in the development of PAs in highly myopic patients. In addition, although no mixed forms were identified within our cohort, the possibility of an unrecognized, subclinical overlap between inflammatory, idiopathic, and neovascular components cannot be excluded. Such potential misclassification may have affected our findings, particularly regarding the estimated progression rates of atrophic lesions and clinical outcomes. A further limitation concerns the heterogeneity within the idiopathic group. Different entities, such as atrophy associated with vascular arcades, perforating vessels, and posterior staphyloma, may have distinct underlying mechanisms and progression patterns. However, the number of cases within each subgroup in our cohort was insufficient to allow for a robust statistical analysis. Consequently, these forms were grouped together based on the absence of both neovascular and inflammatory components. This approach may have masked potential differences within the group. Imaging protocols and follow-up intervals slightly differed between the participating centers, but the criteria for the diagnosis of PAs and their precursors and sequels were uniform. An additional limiting factor of our study was that our analysis was restricted to the posterior pole, without a systematic evaluation of peripheral retinal changes. It might have led to an underdetection of more peripherally located atrophic or precursor lesions and, consequently, to an incomplete characterization of the disease extent. Moreover, histoplasmin skin testing was not performed, and therefore, we cannot formally exclude presumed ocular histoplasmosis syndrome in all cases. A strength of our investigation was the availability of longitudinal MMI prior to the PA onset, allowing a reconstruction of disease development. Also, our analysis focused on extrafoveal MNVs, excluding foveal lesions, which represent the majority of cases in high myopia. To make this distinction, we applied the MNV classification criteria used in early clinical trials on MNVs, as no criteria currently exist to clearly differentiate PAs from MA, which might have introduced potential biases.
In summary, PAs in highly myopic eyes represented the outcome of 3 different pathogenic pathways: inflammatory, neovascular, and idiopathic. While late-stage atrophy converged toward a common appearance, inflammatory, neovascular, and idiopathic forms differed in their structural features, complication rates, and atrophy progression. Recognizing these pathways may help refine risk stratification and guide follow-up strategies in patients with pathologic myopia.
Declaration of Generative AI and AI-Assisted Technologies in the Writing Process
During the preparation of this work, the authors used ChatGPT (OpenAI) for linguistic purposes. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Manuscript no. XOPS-D-26-00310
Footnotes
Supplemental material available at www.ophthalmologyscience.org.
Disclosure(s):
All authors have completed and submitted the ICMJE disclosures form.
The authors have no proprietary or commercial interest in any materials discussed in this article.
Support for Open Access publication was provided by Fondation Adolphe de Rothschild Hospital.
HUMAN SUBJECTS: Human subjects were included in this study The study received approval from the institutional review boards of all participating centers. All research adhered to the tenets of the Declaration of Helsinki. All participants provided informed consent.
No animal subjects were used in this study.
Author Contributions:
Conception and design: Del Fabbro
Data collection: Del Fabbro, Motloch, Philippakis
Analysis and interpretation: Del Fabbro, Jonas, Panda-Jonas, Cicinelli, Battaglia Parodi, Miserocchi, Couturier, Philippakis
Obtained funding: N/A
Overall responsibility: Del Fabbro, Jonas, Cicinelli, Battaglia Parodi, Couturier, Philippakis
Supplementary Data
Flowchart of study eye selection and exclusion criteria.
Decision tree model for etiologic classification of patchy atrophy. Decision tree derived from supervised classification analysis illustrating the hierarchical predictors associated with idiopathic, inflammatory, and neovascular patchy atrophy. The model was trained on the overall cohort and recursively partitions patient population according to clinical and imaging variables to maximize class separation. At each node, the splitting variable and threshold are reported, along the predicted class, and the corresponding class probabilities (Idiopathic/Neovascular/Inflammatory). Terminal nodes represent the final classification output.
References
- 1.Holden B.A., Fricke T.R., Wilson D.A., et al. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology. 2016;123:1036–1042. doi: 10.1016/j.ophtha.2016.01.006. [DOI] [PubMed] [Google Scholar]
- 2.Chen Y., Xiong R., Zhang J., et al. Ten-year change in visual function and incidence of visual impairment in highly myopic children and adults. Invest Ophthalmol Vis Sci. 2025;66:2. doi: 10.1167/iovs.66.1.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tideman J.W.L., Snabel M.C.C., Tedja M.S., et al. Association of axial length with risk of uncorrectable visual impairment for Europeans with myopia. JAMA Ophthalmol. 2016;134:1355. doi: 10.1001/jamaophthalmol.2016.4009. [DOI] [PubMed] [Google Scholar]
- 4.Van Newkirk M.R. The Hong Kong vision study: a pilot assessment of visual impairment in adults. Trans Am Ophthalmol Soc. 1997;95:715–749. [PMC free article] [PubMed] [Google Scholar]
- 5.Cedrone C., Nucci C., Scuderi G., et al. Prevalence of blindness and low vision in an Italian population: a comparison with other European studies. Eye. 2006;20:661–667. doi: 10.1038/sj.eye.6701934. [DOI] [PubMed] [Google Scholar]
- 6.Fang Y., Yokoi T., Nagaoka N., et al. Progression of myopic maculopathy during 18-year follow-up. Ophthalmology. 2018;125:863–877. doi: 10.1016/j.ophtha.2017.12.005. [DOI] [PubMed] [Google Scholar]
- 7.Carlà M.M., Boselli F., Giannuzzi F., et al. Longitudinal progression of myopic maculopathy in a long-term follow-up of a European cohort. Ophthalmol Retina. 2025;9:774–786. doi: 10.1016/j.oret.2025.02.015. [DOI] [PubMed] [Google Scholar]
- 8.Ohno-Matsui K., Kawasaki R., Jonas J.B., et al. International photographic classification and grading system for myopic maculopathy. Am J Ophthalmol. 2015;159:877–883.e7. doi: 10.1016/j.ajo.2015.01.022. [DOI] [PubMed] [Google Scholar]
- 9.Ohno-Matsui K., Jonas J.B., Spaide R.F. Macular Bruch membrane holes in highly myopic patchy chorioretinal atrophy. Am J Ophthalmol. 2016;166:22–28. doi: 10.1016/j.ajo.2016.03.019. [DOI] [PubMed] [Google Scholar]
- 10.Hayashi K., Shimada N., Moriyama M., et al. Two-year outcomes of intravitreal bevacizumab for choroidal neovascularization in Japanese patients with pathologic myopia. Retina. 2012;32:687–695. doi: 10.1097/IAE.0b013e3182278bae. [DOI] [PubMed] [Google Scholar]
- 11.Hampton G.R., Kohen D., Bird A.C. Visual prognosis of disciform degeneration in myopia. Ophthalmology. 1983;90:923–926. doi: 10.1016/s0161-6420(83)80018-9. [DOI] [PubMed] [Google Scholar]
- 12.Cicinelli M.V., La Franca L., De Felice E., et al. Long-term incidence and risk factors of macular fibrosis, macular atrophy, and macular hole in eyes with myopic neovascularization. Ophthalmol Retina. 2022;6:1231–1240. doi: 10.1016/j.oret.2022.06.009. [DOI] [PubMed] [Google Scholar]
- 13.Hady S.K., Xie S., Freund K.B., et al. Prevalence and characteristics of multifocal choroiditis/punctate inner choroidopathy in pathologic myopia eyes with patchy atrophy. Retina. 2022;42:669–678. doi: 10.1097/IAE.0000000000003383. [DOI] [PubMed] [Google Scholar]
- 14.Cicinelli M.V., Del Fabbro S., Ramtohul P., et al. Idiopathic multifocal choroiditis/punctate inner choroidopathy as a secondary inflammatory reaction to lacquer cracks: a structural and temporal analysis. Am J Ophthalmol. 2025;282:263–277. doi: 10.1016/j.ajo.2025.11.019. [DOI] [PubMed] [Google Scholar]
- 15.Flitcroft D.I., He M., Jonas J.B., et al. IMI – defining and classifying myopia: a proposed set of standards for clinical and epidemiologic studies. Invest Ophthalmol Vis Sci. 2019;60:M20. doi: 10.1167/iovs.18-25957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Laser photocoagulation of subfoveal neovascular lesions of age-related macular degeneration. Arch Ophthalmol. 1993;111:1200. doi: 10.1001/archopht.1993.01090090052019. [DOI] [PubMed] [Google Scholar]
- 17.Gangaputra S., Agarwal A., Norel J.O., et al. Evidence and consensus-based imaging guidelines in multifocal choroiditis with panuveitis and punctate inner choroiditis—Multimodal imaging in Uveitis (MUV) taskforce report 5. Am J Ophthalmol. 2025;276:272–285. doi: 10.1016/j.ajo.2025.04.018. [DOI] [PubMed] [Google Scholar]
- 18.Jabs D.A., Brezin A.P., Levinson R.D., et al. Classification criteria for punctate inner choroiditis. Am J Ophthalmol. 2021;228:275–280. doi: 10.1016/j.ajo.2021.03.046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Classification criteria for punctate inner choroiditis. Am J Ophthalmol. 2021;228:275–280. doi: 10.1016/j.ajo.2021.03.046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Sadda S.R., Guymer R., Holz F.G., et al. Consensus definition for atrophy associated with age-related macular degeneration on OCT. Ophthalmology. 2018;125:537–548. doi: 10.1016/j.ophtha.2017.09.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Csincsik L., Cheung C.M.G., Bannon F., et al. Agreement between color, fluorescein angiography, and spectral domain optical coherence tomography in the detection of macular fibrosis in neovascular age-related macular degeneration. Am J Ophthalmol. 2025;272:126–135. doi: 10.1016/j.ajo.2025.01.011. [DOI] [PubMed] [Google Scholar]
- 22.Del Fabbro S., Arrigo A., Bianco L., et al. Dome-shaped macula curvature: a new quantitative metric to assess dome-shaped macula subtypes. Transl Vis Sci Technol. 2025;14:9. doi: 10.1167/tvst.14.8.9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ohno-Matsui K. Proposed classification of posterior staphylomas based on analyses of eye shape by three-dimensional magnetic resonance imaging and wide-field fundus imaging. Ophthalmology. 2014;121:1798–1809. doi: 10.1016/j.ophtha.2014.03.035. [DOI] [PubMed] [Google Scholar]
- 24.Xu X., Fang Y., Uramoto K., et al. Clinical features of lacquer cracks in eyes with pathologic myopia. Retina. 2019;39:1265–1277. doi: 10.1097/IAE.0000000000002168. [DOI] [PubMed] [Google Scholar]
- 25.Zhang X., Zuo C., Li M., et al. Spectral-domain optical coherence tomographic findings at each stage of punctate inner choroidopathy. Ophthalmology. 2013;120:2678–2683. doi: 10.1016/j.ophtha.2013.05.012. [DOI] [PubMed] [Google Scholar]
- 26.Moukadem O., Motloch K., Couanon M., et al. Incidence and enlargement rate of myopic macular neovascularization-related atrophy in highly myopic eyes. Retina. 2025;46:622–629. doi: 10.1097/IAE.0000000000004731. [DOI] [PubMed] [Google Scholar]
- 27.Watzke R.C., Packer A.J., Folk J.C., et al. Punctate inner choroidopathy. Am J Ophthalmol. 1984;98:572–584. doi: 10.1016/0002-9394(84)90243-5. [DOI] [PubMed] [Google Scholar]
- 28.Essex R.W. Punctate inner choroidopathy. Arch Ophthalmol. 2010;128:982. doi: 10.1001/archophthalmol.2010.157. [DOI] [PubMed] [Google Scholar]
- 29.Gerstenblith A.T., Thorne J.E., Sobrin L., et al. Punctate inner choroidopathy. Ophthalmology. 2007;114:1201–1204.e4. doi: 10.1016/j.ophtha.2006.10.047. [DOI] [PubMed] [Google Scholar]
- 30.Del Fabbro S., Perna L., Introini U., et al. Macular complications predict visual decline in punctate inner choroidopathy. Ophthalmology. 2025;133:258–291. doi: 10.1016/j.ophtha.2025.10.012. [DOI] [PubMed] [Google Scholar]
- 31.Zhang X.J., Chen X.N., Tang F.Y., et al. Pathogenesis of myopic choroidal neovascularization: a systematic review and meta-analysis. Surv Ophthalmol. 2023;68:1011–1026. doi: 10.1016/j.survophthal.2023.07.006. [DOI] [PubMed] [Google Scholar]
- 32.Jonas J.B., Jonas R.A., Bikbov M.M., et al. Myopia: histology, clinical features, and potential implications for the etiology of axial elongation. Prog Retin Eye Res. 2023;96 doi: 10.1016/j.preteyeres.2022.101156. [DOI] [PubMed] [Google Scholar]
- 33.Benahmed R., Dormegny L., Gaudric A., et al. Perivascular chorioretinal atrophy: an unusual feature in pathologic myopia eyes. Am J Ophthalmol. 2025;271:498–506. doi: 10.1016/j.ajo.2024.12.022. [DOI] [PubMed] [Google Scholar]
- 34.Querques G., Corvi F., Balaratnasingam C., et al. Lacquer cracks and perforating scleral vessels in pathologic myopia: a possible causal relationship. Am J Ophthalmol. 2015;160:759–766.e2. doi: 10.1016/j.ajo.2015.07.017. [DOI] [PubMed] [Google Scholar]
- 35.Chen Y.-C., Chen Y.-L., Chen S.-N. Chorioretinal atrophy in punctate inner choroidopathy/multifocal choroiditis: a five-year follow-up study. Ocul Immunol Inflamm. 2022;30:270–275. doi: 10.1080/09273948.2020.1869269. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Flowchart of study eye selection and exclusion criteria.
Decision tree model for etiologic classification of patchy atrophy. Decision tree derived from supervised classification analysis illustrating the hierarchical predictors associated with idiopathic, inflammatory, and neovascular patchy atrophy. The model was trained on the overall cohort and recursively partitions patient population according to clinical and imaging variables to maximize class separation. At each node, the splitting variable and threshold are reported, along the predicted class, and the corresponding class probabilities (Idiopathic/Neovascular/Inflammatory). Terminal nodes represent the final classification output.