To describe a series of chorioretinal folds (CRFs) representing a clinical sign that may be associated with multiple systemic, orbital, and ophthalmologic disorders. We report the associations with systemic disease and describe 3 stages of a CRF-related maculopathy.
Observational, retrospective case series.
We reviewed 57 affected eyes from 40 patients with the clinical sign of CRF from 1 of 2 academic institutions. A careful review of the medical histories and systemic diagnostic evaluations were conducted. Imaging studies were conducted.
The mean age at diagnosis was 64 ± 17 years. Most eyes (n = 18) were hyperopic (+2.60 ± +2.90 diopters). There were 20 patients (50%) with some form of autoimmune disorder. Overall, the mean presenting visual acuity was 20/50, declining slightly to 20/60 over 19 ± 30 months. Ten eyes had stage 3 CRF-related maculopathy, more common in older individuals with more chronic CRFs. Four stage 3 eyes had associated choroidal neovascularization, and these eyes had 20/60 presenting visual acuity that decreased to 20/100 over approximately 1.5 years. Stage 3 eyes without choroidal neovascularization had a mean presenting visual acuity of 20/40 that decreased to 20/65 over 2.1 years.
CRFs are associated with numerous ophthalmic and systemic disorders. A careful medical history and evaluation are essential. We describe 3 stages of a unique CRF-related maculopathy. Stage 3 resembles occult choroidal neovascularization, occurs primarily in older individuals with chronic CRFs, and is accompanied by a slow deterioration in central acuity.
Chorioretinal folds (CRFs) were first described in 1884 by Nettleship in a patient with papilledema. Birch-Hirschfeld and Siegfried demonstrated with histopathologic analysis that the folds extended into the neurosensory retina. CRFs represent a relatively uncommon clinical sign that should prompt the clinician to consider numerous potential causes, ranging from minor hyperopia to serious systemic disorders such as neoplastic, infectious, or immunologic disorders.
In the 1950s, the suspicion of systemic disease associations with CRFs was heightened and CRFs were considered to represent an ominous sign. By 1959 and 1962, respectively, Hedges and Leopold and then Wolter pointed out that CRFs could be the presenting sign of an orbital tumor indenting the sclera. Over time, clinicians have become more adept at diagnosing and detecting CRFs during funduscopic examination and have associated this important clinical sign with ocular, orbital, and systemic conditions. An important, and some may argue obvious, clarification is that CRFs represent a clinical sign, rather than a specific diagnosis.
The disease-related categories associated with CRFs include inflammatory, neoplastic, infiltrative, and infectious. The list of specific diagnoses that have been associated with CRFs has increased along with improvements in our diagnostic capabilities, thereby, leaving the exclusionary category of idiopathic CRFs decreasing proportionately. Common diagnoses that have been associated with CRFs include thyroid eye disease, posterior scleritis, tumors, hypotony, uveal effusion syndrome, hyperopia, scleral buckling surgery, and age-related macular degeneration (AMD). There is sufficient evidence that as soon as the clinician recognizes the presence of CRFs as a clinical sign, a prompt systemic investigation for a specific diagnosis or cause is mandatory.
Anatomically, CRFs represent undulations of the inner choroid, the Bruch membrane, and overlying retinal pigment epithelium (RPE), as well as secondarily affecting the overlying neurosensory retina. In 1972, Newell proposed that CRFs developed as a result of the strong connection between Bruchs membrane and the underlying choriocapillaris. As the choroid swells or expands, the overlying Bruchs membrane is forced into folds, thus leading to the clinical appearance of CRFs. Friberg elegantly provided a biomechanical analysis that further explained the evolution of CRFs. He proposed that CRFs develop as a result of the stress-and-strain relationship that occurs between the sclera and choroid, with a reduction in the area of the inner lining of the sclera, resulting in a buckling force affecting the choroid from either scleral thickening or shrinkage. Furthermore, Friberg suggested that CRFs represent a potential risk factor for RPE atrophy and potentially irreversible injury to the Bruch membrane, leading to an angiographic appearance that resembles angioid streaks.
Herein, we present a series of patients with the clinical sign of CRFs that were reviewed from 2 tertiary academic centers (the University of Minnesota and Emory University). The associated systemic and ophthalmologic diagnoses represent broad ranging causes that may contribute to the formation of CRFs. Furthermore, we describe a specific maculopathy (CRF maculopathy) that evolves in selected cases of chronic CRFs, especially in older patients.
The Institutional Review Boards of the University of Minnesota and Emory University approved the retrospective review of patient data for this study. We reviewed the medical records of consecutive patients with the clinical sign of CRFs. Forty patients were identified with the clinical sign from either the medical record database or by searching the fundus photographic imaging database. Demographic information, including age at presentation, ocular and medical history, duration of follow-up, potential causes, and associated diagnoses (either systemic, ophthalmologic, or both), were recorded for each patient. The best-corrected Snellen visual acuity, funduscopic features, refraction, axial length (when available), clinical diagnosis requested with the imaging studies (fundus photography, ocular coherence tomography, fluorescein angiography [FA]), and specific treatment methods also were noted. All statistical analysis was performed using a paired Student t test. Statistical significance was defined as P < .05.
Fifty-seven eyes of 40 patients were identified with a history of CRFs. Seventeen were bilateral, whereas 23 (40%) were unilateral. The mean age ± standard deviation at presentation was 64 ± 17 years, with ages ranging from 27 to 96 years. Other demographic features collected are listed in Table 1 . Nineteen patients (48%) had a history of hypertension (although the presence of hypertension was not documented in 21 of our records) and 11 patients (28%) had a history of autoimmune disease. Eighteen of those in whom a refraction was documented were hyperopic with a mean refraction of +2.4 diopters. Eleven patients (28%) had clinical features of AMD. The mean follow-up was 19 ± 30 months (range, 0 to 156 months).
|Total no. of patients||40|
|Total no. of eyes||57|
|Mean ± SD||19 ± 30|
|Range||0 to 156|
|Mean age ± SD (y)||63 ± 17|
|Mean refraction ± SD (diopters)||+2.60 ± 2.90|
|Age-related macular degeneration||11|
|Uveal effusion syndrome||4|
|Thyroid eye disease||3|
|Inflammatory bowel disease||3|
|Systemic lupus erythematosus||1|
Selected CRFs should be considered secondary CRFs and are related to a specific abnormality located at the chorioretinal interface such as a scar or fibrosis or a choroidal neovascular complex ( Figure 1 ). Six eyes were identified as having secondary folds that resulted from scarring or fibrosis related to exudative AMD (n = 4), from other chorioretinal scarring (n = 1), or from a scleral buckle (n = 1).
Primary CRFs involve pathologic features that affect the choroid or sclera directly and typically are arranged as linear or parallel folds that generally extended directly through the central macula. In 19 eyes, the CRFs projected mostly along the temporal arcades, or less commonly radiated from the optic nerve (5 eyes). Others had varying and more random distribution of CRFs (8 eyes). A significant portion of eyes (n = 14) demonstrated a pigmentary maculopathy ( Figure 2 ). Two patients had a submacular hemorrhage, presumably from a rupture of the Bruch membrane that resulted directly from the folds that were not clearly associated with angiogenesis or neovascularization.
Multiple conditions or diagnoses were associated with the CRFs. Cases identified in our series as listed in decreasing frequency include (note that there is overlap): hypertension (n = 19), hyperopia (n = 18), idiopathic or unidentified etiologies (n = 6), AMD (secondary CRFs; n = 11), autoimmune disease (n = 11), rheumatoid arthritis (n = 5), uveal effusion syndrome (n = 4), posterior scleritis (n = 4), hypotony (n = 3), thyroid eye disease (n = 3), choroidal infiltrates (ie, lymphoma; n = 2), orbital mass (n = 2), inflammatory bowel disease (n = 3), and several isolated associations that included vitiligo, polymyalgia rheumatic, systemic lupus erythematosus, Lyme disease, psoriasis, multiple sclerosis, and ethmoid sinusitis ( Table 1 ). Overall, nearly half (20 of the 40 patients) had some form of autoimmune disorder. Note that some had overlapping conditions such as rheumatoid arthritis or scleritis. Twelve of our patients were pseudophakic at presentation, and information about their specific refractive state before surgery was not available. Therefore, it is possible that some patients in the idiopathic group may represent additional cases with hyperopia.
Excluding patients with secondary CRFs, a comparison of the visual acuity at the initial visit to the latest follow-up visit (mean ± standard deviation, 19 ± 30 months) demonstrated a decline of 1 line of Snellen acuity (20/50 to 20/60, respectively; P = .01). Therefore, a decline in best-corrected visual acuity is a slow process in primary CRFs.
FA results were available for 22 patients. Three progressive angiographic stages were observed. First, in stage 1, there are alternating bands of hyperfluorescence and hypofluorescence characteristic of choroidal folds (n = 8; Figure 3 ). Next, in stage 2, there are beginning to form areas of staining that correspond to early breakdown of the RPE along with breaks in the Bruch membrane (n = 6; Figure 2 ). Finally, in stage 3, the CRF-related maculopathy is best exemplified in Figures 4 and 5 . In the more advanced stage 3, the clinical appearance usually is associated with either a yellow, luteal macular appearance or atrophy of the RPE; plus, there are more prominent areas of RPE hyperplasia. On FA, the early phase has a stippled pattern of hypofluorescence (n = 8), whereas in the later frames of the angiogram, there is increasing stippled hyperfluorescence with both staining and mild leakage. This pattern also has been described in occult choroidal neovascularization (CNV), referred to as late leakage of undetermined source by the Macular Photocoagulation Study ( Figures 4 and 5 ).
Ten eyes of 8 patients had stage 3 macular changes. The mean age of this group was 68 years, as compared with patients with stage 1 changes (56 years; P < .01) or stage 2 changes (64 years; P = .2), respectively. In stage 3 patients, the mean Snellen visual acuity was 20/60 at the initial visit and 20/100 at the latest follow-up visit. Half of the stage 3 patients had CRFs extending through the center of the macula, whereas the remainder had folds along the temporal arcades. The cause of the more chronic CRFs with stage 3 CRF-related maculopathy included rheumatoid arthritis, posterior scleritis, and hyperopia (usually > 3 diopters).
Six eyes of 4 patients with stage 3 maculopathy did not require anti-CNV therapy. In this group, the initial visual acuity was 20/40 and the final visual acuity was 20/65 at an average follow-up of 25 months ( P = .06). These data suggest that this group has a poor prognosis with gradual loss in central acuity as soon as stage 3 is reached. Four stage 3 eyes were suspected to have active CNV and were treated. Three of the 4 did not have evidence of CNV at presentation. Two of the 4 had small subretinal hemorrhages at or near the fovea. In the treatment group, mean visual acuity at the initial visit was 20/80, yet it decreased to 20/160 at an average follow-up of 19 months. Statistical analysis did not reach a level of significance because the number in this group was too small. None of the eyes that were treated had a measurable improvement in visual acuity.
Of those treated with stage 3 changes, 2 patients received a periocular corticosteroid injection without an improvement in macular status or central visual acuity. One patient, managed in the era before anti–vascular endothelial growth factor (anti-VEGF), was treated with submacular surgery and experienced a recurrence or persistence of the CNV within 1 month of surgery. Photodynamic therapy then was performed to treat the recurrent CNV successfully. Nevertheless, the visual acuity failed to improve. In the other 3 stage 3 eyes, the treatments performed did not result in a significant improvement in the central visual acuity or in angiographic appearance. During the anti-VEGF era, a single intravitreal bevacizumab injection was administered in 1 stage 3 eye with active leakage and after a failed periocular corticosteroid injection. The angiogram performed 1 month after injection demonstrated a clear resolution of the leakage pattern and the eye did not require further therapy. However, there was no improvement in visual acuity. Separately, an intravitreal ranibizumab was administered to a patient with stage 3 changes and possible CNV who also had an accompanying subretinal hemorrhage. This lesion also regressed after a single injection and did not recur during a 22-month follow-up interval. There were no changes in visual acuity (see Patient 1 below; Figure 6 ).