To describe choroidal findings in dome-shaped macula associated with high myopia using fluorescein angiography (FA), indocyanine green angiography (ICGA), and spectral-domain optical coherence tomography (SD OCT), and to elucidate the mechanism and natural course of serous retinal detachment (RD) associated with dome-shaped macula.
Retrospective, observational case series.
We reviewed longitudinal imaging results of 52 highly myopic eyes with dome-shaped macula. Changes on FA and ICGA were assessed. Retinal, choroidal, and scleral thicknesses and bulge height were measured on SD OCT.
Serous RD was the most common abnormality associated with dome-shaped macula, detected by SD OCT in 44% of the cases with no associated choroidal neovascularization. Significant differences in the proportion of eyes with pinpoint leakage on FA ( P < .001), punctate hypercyanescence on ICGA ( P < .001), and pigment epithelium detachment on SD OCT ( P < .001) were noted inside the inward bulge of the staphyloma between eyes with and without serous RD. Serous RD was not associated with hyperpermeability areas on ICGA. Eyes with serous RD had thicker choroid ( P = .004) and tended to have thicker sclera ( P = .067) and greater bulge height ( P = .079). Choroidal thickness, scleral thickness, and bulge height were positively correlated ( P < .01). All eyes presented a fluctuating course of serous RD during follow-up. Worsening of serous RD was associated with appearance of new punctate hypercyanescent spots on ICGA and leaking points on FA ( P < .001 and P = .016, respectively).
Serous RD in dome-shaped macula was likely caused by choroidal vascular changes, similar to central serous chorioretinopathy, but specifically confined in the inward bulge of the staphyloma and secondary to excessive scleral thickening. Serous retinal detachment showed fluctuating changes over time, with alternating active and inactive stages. Angiographic findings in dome-shaped macula suggest the choroid as a target for possible treatment strategies.
Dome-shaped macula was first described by Gaucher and associates in 2008 by the use of time-domain optical coherence tomography (OCT) as an inward convexity or bulge of the macula that occurred in approximately 10% of highly myopic eyes within the concavity of the posterior staphyloma. Subsequently, enhanced depth imaging optical coherence tomography (EDI OCT) showed that dome-shaped macula is the result of a localized variation in scleral thickness in the macular area. By using swept source OCT, which allowed deeper tissue penetration into the choroid and even sclera, Ellabban and associates reported the precise topography of the posterior pole in eyes with dome-shaped macula, describing a horizontal ridge formed within the posterior staphyloma by uneven thinning of the sclera. More recently, Caillaux and associates described 3 morphologic dome-shaped macula patterns according to spectral-domain OCT (SD OCT) features: round domes (without predominant axis), horizontally oriented oval-shaped domes, and vertically oriented oval-shaped domes.
Vision-threating macular complications typically described in highly myopic patients, including choroidal neovascularization (CNV), retinal pigment epithelial (RPE) changes, macular holes, and foveoschisis, have been reported to be associated with dome-shaped macula as well. Serous retinal detachment (RD) without CNV is a well-established complication responsible for vision loss in dome-shaped macula, but is rarely reported in highly myopic eyes without the presence of dome-shaped macula. To date, however, limited information is available on the role played by choroid dynamics based on fluorescein (FA) and indocyanine green (ICGA) angiography in the pathogenesis of serous RD associated with dome-shaped macula. Interestingly, this complication presents similar ophthalmoscopic and angiographic findings to central serous chorioretinopathy (CSC), as well as tilted disc syndrome.
In the study described herein, we aimed to analyze macular changes in highly myopic eyes with dome-shaped macula imaged by FA and ICGA angiography combined with SD OCT in order to elucidate the mechanism of serous RD development. We also aimed to study the long-term clinical course of dome-shaped macula by analyzing morphologic and angiographic changes over time.
We retrospectively reviewed charts and imaging studies of highly myopic patients diagnosed with dome-shaped macula between October 2009 and September 2013 at the Ophthalmological Unit, Ca’ Granda Foundation-Ospedale Maggiore Policlinico, or at the Eye Clinic, “Luigi Sacco” Hospital, 2 tertiary care centers specialized in the diagnosis and treatment of retinal diseases in Milan, Italy. Patients had been referred to us for diagnosis of visual complains or imaging of suspected fundus anomalies diagnosed in an examination performed routinely. To be included in the chart and imaging review, eyes must have had: (1) refractive error of 6.0 diopters or more, axial length of 26.5 mm or more, or both; (2) unilateral or bilateral dome-shaped macula configuration on OCT according to the description of Gaucher and associates ; and (3) posterior staphyloma resembling Curtin type I or II. We excluded eyes with inferior staphyloma (type V ) alone or associated with tilted disc syndrome in which the macula lays on the edge or slope of the inferior staphyloma; eyes with poor image quality owing to media opacities; and eyes with history of CSC, retinal vascular diseases, or major retinal surgery. Patients with a history or receiving medications potentially at risk of serous RD (eg, corticosteroids) were excluded from the study as well. The study and data accumulation were in conformity with Italian laws. The study was in adherence to the tenets of the Declaration of Helsinki.
Imaging Protocol and Analysis
All patients were scanned with a confocal scanning laser ophthalmoscope (cSLO) after measurement of the best-corrected visual acuity (BCVA) using standard Early Treatment Diabetic Retinopathy Study (ETDRS) charts. Patients underwent multimodal imaging including late-phase FA, late-phase ICGA, and SD OCT scans carried out with the Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany), which allows simultaneous co-localization of posterior structures on bidimensional cSLO images and cross-sectional SD OCT scans with high accuracy. The SD OCT scanning protocol included 9-mm horizontal and vertical scans centered on the fovea in EDI mode and a 3D raster scan centered on the fovea; the baseline scans were set as reference for the subsequent scans that were taken usually every 3 months. Scans were performed between 8 AM and 11 AM for logistical reasons.
Two trained physicians (F.V., L.D.A.) performed imaging analysis to identify fundus abnormalities on multimodal imaging. On late-phase FA, the physicians evaluated presence of leaking points. On late-phase ICGA, they evaluated presence of punctate hypercyanescent spots, areas of hyperpermeability of the choriocapillaris, and subfoveal hypocyanescent oval area. In case of disagreement, a third physician (G.B.) was consulted to achieve an acceptable result. On SD OCT scans, the physicians evaluated presence of macular anomalies or complications (eg, macular holes, foveoschisis, foveal serous RD, fibrovascular RPE detachment suggesting a CNV), subfoveal retinal thickness, choroidal thickness in the fovea and at 1500 μm superiorly and inferiorly to the fovea, subfoveal scleral thickness, and macular bulge height and orientation.
To measure the choroidal thickness, the line corresponding to the internal limiting membrane automatically placed by the built-in automated segmentation software of the SD OCT device was manually moved to the outer part of the hyperreflective line corresponding to the base of the RPE. The line corresponding to the basement membrane was moved to the posterior edge of the choroid, as demarcated by the hyperreflective margin line corresponding to the chorioscleral interface. This method allowed automatic measurement of the choroidal thickness along the 9-mm line scans using the built-in retinal thickness software, with high reproducibility as previously shown. A similar approach was used to measure the subfoveal scleral thickness as well. The macular bulge height was measured on horizontal and vertical SD OCT scans by using the caliper tool of the device between the outer border of the RPE at the fovea and the line tangent to the outer border of the RPE at the bottom of the staphyloma, using a previously described method.
The statistical analyses, including the Fisher exact test, Mann-Whitney U test, and Spearman rank correlation coefficient test, were performed using SPSS statistical software version 20 (SPSS Inc, Chicago, Illinois, USA). The alpha level (type I error) was set at 0.05.
The study included 52 eyes with dome-shaped macula of 32 consecutive white highly myopic patients, whose characteristics are summarized in Table 1 . The dome-shaped macula was bilateral in 20 of 32 patients (62.5%). Mean age was 56.7 ± 15.4 years (range, 20–82 years). Mean spherical equivalent was −14 ± 5.6 diopter (D) (range, −3 to −25 D). Mean axial length was 29.02 ± 1.18 mm (range, 27.59–31.51 mm). Mean BCVA was 0.32 ± 0.24 logMAR.
|Sex||23 female/9 male|
|Age (y), mean ± SD (range)||56.7 ± 15.4 (20–82)|
|Spherical equivalent (D), mean ± SD (range)||−14.0 ± 5.6 (−3 to −25)|
|Axial length (mm), mean ± SD (range)||29.02 ± 1.18 (27.59–31.51)|
|BCVA (logMAR), mean ± SD (range)||0.32 ± 0.24 (0–1.0)|
|Macular abnormalities on SD OCT|
|Foveal SRD||17/52 (32.6%)|
|Extrafoveal schisis||5/52 (9.6%)|
|Lamellar MH||1/52 (2.0%)|
|Dome-shaped macula patterns|
|Horizontally oriented, oval||39/52 (75.0%)|
|Vertically oriented, oval||12/52 (23.1%)|
|Central, round||1/52 (1.9%)|
Seventeen eyes out of 52 (32.7%) had foveal serous RD at baseline SD OCT scan with no associated vitreomacular tractions or CNV. Other macular abnormalities or complications detected by SD OCT included CNV (13 eyes), extrafoveal schisis (5 eyes), foveoschisis (2 eyes), and lamellar macular hole (1 eye). Three-dimensional OCT imaging of the posterior pole showed 39 eyes (75%) with a horizontally oriented oval-shaped dome (ie, showing a convex configuration on the vertical scan through the fovea, while the horizontal scan showed an almost flat RPE line ), 12 eyes (23%) with a central round dome, and only 1 eye (2%) with a vertically oriented oval-shaped dome.
Comparison Between Eyes With and Without Serous Retinal Detachment
Among the 39 eyes without a CNV, 17 (44%) showed serous RD at baseline SD OCT examination. No differences in age ( P = .586), axial length ( P = .358), and BCVA ( P = .226) were found between eyes with serous RD and eyes without serous RD. Among eyes with and without serous RD ( Table 2 ) we found a statistically significant difference in the proportion of eyes with pinpoint leakage on FA ( P < .001, Fisher exact test), punctate choroidal hypercyanescent spots on ICGA ( P < .001), and oval hypocyanescent area on ICGA ( P = .033) ( Figure 1 ). Among the 17 eyes with serous RD, a small nonfibrovascular pigment epithelium detachment (PED) was detected in 15 eyes (88%) ( Figure 1 ); there was a statistically significant difference in the proportion of eyes with a PED among eyes with and without serous RD ( P < .001). The presence of a PED was significantly correlated with the presence of pinpoint leakage on FA ( P < .001) and punctate choroidal hypercyanescent spots on ICGA ( P < .001). Seventeen out of 39 eyes (44%) had RPE atrophic changes at baseline. No differences were found in the proportion of eyes showing RPE atrophic changes on FA ( P = .092), or hyperpermeability areas on late-phase ICGA ( P = .184), or a particular orientation of the dome ( P > .05) between eyes with and without serous RD.
|Test||Finding||Serous Retinal Detachment|
|Yes (n = 17)||No (n = 22)||P Value (Fisher Exact Test)|
|FA||Pinpoint leakage||Yes (n = 15)||15 (88%)||0 (0%)||<.001|
|No (n = 24)||2 (12%)||22 (100%)|
|RPE atrophic changes||Yes (n = 17)||7 (32%)||10 (59%)||.092|
|No (n = 22)||15 (68%)||7 (41%)|
|ICGA||Punctate hypercyanescent spots||Yes (n = 16)||15 (88%)||1 (5%)||<.001|
|No (n = 23)||2 (12%)||21 (95%)|
|Focal choroidal hyperpermeability||Yes (n = 2)||2 (12%)||0 (0%)||.184|
|No (n = 37)||15 (88%)||22 (100%)|
|Subfoveal oval hypofluorescent area||Yes (n = 6)||5 (29%)||1 (5%)||.033|
|No (n = 33)||12 (71%)||21 (95%)|
|SD OCT||Small PED||Yes (n = 20)||15 (88%)||5 (23%)||<.001|
|No (n = 19)||2 (12%)||17 (77%)|
|Horizontally oriented, oval dome||Yes (n = 38)||16 (94%)||22 (100%)||.436|
|No (n = 1)||1 (6%)||0 (0%)|
|Vertically oriented, oval dome||Yes (n = 12)||8 (47%)||4 (18%)||.053|
|No (n = 27)||9 (53%)||18 (82%)|
|Central, round dome||Yes (n = 11)||7 (41%)||4 (18%)||.114|
|No (n = 28)||10 (59%)||18 (82%)|
The comparison of the 2 groups of eyes with and without serous RD by Mann-Whitney U test indicated that eyes with serous RD had a thicker choroid in the fovea ( P = .004) and 1500 μm inferiorly to the fovea ( P = .014) compared to eyes without serous RD ( Table 3 ). They also tended to have a greater subfoveal scleral thickness ( P = .067) and a greater macular bulge height ( P = .079). The subfoveal choroidal thickness was positively correlated with the macular bulge height ( P < .001, Spearman rank correlation coefficient test) ( Figure 2 ) and with the scleral thickness ( P = .002), but was not associated with BCVA ( P = .660). A positive correlation was also found between the bulge height and the subfoveal scleral thickness ( P = .004).
|All Eyes (n = 39)||Serous Retinal Detachment|
|No (n = 22)||Yes (n = 17)||P Value (Mann-Whitney U Test)|
|Choroidal thickness (μm)|
|Fovea||132.7 ± 79.1||100.5 ± 61.5||174.3 ± 79.8||.004|
|Superior 1500||135.7 ± 88.0||114.3 ± 75.4||165.1 ± 95.3||.092|
|Inferior 1500||131.2 ± 93.3||93.7 ± 57.4||182.8 ± 107.4||.014|
|Foveal scleral thickness (μm)||577.4 ± 137.3||547.2 ± 147.8||653.1 ± 58.1||.067|
|Macular bulge height (μm)||371.2 ± 217.7||322.8 ± 228.6||425.2 ± 190.7||.079|