Idiopathic serous epithelial detachments (SPEDs) and choroidal thickness in UK Biobank (UKBB)

Catherine Jamison, Barbra Hamill, Michael J Quinn, Tunde Peto, Paul Foster

Research output: Contribution to conferencePaper


Idiopathic SPEDs are thought to represent an early stage of central serous chorioretinopathy (CSCR), which may or may not progress into acute or chronic CSCR, a disease typically affecting young/middle-aged individuals. CSCR is more common in males, those with type-1 personalities, and is associated with the use of corticosteroids and anti-psychotic medication. The choroidal circulation is altered in CSCR, with increased choroidal vascular permeability, thickened choroid and dilated choroidal vessels. Acute CSCR presents as sub-retinal fluid, with or without SPEDs, and disruption of the outer retinal layers. It will often resolve spontaneously, although it recurs in up to 50% of cases, and can lead to secondary CNV or macular oedema. If fluid or RPE disruption occurs at the fovea irreversible visual loss may occur. Clinically, patients tend to present when visual changes occur, therefore generally at the acute stage.
The grading of multiple subjects with idiopathic SPEDs during the ocular component of UKBB presented an opportunity to carry out a more detailed grading on these subjects. In particular, grading characterised SPED, choroidal thickness and vessel diameter compared to subjects with no CSCR pathology.

UKBB is a biomedical database comprising lifestyle, biomedical and genetic data from over 500,000 UK participants aged 40-69. The ocular component included colour and OCT images of volunteers. OCT images were obtained using spectral domain OCT (3D OCT-1000 Mark II, Topcon, Japan), captured in a darkened room without pupil dilation using a 3D macular volume scan (512 horizontal A-scans/B-scan; 128 B-scans in a 6×6mm raster-pattern). No EDI scans were taken. One colour photograph covering the disc and macula was acquired using the same equipment. The images were graded masked, with no information about subjects available to the graders.

Individuals graded as having an idiopathic SPED in either eye during the UKBB image analysis were selected for more detailed grading. This was predominantly OCT based and included; the number of SPEDs, the height, base diameter and location of the largest SPED, including whether sub-, juxta- or extra-foveal, superior or inferior, and nasal or temporal. Where gradable, sub-foveal choroidal thickness (SFCT) was measured directly below the foveal dip, maximum choroidal thickness (MCT), using any scan, and greatest (vertical) choroidal vessel diameter (CVD) on the foveal scan.
Those with characteristic acute stage CSCR changes in either eye were excluded from this sub-study, as were any subjects with AMD or any other notable pathology. Four eyes showed signs of past CSCR in the fellow-eye and those were included. Colour images were used to confirm OCT findings, and rule out any other pathology outside the OCT-grid. If any additional distinct SPEDs were seen outside the OCT-grid on the colour images, these were included in the SPED number.
A thickness measurement control-group of an equal number of eyes recorded as having no abnormalities, were graded using the same method. Paired T-tests were used for comparison between SPED-subject and normal measurements, using SPSS v26.

68527 participant image-sets were supplied, 99% of colour/OCTs were available, equating to 67995 colour/67935 OCTs for the right eye, and 67623 colours/67591 OCTs for the left eye. Approximately 90% of colour images and 99% of OCTs were gradable quality.
109 eyes (85 patients) had at least one idiopathic SPED, 12 patients had bilateral SPEDs. This equates to 0.08% of the gradable study population having unilateral SPEDs, and 0.009% having bilateral. In 76% of cases one SPED was present, 17% had two SPEDs, 4% three, 2% two and 2% over 5. Mean SPED height was 100.23µm (range 23-501) and mean base diameter 483.97µm (119-1961).
80% of SPEDs were extra-foveal, 21% juxta-foveal (within or touching the foveal avascular zone) and 8% sub-foveal. 45.9% of SPEDs were located in the superior, 40.4% inferior and 13.8% located centrally. 56% were located nasally, 37.6% temporally, and 6.4% centrally.
SPED-eyes had a mean SFCT of 355µm (141-567), mean MCT of 414 µm (190-567) and mean CVD of 191 µm (108-439). For normal eyes SFCT, MCT and CVD were; 265 (111-443), 320 (176--458) and 165 (93-267) respectively. There was significant difference between the two groups for; SFCT (n=88, p<0.001), MCT (n=99, p<0.001), and CVD (n=84, p=0.001).

Choroidal thickness at the fovea, and maximum choroidal thickness of those with SPEDs were significantly higher than those of individuals with no pathology. Additionally vertical diameter of the largest (gradable) vessel on the foveal scan was significantly greater in those with SPEDs, signifying choroidal changes and vessel dilation in these subjects.
There was a slight difference between superior and inferior location of SPEDs at 40 and 46% respectively, however there seems to be a propensity towards nasal occurrence of SPEDs with 56% occurring nasally, rather than 38% temporally.
Idiopathic SPEDs often resolve themselves and unless sub-foveal are unlikely to cause any vision changes. Only 8% of SPEDs were sub-foveal and 21% juxta-foveal, therefore it is likely that most of these individuals were unaware of any eye problems at the time of image collection.

Financial Disclosure: The Macular Society and Moorfields Eye Charity provided funding for the eye and vision component of UK Biobank
Original languageEnglish
Publication statusPublished - 10 Sept 2021
EventEuropean Society of Retina Specialists (EURETINA) Virtual Congress 2021 -
Duration: 09 Sept 202112 Sept 2021


ConferenceEuropean Society of Retina Specialists (EURETINA) Virtual Congress 2021
Abbreviated titleEURETINA 2021 Virtual


Dive into the research topics of 'Idiopathic serous epithelial detachments (SPEDs) and choroidal thickness in UK Biobank (UKBB)'. Together they form a unique fingerprint.

Cite this