Clinicopathologic Reports, Case Reports, and Small Case Series
June 2004

Central Serous Chorioretinopathy Associated With Periocular CorticosteroidInjection Treatment for HLA-B27–Associated Iritis

Author Affiliations



Copyright 2004 American Medical Association. All Rights Reserved.Applicable FARS/DFARS Restrictions Apply to Government Use.2004

Arch Ophthalmol. 2004;122(6):926-928. doi:10.1001/archopht.122.6.926

Central serous chorioretinopathy (CSCR) is characterized by accumulationof fluid under the neurosensory retina and/or retinal pigment epithelium,resulting in a serous detachment that often involves the macula. Althoughthe exact mechanism producing CSCR is unknown, increased adrenergic stimulationhas been hypothesized to play a role.1 Furthermore,elevated corticosteroid levels after systemic administration or secondaryto endogenous sources have been implicated in the causation or exacerbationof CSCR.2,3 Subtenon localcorticosteroid injection is effective in the treatment of certain forms ofuveitis. This report details a case of CSCR that developed after a singlelocal subtenon corticosteroid injection to treat HLA-B27–associatediritis and was confirmed by optical coherence tomography (OCT).

Report of a Case

A healthy 37-year-old man was examined because of a 10-day history ofprogressive blurred vision, photophobia, and floaters in the left eye. Medicaland social histories were noncontributory. The patient had had an upper respiratorytract infection 2 weeks before the examination. A review of systems was negativefor gastrointestinal, genitourinary, dermatological, or musculoskeletal symptoms,including joint or back pain.

Best-corrected visual acuity measured 20/20 OD and 20/50 OS. Intraocularpressure was 21 mm Hg OD and 16 mm Hg OS. Results of an examination of theanterior and posterior segment of the right eye were normal. However, theanterior segment of the left eye showed 2+ conjunctival injection and mildkeratitic precipitates. There was a 2+ anterior-chamber cellular reactionwith a 1-mm hypopyon, engorged iris vessels, and fibrinous iris posteriorsynechiae that were released after pupillary dilation. Binocular and indirectophthalmoscopy of the left eye showed a normal optic nerve, macula, retinalvasculature, and periphery. There was no evidence of retinal or vitreous inflammation,vasculitis, or cystoid macular edema (CME). The fovea was well visualizedafter pupillary dilation, with a normal and distinct foveal reflex.

Although this was the first episode of iritis, systemic evaluation wasperformed because of the fibrinous anterior-chamber reaction and hypopyon.HLA-B27 iritis was suspected and subsequently confirmed with positive serotypingresults. All others tests produced normal findings, including a complete bloodcell count, an erythrocyte sedimentation rate, a chest x-ray, and fluorescenttreponemal antibody absorption, angiotensin-converting enzyme, lysozyme, antinuclearantibody, and purified protein derivative of tuberculin tests. Treatment with1% prednisolone acetate every hour and cyclopegic eye drops was commenced,and a 1.0-mL periocular injection of triamcinolone acetonide corticosteroid(40 mg/mL) was given in the subtenon space of the left eye. Within 1 week,there was marked therapeutic response with complete resolution of the hypopyonand fibrin deposition and partial improvement of acuity to 20/40 OS. The occasionalresidual anterior-chamber inflammatory cell (<1 per high-power field) wasnoted. The ocular media was clear. Macular biomicroscopy revealed the newdevelopment of subretinal fluid and a serous pigment epithelial detachmentat the fovea. Fluorescein angiography confirmed an enlarging pinpoint spotof hyperfluorescence, as may be observed with CSCR (Figure 1). The fluorescein leakage was not petalloid and did notdemonstrate CME. Optical coherence tomography confirmed the subretinal locationof this fluid collection, which was consistent with the diagnosis of CSCR(Figure 2A). The topical corticosteroiddrops were rapidly tapered and discontinued over 5 days, and the patient wasfollowed up with serial clinical and OCT examinations. He showed progressivereduction in subretinal fluid, with gradual improvement in visual acuity.By 12 weeks after the initial examination, the fluid had resolved and visualacuity recovered to 20/20 OS. There were no sequelae of ocular inflammation,and OCT confirmed resolution of the foveal contour toward normal (Figure 2B).

Figure 1.
Image not available

Fluorescein angiography showsa focus of pinpoint hyperfluorescence (A) that enlarges in the later frames(B) and is consistent with the diagnosis of central serous chorioretinopathy.

Figure 2.
Image not available

A, Cross-sectional optical coherencetomography demonstrates a collection of fluid under the fovea, confirmingthe diagnosis of central serous chorioretinopathy after a subtenon injectionof corticosteroid. B, Optical coherence tomography shows resolution of thefluid and return of the foveal depression.


Central serous chorioretinopathy is considered an idiopathic disorderand the exact pathogenesis is uncertain. Speculation regarding adrenergicstimulation as a mechanistic cause has been based on the association of CSCRwith emotional stress, type A personality, and pregnancy.1 Hypercortisolismhas also been implicated in the etiology, as increased endogenous cortisollevels and exogenous corticosteroid use have been associated with the developmentor exacerbation of CSCR.1 This mechanismis presumed to act via damage to the choroidal circulation. Such damage mayresult from increased catecholamine-mediated vasoconstriction in the choroid,or the choroid may become hyperpermeable because of increased capillary fragilitycaused by elevated cortisol levels.2 Alternatively,corticosteroids may damage retinal pigment epithelial cells directly throughinhibition of extracellular matrix components and decreased fibroblastic activity.3 Although the exact sequence of events in this cause-effectrelationship is not understood, evidence from multiple studies2,3 suggeststhat the role of corticosteroids in this disorder is not coincidental.

Exogenous corticosteroids that are inhaled or that are administeredorally, intravenously, or by epidural or, most recently, intra-articular injectionhave been associated with development of CSCR.2,3 Toour knowledge, this is the initial report detailing the onset of CSCR followingperiocular corticosteroid injection with confirmation by OCT imaging. Periocularcorticosteroids may be administered for the treatment of uveitis as well aspostoperative CME. The presence of intraocular inflammation secondary to uveitismay affect the retinal pigment epithelial barrier and increase susceptibilityto local periocular corticosteroid effects. A similar mechanism has been proposedin a small series that reported CSCR following systemic corticosteroid therapyfor uveitis that occurred secondary to bird-shot chorioretinopathy, Vogt-Koyanagi-Haradadisease, and scleritis.4

A variety of contributing factors may account for the paucity of reportslinking CSCR and the common procedure of periocular corticosteroid injection.In eyes with uveitis, clinical features such as synechiae, pupil constriction,or media opacity severe enough to warrant periocular corticosteroid administrationmay preclude visualization of distinct macular details. Subtle fluid fromCSCR that is related to a periocular corticosteroid injection may be interpretedas CME or may be overshadowed by features of coexisting CME. Visual symptomsfrom secondary CSCR may be attributed to the primary diagnosis of uveitis,and thus the fundus examination that is required to detect this entity maybe omitted. Optical coherence tomographic imaging may be especially helpfulin this scenario to differentiate between the features of CSCR and CME. Itis unlikely that the topical corticosteroid drops played a role in the developmentof CSCR because of the relatively brief time of administration and the lackof posterior-segment penetration, especially in this patient with phakic eyes.In contrast, periocular corticosteroids are injected directly behind the globeto induce a posterior pole effect.

Because almost every other route of corticosteroid administration, includinglocal intra-articular injection, has been linked with CSCR thus far, it appearslogical that the periocular depot corticosteroid injection would be associatedwith development of this disorder.2,3 Theperiocular route should be included in the suggested etiologic associationbetween corticosteroid therapy and development or exacerbation of CSCR.

The authors have no relevant financial interest in this article.

Corresponding author: Caroline R. Baumal, MD. FRCSC, Department ofVitreoretinal Surgery, New England Eye Center, 750 Washington St, Box 450,Boston, MA 02111 (e-mail:

Garg  SPDada  TTalwar  DBiswas  NR Endogenous cortisol profile in patients with central serous chorioretinopathy. Br J Ophthalmol. 1997;81962- 964
Wakakura  MIshikawa  S Central serous chorioretinopathy complicating systemic corticosteroidtreatment. Br J Ophthalmol. 1984;68329- 331
Carvalho-Recchia  CAYannuzzi  LANegrao  S  et al.  Corticosteroids and central serous chorioretinopathy. Ophthalmology. 2002;1091834- 1837
Schalenbourg  ALeys  ADe Courten  CCoutteel  CHerbort  CP Corticosteroid-induced central serous chorioretinopathy in patientswith ocular inflammatory disorders. Klin Monatsbl Augenheilkd. 2002;219264- 267