Acta Ophthalmologica 2016
both healthy and disease states. Associations between subfoveal choroidal thickness (SFCT) and age, refractive error (RE) and axial length (AL) have been demonstrated, although ﬁndings are inconsistent and variability in SFCT cannot be fully explained by these factors alone. The high vascular structure of the choroid suggests that some variation in SFCT may be the result of changes in the choroidal blood ﬂow. This study investigated the eﬀects of ocular perfusion pressure (OPP), blood pressure [systolic (SBP) and diastolic (DBP)] and intra-ocular pressure (IOP) on SFCT in normal individuals and also the eﬀect of age, gender, RE and AL. Participants were healthy volunteers, recruited from staﬀ and students from the Ophthalmology Department at St James’s University Hospital, Leeds. Inclusion criteria were as follows: age from 18 to 60 years, no history of ocular disease including ocular surgery and laser, myopia less than 6D and written, informed consent. Ethical approval was obtained from the Leeds Institute of Health Sciences and Leeds Institute of Genetics, Health and Therapeutics and Leeds Institute of Molecule Medicine Joint Ethics Committee (HSLTLM/11/043). OPP formula: OPP = 2/3[DBP + 1/3(SBPDBP)]IOP. All data were collected from participants on the same day, as close in time to OCT imaging as possible. Mean age of the 46 participants was 35.83 (range: 22–60) years, and 21 were male. Mean SFCT was 313.1 lm (SD = 77.0 lm) in the right eyes. Data for right eyes were used for analysis. There was a signiﬁcant negative correlation between SFCT and the following variables on univariate analysis: AL (p = 0.001, r = 0.47, r2 = 0.22), SBP (p = 0.041, r = 0.30, r2 = 0.09), OPP (p = 0.045, r = 0.29, r2 = 0.08) and a non-signiﬁcant trend suggesting association between SFCT and DBP (p = 0.053, r = 0.28, r2 = 0.08), RE (p = 0.055, r = 0.28, r2 = 0.08) (Fig. 1). Using multiple regression analysis, only AL was found to be a signiﬁcant independent variable (p = 0.024, r2 = 0.265). No association seen with other variables. This study identiﬁed a modest, negative correlation between SFCT and SBP. SBP has been shown to
inﬂuence CT, although limited consensus exists. Previous reports state acute rises in blood pressure do not cause signiﬁcant changes in CT, and therefore, blood pressure-associated variation in CT may occur over a relatively greater time scale (Alwassia et al. 2013). A modest, negative association between SFCT and OPP was found. This ﬁnding has been reported previously (Kim et al. 2012). Kim et al. (2012) proposed that relatively thicker choroids may require a lower OPP to maintain ocular blood ﬂow, and in relatively thinner choroids, a higher OPP may be required to compensate for potential reductions in choroidal blood ﬂow. Poor ocular blood ﬂow, secondary to reduced OPP, has been shown to increase the risk of ocular ischaemia, with the fovea at particular risk (Levin et al. 2011; Schmidl et al. 2011). Given the association between SFCT and a range of physiological variables, care needs to be taken in attributing variation in CT as the basis for ocular disease. Rishi et al. (2013) report a study of polypoidal choroidal vasculopathy (PCV) compared to agematched controls and eyes with other forms of neovascular AMD. Eyes with PCV had higher mean OPP than controls and eyes with AMD. This suggests diﬀerences in SFCT between diseased and healthy eyes may more likely be the result of physiological variation in ocular and systemic factors. The studied variables here and others need further study to explain fully the variation in SFCT seen in normal individuals and in others with systemic vascular disease.
Rishi P, RishI E, Mathur G et al. (2013): Ocular perfusion pressure and choroidal thickness in eyes with polypoidal choroidal vasculopathy, wet-age-related macular degeneration, and normals. Eye 27: 1038–1043. Schmidl D, Garhofer G & Schmetterer L (2011): The complex interaction between ocular perfusion pressure and ocular blood ﬂow - Relevance for glaucoma. Exp Eye Res 93: 141–155.
Correspondence: Martin McKibbin Department of Ophthalmology St. James’s University Hospital Leeds LS9 7TF, UK Tel: +0113 206 6429 Fax: +113 206 5084 Email: [email protected]
Provisional data presented as a poster at Royal College of Ophthalmologists Annual Congress 2013.
Cytomegalovirus retinitis following intravitreal dexamethasone implant in a patient with central retinal vein occlusion Lorenzo Vannozzi,1 Daniela Bacherini,1 Andrea Sodi,1 Enrico Beccastrini,2 Giacomo Emmi,3 Andrea Giorni1 and Ugo Menchini1 1 Department of Surgery and Translational Medicine, Eye Clinic, University of Florence, Florence, Italy; 2 Internal Medicine Unit, Santa Maria Annunziata Hospital, Florence, Italy; 3 Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy
References Alwassia AA, Adhi M, Zhang JY et al. (2013): Exercise-induced acute changes in systolic blood pressure do not alter choroidal thickness as measured by a portable spectraldomain optical coherence tomography device. Retina 33: 160–165. Kim M, Kim SS, Kwon HJ et al. (2012): Association between choroidal thickness and ocular perfusion pressure in young, healthy subjects: enhanced depth imaging optical coherence tomography study. Invest Ophthalmol Vis Sci 53: 7710–7717. Levin LA, Nilsson SFE, Ver Hoeve J et al. (2011): Adler’s physiology of the eye, 11th edn. Philadelphia: Saunders/Elsevier.
Editor, ytomegalovirus (CMV) retinitis represents a sight-threatening opportunistic ocular infection in patients aﬀected by immunocompromising diseases. We report a case of CMV retinitis following an intravitreal injection of dexamethasone implant (OzurdexÒ) in a 54-year-old Caucasian male referred to our centre for relapsing bilateral central retinal vein occlusions (CRVO). At our ﬁrst evaluation, best-corrected visual acuity (BCVA) was 20/200 in the
Acta Ophthalmologica 2016
Fig. 1. (A) Fundus photography of the right eye before intravitreal injection of dexamethasone implant, showing vascular retinal anomalies after central retinal vein occlusion; (B) OCT showing macular oedema; (C) late-phase ﬂuorescein angiography showing diﬀuse macular leakage corresponding to macular oedema. (D, E, F) 3 months after steroid implant, areas of retinitis, diﬀuse retinal arteritis and vitreous opacity developed; the dexamethasone intravitreal implant was still visible.
right eye (RE) and 20/100 in the left eye (LE); clinical examination revealed bilateral non-ischaemic CRVO sequelae treated with laser photocoagulation and a recurrence of macular oedema in RE. The patient’s clinical history revealed arterial hypertension and a mild hyperhomocysteinaemia associated with a homozygous methylenetetrahydrofolate reductase gene mutation. He had previously been treated in another hospital and had been undergoing treatment with antiplatelet and immunosuppressive therapy (mycophenolate mofetil) for 2 years for presumed retinal vasculitis. However, his clinical history gave no evidence of systemic autoimmune disease: only low titles of type III cryoglobulins were found. Moreover, fundus examination did not show vitritis, vessel sheathing or other clinical features deﬁnitely suggesting the diagnosis of retinal vasculitis. In November 2011, dexamethasone intravitreal implant was administered for the treatment of macular oedema in RE. Initially, the patient’s follow-up was uncomplicated, but 3 months after the injection he complained of rapid vision decrease and ﬂoaters. In our examination, RE showed a visual acuity reduced to light perception, intra-
ocular pressure (IOP) had increased to 37 mmHg, and granulomatous keratic precipitates in the anterior chamber with grade 2+ ﬂare and 2+ inﬂammatory cells were detected (Fig. 1). Fundus examination revealed 2+ vitritis, papillitis, diﬀuse retinal arteritis and an inferotemporal area of retinitis. A fragment of the dexamethasone intravitreal implant was still visible. A presumptive diagnosis of CMV retinitis was conﬁrmed by polymerase chain reaction (PCR) of the aqueous sample, which was positive for CMV DNA and negative for the herpes simplex virus, herpes zoster virus and toxoplasmosis. Serologic tests for herpes zoster virus (HZV), herpes simplex virus (HSV), toxoplasmosis and syphilis (VDRL) and a test for HIV were negative. The PCR on a blood sample for CMV was positive, CMV IgM antibody was negative, and IgG was positive. Medical treatment began immediately with intravenous ganciclovir (250 mg twice daily for 3 weeks) followed by oral valganciclovir (900 mg twice daily for 8 weeks). The patient declined any surgical intervention: thus, no consent was obtained for an intravitreal ganciclovir injection, Ozurdex implant removal or vitrectomy. After consulting the immunologist,
mycophenolate mofetil was immediately discontinued. At 2 months, RE vision was unchanged while the area of retinitis became smaller but did not disappear completely; therefore, oral valganciclovir was continued. In December 2012, RE iris neovascularization developed and laser treatment was performed in the visible retinal areas. Two months later, an inferior retinal detachment developed, but the patient refused any surgical procedure considering the poor visual prognosis. During the following 9 months, the clinical picture remained unchanged and visual acuity was limited to light perception in the RE. The LE remained stable with no complications and a visual acuity of 20/100. Some cases of CMV retinitis following intravitreal steroid injections have already been reported (Ufret-Vincenty et al. 2007; Park & Byeon 2008; Welling et al. 2012). Two cases occurred after retinal vein occlusion, but we are unaware of any previous reports of CMV retinitis after an intravitreal dexamethasone implant. We can speculate that in patients receiving systemic immunomodulators, dexamethasone implant may cause further local immunosuppression capable of allowing the
Acta Ophthalmologica 2016
CMV to replicate and eventually cause retinitis. Takakura et al. (2014) reports on a relatively large series of viral retinitis following local steroid administration, but in this paper the injected steroid was triamcinolone or ﬂuocinolone and the contribution of systemic immunosuppression remained unclear. Meanwhile, exclusion criteria of the GENEVA Study did not include systemic immunosuppressive treatment but only the use of systemic steroids (Haller et al. 2010). Moreover, it is possible that in retinal vein occlusion, the retinal blood ﬂow stasis and the breakdown in the blood–retina barrier may increase the susceptibility of ocular tissues to virus penetration. In conclusion, ophthalmologists using intravitreal steroids in iatrogenic compromised patients should be aware of the potential risk of CMV retinitis.
References Haller JA, Bandello F, Belfort R Jr et al. (2010): Randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Ophthalmology 117: 1134–1146. Park YS & Byeon SH (2008): Cytomegalovirus retinitis after intravitreous triamcinolone injection in a patient with central retinal vein occlusion. Korean J Ophthalmol 22: 143–144. Takakura A, Tessler HH, Goldstein DA et al. (2014): Viral retinitis following intraocular or periocular corticosteroid administration: a case series and comprehensive review of the literature. Ocul Immunol Inﬂamm 22: 175–182. Ufret-Vincenty RL, Singh RP, Lowder C & Kaiser PK (2007): Cytomegalovirus retinitis after ﬂuocinolone acetonide (RetisertTM) implant. Am J Ophthalmol 143: 334–335. Welling JD, Tarabishy AB & Christoforidis JB (2012): Cytomegalovirus retinitis after central retinal vein occlusion in a patient on systemic immunosuppression: does venooclusive disease predispose to cytomegalovirus retinitis in patients already at risk? Clin Ophthalmol 6: 601–603.
Correspondence: Daniela Bacherini Department of Surgery and Translational Medicine Eye Clinic, University of Florence Largo Brambilla 3, Florence 50134 Italy Tel: +393392037649 Fax: +390554377749 Email: [email protected]
Early panretinal abnormalities on fundus autofluorescence and spectral domain optical coherence tomography after intravitreal ocriplasmin Giulio Barteselli,1,2 Elisa Carini,1 Alessandro Invernizzi,1,3 Roberto Ratiglia1 and Francesco Viola1 1 Ophthalmological Unit, Department of Clinical Sciences and Community Health, Ca’ Granda Foundation-Ospedale Maggiore Policlinico, University of Milan, Milan, Italy; 2Genentech Inc, South San Francisco, California, USA; 3 Eye Clinic, Department of Biomedical and Clinical Sciences ‘Luigi Sacco’, Luigi Sacco Hospital, University of Milan, Milan, Italy
Editor, ntravitreal ocriplasmin, a recombinant truncated form of plasmin with proteolytic activity against laminin and ﬁbronectin (Hermel et al. 2010), has recently been approved as treatment strategy for vitreomacular traction (VMT). However, concerns have been raised regarding its ocular safety (Kim 2014). Clinical trials reported adverse events that included transient visual loss, dyschromatopsia and photopsias, usually associated with decreased amplitudes on electroretinography (ERG) (Stalmans et al. 2012). Spectral domain optical coherence tomography (SD-OCT) abnormalities have been detected within the macula, and ERG studies have demonstrated diﬀuse retinal dysfunction related to presumed ocriplasmin toxicity (Fahim et al. 2014; Tibbetts et al. 2014). Herein, we demonstrate that early structural abnormalities to the photoreceptors as detected by fundus autoﬂuorescence (FAF) and SD-OCT may aﬀect not only the macula but also the peripheral retina as well. A 69-year-old woman presented with visual acuity of 20/32 and metamorphopsia in her right eye. Fundus examination and SD-OCT revealed VMT with small detachment of the
foveal cones; FAF was normal (Fig. 1A,E). The day after uneventful intravitreal ocriplasmin (0.125 mg/ 0.1 ml), her vision decreased to 20/50 and she complained of dyschromatopsia. SD-OCT demonstrated the appearance of subfoveal ﬂuid and multifocal shallow subretinal detachments (Fig. 1J–L), as well as focal areas of attenuation/thinning of the ellipsoid zone line and disappearance of the interdigitation zone line of the photoreceptors. Photoreceptor abnormalities corresponded to patchy areas of abnormally increased FAF signal within the macula (Fig. 1B,F) as well as to radial lines of abnormally increased FAF signal towards the retinal periphery (Fig. 1I). At day 7, her vision improved to 20/40 and dyschromatopsia was resolved. SD-OCT demonstrated the released VMT, decreased subfoveal ﬂuid and shaggy photoreceptors in the fovea (Fig. 1C,G). Persisting attenuation/thinning of the ellipsoid zone line and further disappearance of the interdigitation zone line were noted. After 4 months, FAF and SDOCT abnormalities disappeared completely (Fig. 1D,H). Clinical trials that compared intravitreal ocriplasmin to placebo in treatment of VMT indicated that visual symptoms were greater in patients receiving ocriplasmin (Stalmans et al. 2012). In addition, decreased amplitudes in all ERG variables were noted, indicating the presence of panretinal dysfunction that persisted for several months (Fahim et al. 2014; Tibbetts et al. 2014). Our imaging study supports these functional ﬁndings; indeed, structural damage to the photoreceptors was not conﬁned within the macula, but was actually widespread throughout the periphery. These abnormalities may be due to a protease eﬀect on laminin, which is present not only in the vitreous but also throughout the retina, including the interphotoreceptor matrix (IPM) (Fahim et al. 2014). Interestingly, in our case, there was greater and broader damage to the interdigitation zone line compared to the ellipsoid zone line. Normal reﬂectivity of the interdigitation zone line arises from well-oriented and aligned interdigitations of apical processes of the retinal pigment epithelium (RPE) with outer segments of the photoreceptors. We hypothesize that lysis of the IPM caused by ocriplasmin may create