Patients with a history of diabetes, indwelling lines/catheters, hyperalimentation, or immunocompromise are at increased risk for fungal bloodstream infections.1–3 Fungemic patients are subject to a variety of pathologic manifestations related to fungal dissemination and end organ damage, including chorioretinitis and endophthalmitis via hematogenous seeding of small retinal and choroidal capillaries. As has been demonstrated pathologically, localized ocular fungal proliferation can progress to focal or multifocal inflammatory lesions involving the retina and/or choroid along with subsequent vitreous seeding, which can then lead to frank endophthalmitis.4
As most patients with fungemia are systemically ill and require hospitalization, these patients are screened for ocular fungal involvement and monitored as inpatients with serial dilated fundus examinations.5–8 Few reports have examined spectral-domain optical coherence tomography (SD-OCT) findings in posterior fungal eye disease.9–13 SD-OCT can visualize the chorioretinal interface with high resolution and has been described in three published case reports to assess disease stage and monitor response to treatment.9–13 The OCT images correlate well with earlier pathologic analyses in eyes with fungal chorioretinitis and endophthalmitis.14–16 With the paucity of OCT data in the literature, the purpose of the current study is to describe OCT findings in patients with endogenous fungal chorioretinitis and endophthalmitis.
Patients and Methods
This multicenter, retrospective, observational case series was approved by the Institutional Review Boards at Wills Eye Hospital and William Beaumont Hospital prior to initiation of the study. Potential subjects were identified by screening OCT billing data for “chorioretinitis” (ICD-9 363.20 and ICD-10 H30.92) and “endophthalmitis” (ICD-9 360.19 and ICD-10 H44.19) and cross-referencing this data with patient charts with evaluations from April 4, 2012, to August 19, 2016. Chorioretinitis was defined as white chorioretinal infiltrates not associated with vitreal cells or “fluff balls.” Endophthalmitis was defined as vitreous extension of chorioretinal infiltrates, associated with vitreous cells or “fluff balls.” Fundus photos were available and reviewed in all study eyes. SD-OCT imaging was obtained with the Spectralis HRA-OCT (Heidelberg Engineering, Heidelberg, Germany) and Cirrus HD-OCT 4000 (Carl Zeiss Meditec, Dublin, CA).
Clinical and imaging (OCT and fundus photography) data were reviewed. Specific data collected included: retinal and choroidal anatomy, reflective qualities of subretinal pigment epithelium (RPE), inner and outer retinal lesions, vitreous opacities, and posterior shadowing.
During the 4-year study period, 12 eyes of seven patients who had SD-OCT imaging and were diagnosed with fungal chorioretinitis or endophthalmitis (mean age: 48 years; range: 30 years to 72 years) were identified. The clinical characteristics of each patient are summarized in Table 1. Six eyes (four patients) were diagnosed with fungal chorioretinitis and six eyes (five patients) were diagnosed with fungal endophthalmitis (both chorioretinal and vitreous involvement). Two of seven patients (28.6%) had bilateral chorioretinitis, whereas one patient (14.3%) had bilateral endophthalmitis. Presenting visual acuity (VA) ranged from 20/20 to 20/400 in eyes with chorioretinitis, and from 20/50 to counting fingers in eyes with endophthalmitis. Five of seven patients (71.4%; six eyes) had known candidemia at the time of initial examination. Six of seven patients (83.3%; 10 eyes) were treated with both intravitreal and systemic antifungal medication and one patient (16.7%; two eyes) was treated with systemic antifungals alone. Follow-up clinical data and imaging were available for six of 12 eyes (50%; average: 8.2 months; range: 2 months to 25 months).
Clinical Features of Patients With Fungal Chorioretinitis or Endophthalmitis
By OCT, two patterns of posterior ocular involvement were identified: (1) chorioretinal infiltration (10 eyes) (Figures 1 and 2), and (2) inner retinal/retinal vascular infiltration without choroidal involvement (five eyes) (Figure 3). Six of 12 eyes (50%) demonstrated more than one type of infiltration. Four eyes had chorioretinal and retinal vascular infiltration. Four of 12 eyes (25%) demonstrated hyperreflective superficial retinal lesions with associated shadowing and inner segment/outer segment (IS/OS) disruption deep to the lesions. Three of six eyes (50%) with endophthalmitis based on clinical examination had vitreal extension evident on OCT. In one case, the vitreous opacities were focal and well-defined (Case 2); in another (Case 6), the vitreous infiltration was large and ill-defined, emanating from the optic nerve; and in another (Case 7), there were multifocal lesions of vitreous infiltration consistent with the “string of pearls” appearance found on clinical examination.
Optical coherence tomography (OCT) demonstrating progressive stages of chorioretinal infiltration leading to endophthalmitis (A) Case 1. Enhanced depth imaging OCT (EDI-OCT) of the right eye disclosing a subretinal pigment epithelium (sub-RPE) infiltrate and adjacent lesions with RPE breakthrough (asterisks). There is also loss of choroidal vascular detail. (B) Case 1. EDI-OCT imaging of the right eye demonstrating a hyperreflective inner retinal lesion with posterior shadowing and loss of inner segment/outer segment (IS/OS) integrity below the lesion. (C) Case 2. OCT of the right eye with a hyperreflective vitreous opacity above the inner retina and posterior shadowing. There is diffuse IS/OS irregularity and subretinal hyperreflective material (SRHM) with a focal, circular absence of SRHM visible on the near-infrared image.
Optical coherence tomography (OCT) imaging before and after treatment for fungal chorioretinitis: subfoveal infiltration. (A) Case 3: Enhanced depth imaging OCT imaging of the right eye with a large subfoveal infiltrate associated with trace subretinal fluid and loss of choroidal detail and inner segment/outer segment (IS/OS) integrity. (B) Two-month follow-up OCT demonstrating consolidation of infiltration with development of subfoveal hyperreflective material with reconstitution of perifoveal IS/OS signal.
Optical coherence tomography (OCT) imaging before and after treatment for fungal chorioretinitis: inner retinal/retinal vascular infiltration. (A) Case 7. OCT of the right eye with two inner retinal hyperreflective lesions, which extend from the nerve fiber layer to the outer plexiform layer and overlying shadowing of vitreous opacities, consistent with vascular breakthrough of a fungal embolus (asterisk). (B) OCT of the same eye 2 months after demonstrating decreased size of the retinal infiltrates with improved, but mildly disrupted retinal architecture. (C) Case 6. OCT of the left eye with large spherical hyperreflective ring with hyporeflective center and significant shadowing over an elevated optic nerve. Examination revealed a central retinal vascular lesion with vitreous breakthrough. (D) After antifungal treatment, the hyperreflective lesion is consolidated and decreased in size with shadowing due to overlying vitreous opacity. Optic nerve elevation is also improved.
Other findings on presenting OCT included optic nerve edema in one of 12 eyes (8%), subretinal fluid in two of 12 eyes (17%), and epiretinal membrane noted at presentation and follow-up in three of 12 eyes (25%). Six of the 12 eyes (50%) had follow-up OCT imaging performed and all six eyes demonstrated decreased size of chorioretinal and retinal vascular lesions after antifungal therapy (Figures 2, 3, and 4). Furthermore, all eyes with chorioretinal infiltration demonstrated improved choroidal detail and decreased choroidal thickness following treatment (Figure 4).
Optical coherence tomography (OCT) imaging before and after treatment for fungal chorioretinitis: choroidal detail. (A). Case 4. OCT demonstrating subfoveal infiltration with underlying hyperreflective choroid, choroidal thickening, and loss of choroidal detail. (B) After treatment, the size of the chorioretinal infiltrate is decreased. Choroidal detail is improved, and choroidal thickness appears decreased after treatment.
Three of seven patients (42.8%) had baseline OCT imaging prior to initiation of anti-fungal treatment, whereas three of seven patients (42.8%) had their last follow-up OCT images obtained following cessation of antifungal treatment. VA on final follow-up ranged from 20/20 to 20/400 in the patients with chorioretinal infiltration without subfoveal involvement, 20/40 to counting fingers in patients with chorioretinal infiltration with subfoveal involvement, and 20/25 to 20/40 with inner retinal/retinal vascular infiltration.
Depending on species virulence and response to treatment, sequelae of endogenous fungal chorioretinitis and endophthalmitis range from mild to severe, with Candida species representing the most common causative fungus. Our current understanding of the pathophysiology of endogenous ocular fungal involvement is based on detailed histologic analyses done prior to invention of OCT.4,16,17 The advent of SD-OCT has more recently allowed clinicians to correlate these histologic findings to in vivo imaging in a few case reports. SD-OCT allows visualization of cross-sectional retinal and choroidal microarchitecture with a resolution on the order of 3 μm.5,6,8 The current retrospective study examined OCT findings in endogenous fungal eye disease and expands on these case reports to further characterize choroidal and retinal fungal infiltration.
To the best of our knowledge, there have been no prior published case series of OCT findings in endogenous fungal chorioretinitis and endophthalmitis. However, the few case reports available have provided important insights. Lavine and Mititelu reported multimodal imaging findings in a case of C. albicans chorioretinitis progressing to endophthalmitis.11 In their report, the patient presented with a RPE elevation in the central macula with an adjacent hyperreflective inner retinal lesion. With systemic treatment, the lesion regressed, leaving persistently elevated, discontinuous RPE with loss of IS/OS, external limiting membrane, outer nuclear, and outer plexiform layers. These findings suggested that the infection resulted from choroidal infiltration via the short posterior ciliary arteries with resultant breakthrough into the retina, rather than via the central retinal artery. Pathologic reports have found analogous progression of endogenous ocular fungal infection in enucleated specimens.4,14 In some eyes, endophthalmitis appears to occur as an extension of this chorioretinal infiltration based on histopathology and OCT. In a case of endogenous fungal endophthalmitis, Cho et al. described early hyperreflective chorioretinal OCT lesions involving the RPE and outer retina with progression inner retinal and vitreous involvement.10
In total, two patterns of posterior fungal involvement were found on OCT, of which one has not been previously described. The first pattern, chorioretinal infiltration, appears consistent with the reports mentioned above, in which solitary or multiple infiltrates arise in the choroid beneath the RPE (Figure 1A) appear to progress through the RPE into the outer retina (Figure 1B), which can eventually disrupt all retinal layers with extension into the vitreous (Figure 1C). This pattern (both with and without vitreous breakthrough) was noted in nine of 12 (75%) of the eyes in our review. Histopathologic examination of postmortem eyes have demonstrated similar evidence of fungal infiltration progressing inward from the sub-RPE space through the internal limiting membrane into the vitreous cavity.13,14 Subfoveal chorioretinal infiltration, a centrally localized variant of chorioretinal infiltration, was found in four of 12 eyes (33.3%) and was associated with worse presenting and final VAs compared to eyes with extrafoveal chorioretinal infiltration and retinal vascular infiltration. The second pattern, inner retinal/retinal vascular infiltration, occurred in four of 12 eyes (33.3%) and demonstrated superficial retinal whitening on clinical exam, which appeared as hyperreflective inner retinal lesions on OCT overlying or immediately adjacent to superficial retinal vessels without choroidal involvement. This pattern of involvement could represent a focal vasculitis with adjacent retinitis, rather than the classically described chorioretinitis. These lesions might also represent nerve fiber layer infarctions due to sepsis related ischemia, underlying hematologic abnormalities, or fungal emboli. It is also plausible they could represent migration of fungal emboli to the inner retina from the choroid. However, the lack of significant RPE changes below the lesions and direct association with retinal vessels on clinical exam and fundus photography strongly argues against this. In one patient (Case 2), underlying proliferative T-cell leukemia with associated anemia may have caused the inner retinal hyperreflectivity consistent with nerve fiber layer infarction. However, the three other patients with inner retinal hyperreflectivity on OCT in our study did not have hematologic abnormalities. Furthermore, in two patients (Cases 6 and 7), the OCT findings were limited to the inner retina, there were no outer retinal findings, and these patients had vitreous opacities over the lesion. These findings consistent with inner retinal infiltration correlate well with the pathologic analysis by Griffin et al., in which four enucleated eyes with endogenous fungal ocular disease demonstrated lesions confined to the retina, which appeared to originate from the deep capillary plexus via the central retinal artery.4 The most striking evidence of fungal emboli entering retinal vasculature was found in Case 6 (Figure 3), where a 30-year-old man with recent intravenous drug use presented with a large white central retinal vascular lesion with vitreous breakthrough and vitritis.
Treatment with systemic with or without intravitreal antifungals was associated with improvement in visual function and anatomic improvement on OCT in (four of six eyes; 66.7%). Late findings of posterior fungal involvement that have been previously described include epiretinal membranes; macular holes; choroidal neovascularization; retinal detachment; and development of a blind, painful eye.18–21 In the current series, three patients had epiretinal membranes noted at presentation and follow-up imaging. One patient (Case 7) developed cystoid macular edema on final OCT. No patients in the current series developed macular holes, choroidal neovascularization, retinal detachments, or blind, painful eyes.
Some visual disability persisted in most eyes (four of six eyes; 66.7%), likely due to compromised outer retinal anatomy. Subfoveal infiltration was associated with poorer final VA compared to eyes with extrafoveal chorioretinal infiltration and superficial retinal/retinal vascular infiltration as these patients were left with significant loss of central outer retinal anatomy on final examination. One eye with extrafoveal chorioretinal infiltration had final visual acuity 20/400, which was thought to be due to the patient's underlying strabismic amblyopia in that eye. Another eye with subfoveal infiltration had final VA of 20/40 with apparent reconstitution of perifoveal outer retinal anatomy following initiation of treatment (Case 3; Figure 2).
Strengths of the current study include a relatively high number of patients compared to previous reports and the presence of follow-up imaging for most eyes. Weaknesses include its retrospective nature, heterogeneity of presentations and follow-up, and the lack of direct histopathologic correlation of OCT findings for study eyes. Additionally, the patients in our report were systemically well enough to undergo outpatient OCT imaging, which may have led to underestimation of the potential range of disease severity, potentially limiting generalizability of the results.
In conclusion, the current study describes patterns of infiltration on OCT in patients with endogenous fungal chorioretinitis and endophthalmitis. Based on these data, fungal infiltration can occur from both the choroidal and retinal vasculature. OCT imaging of eyes with fungal chorioretinitis and endophthalmitis may assist clinicians to prognosticate visual outcomes based on presenting macular anatomy, to monitor response to treatment, and, in conjunction with infectious disease specialists, to decide when antifungal treatment can be safely ceased.
- Lingappan A, Wykoff CC, Albini TA, et al. Endogenous fungal endophthalmitis: Causative organisms, management strategies, and visual acuity outcomes. Am J Ophthalmol. 2012;153(1):162–166 doi:10.1016/j.ajo.2011.06.020 [CrossRef]
- Aguilar GL, Blumenkrantz MS, Egbert PR, McCulley JP. Candida endophthalmitis after intravenous drug abuse. Arch Ophthalmol. 1979;97(1):96–100. doi:10.1001/archopht.1979.01020010036008 [CrossRef]
- Wenzel RP. Nosocomial candidemia: Risk factors and attributable mortality. Clin Infect Dis. 1995;20(6):1531–1534 doi:10.1093/clinids/20.6.1531 [CrossRef]
- Griffin JR, Pettit TH, Fishman LS, Foos RY. Blood-borne candida endophthalmitis: A clinical and pathologic study of 21 cases. Arch Ophthalmol. 1973;89(6):450–456. doi:10.1001/archopht.1973.01000040452002 [CrossRef]
- Adam MK, Vahedi S, Nichols MM, et al. Inpatient ophthalmology consultation for fungemia: Prevalence of ocular involvement and necessity of funduscopic screening. Am J Ophthalmol. 2015;160(5):1078–1083. doi:10.1016/j.ajo.2015.07.033 [CrossRef]
- Dozier CC, Tarantola RM, Jiramongkolchai K, Donahue SP. Fungal eye disease at a tertiary care center: The utility of routine inpatient consultation. Ophthalmology. 2011;118(8):1671–1676. doi:10.1016/j.ophtha.2011.01.038 [CrossRef]
- Donahue SP, Greven CM, Zuravleff JJ, et al. Intraocular candidiasis in patients with candidemia. Clinical implications derived from a prospective multicenter study. Ophthalmology. 1994;101(7):1302–1309. doi:10.1016/S0161-6420(94)31175-4 [CrossRef]
- Paulus YM, Cheng S, Karth PA, Leng T. Prospective trial of endogenous endophthalmitis and chorioretinitis rates, clinical course, and outcomes in patients with fungemia. Retina. 2016;36(7):1357–1363. doi:10.1097/IAE.0000000000000919 [CrossRef]
- Mahendradas P, Avadhani K, Yadav NK, et al. Role of Spectralis HRA+OCT spectral domain optical coherence tomography in the diagnosis and management of fungal choroidal granuloma. Ocul Immunol Inflamm. 2010;18(5):408–410. doi:10.3109/09273948.2010.498656 [CrossRef]
- Cho M, Khanifar AA, Chan RVP. Spectral-domain optical coherence tomography of endogenous fungal endophthalmitis. Retin Cases Brief Rep. 2011;5(2):136–140. doi:10.1097/ICB.0b013e3181cc2146 [CrossRef]
- Lavine JA, Mititelu M. Multimodal imaging of refractory Candida chorioretinitis progressing to endogenous endophthalmitis. J Ophthalmic Inflamm Infect. 2015;5(1):54. doi:10.1186/s12348-015-0054-z [CrossRef]
- Adam MK, Rahimy E. Enhanced depth imaging optical coherence tomography of endogenous fungal chorioretinitis. JAMA Ophthalmol. 2015;133(11):e151931. doi:10.1001/jamaophthalmol.2015.1931 [CrossRef]
- Querques G, Modorati G, Miserocchi E, Baruffaldi Preis F, Bandello F. Bilateral endogenous endophthalmitis causes by Candida albicans after breast implant surgery. JAMA Ophthalmol. 2016;134(4):467–469. doi:10.1001/jamaophthalmol.2016.0015 [CrossRef]
- Rao NA, Hidayat AA. Endogenous mycotic endophthalmitis: Variations in clinical and histopathologic changes in candidiasis compared with aspergillosis. Am J Ophthalmol. 2001;132(2):244–251. doi:10.1016/S0002-9394(01)00968-0 [CrossRef]
- Kawanishi Y, Morinobu T, Shirakawa K, Ueno H. Histopathological studies of endogenous fungal endophthalmitis. Folia Ophthalmologica Japonica. 1984;38(2):204–211.
- Edwards JE Jr, Montgomerie JZ, Foos RY, Shaw VK, Guze LB. Experimental hematogenous endophthalmitis caused by Candida albicans. J Infect Dis. 1975;131(6):649–657. doi:10.1093/infdis/131.6.649 [CrossRef]
- Zakka KA, Foos RY, Brown WJ. Intraocular coccidioidomycosis. Surv Ophthalmology. 1978;22(5):313–321 doi:10.1016/0039-6257(78)90176-5 [CrossRef]
- Wålinder PE, Kock E. Endogenous fungus endophthalmitis. Acta Ophthalmol (Copenh). 1971;49(2):263–272. doi:10.1111/j.1755-3768.1971.tb00950.x [CrossRef]
- Jampol LM, Sung J, Walker JD, et al. Choroidal neovascularization secondary to Candida albicans chorioretinitis. Am J Ophthalmol. 1996;121(6):643–649. doi:10.1016/S0002-9394(14)70630-0 [CrossRef]
- Pesin SR, Thomas MA, Smith ME. Combined rhegmatogenous-traction retinal detachment following successful treatment of Candida chorioretinitis. Arch Ophthalmol. 1992;110(8):1051–1052. doi:10.1001/archopht.1992.01080200031013 [CrossRef]
- Kusaka S, Hayashi N, Ohji M, et al. Macular hole secondary to fungal endophthalmitis. Arch Ophthalmol. 2003;121(5):732–733. doi:10.1001/archopht.121.5.732 [CrossRef]
- McDonald HR, De Bustros S, Sipperley JO. Vitrectomy for epiretinal membrane with Candida chorioretinitis. Ophthalmology. 1990;97(4):466–469. doi:10.1016/S0161-6420(90)32559-9 [CrossRef]
Clinical Features of Patients With Fungal Chorioretinitis or Endophthalmitis
|Case #||Age/Gender||Fungemia Risk Factors||Retinal Comorbid Conditions||Eye(s), Presentation||Pattern(s) of Involvement||Initial VA||Final VA||Time to Follow-Up||Antifungal Treatment at Time of Initial OCT||Blood Culture Results||Treatment|
|1||36/M||TPN, right atrial vegetation||None||OD CR; OS CR||Chorioretinal (extrafoveal) retinal vascular||20/25; 20/25||n/a||Lost to follow up||Yes||C. albicans||Systemic: micafungin, fluconazole. Ocular: none|
|2||52/F||T-cell leukemia||None||OD E; OS CR||Chorioretinal (subfoveal) retinal vascular||CF; 20/70||CF; 20/40||3 weeks||No||C. dubilensis||Systemic: fluconazole. Ocular: voriconazole|
|3||64/F||Epiglottis cancer, chemotherapy||Wet AMD||OU E||Chorioretinal (extrafoveal, subfoveal)||20/400; CF||20/40; CF||25 months||No||C. dubilensis||Systemic: fluconazole; Ocular: voriconazole|
|4||36/F||TPN||None||OD CR; OS E||Chorioretinal (extrafoveal)||20/20; 20/60||20/20; 20/20||4 days||Yes||C. albicans||Systemic: fluconazole; Ocular: fluconazole|
|5||72/F||Uterine carcinoma; chemotherapy; TPN||Strabismic amblyopia OD||OU CR||Chorioretinal (extrafoveal)||20/400; 20/40||20/400; 20/30||8 weeks||Yes||C. albicans||Systemic: fluconazole Ocular: voriconazole|
|6||30/M||Intravenous drug use||None||OS E||Retinal vascular||20/400||20/70||2 months||No||N/A||Systemic: fluconazole; Ocular: voriconazole|
|7||51/M||HIV off HAART||None||OD E||Retinal vascular||20/50||20/25||4 months||No||No growth||Systemic: fluconazole; Ocular: voriconazole|