Imaging Ocular Toxoplasmosis

Toxoplasmosis occurs following infection with Toxoplasma gondii, an obligate intracellular parasite.^1–3 The prevalence of the protozoan in soil samples, wild and domesticated animals, and humans is high – with over one-third of the world’s population believed to be infected.⁴ Transmission of T. gondii is typically foodborne, but can occur from animal to human via oocyst ingestion, from mother to child during gestation, or rarely as a result of contaminated organ transplantation or blood transfusion. Retinal involvement has been estimated to occur in approximately 2% of those infected and is associated with a lifelong risk of recurrence and progressive vision loss. The diagnosis of toxoplasmic retinochoroiditis (TRC) is most commonly made clinically with the identification of moderate to severe vitreous inflammation associated with a focal retinochoroiditis, often with an adjacent or nearby retinochoroidal scar. Serologic testing, polymerase chain reaction-based analysis of ocular fluids, and multimodal imaging (MMI)^5–7 can be useful, particularly in atypical presentations⁸ or when direct visualization of the retina is limited. This issue of Ocular Immunology & Inflammation contains four original articles^9–12 addressing the role of ocular imaging in the diagnosis and management of TRC.

Zicarelli et al⁹ retrospectively reviewed fundus photographs (FP), optical coherence tomography angiography (OCTA), indocyanine green angiography (ICGA), and fluorescein angiography (FA) images in 15 eyes of 15 patients with active TRC seen in tertiary referral clinics in Milan and Abu Dhabi between 2018 and 2019. The diagnosis of TRC was largely clinical, with serologic support as indicated. The primary measures were lesion area using various imaging modalities and detection of so-called ‘satellite dark dots’ (SDD) on OCTA and ICGA imaging of the choroid. Retinal mean lesion area as measured on FP, OCTA of the retinal circulation (OCTA ret), and FA were all quite similar at approximately 4 mm² (p = .90), whereas choroidal mean lesion area was consistently about twice that size on OCTA of the choriocapillary (OCTA cc) and deeper choroid (OCTA ch), and on ICGA (~6.5 mm²; p = .95). Roughly half of the subjects with active TRC showed SDD (7/15; 46.7%). The authors concluded that choroidal imaging techniques should be considered in eyes with presumed or possible TRC to provide a comprehensive assessment of activity within the choroid. Although not mentioned specifically by the authors, such imaging can also aid in the diagnosis of clinically atypical cases, since the degree of choroidal involvement seen commonly in TRC would be unusual for those disorders most often in the differential diagnosis, including necrotizing herpetic retinitis (NHR), syphilis, and primary vitreoretinal lymphoma (PVRL). Both OCT and OCTA offer convenient and non-invasive means of measuring involvement of the choroid underlying active lesions within the posterior pole.

Ebrahimiadib et al¹⁰ retrospectively reviewed the retinal and choroidal OCT findings of 32 eyes in 31 patients with active (n = 14), partially active (n = 13), and inactive (n = 16) TRC lesions seen at a tertiary referral clinic in Iran between 2018 and 2019. The diagnosis of TRC was largely clinical, with serologic support as indicated. Among the active lesions common findings including punctate vitreous opacities consistent with vitreous inflammation (100%), localized retinal thickening with loss of retinal layer details (termed ‘smudge effect’; 100%), choroidal shadowing (12/13; 92%), serous retinal detachment (9/13; 64%), and localized loss of underlying choroidal vascular architecture (8/12; 61.5%). Among the partially active lesions frequent findings included opacities consistent with vitreous inflammation (8/10 imaged; 80.0%), persistent retinal thickening with choroidal shadowing (9/13; 69%) and continued choroidal thickening with loss of choroidal vascular architecture (5/12; 41.6%). Only two eyes (15.3%) had persistent SRD. Among inactive lesions, persistent vitritis was present, but less common (6/14; 43.0%), with consistent retinal thinning (100%), disorganization of the normal retinal pigment epithelium (RPE)-Bruch’s membrane interface (100%), and choroidal thinning (12/16; 75.0%). The authors concluded that OCT imaging can be used to assess where in disease evolution a given TRC lesion may be on any given visit, and that features typical of TRC can be used to differentiate such infections from NHR, syphilis, PVRL and other causes of retinitis, as noted above.

Kelgoankar et al¹¹ retrospectively described the clinical and MMI findings of six eyes of six patients with punctate inner retinal toxoplasmosis (PIRT) seen in a tertiary referral clinic in India between 2014 and 2020. The diagnosis was made clinically with positive serologic support and after ruling out other causes of focal retinitis as indicated. The lesions were differentiated anatomically from the better known punctate outer retinal toxoplasmosis (PORT) lesions based on primary involvement of the inner versus outer retina on OCT, with PIRT lesions extending to the outer plexiform layer and/or sparing the outer nuclear layer. Lesions were unilateral in all but one eye, were yellow-white in color, small in diameter (mean 2.41 mm; range 1.6–3.2 mm), and tended to result in minimal elevation of the inner retina. All eyes also had more typical TRC lesions, either active or inactive, often in close proximity. Three eyes had co-existing PORT lesions. Vitreous inflammation tended to be minimal. Fundus autofluorescence (FAF) tended to be unremarkable and distinguished from PORT, which tends to show hypo-FAF. Fluorescein angiography showed slowly progressive hyperfluorescence with leakage consistent with active retinitis. While some inner retinal thinning was noted with resolution, this tended to be mild. Three eyes with PIRT subsequently developed a more typical full-thickness TRC lesion at or near the same site. The authors concluded that PIRT is a not uncommon component of the full TRC lesion spectrum and that MMI can be useful to confirm the diagnosis, as with PORT.¹³

Conrady et al¹² retrospectively described eight patients who developed active TRC following exposure to wild game. Patients were collected from tertiary referral centers in Michigan, Boston, Miami, and Pennsylvania. All eight subjects were men with a mean age of 56 years (range 29 to 71 years), and in each case the diagnosis was made clinically with serologic support as indicated. Unilateral TRC occurred two to eight weeks after hunting, cleaning (3/8), or consuming (5/8) wild deer (7/8) or bear (1/8), including one patient who cut himself while cleaning the deer. One patient was chronically immunosuppressed following renal transplantation and four were 60 years of age or older and so relatively immunosuppressed. Two patients had two contiguous active lesions. Positive anti-T. gondii IgM antibodies were identified in three subjects. The authors noted that three of five subjects who developed TRC following consumption of undercooked meat were taking proton pump inhibitors for gastroesophageal reflux and suggested that such treatment may have increased susceptibility to systemic T. gondii infection by raising the pH of the gastrointestinal tract. The authors reminded us of the high rate of toxoplasmosis among many species of wild game and recommended that hunters take appropriate precautions when cleaning and consuming such meats. Of note, the authors saw no need to mention or show posterior segment images in six of the eight cases, supporting the notion that TRC remains a clinical diagnosis in most patients.

Together, these studies demonstrate the utility of MMI, and particularly OCT, in select patients suspected of having TRC. This is especially true when the direct view of the posterior segment is limited, the findings are atypical, and/or the more anatomically defined diagnoses of PORT or PIRT are being considered. Involvement of the underlying choroid on OCT can be used to help differentiate full-thickness TRC from other causes of necrotizing and non-necrotizing retinitis.