Color fundus photography of a 37-year-old, Asian, myopic woman, right (a) and left (b) eyes, demonstrating characteristic features of punctate inner choroidopathy
Watzke and colleagues reported that the lesions were hyperfluorescent in the early arterial phase of fluorescein angiography (FA) and exhibited staining in the arteriovenous phase (Watzke et al. 1984). More lesions appearing as tiny hyperfluorescent lesions that scattered in the posterior pole are observed using FA compared to clinical fundus examination. Areas of hyperfluorescence correspond to the lesions regardless of the presence of CNV and persist through the early and late venous phase; nevertheless, some lesions may demonstrate blocked fluorescence in the arterial phase. On the other hand, the presence of CNV is characterized by hyperfluorescence in the early phase and leakage in the late phase (Fig. 3.2) (Olsen et al. 1996). Leakage of fluorescein into a serous neurosensory retinal detachment is also possible (Watzke et al. 1984). In later stages of disease, punctate window defects are seen as the RPE atrophies.
Indocyanine green angiography (ICGA) can detect many more lesions, which appear as hypofluorescent areas in both early and late phases; as in the case of other white dot syndromes, these hypofluorescent areas correspond to the choroidal lesions that represent localized areas of hypoperfusion (Levy et al. 2005).
Several choroidal vessels revealed localized hyperfluorescent points close to the vessel wall/border, which indicates the presence of associated vasculitis. Larger choroidal vessels crossed these hypofluorescence areas, which suggest that the vasculitis is limited to smaller choroidal vessels and the choriocapillaris (Tiffin et al. 1996).
Spectral-domain optical coherence tomography (SD-OCT) reveals RPE elevation that fluctuates with disease activity and sub-RPE hyperreflective signals with intact Bruch’s membrane (BM) (Fig. 3.3). Photoreceptor-associated bands are not visible during the active phase but are readily detected when lesions are clinically stable. This finding may aid in monitoring the clinical course in patients with PIC (Channa et al. 2012).
Zhang and colleagues suggest that five stages (I–V) of choroidal vasculitis may occur during the evolution of PIC lesions and show that OCT findings vary depending on the stage (Zhang et al. 2011, 2013). In stage I, “presumed choroidal vasculitis,” there is minimal irregularity in the outer nuclear layer because the involved level is confined to the inner choroid. In stage II, “obliteration of choriocapillaris,” the lesion is represented by a focal hyperreflective elevation of the RPE with corresponding disruption of the photoreceptor ellipsoid zone (Fig. 3.4). In stage III, “inflammatory infiltration,” the nodular lesions with moderate reflectivity break through the RPE, forming a hump-shaped lesion beneath the outer plexiform layer. This may be associated with subsequent disruption of BM and involvement of the outer retina (Fig. 3.4). In stage IV, “atrophy,” lesions regress in a retrograde manner with tissue loss from the photoreceptor layer and inner choroid. This results in a V-shaped herniation of the outer plexiform layer and inner retina into the choroid (Fig. 3.5). In stage V, “progressive atrophy and pigmentation,” there is loss of photoreceptors around the lesion. This is often seen as extensive attenuation of the external limiting membrane, photoreceptor ellipsoid zone, and interdigitation zone, adjacent to the PIC lesions (Fig. 3.6).
Spaide and colleagues reported that acute lesions appeared as nodular collections under RPE and ruptured into the outer retina with outpouring of infiltrate (Spaide et al. 2013). They point out that the differentiation between active inflammatory lesions and CNV may not be possible even with multimodal imaging, because both can cause infiltrative lesions with breakdown in the blood–retina barrier. However, the use of OCT angiography (OCTA) to evaluate PIC lesions may enable differentiation between active inflammatory lesions and CNV, even in patients in whom FFA was inconclusive (Fig. 3.7) (Levison et al. 2017).
In a study using enhanced depth imaging (EDI)-OCT, about 20% of clinically inactive PIC patients showed localized RPE elevation with an underlying hyporeflective space, which has been previously described as a sign of activity (Levison et al. 2017). The authors suggested that this may represent subclinical PIC. Choroidal thickness can be used to monitor the stage of disease activity. Choroidal thickness increases throughout the active phase, then significantly decreases at later stages due to atrophy of outer retinal layers, reaching a lower minimum level than that attained during the early stages of the disease (Zhang et al. 2013). The use of serial quantitative assessment of retinal thickness maps on SD-OCT for detecting flare-up of PIC lesions and monitoring treatment response has also been described (Madhusudhan et al. 2016).
Fundus autofluorescence (FAF) imaging can distinguish between the different phases of PIC lesions (Arcinue et al. 2015). Active lesions demonstrated hypoautofluorescent spots with hyperautofluorescent margins (Fig. 3.8). This hyperautofluorescent margin resulted from the gradual loss of the photoreceptors around the active lesions with intact RPE. Both subclinical lesions and atrophic lesions appeared hypoautofluorescent, but the subclinical lesions appeared more distinctive on near-infrared FAF imaging than on blue FAF imaging. The hypoautofluorescence of subclinical lesions was the result of small clusters of damaged RPE cells overlying focal chorioretinal lesions, whereas the hypoautofluorescence of atrophic lesions was the result of a lack of RPE cells carrying fluorophores. The progression of PIC can be predicted noninvasively using FAF, which reflects the progression of RPE atrophy (Hua et al. 2014). In stage II PIC lesions, defined by Zhang and colleagues as focal elevation of the RPE and corresponding disruption of EZ (Zhang et al. 2011, 2013), the mean progression rate of RPE atrophy was 3.735 mm2/year, which was greater than that of laser scars in patients with nonproliferative diabetic retinopathy cases (0.127 mm2/year).
Management
The management of PIC is challenging due to the variability in disease severity and the cause of vision loss between patients or even in the same patient at different time points. Intervention is generally indicated to treat new or active inflammatory lesions, particularly those threatening the fovea or secondary CNV. The condition of the fellow eye should also be considered while formulating treatment strategies. Treatment options include local and systemic corticosteroids, systemic immunomodulatory drugs, intravitreal anti-vascular endothelial growth factory (VEGF), photodynamic therapy (PDT), argon laser, submacular surgery, and combination therapy.
Observation
No treatment is advised for the majority of patients without evidence of CNV or inflammatory lesions very close to fixation, especially those who show a self-limiting course with good visual prognosis (Amer and Lois 2011). Essex and colleagues reported that 66% of eyes with PIC lesions but no CNV at baseline remained unchanged without new PIC lesions or CNV lesions during a mean follow-up of 4.5 years (Essex et al. 2010). In this patient cohort, VA was also well preserved, with a mean VA of 0.11 logMAR to 0.09 logMAR.
Medical Therapy
Corticosteroids: Systemic or local (periocular or intravitreal) administration of corticosteroids is thought to effectively inhibit the critical immune and inflammatory pathways in PIC and PIC-associated CNV, and steroids are thought to have an antiangiogenic role (Levy et al. 2005; Flaxel et al. 1998; Brueggeman et al. 2002). In the Gerstenblith survey, 60% of PIC patients underwent treatment with systemic corticosteroids, 22% with intraocular corticosteroids, and 10% with periocular corticosteroids (Gerstenblith et al. 2007). The effectiveness of corticosteroids in controlling PIC-associated CNV was reported by several investigators (Flaxel et al. 1998; Levy et al. 2005) who postulated that the use of corticosteroids might not alter outcomes but may result in faster recovery of vision.
Other Immunomodulatory Therapy: Corticosteroid-sparing immunosuppressants should be considered in patients who require maintenance doses greater than 7.5 mg of prednisolone per day or when there are specific contraindications for the ongoing corticosteroid therapy. Mycophenolate mofetil (CellCept®; Genentech, San Francisco, CA) is commonly used (Galor et al. 2008; Ehlers et al. 2011; Turkcuoglu et al. 2011), and the use of other agents such as sirolimus (rapamycin), interferon beta-1A, and thalidomide has been reported in some cases (Nussenblatt et al. 2007; Cirino et al. 2006; Ip and Gorin 1996).
Intravitreal Anti-VEGF Therapy
Several studies have evaluated the safety and efficacy of anti-VEGF therapy in CNV secondary to PIC, specifically bevacizumab (Mansour et al. 2009; Zhang et al. 2012; Chan et al. 2008) and ranibizumab (Chan et al. 2008; Rouvas et al. 2011). Most of the studies reported favorable anatomical and functional outcomes. Figure 3.9 shows a representative case with CNV secondary to PIC treated with bevacizumab.