Possible vitreous involvement in a case with rapidly progressing choroidal neovascularization.

A 65-year-old man with subfoveal choroidal neovascularization (CNV) underwent photodynamic therapy (PDT). Despite the sequential treatments, the CNV grew larger and finally penetrated the retina. Vitreous adhesion was observed at the edge of the supraretinal fibrotic tissue. The case highlighted the possible unexpected side-effect of PDT. The upregulation of the vascular endothelial growth factor or the enhanced vitreous traction was considered to be responsible for the event.

Photodynamic therapy (PDT) with verteporfin was the first-line treatment for subfoveal choroidal neovascularization (CNV) secondary to age-related macular degeneration (AMD) before the era of anti-vascular endothelial growth factor (VEGF) therapy.[1] However, a certain percentage of cases are resistant to the treatment and recurrences are sometimes observed.[2] Here, we report a case with massive CNV growth following a series of PDT. The case highlights the possible vitreous involvement in CNV aggravation as having been suggested previously[3] after the treatment.

Case Report

A 65-year-old man presented with metamorphopsia and visual loss in the left eye (OS). His corrected visual acuity was 1.0 (20/20) in the right eye (OD) and 0.4 (20/50) OS. Funduscopic examination revealed subfoveal CNV, serous retinal detachment and pigment epithelium detachment OS [Fig. 1a]. Fluorescein angiography and optical coherence tomography (OCT) confirmed the presence of subfoveal classic CNV [Fig. 1a]. He underwent PDT with verteporfin. The treatment was temporarily effective. Fluorescein leakage from the CNV and the serous retinal detachment regressed. However, he noticed deteriorating metamorphopsia and visual loss five months after the treatment. Funduscopic examination and angiography confirmed the exacerbation of CNV [Fig. 1b]. He was treated with sequential PDT.

Figure 1

Fundus photographs, fluorescein angiography, and OCT images of a 65-year-old man with subfoveal AMD. At the first visit subfoveal CNV accompanying the serous retinal detachment and the pigment epithelium detachment was confirmed (a). The CNV grew larger after the first PDT (b). The CNV penetrated the retina five months later. Vitreous traction was observed at the edge of fibrotic lesion (c). The CNV was stabilized after about one and a half year involving six times of PDT (d)

Although the second PDT temporarily stabilized the CNV, the CNV aggravated again in three months. The OCT image revealed that the CNV grew further into the retina as a mass and the vitreoretinal adhesions were noted in both eyes [Fig. 1c]. Although he received PDT two more times, CNV grew prominently. After 16 months with the intervention of PDT six times the CNV gradually turned into fibrous tissue and his visual acuity ended up in 0.04 (20/500) OS [Fig. 1d].

Discussion

The present case showed massive CNV growth after PDT, suggesting a possible unexpected side-effect of PDT. At least two mechanisms were postulated.

The first was up-regulated VEGF after PDT. VEGF is a potent angiogenic factor and is known to be associated with the development and growth of CNV. Although PDT is effective for the treatment of AMD, there are some concerns about VEGF up-regulation after PDT.[4] In the present case, it was possible that PDT induced hypoperfusion of the normal choroid, which enhanced VEGF up-regulation and induced rapid CNV growth.

The second hypothesis included the inflammation and the subsequent vitreous traction. OCT images, taken when the CNV penetrated the retina, showed the vitreo-macular traction on the margin of the CNV mass, and in the scarring phase, thickening of the posterior vitreous membrane continuous with the CNV scar tissue was observed [Fig. 1d]. The abnormal vitreo-macular traction is frequently observed in AMD patients and is suggested to be involved in the development or progression of the disease.[35] In addition, it is suggested that the local inflammation induced by AMD causes abnormal adhesion.[6] In the present case, CNV could be connected to the posterior vitreous membrane through the intraretinal connective tissue. The inflammation following PDT might have induced retraction of the prefoveolar vitreous cortex, which pulled up the CNV. Although the present case did not show clearly whether the pathologic vitreoretinal adhesion is the cause or the result of the CNV deterioration, vitreous traction should have played an important role in the clinical course.

The present case indicates the possibility of CNV aggravation after PDT. Since anti-VEGF therapy was not available at the observed period, PDT seemed the only option. Considering the supposed VEGF up-regulation after PDT and the superior effect of anti-VEGF agents, PDT should be indicated cautiously. In particular, in case of the presence of strong vitreous-retinal adhesion and traction, CNV may be less susceptible to PDT.