Vision Loss after PDT

In the last few years, several ocular factors have been suggested to explain the variable outcome of photodynamic therapy with verteporfin for treatment of subfoveal choroidal neovascularization secondary to age-related macular degeneration (AMD) (e.g., baseline visual acuity, choroidal neovascularization size and/or composition), but the possible role of thrombophilia in photodynamic therapy effects and side effects has not been considered. Acute severe vision decrease (ASVD) has been observed after standardized photodynamic therapy in about 1.0% of predominantly classic choroidal neovascularizations and in 3.0% to 4.5% of occult or minimally classic lesions. Although the possibility of a nonselective chorioretinal nonperfusion has been suggested in a few patients, the chance that ASVD was related to an abnormal occlusive response of the macular vasculature to photodynamic therapy cannot be definitely ruled out. In the course of our routine clinical practice, we have observed only one case of ASVD after photodynamic therapy, with poor visual outcome. This adverse event occurred in a 68-year-old white woman suffering from AMD and subfoveal occult choroidal neovascularization with a minimal classic component (Fig 1 [available at http://aaojournal.org]). Fourteen days after photodynamic therapy, fundus examination, fluorescein angiography, and indocyanine green angiography documented increased subretinal fluid, delayed filling of the capillary perifoveal net, marked reduction of choroidal neovascularization leakage, and a persistent hypofluorescent nonperfusion pattern of the choroidal vasculature in the light-exposed area (Fig 2 [available at http://aaojournal.org]). A careful reanalysis of all data concerning both the patient and the photodynamic application ruled out the possibility of an accidental verteporfin overdosage, as well as different procedural mistakes and/or technical malfunction. A retrospective anamnesis pointed out that the patient correctly followed all the usual pre– and post–photodynamic therapy advice, revealing that she consumed a protein breakfast 5 hours before the photodynamic therapy execution. These findings prompted us to investigate the possibility of an excessive, individual thrombotic response to photodynamic therapy. The following hemostasis changes were identified: (1) moderate increase of activated partial thromboplastin time (37.5 seconds) and plasma fibrinogen level (488 mg/dl), (2) significant alterations of the ProC Global test (Agkistrodon contortrix snake venom ratio, 0.70 [normal value > 0.85]) and clot longevity properties (decrease of fibrinolytic capacity, 35% less than the normal controls), (3) marked increase of plasma homocysteine concentration after methionine load test (absolute increment of fasting total plasma homocysteine, 51.6 mic-mol/l; absolute difference between post–methionine load and fasting total plasma homocysteine, 41.5 mol/l; percentage difference over fasting total plasma homocysteine, 507%), and (4) several hyperhomocysteinemic gene polymorphisms (GG homozygosity for the 66 methionine synthase reductase, CT heterozygosity for the 677 methylenetetrahydrofolate reductase, and AG heterozygosity for the 2756 methionine synthase). Other laboratory analyses, including erythrocyte sedimentation rate, glucose tolerance test, folate, vitamin B12 and vitamin B6 plasma levels, fasting total plasma homocysteine, lipoprotein profile, complete blood cell count, lupus anticoagulant, prothrombin time, antithrombin, and protein S and protein C levels were within the normal range. Assessments for the detection of other mutations associated with increased thrombophilic risk—that is, methylenetetrahydrofolate reductase A1298C, factor V Leiden R506Q, prothrombin G20210A, factor XIII Val34Leu, 68–base pair insertion polymorphism in exon 8, and cystathionine-beta-synthase G919A—did not show any abnormality. Ophthalmologic, hematologic, and genetic findings were suggestive of a prolonged, nonselective full-thickness choroidal vaso-occlusion due, perhaps, to abnormal photothrombotic perturbation in the photodynamic therapy–treated area (Fig 2C, D [available at http://aaojournal.org]). Poor overall anticoagulant and fibrinolytic capacity, together with acute (induced by a protein breakfast in a patient with a marked increase of plasma homocysteine concentration after a methionine load test) and chronic (due to the carrier status for methionine synthase reductase, methylenetetrahydrofolate reductase, and methionine synthase polymorphisms) hyperhomocysteinemia, could have led synergistically to an individual lack of photodynamic therapy selectivity. This hypothesis is based on the overlapping between the photodynamic therapy mechanism of action and the hyperhomocysteinemic modifications of vascular endothelium. During photodynamic therapy, in the endothelial-cell membrane of choroidal neovascularization the activated photosensitizer triggers localized oxidative changes, followed by platelet adhesion/degranulation, release of thromboxane A2, and prostaglandins, as happened in the course of the vascular damage related to elevated homocysteine plasma concentration. Thus, the hyperhomocysteinemia could be responsible for nonselective verteporfin binding to the entire macular vasculature, owing to an alteration of the endothelium receptor expression. Ours is the first report of ASVD after photodynamic therapy in a patient with a thrombophilic diathesis. We postulate that the exposure to an environmental triggering event (i.e., photodynamic therapy) may be the precipitating factor for the occurrence of nonselective full-thickness choroidal thrombosis. Retrospective investigations may be useful to ascertain the presence of a significant risk correlation between thrombophilic disorders and severe photodynamic therapy adverse events. Moreover, further prospective studies are required to establish whether the photodynamic therapy effectiveness/safety profile is determined not only by the container (e.g., choroidal neovascularization size and/or composition), but also by its contents (e.g., hematologic factors).