Objectives: To compare topographic features of surgically excised subfoveal choroidal neovascularization with preoperative and postoperative fluorescein angiographic features from Submacular Surgery Trials (SST) patients, and to compare histological and angiographic features with preoperative and postoperative visual acuities (VAs).
Methods: Patients enrolled in the SST Groups N, B, and H trials between October 1999 and September 2001 and assigned to the surgery arm had surgically removed choroidal neovascularization sent to the SST Pathology Center. Grossly intact specimens were sectioned serially for 2-dimensional reconstruction and were assigned to growth pattern groups based on topographic mapping of the location of cellular components relative to the retinal pigment epithelium (RPE): sub-RPE, subretinal, combined, or indeterminate. These features were compared with preoperative fluorescein angiographic features. The histological choroidal neovascularization growth pattern was compared with preoperative VAs.
Results: Two-dimensional reconstructions of surgically excised choroidal neovascularization could not be matched point for point to fluorescein angiographic features. Among the 52 specimens selected, the growth pattern could be determined by 2-dimensional reconstruction in 34 instances (65%), including 28 (80%) of 35 Group N specimens, 2 (40%) of 5 Group B specimens, and 4 (33%) of 12 Group H specimens. Among the choroidal neovascularization growth patterns that could be determined from specimens submitted, the majority of Group N specimens were combined, and the majority of Group H specimens were subretinal. In most instances for Group B specimens, the growth pattern was indeterminate. The postoperative abnormalities on fluorescein angiography were generally larger than measurements of excised specimens. The subretinal growth pattern was associated with the smallest decrease in 3-month postoperative average VA.
Conclusions: Among the 52 specimens from the SST with adequate tissue to try to evaluate the growth pattern of the neovascular lesion, 34 had patterns that could be determined. The subretinal growth pattern tended to correspond with lesions that were classic choroidal neovascularization without occult choroidal neovascularization on fluorescein angiography. The most favorable visual outcome occurred if the choroidal neovascularization had a subretinal growth pattern. The combined growth pattern tended to correspond with lesions on fluorescein angiography that had both classic choroidal neovascularization and occult choroidal neovascularization. The conclusions must be viewed within the context that 18 (35%) of the 52 participants could not be characterized for histologic growth pattern.