There was a weak association between active disease and presence of choroidal lesions. Among 50 eyes identified clinically as having active disease, 38 (76%) had choroidal lesions; among the remaining 122 eyes, 74 (60.7%;
P = 0.055, chi-squared test) had choroidal lesions.
Table 3 shows further associations between choroidal findings and selected demographic and clinical characteristics.
Table 4 summarizes correlations between various choroidal findings. Thinner choroids tended to be in older participants, and thicker choroids were associated with increased hyperopia. A longer disease duration was associated with any type of choroidal lesion (OR: 1.08;
P = 0.03) and with thin choroids (OR: 1.10;
P = 0.01). Vitreous haze ≥0.5+ (clinically active disease) was associated with any type of choroidal lesion (OR: 4.43;
P = 0.02), with macular choroidal lesions (OR: 3.82;
P = 0.03), and with both thin (OR: 3.08;
P = 0.03) and thick (OR: 2.75;
P = 0.03) choroids but not with the presence of a suprachoroidal space (OR: 0.95;
P = 0.92). Macular edema was a risk factor for thick choroids (OR: 5.47;
P = 0.02), presence of any choroidal lesions (OR: 3.00;
P = 0.02), and macular lesions (OR: 3.65;
P = 0.05). Interestingly, macular edema appeared to be associated with the absence of a suprachoroidal space (OR: 0.22;
P = 0.02). The presence of any choroidal lesion and of macular lesions were strongly associated with thick choroids (OR: 3.89;
P = 0.001 and OR: 4.56;
P < 0.001, respectively) and with absence of a suprachoroidal space (OR: 0.43;
P = 0.02 and OR: 0.39;
P = 0.01, respectively). Presence of a suprachoroidal space was associated with having a normal choroidal thickness, defined as being within 1 SD of the mean (OR: 3.19;
P = 0.03). Relationships between choroidal findings and signs of intraocular inflammation remained essentially the same after adjustment for age, duration of disease, current immunomodulatory therapy, and current corticosteroid use in multivariate analyses (
Table 3).