One hundred and two patients were coded as receiving PDT for CSCR between 2009 and 2019. Fifty-three cases (55 patients) were excluded for the following reasons: unclear diagnosis and/or suspected choroidal neovascularization (10 patients), insufficient data or follow up shorter than 3 months (38 patients), not central PDT (3 patients), previously treated (4 patients).
Forty-seven eyes (47 patients) were included in the final analysis, 21 eyes in the early treatment group and 26 eyes in the delayed treatment group. The baseline characteristics of the two groups are detailed in table 1.
There was a significantly higher percentage of women in the long duration group (4.8% vs 38.5%, P=0.007) but the mean age was similar. The BCVA at baseline was significantly worse in the delayed treatment group (0.500±0.26 vs 0.331±0.24, P=0.042). The delayed treatment group had longer follow-up duration post PDT treatment (Table 1), but the difference between the two groups was not statistically significant (P=0.195).
Prior to PDT treatment, CRT was higher among subjects in the early treatment group as compared to those in delayed treatment group, almost reaching statistical significance (P=0.052). No other significant differences were observed in baseline OCT scans or PDT treatment parameters. Most treatments in both groups were performed using half fluence (66%). There was also no difference between the groups in terms of recurrent treatments (1.5 vs 1.2). Prior oral treaments or intravitreal injections had no significant effect on the PDT treatment outcomes.
The PDT treatment was successful (i.e. absorption of SRF – P<0.001) in 76% and 77% of eyes in the early and delayed treatment groups respectively. A statistically significant resolution of SHRM and RPE irregularities was also observed in the delayed treatment group (P<0.001). Both groups showed significant improvement in CRT measurements at the 3-months follow up which persisted to the last follow-up visit (Fig. 1). Both CRT and max retinal thickness decreased significantly after the PDT treatment (P<0.001) and the gap between the two groups narrowed (figure 1a and 1b respectively).
In the subgroup of 29 patients (9 early and 20 delayed) with available manual measurements, the retina was significantly thinner in the delayed treatment group (193.3±49.2 microns vs 132.3±61.4, P=0.014, figure 2a) yet there was non-statistically significant difference in SRF height (144.4±85.0 vs 132.2±77.2, P=0.570, figure 2b). Manual CRT measurements also had a significantly negative correlation with BCVA at baseline and at the 3-months and last follow-up visit (Spearman's rho -0.498, P=0.006; -0.691, P<0.001; -0.533, P=0.003 respectively).
Overall, BCVA improved significantly by nearly 1 line at last follow-up (from 0.42±0.26 to 0.36±0.33 LogMAR, P=0.017). BCVA improved in both groups (figure 3). In sub-analysis of only successful cases, BCVA improved from 0.44±0.28 to 0.36±0.36 (P=0.013), however the improvement was statistically significant only in the delayed treatment group (from 0.52±0.28 to 0.43±0.42, P=0.046) and the early treatment group had a non-significant change (from 0.34±0.26 to 0.26±0.24, P=0.126).
Compared to the early treatment group, the rate of improvement in BCVA was higher in the delayed treatment group at the 3-months follow-up visit (15/26, 57.7% vs. 5/21, 23.8%; P=0.014) but the difference did not reach statistical significance at the last follow-up visit analysis (15/27, 57.7% vs. 8/21, 38.1%; P=0.181). Figure 4 depicts the rate of clinically significant improvement/deterioration in BCVA (i.e. equivalent to 3 ETDRS lines) at the 3-months (a) and at the last follow-up (b) visits. Most of the subjects in both groups did not show a clinically significant change in BCVA and the rates of improvement were quite similar.
Overall, the BCVA at each visit had a significant negative correlation with CRT and max retinal thickness at the 3-month follow-up visit (Spearman's rho -0.359, P=0.014 and -0.401, P=0.021 respectively) but not at the last follow-up visit (-0.177, P=0.235; -0.246, P=0.155).