Table 2.
Overview of the visual acuity outcomes of the Protocol T.
| Visual acuity letter score and Snellen equivalent | Aflibercept | Bevacizumab | Ranibizumab | Aflibercept vs. bevacizumab | Aflibercept vs. ranibizumab | Ranibizumab vs. bevacizumab | |||
|---|---|---|---|---|---|---|---|---|---|
| Difference (95% CI) | p valuea | Difference (95% CI) | p valuea | Difference (95% CI) | p valuea | ||||
| Letter score of <69, equivalent to 20/50 or worse, at baseline | |||||||||
| Number of eyes | 102 | 102 | 101 | ||||||
| Visual acuity at baseline | |||||||||
| Mean (SD) letter score | 56.2 (11.1) | 56.6 (10.6) | 56.5 (9.9) | ||||||
| Approximate Snellen equivalent | 20/80 | 20/80 | 20/80 | ||||||
| Visual acuity at 1 year | |||||||||
| Mean (SD) letter score | 75.2 (10.9) | 68.5 (13.6) | 70.7 (12.0) | ||||||
| Approximate Snellen equivalent | 20/32 | 20/40 | 20/40 | ||||||
| Change from baseline in letter score | |||||||||
| Mean (SD) improvement | 18.9 (11.5) | 11.8 (12.0) | 14.2 (10.6) | 6.5 (2.9 to 10.1) | <0.001 | 4.7 (1.4 to 8.0) | 0.003 | 1.8 (−1.1 to 4.8) | 0.21 |
| Improvement of ≥10 letters, n (%) | 79 (77) | 61 (60) | 70 (69) | 17 (2 to 31) | 0.02 | 10 (−4 to 23) | 0.20 | 7 (−6 to 20) | 0.28 |
| Worsening of ≥10 letters, n (%) | 1 (1) | 4 (4) | 2 (2) | −3 (−7 to 2) | 0.56 | −1 (−5 to 3) | 0.56 | −1 (−6 to 3) | 0.56 |
| Improvement of ≥15 letters, n (%) | 68 (67) | 42 (41) | 50 (50) | 24 (9 to 39) | <0.001 | 18 (4 to 32) | 0.008 | 6 (−7 to 19) | 0.34 |
| Worsening of ≥15 letters, n (%) | 1 (1) | 2 (2) | 2 (2) | 0 (−3 to 3) | 0.85 | −1 (−4 to 2) | 0.85 | 1 (−3 to 4) | 0.85 |
|
| |||||||||
| Letter score of 78 to 69, equivalent to 20/32 to 20/40, at baseline | |||||||||
| Number of eyes | 106 | 104 | 105 | ||||||
| Visual acuity at baseline | |||||||||
| Mean (SD) letter score | 73.5 (2.6) | 72.8 (2.9) | 73.4 (2.7) | ||||||
| Approximate Snellen equivalent | 20/32 | 20/40 | 20/40 | ||||||
| Visual acuity at 1 year | |||||||||
| Mean (SD) letter score | 81.4 (8.3) | 79.9 (10.1) | 81.6 (6.8) | ||||||
| Approximate Snellen equivalent | 20/25 | 20/25 | 20/25 | ||||||
| Change from baseline in letter score | |||||||||
| Mean (SD) improvement | 8.0 (7.6) | 7.5 (7.4) | 8.3 (6.8) | 0.7 (−1.3 to 2.7) | 0.69 | −0.4 (−2.3 to 1.5) | 0.69 | 1.1 (−0.9 to 3.1) | 0.69 |
| Improvement of ≥10 letters, n (%) | 53 (50) | 47 (45) | 52 (50) | 6 (−9 to 21) | 0.82 | 0 (−13 to 14) | 0.95 | 6 (−10 to 21) | 0.82 |
| Worsening of ≥10 letters, n (%) | 4 (4) | 2 (2) | 1 (1) | 2 (−3 to 6) | 0.54 | 3 (−1 to 7) | 0.54 | −1 (−4 to 2) | 0.54 |
| Improvement of ≥15 letters, n (%) | 19 (18) | 17 (16) | 16 (15) | 2 (−7 to 11) | 0.73 | 4 (−5 to 12) | 0.73 | −2 (−10 to 7) | 0.73 |
| Worsening of ≥15 letters, n (%) | 2 (2) | 1 (1) | 1 (1) | 1 (−2 to 4) | 0.99 | 1 (−2 to 4) | 0.99 | 0 (−3 to 3) | 0.99 |
aTreatment group comparisons were performed with ANCOVA models adjusted for continuous baseline visual acuity or from binomial regression models adjusted for categorical baseline visual acuity (adapted from Diabetic Retinopathy Clinical Research Network et al. [29]). CI: confidence interval; SD: standard deviation; n = number.