TABLE 3.
ORs with 95% CrIs of network meta-analysis for the proportion of patients with a gain of at least 15 ETDRS letters.
| At 1-year follow-up | ||||||
|---|---|---|---|---|---|---|
| IVT-AFL | 2.05 ( 1.18, 3.58 ) | 2.85 ( 1.24, 6.41 ) | 2.89 ( 1.67, 5.08 ) | 4.72 ( 1.56, 14.47 ) | 5.48 ( 3.87, 7.97 ) | 9.16 ( 2.99, 32.83 ) |
| 1.55 ( 1.11, 2.17 ) | IVR | 1.4 (0.5, 3.74) | 1.41 (0.86, 2.32) | 2.31 (0.78, 6.89) | 2.69 ( 1.37, 5.23 ) | 4.44 ( 1.71, 14.23 ) |
| 2.78 ( 1.23, 6.04 ) | 1.79 (0.77, 4.03) | IVC | 1.01 (0.38, 2.77) | 1.66 (0.41, 6.71) | 1.93 (0.94, 4.11) | 3.23 (0.79, 14.85) |
| 1.85 ( 1.26, 2.73 ) | 1.19 (0.82, 1.75) | 0.67 (0.28, 1.61) | IVB | 1.63 (0.62, 4.38) | 1.9 (0.98, 3.7) | 3.15 ( 1.08, 11.07 ) |
| 3.03 ( 1.06, 8.71 ) | 1.95 (0.69, 5.57) | 1.1 (0.3, 4.12) | 1.63 (0.62, 4.41) | Dexamethasone implant | 1.17 (0.36, 3.78) | 1.95 (0.45, 9.29) |
| 5.35 ( 3.93, 7.41 ) | 3.45 ( 2.39, 5.09 ) | 1.93 (0.94, 4.1) | 2.89 ( 1.83, 4.63 ) | 1.77 (0.6, 5.19) | Laser | 1.66 (0.51, 6.25) |
| 6.9 ( 2.51, 22.69 ) | 4.43 ( 1.72, 13.93 ) | 2.51 (0.7, 10.25) | 3.73 ( 1.33, 12.46 ) | 2.3 (0.54, 10.73) | 1.29 (0.46, 4.29) | Placebo |
| At 2-year follow-up | ||||||
| IVT-AFL | 1.1 (0.64, 1.87) | - | 1.41 (0.86, 2.29) | - | 3.72 ( 2.63, 5.37 ) | 4.28 ( 2.22, 8.32 ) |
| 1.18 (0.8, 1.73) | IVR | - | 1.28 (0.74, 2.22) | - | 3.4 ( 1.88, 6.19 ) | 3.89 ( 2.67, 5.82 ) |
| - | - | IVC | - | - | - | - |
| 1.38 (0.95, 2.01) | 1.17 (0.78, 1.75) | - | IVB | - | 2.64 ( 1.57, 4.55 ) | 3.03 ( 1.56, 5.97 ) |
| 3.72 ( 2.64, 5.32 ) | 3.16 ( 1.94, 5.19 ) | - | 2.7 ( 1.71, 4.32 ) | Dexamethasone implant | - | - |
| - | - | - | - | - | Laser | 1.15 (0.56, 2.33) |
| 4.6 ( 2.67, 8 ) | 3.89 ( 2.66, 5.83 ) | - | 3.34 ( 1.92, 5.85 ) | - | 1.23 (0.66, 2.32) | Placebo |
Results of the network meta-analysis for all the population were in the lower triangle, and the estimation was calculated as the treatment in column compared with the treatment in row. ORs higher than 1 favor the treatment in the column. Results of the network meta-analysis for the population with worse baseline VA were in the upper triangle, and the estimation was calculated as the treatment in the row compared with the treatment in the column. ORs higher than 1 favor the treatment in row. Statistically, significance was presented in bold italic format. To obtain ORs for comparisons in the opposite direction, reciprocals should be taken.
CrI, credible interval; ETDRS, early treatment diabetic retinopathy study; IVB, intravitreal bevacizumab; IVC, intravitreal conbercept; IVR, intravitreal ranibizumab; IVT-AFL, intravitreal aflibercept; OR, odds ratio.