Table 6.
Change in Retinal Thickness from Baseline to 14-Week Visit*
| Change in OCT Central Subfield Thickness | Sham+ Focal/Grid/PRP Laser | Ranibizumab+ Focal/Grid/PRP Laser | Triamcinolone+ Focal/Grid/PRP Laser |
|---|---|---|---|
|
| |||
| N = 115 | N = 100 | N = 103 | |
| Overall Change† | |||
| Thickness(μm) Median (25th, 75th percentile) | 362 (287, 484) | 312 (259, 453) | 265 (230, 304) |
| Change from baseline (μm) Mean±standard deviation | −5±113 | −39±127 | −92±115 |
| Change from baseline (μm) Median (25th, 75th percentile) | 0 (−80, +70) | −26 (−92, +15) | −75 (−168, −17) |
| Difference in mean change from sham+focal/grid/PRP laser (95% CI) [P Value] ‡ | −35 (−64, −6) [P = 0.007] |
−100 (−128, −71) [P < 0.001] |
|
| Thickness ≥10% increase with at least a 25 μm increase from baseline, no. (%) | 44 (38%) | 17 (17%) | 10 (10%) |
| Relative risk (95% CI) [P Value] ‡ for comparison with sham+focal/grid/PRP laser | 1.0 | 0.44 (0.25, 0.79) [P = 0.002] |
0.24 (0.12, 0.48) [P < 0.001] |
| Thickness <250 with at least a 25 μm decrease from baseline, no. (%) | 12 (10%) | 17 (17%) | 28 (27%) |
| Relative risk (95% CI) [P Value] ‡ for comparison with sham+focal/grid/PRP laser | 1.0 | 2.07 (0.96, 4.47) [P = 0.04] |
3.15 (1.56, 6.36) [P < 0.001] |
| LogOCT, no (%) || | |||
| Two or more step improvement | 9 (8%) | 12 (12%) | 28 (27%) |
| At least 1, but less than 2 step improvement | 17 (15%) | 20 (20%) | 27 (26%) |
| Less than 1 step improvement and less than 1 step worsening | 67 (58%) | 58 (58%) | 45 (44%) |
| At least 1 step but less than 2 step worsening | 18 (16%) | 7 (7%) | 2 (2%) |
| Two or more step worsening | 4 (3%) | 3 (3%) | 1 (1%) |
|
| |||
| Baseline thickness <400 μm | N = 73 | N = 60 | N = 64 |
|
| |||
| Thickness(μm) Median (25th, 75th percentile) | 315 (254, 377) | 273 (246, 320) | 255 (225, 282) |
| Change from baseline (μm) Mean±standard deviation | +31±95 | −12±70 | −35±71 |
| Change from baseline (μm) Median (25th, 75th percentile) | +28 (−20, +93) | −9 (−61, +15) | −23 (−78, +11) |
| Thickness ≥10% increase with at least a 25 μm increase from baseline, no. (%) | 35 (48%) | 10 (17%) | 8 (13%) |
| Thickness <250 with at least a 25 μm decrease from baseline | 11 (15%) | 14 (23%) | 18 (28%) |
| LogOCT, no. (%) || | |||
| Two or more step improvement | 4 (5%) | 3 (5%) | 7 (11%) |
| At least 1, but less than 2 step improvement | 6 (8%) | 12 (20%) | 14 (22%) |
| Less than 1 step improvement and less than 1 step worsening | 42 (58%) | 40 (67%) | 40 (63%) |
| At least 1 step but less than 2 step worsening | 17 (23%) | 4 (7%) | 2 (3%) |
| Two or more step worsening | 4 (5%) | 1 (2%) | 1 (2%) |
|
| |||
| Baseline thickness ≥400 μm | N = 42 | N = 40 | N = 39 |
|
| |||
| Thickness(μm) Median (25th, 75th percentile) | 498 (395, 570) | 469 (369, 547) | 293 (247, 391) |
| Change from baseline (μm) Mean±standard deviation | −67±116 | −80±175 | −186±111 |
| Change from baseline (μm) Median (25th, 75th percentile) | −77 (−171, +37) | −81 (−184, −4) | −193 (−246, −132) |
| Thickness ≥10% increase with at least a 25 μm increase from baseline, no. (%) | 9 (21%) | 7 (18%) | 2 (5%) |
| Thickness <250 with at least a 25 μm decrease from baseline | 1 (2%) | 3 (8%) | 10 (26%) |
| LogOCT, no. (%) || | |||
| Two or more step improvement | 5 (12%) | 9 (23%) | 21 (54%) |
| At least 1, but less than 2 step improvement | 11 (26%) | 8 (20%) | 13 (33%) |
| Less than 1 step improvement and less than 1 step worsening | 25 (60%) | 18 (45%) | 5 (13%) |
| At least 1 step but less than 2 step worsening | 1 (2%) | 3 (8%) | 0 |
| Two or more step worsening | 0 | 2 (5%) | 0 |
Visits occurring between 70 and 153 days (between 10 and 22 weeks) from randomization were included as 14-week visits. When more than 1 visit occurred in this window, data from the visit closest to the 14-week target date were used.
Missing (or ungradeable) data as follows for the sham+focal/grid/PRP laser group, ranibizumab+focal/grid/PRP laser group, and triamcinolone+focal/grid/PRP laser groups, respectively: 3, 3, and 2.
Adjusted for baseline optical coherence tomography (OCT) retinal thickness and visual acuity, number of planned panretinal photocoagulation (PRP) sittings, and correlation between 2 study eyes. Confidence intervals (CI) are adjusted for multiple comparisons.
Logarithmic transformation of OCT central subfield thickness (LogOCT) is calculated by taking the log base 10 of the ratio of the central subfield thickness divided by 200 and rounding to the nearest hundredth. The change is the change in the log values.