Cochrane · systematic review · CD007419
Anti‐vascular endothelial growth factor for diabetic macular oedema: a network meta‐analysis
Virgili, G, Curran, K, Lucenteforte, E, et al. · Cochrane Database of Systematic Reviews · 2023 · Cochrane Library
SoF tableOnly the outcomes this review put in its Summary of Findings table are listed (the Cochrane Handbook caps them at about seven per comparison). Its other analyses are not shown here.
Questions
37
current version
GRADE rated
37
No studies
0
Included studies
NR
No studies 0 · 0%Ungraded 0 · 0%Very low 7 · 18.9%Low 14 · 37.8%Moderate 14 · 37.8%High 2 · 5.4%
Ungraded means an effect was reported without a GRADE rating; it is not a gap. No studies means the review found nothing to answer the question. The six categories are a display order, not a score.
Questions and results
Current version onlySummary of findings: mean change in best‐corrected visual acuity from baseline to 24 months
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Aflibercept | no estimate extracted | 3 studies · 641 participants | Moderate Downgraded: imprecision |
Bevacizumab | no estimate extracted | 3 studies · 261 participants | Low Downgraded: imprecision, indirectness |
Brolucizumab | no estimate extracted | 1 study · 154 participants | Low Downgraded: publication bias |
Ranibizumab deferred laser | no estimate extracted | 1 study · 139 participants | Low Downgraded: imprecision |
Ranibizumab prompt laser | no estimate extracted | 3 studies · 287 participants | Very low Downgraded: risk of bias, inconsistency |
Laser | no estimate extracted | 5 studies · 598 participants | Low Downgraded: risk of bias, imprecision |
Summary of findings: mean change in best‐corrected visual acuity from baseline to 12 months
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Faricimab (PTI) | no estimate extracted | 1 study · 632 participants | Moderate Downgraded: imprecision |
Aflibercept | no estimate extracted | 5 studies · 1592 participants | Moderate Downgraded: imprecision |
Brolucizumab | no estimate extracted | 1 study · 368 participants | Moderate Downgraded: imprecision |
Conbercept | no estimate extracted | 1 study · 125 participants | Low Downgraded: risk of bias, imprecision |
Bevacizumab | no estimate extracted | 5 studies · 373 participants | Moderate Downgraded: risk of bias |
Ranibizumab deferred laser | no estimate extracted | 1 study · 188 participants | High |
Ranibizumab prompt laser | no estimate extracted | 7 studies · 746 participants | Moderate Downgraded: risk of bias |
Laser | no estimate extracted | 13 studies · 1296 participants | Moderate Downgraded: indirectness |
Summary of findings: gain of three or more ETDRS lines from baseline to 24 months
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Aflibercept | no estimate extracted | 2 studies · 487 participants | Moderate Downgraded: imprecision |
Bevacizumab | no estimate extracted | 1 study · 185 participants | Low Downgraded: imprecision |
Ranibizumab prompt laser | no estimate extracted | 2 studies · 170 participants | Very low Downgraded: risk of bias, imprecision |
Ranibizumab deferred laser | no estimate extracted | 1 study · 139 participants | Low Downgraded: imprecision |
Laser | no estimate extracted | 4 studies · 559 participants | High |
Summary of findings: mean change in central retinal thickness from baseline to 24 months
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Aflibercept | no estimate extracted | 3 studies · 640 participants | Moderate Downgraded: imprecision |
Brolucizumab | no estimate extracted | 1 study · 154 participants | Moderate Downgraded: imprecision |
Bevacizumab | no estimate extracted | 2 studies · 261 participants | Moderate Downgraded: imprecision |
Ranibizumab deferred laser | no estimate extracted | 1 study · 136 participants | Moderate Downgraded: imprecision |
Ranibizumab prompt laser | no estimate extracted | 1 study · 136 participants | Moderate Downgraded: imprecision |
Laser | no estimate extracted | 4 studies · 563 participants | Moderate Downgraded: imprecision |
Summary of findings: all‐cause mortality at longest available follow‐up
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Aflibercept | no estimate extracted | 10 studies · 2644 participants | Very low Downgraded: imprecision, indirectness |
Bevacizumab | no estimate extracted | 4 studies · 305 participants | Low Downgraded: imprecision |
Brolucizumab | no estimate extracted | 1 study · 368 participants | Low Downgraded: imprecision |
Conbercept | no estimate extracted | 1 study · 125 participants | Very low Downgraded: risk of bias, imprecision |
Faricimab | no estimate extracted | 2 studies · 1262 participants | Low Downgraded: imprecision |
Ranibizumab | no estimate extracted | 11 studies · 2065 participants | Low Downgraded: imprecision |
Summary of findings: Antiplatelet Trialists Collaboration arterial thromboembolic events at longest available follow‐up
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Aflibercept | no estimate extracted | 10 studies · 2701 participants | Very low Downgraded: imprecision, indirectness |
Bevacizumab | no estimate extracted | 3 studies · 310 participants | Low Downgraded: imprecision |
Brolucizumab | no estimate extracted | 1 study · 369 participants | Low Downgraded: imprecision, indirectness |
Conbercept | no estimate extracted | 1 study · 125 participants | Very low Downgraded: imprecision, indirectness |
Faricimab | no estimate extracted | 2 studies · 1262 participants | Low Downgraded: imprecision, indirectness |
Ranibizumab | no estimate extracted | 11 studies · 1501 participants | Very low Downgraded: imprecision, indirectness |
Antiangiogenic therapy versus control
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Antiangiogenic therapy vs control | no estimate extracted | NR | High |
Visual acuity change at 1 year Measured on the logMAR scale, range −0.3 to 1.3. Higher values represent worse visual acuity. Antiangiogenic therapy vs control | no estimate extracted | NR | Moderate |
Central retinal thickness μm (CRT) change at 1 year The aim of treatment is to reduce central retinal thickness so thinner is better. Antiangiogenic therapy vs control | no estimate extracted | NR | High |
Quality of life: NEI‐VFQ composite score at 6 to 12 months An improvement by 5 units is clinically significant. Antiangiogenic therapy vs control | no estimate extracted | NR | Moderate |
All serious systemic adverse events at 1 to 2 years Antiangiogenic therapy vs control | no estimate extracted | NR | High |
Arterial thromboembolic events at 1 to 2 years Antiangiogenic therapy vs control | no estimate extracted | NR | Low |
Death at 1 to 2 years Antiangiogenic therapy vs control | no estimate extracted | NR | Very low |
Ranibizumab versus aflibercept for diabetic macular oedema
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Ranibizumab vs aflibercept | RR 0.75 [0.6, 0.94] | NR | Moderate |
Visual acuity change at 1 year Measured on the logMAR scale, range −1.3 to 1.3. Higher values represent worse visual acuity. Ranibizumab vs aflibercept | no estimate extracted | NR | Moderate |
Central retinal thickness μm (CRT) change at 1 year The aim of treatment is to reduce central macular thickness so thinner is better. Ranibizumab vs aflibercept | no estimate extracted | NR | Low |
Quality of life at 1 year Ranibizumab vs aflibercept | no estimate extracted | NR | No studies |
All serious systemic adverse events at 1 to 2 years Ranibizumab vs aflibercept | RR 0.98 [0.82, 1.19] | NR | High |
Arterial thromboembolic events at 1 to 2 years Ranibizumab vs aflibercept | RR 1.24 [0.48, 3.19] | NR | Very low |
Death at 1 to 2 years Ranibizumab vs aflibercept | RR 1.16 [0.38, 3.58] | NR | Very low |
Ranibizumab versus bevacizumab for diabetic macular oedema
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Ranibizumab vs bevacizumab | RR 1.11 [0.87, 1.43] | NR | Moderate |
Visual acuity change at 1 year Measured on the logMAR scale, range −1.3 to 1.3. Higher values represent worse visual acuity. Ranibizumab vs bevacizumab | no estimate extracted | NR | Moderate |
Central retinal thickness (CRT) change at 1 year The aim of treatment is to reduce central macular thickness so thinner is better. Ranibizumab vs bevacizumab | no estimate extracted | NR | Low |
Quality of life at 1 year Ranibizumab vs bevacizumab | no estimate extracted | NR | No studies |
All serious systemic adverse events at 1 to 2 years Ranibizumab vs bevacizumab | RR 1.04 [0.84, 1.28] | NR | Moderate |
Arterial thromboembolic events at 1 to 2 years Ranibizumab vs bevacizumab | RR 1.17 [0.43, 3.13] | NR | Very low |
Death at 1 to 2 years Ranibizumab vs bevacizumab | RR 0.73 [0.22, 2.37] | NR | Very low |
Antiangiogenic therapy versus control
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Antiangiogenic therapy vs control | no estimate extracted | NR | High |
Visual acuity change at 1 year Measured on the logMAR scale, range −0.3 to 1.3. Higher values represent worse visual acuity. Antiangiogenic therapy vs control | no estimate extracted | NR | Moderate |
Central retinal thickness μm (CRT) change at 1 year The aim of treatment is to reduce central retinal thickness so thinner is better. Antiangiogenic therapy vs control | no estimate extracted | NR | High |
Quality of life: NEI‐VFQ composite score at 6 to 12 months An improvement by 5 units is clinically significant. Antiangiogenic therapy vs control | no estimate extracted | NR | Moderate |
All serious systemic adverse events at 1 to 2 years Antiangiogenic therapy vs control | no estimate extracted | NR | High |
Arterial thromboembolic events at 1 to 2 years Antiangiogenic therapy vs control | no estimate extracted | NR | Low |
Death at 1 to 2 years Antiangiogenic therapy vs control | no estimate extracted | NR | Very low |
Ranibizumab versus aflibercept for diabetic macular oedema
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Ranibizumab vs aflibercept | RR 0.75 [0.6, 0.94] | NR | Moderate |
Visual acuity change at 1 year Measured on the logMAR scale, range −1.3 to 1.3. Higher values represent worse visual acuity. Ranibizumab vs aflibercept | no estimate extracted | NR | Moderate |
Central retinal thickness μm (CRT) change at 1 year The aim of treatment is to reduce central macular thickness so thinner is better. Ranibizumab vs aflibercept | no estimate extracted | NR | Low |
Quality of life at 1 year Ranibizumab vs aflibercept | no estimate extracted | NR | No studies |
All serious systemic adverse events at 1 to 2 years Ranibizumab vs aflibercept | RR 0.98 [0.82, 1.19] | NR | High |
Arterial thromboembolic events at 1 to 2 years Ranibizumab vs aflibercept | RR 1.24 [0.48, 3.19] | NR | Very low |
Death at 1 to 2 years Ranibizumab vs aflibercept | RR 1.16 [0.38, 3.58] | NR | Very low |
Ranibizumab versus bevacizumab for diabetic macular oedema
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Ranibizumab vs bevacizumab | RR 1.11 [0.87, 1.43] | NR | Moderate |
Visual acuity change at 1 year Measured on the logMAR scale, range −1.3 to 1.3. Higher values represent worse visual acuity. Ranibizumab vs bevacizumab | no estimate extracted | NR | Moderate |
Central retinal thickness (CRT) change at 1 year The aim of treatment is to reduce central macular thickness so thinner is better. Ranibizumab vs bevacizumab | no estimate extracted | NR | Low |
Quality of life at 1 year Ranibizumab vs bevacizumab | no estimate extracted | NR | No studies |
All serious systemic adverse events at 1 to 2 years Ranibizumab vs bevacizumab | RR 1.04 [0.84, 1.28] | NR | Moderate |
Arterial thromboembolic events at 1 to 2 years Ranibizumab vs bevacizumab | RR 1.17 [0.43, 3.13] | NR | Very low |
Death at 1 to 2 years Ranibizumab vs bevacizumab | RR 0.73 [0.22, 2.37] | NR | Very low |
Anti‐VEGF versus laser for diabetic macular oedema
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Anti‐VEGF vs laser | RR 3.6 [2.7, 4.8] | 10 studies · 1333 participants | High |
Loss 3+ lines of visual acuity at 1 year Anti‐VEGF vs laser | RR 0.11 [0.05, 0.24] | 7 studies · 1086 participants | High |
Visual acuity at 1 year Anti‐VEGF vs laser | no estimate extracted | 9 studies · 1292 participants | High |
Central macular thickness at 1 year Anti‐VEGF vs laser | no estimate extracted | 8 studies · 1215 participants | High |
Adverse events: Anti‐VEGF compared with control for diabetic macular oedema
Population: people with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Serious systemic adverse events Adverse events: Anti‐VEGF vs control | RR 0.98 [0.83, 1.17] | 15 studies · 2985 participants | Moderate Downgraded: indirectness |
Total ATC thromboembolic events Adverse events: Anti‐VEGF vs control | RR 0.89 [0.63, 1.25] | 14 studies · 3034 participants | Moderate Downgraded: indirectness |
Death Adverse events: Anti‐VEGF vs control | RR 0.88 | 15 studies · 3562 participants | Moderate Downgraded: indirectness |
Anti‐VEGF versus laser for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year Anti‐VEGF vs laser | RR 3.2 [2.07, 4.95] | 5 studies · 556 participants | Moderate Downgraded: imprecision, inconsistency |
Loss 3+ lines of visual acuity at 1 year Anti‐VEGF vs laser | RR 0.13 [0.05, 0.34] | 4 studies · 481 participants | Moderate Downgraded: imprecision, inconsistency |
Cost‐effectiveness studies, UK setting Anti‐VEGF vs laser | no estimate extracted | NR | No studies |
NICE 2011 Anti‐VEGF vs laser | no estimate extracted | 1 participants | Ungraded |
Mitchell 2012 Anti‐VEGF vs laser | no estimate extracted | 1 participants | Ungraded |
Anti‐VEGF compared with sham for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 6 to 12 months Anti‐VEGF vs sham | RR 2.19 [1.36, 3.53] | 3 studies · 497 participants | Moderate Downgraded: imprecision, inconsistency |
Loss 3+ lines of visual acuity at 6 to 12 months Anti‐VEGF vs sham | RR 0.28 [0.13, 0.59] | 2 studies · 411 participants | Moderate Downgraded: imprecision, inconsistency |
Ranibizumab plus laser compared with laser alone for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Gain 3+ lines of visual acuity at 1 year ‐ Prompt photocoagulation Ranibizumab plus laser vs laser alone | RR 2.11 [1.67, 3.13] | 3 studies · 786 participants | Moderate Downgraded: inconsistency |
Gain 3+ lines of visual acuity at 1 year ‐ Deferred photocoagulation Ranibizumab plus laser vs laser alone | RR 1.88 [1.31, 2.7] | 1 study · 481 participants | Moderate Downgraded: inconsistency |
Loss 3+ lines of visual acuity at 1 year ‐ Prompt photocoagulation Ranibizumab plus laser vs laser alone | RR 0.29 [0.13, 0.67] | 2 studies · 708 participants | Moderate Downgraded: inconsistency |
Loss 3+ lines of visual acuity at 1 year ‐ Deferred photocoagulation Ranibizumab plus laser vs laser alone | RR 0.27 [0.1, 0.77] | 1 study · 481 participants | Moderate Downgraded: inconsistency |
Adverse events: Anti‐VEGF compared with control for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Total ATC thromboembolic events Adverse events: Anti‐VEGF vs control | RR 0.85 [0.56, 1.28] | 9 studies · 2159 participants | Low |
Death Adverse events: Anti‐VEGF vs control | RR 0.95 | 9 studies · 2159 participants | Low |
Pegaptanib compared to sham treatment for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Loss of 3 or more lines of visual acuity Pegaptanib vs sham treatment | RR 0.57 [0.18, 1.85] | 1 study · 171 participants | Low |
Gain of 3 or more lines of visual acuity Pegaptanib vs sham treatment | RR 1.84 [0.57, 5.99] | 1 study · 171 participants | Low |
Mean change of visual acuity (logMAR) Pegaptanib vs sham treatment | no estimate extracted | 1 study · 171 participants | Low |
Bevacizumab compared to sham treatment for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Visual acuity (logMAR) Bevacizumab vs sham treatment | no estimate extracted | 1 study · 78 participants | Low |
Bevacizumab compared to photocoagulation for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Loss of 3 or more lines of visual acuity Bevacizumab vs photocoagulation | RR 0.25 [0.04, 1.37] | 1 study · 87 participants | Low |
Gain of 3 or more lines of visual acuity Bevacizumab vs photocoagulation | RR 2.58 [1.06, 6.31] | 1 study · 87 participants | Low |
Mean change of visual acuity (logMAR) Bevacizumab vs photocoagulation | no estimate extracted | 1 study · 87 participants | Moderate |
Bevacizumab compared to triamcinolone for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Loss of 3 or more lines of visual acuity Bevacizumab vs triamcinolone | no estimate extracted | 1 study · 26 participants | Ungraded |
Gain of 3 or more lines of visual acuity Bevacizumab vs triamcinolone | RR 0.09 [0.01, 1.49] | 1 study · 26 participants | Low |
Visual acuity (logMAR) Bevacizumab vs triamcinolone | no estimate extracted | 1 study · 26 participants | Very low |
Bevacizumab compared to bevacizumab/triamcinolone (IVTA) for diabetic macular oedema
Population: patients with diabetic macular oedema
| Outcome | Effect | Evidence | Status / GRADE |
|---|---|---|---|
Loss of 3 or more lines of visual acuity Bevacizumab vs bevacizumab/triamcinolone (IVTA) | RR 0.74 [0.5, 1.09] | 1 study · 82 participants | Low |
Gain of 3 or more lines of visual acuity Bevacizumab vs bevacizumab/triamcinolone (IVTA) | RR 2.49 [0.85, 7.34] | 1 study · 82 participants | Low |
Mean change of visual acuity (logMAR) Bevacizumab vs bevacizumab/triamcinolone (IVTA) | no estimate extracted | 2 studies · 160 participants | Low |