Table 1 An overview of the PICOS and other criteria used for study inclusion and exclusion

Study design

Published and unpublished randomized controlled studies

Systematic or non-systematic reviews and meta-analyses

Dose or frequency comparison studies

Preclinical studies, retrospective prognostic studies, and case reports

Ad-hoc analyses of randomized controlled study data

Editorials, commentaries, letters, and consensus reports

Crossover randomized controlled studies

Pilot studies (if phase not mentioned), phase I and II randomized controlled studies (to be included as second-level evidence, if primary evidence is unavailable)

Controlled observational studies (to be included as second-level evidence, if primary evidence is unavailable)

Separate searches will be performed as required

Single dose of intervention studies

Studies of less than 3 months follow-up

Population

Patients with DME

Interventions

Eylea/VEGF Trap-Eye/aflibercept

Systemic treatments (alone or in combination with intervention)

Anti-VEGF treatments (any including ranibizumab/Lucentis, bevacizumab/Avastin, and pegaptanib/Macugen)

Surgery (alone or in combination with intervention)

Subtenon injections

Intravitreal steroids (any including triamcinolone, fluocinolone acetonide/Iluvien, dexamethasone/Ozurdex, and implants)

Laser treatments

NOTE the intervention should be to treat the DME not to treat cataracts

The above interventions can be included if combined with other treatments (e.g., eye drops) except the exclusions

Comparators

Placebo, best standard care, masked control, sham, and eye drops

Systemic treatments (alone or in combination with intervention)

Any intervention (from those listed as interventions)

Surgery (alone or in combination with intervention)

NOTE: this can be a single treatment/implant

Clinical Outcomes

Number of injections/visits/assessments

BCVA (mean change from baseline, mean average change from baseline, as measured by ETDRS score or Snellen equivalent)

  Loss of ≤ 15, ≥ 15, ≥ 30 ETDRS letters

  Gain of ≥ 0, 10, 15, 30 ETDRS letters

  20/40 vision or better (Snellen chart)

  20/200 or worse (Snellen chart)

Reduction in laser use

Anatomical changes (e.g., change in CNV and lesion area, central foveal thickness, and fluid on OCT)

Health-related quality of life (EQ-5D, NEI VFQ-25, and other scales)

Treatment discontinuations

Serious AE (all serious AE, all ocular serious AE, death, endophthalmitis, uveitis, retinal tear, diabetic macular/retinal edema, reduced visual acuity, vitreous hemorrhage, corneal abrasion, and any others)

AE (all AE, all ocular AE, all non-ocular AE, retinal detachment, retinal ischemia, lens damage, all grades of ocular inflammation, eye pain, increased ocular pressure, retinal degradation, macular edema, cataract, neovascularization, and any others)

Serious non-ocular AE (all, non-fatal cardiac infarction, non-fatal stroke, non-ocular hemorrhage, hypertension, serious systemic events, arterial thrombotic events, and venous thrombotic events)

Language

Any

Additional criteria necessary for inclusion in indirect and mixed treatment analysis

Studies that were connected by one arm only and did not form a closed network, unless they included comparators of interest

Studies that formed loops but did not lie along the path between IVT-AFL 2q8 versus comparators of interest (IVR 0.5 mg PRN, or implants of dexamethasone 0.7 mg or fluocinolone acetonide 0.2 μg/day)

Studies that did not report 12-month outcomes