Table 2

Study quality

Study (author and year)Adequate sequence generationAllocation concealmentMaskingIncomplete outcome data addressedFree of selective reportingFree of other bias (eg, similarity at baseline, power assessment)Funder
Anti-VEGFs
Ranibizumab
 READ-2 Study28 47UnclearUnclearUnclearYes (91.3% completion)YesComparison groups similar at baseline; power analysis not mentionedJuvenile Diabetes Research Foundation, Genentech Inc
 RESOLVE Study (Massin et al)36YesYesYes (patients and outcome assessors)Yes (82% completion in sham arm, 90.2% with ranibizumab)YesComparison groups similar at baseline; power analysis unclearNovartis Pharma, Switzerland
 RESTORE Study (Mitchell et al)24YesUnclearYes (patients, outcome assessors)Yes (87.3–88.3% completion)YesComparison groups similar at baseline; power analysis carried out (power adequate for VA changes)Novartis Pharma, Switzerland
 RISE and RIDE (Nguyen et al)38YesYesYes (patients, treating physician masked to assigned dose of ranibizumab)Yes (2 year study completed by 83.3% of patients in RISE and by 84.6% in RIDE)YesComparison groups similar at baseline; ITT analysis; power analysis carried out (power adequate for primary endpoint)Genentech Inc
Bevacizumab
 BOLT Study (Michaelides et al)23 52YesUnclearPartial (outcome assessors, not patients)Yes (97.5% completion)YesComparison groups similar at baseline (except laser group had longer duration of clinically significant DMO); power analysis carried out (power adequate for VA changes)Moorfields Special Trustees, National Institute for Health Research
 Faghihi et al53YesUnclearYes (patientYes (100% completion)YesComparable groups at baselineNot specified
 Lam et al35YesYesYes (patients and technicians assessing BCVA, OCT and IOP)Yes (92.3% follow-up at 6 months)YesComparison groups similar at baseline; power analysis carried out (power adequate for CMT changes)Supported in part by the Action for Vision Eye Foundation Hong Kong (charity)
Pegaptanib
 Cunningham et al/Adamis et al39 57YesUnclearYes (patients and outcome assessors)Yes (95% completion)YesComparison groups similar at baseline; acknowledge lack of power to detect differences between doses of pegaptanibEyetech Pharmaceuticals Inc, New York, and Pfizer Inc, New York
 Sultan et al40YesUnclearYes (patients and outcome assessors)Yes (69.9–73.8% completion)YesComparison groups similar at baseline; power analysis carried out (power adequate for VA changes)Pfizer Inc, New York
Aflibercept
 Da Vinci et al30 58Unclear (predetermined randomisation scheme)UnclearYes (patients)Yes (85% completion)YesComparison groups similar at baseline, power calculation completedRegeneron Pharmaceuticals, Inc, New York
Steroids
Dexamethasone
 Haller et al59YesUnclearYes (patients to dexamethasone dose, outcome assessors)Yes (92% completion)YesComparison groups similar at baseline; power analysis carried out, but study not powered to detect differences in subgroupsOculex Pharmaceuticals Inc
Fluocinolone
 FAME Study  (Campochiaro et al)29 60UnclearUnclearPartial (patients, masking of outcome assessment not mentioned)Yes (drop-out rate 19.0–22.7%)YesComparison groups similar at baseline; power analysis not mentionedAlimera Sciences Inc, Atlanta, Georgia; Psivida Inc, Watertown, Massachusetts
 Pearson et al43YesUnclearThird party masked design (patient and investigator not masked)No losses to follow-upYesDemographic characteristics were similar between implant and SOC groups; power calculation done, study adequately poweredBausch & Lomb Inc, Rochester, New York
Triamcinolone
 DRCR Network 2008 22 61 63 64YesUnclearPartial (patients to triamcinolone dose, outcome assessors not formally masked but generally not aware of participant’s study group)Yes (81–86% completion)YesComparison groups similar at baseline; power analysis carried out (power adequate for VA changes)Cooperative agreement from the National Eye Institute, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Department of Health and Human Services
 Gillies et alSutter et al32 136–138YesYesYes (patients, outcome assessors)Yes (91% completion intervention, 83% control)YesComparison groups similar at baseline (but limited demographic data); power analysis carried out (power adequate for VA changes)Sydney Eye Hospital Foundation and Juvenile Diabetes Research Foundation, New York
 Gillies et al33YesYesYes (patients, outcome assessors)Yes (84.5% completion)YesPower analysis carried out (power adequate for VA changes)National Health and Medical Research
Council, Canberra, Australia, and the Sydney Eye Hospital Foundation
Sydney, Australia
 Lam et al34YesYesPartial (outcome assessors)No losses to follow-upYesComparison groups similar at baseline; power analysis carried out (power adequate for CMT changes)Action for Vision Foundation, Hong Kong
 Ockrim et al/Sivaprasad et al42 62YesUnclearUnclearYes (94% completion)YesComparison groups similar at baseline; power analysis carried out (power adequate for VA changes)Special Trustees of Moorfields Eye Hospital
Active comparator trials
 Ahmadieh et al31YesYesYes (patients and outcome assessors)UnclearYesCMT lower in control group at baseline (p<0.05), other baseline values similar; power analysis carried out (power adequate for CMT changes)Not reported
 DRCR Network 21 46YesUnclearYes (patients, except deferred laser group; outcome assessors); masking discontinued after the first yearYes (1 year completion for 91–95% of eyes)YesComparison groups similar at baseline; power analysis carried out (power adequate for VA changes)Cooperative agreement from the National Eye Institute, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health and Human Services; Ranibizumab provided by Genentech, triamcinolone provided by Allergan Inc; companies also provided funds to defray the study's clinical site costs
 Lim et al55YesUnclearYes (investigators only)Yes (7.5% drop out after enrolment)YesGroups similar at baseline. The bevacizumab group received more injectionsNot reported
 Soheilian et al37 41YesYesYes (patients and outcome assessors)Unclear (36 week completion for 76–88%)YesCMT significantly lower and VA significantly better in MPC group at baseline, other baseline values similar; power analysis carried out (power adequate for VA changes)Ophthalmic Research Centre, Labbafinejad Medical Center, Tehran
  • MPC, macular photocoagulation.