Kaya 2005.

Methods	Country where data collected: Turkey Parallel group trial Unit of randomisation: participant Unit of analysis: ulcer Duration: trial duration not reported; reported treatment duration ranged up to 106 days; primary outcome reported at durations from 21 to 85 (arm 1) and 15 to 83 (arm 2) days.
Participants	Inclusion criteria: spinal‐cord injury patients with pressure ulcers Exclusion criteria: not reported N = 27 (49 ulcers) N males: 24/27 (group distribution NR) Mean age (years): 35.3 vs 29.7 years Ulcer size (cm²) 4.13 vs 6.45 Ulcer stage: I: 6/25 vs 6/24, II: 17/25 vs 17/24; III: 2/25 vs 1/24 Ulcer location: sacral 6/25 vs 7/24, ischia 6/25 vs 3/24, heel 6/25 vs 2/24, greater trochanter 3/25 vs 6/24, iliac crest 0/25 vs 4/24, knee 1/25 vs 2/24, head of fibula 0/25 vs 2/24, lateral malleolus 2/25 vs 0/24, dorsum of foot 0/25 vs 1/24. All ulcers were non‐infected; patients were hospitalised
Interventions	Intervention arm 1: hydrogel‐type dressing (Elasto‐Gel) changed every 4 days or more frequently if membrane contaminated or non‐occlusive Intervention arm 2: povidone‐iodine‐soaked gauze changed daily Cointerventions: necrotic areas debrided
Outcomes	Primary outcome: proportion of wounds completely healed
Notes	Funding: NR
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: “We prospectively studied 27 hospitalised patients with spinal‐cord injury (24 males and three females) who had a total of 49 pressure ulcers. Each patient was randomly assigned to one of two groups.” Comment: how randomised unclear
Allocation concealment (selection bias)	Unclear risk	Quote: “We prospectively studied 27 hospitalised patients with spinal‐cord injury (24 males and three females) who had a total of 49 pressure ulcers. Each patient was randomly assigned to one of two groups.” Comment: no information on concealment of allocation
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: “Ulcers were graded using the NPUAP system to ensure consistency in both groups. Surface area was used as an indicator of healing, and measured in cm².” Comment: no indication if assessment was blinded to allocation
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: “For ulcers that healed during the hospital stay, the rate of healing (cm²/days) was calculated as the initial surface area (cm²) divided by healing time (days). Where patients were discharged before healing was complete, it was calculated by subtracting the ulcer surface area at the most recent examination from the baseline surface area, then dividing this by the treatment time (days) while in hospital.” Comment: outcome data reported for all randomised patients, but it is unclear how many patients’ data were estimated
Selective reporting (reporting bias)	Unclear risk	Quote: "For each lesion, we recorded location, rate of healing, healing time and treatment time. Healing time (days) was defined as the time from the start of treatment to when 100% epithelialisation was observed." Comment: outcomes appear consistently listed or reported throughout text. Not clear what is primary and secondary outcome.
Other bias	High risk	Quote: “We prospectively studied 27 hospitalised patients with spinal‐cord injury (24 males and three females) who had a total of 49 pressure ulcers .” Comment: randomisation was conducted at a patient level but the analysis was carried out at the level of the ulcer; it did not appear that paired data (multiple ulcers from individual participants) were accounted for in the analysis.