Felzani 2011.

Study characteristics
Methods	Research design: RCT, double‐blind, single‐centre Care setting: hospitalised patients Country of origin: Italy Publication source:Advances in Therapy, Springer Healthcare Year of publication: 2011 Duration of follow‐up: 15 days Sources of funding: not described Unit of randomisation: participant Unit of analysis: ulcers Inclusion criteria: hospitalised patients of both sexes; aged above 18 years, with a foreseen hospitalisation period of longer than 15 days, with grade 1 to 3 decubitus ulcers Exclusion criteria: patients that could not co‐operate with the hygienic measures to be adopted for the treatment of sores and those with a history of intolerance to hyaluronic acid; need of concomitant local and/or general antibiotic therapy for skin lesions or for systemic diseases
Participants	Number of participants: 59 participants were recruited, and 50 participants with 54 pressure ulcers were included in analysis. Participants to be treated were divided into 3 groups based on ulcer stage: first group (stage I), 20 participants; second group (stage II), 20 participants; third group (stage III), 10 participants. Among participants in the third group, 2 participants had 2 lesions, and 1 participant had 3 lesions. Therefore, 14 decubitus ulcers were treated. Male gender: 42% (n = 21) Age: 56 ± 7 years
Interventions	Details of the intervention Group 1: hyaluronic acid (Lys‐HA; Lysial, Fatai‐Nyl Srl; Jasper LLC, Lugano, Switzerland) Group 2: sodium hyaluronate Dressing procedure: wounds were initially thoroughly cleaned with saline. Blood clots, foreign bodies, and excess necrotic tissue were removed with gauze. Macerated skin borders were surgically removed. After these cleaning operations, the cream was applied as a thin layer across the ulcer surface. For the secondary medication, fat gauzes were preferred for direct contact with the wound, whereas final dressing was performed with sterile gauzes. Dressing changes were made daily during the first week of the study, and every other day during the second week. Co‐interventions: nutrition supplements and patient mobilisation and turning were provided according to the standard of care Duration of treatment: 15 days
Outcomes	Primary outcomes of the review: complete wound healing for stage III ulcers Secondary outcomes: time necessary to reach 50% lesion size regression
Notes	Losses to follow‐up: 100% of participants completed the treatment
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "This single‐centre randomised controlled trial (RCT) was conducted double‐blinded" Comments: did not specify the method of randomisation
Allocation concealment (selection bias)	Unclear risk	No information provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "The products were provided in identical containers, the only difference being the batch number
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The nursing team was unaware of the study treatment allocation." "Lesion analysis was performed in a blinded manner by expert specialized investigators and the following quantitative criteria were examined: lesion size (area), and regression time of 50% of lesion size"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "specifically, the first group (stage 1 decubitus ulcers) was initially formed by 25 patients; 5 participants were considered as not assessable since treatment was suspended. In 3 cases, this was because patients left the institute, in the other 2 cases it was because worsening of their condition due to underlying disease requiring antibiotic therapy. The second group (stage 2 decubitus ulcers) was initially formed by 24 patients; of those, 2 were considered as not assessable due to spontaneous suspension of study treatment, and 2 were excluded due to worsening of their condition requiring antibiotic therapy." Comment: participants withdrawal were justified and balanced between groups
Selective reporting (reporting bias)	High risk	Authors did not present mean (or corrected mean by covariate) and a measure of variability such as standard deviation for ulcer area or degree of area changes nor measure of variability for time to reach 50% lesion size regression. Complete ulcer healing was reported only for stage III wounds.
Other bias	High risk	Data analysis was based on number of ulcers that exceed the number of randomised participants.