Hyland 2015.

Study characteristics
Methods	2‐arm randomised controlled trial Conducted in Sydney, Australia Follow‐up: 6 months
Participants	61 paediatric participants with acute partial‐thickness burns TBSA 3% to 4% Inclusion criteria: children less than 16 years of age with acute partial‐thickness burns undergoing debridement and SSG Exclusion criteria: children with full‐thickness burns, facial burns, and those requiring delayed or staged skin grafting
Interventions	Intervention arm: Versajet II Exact (Smith+Nephew) hydrosurgical system wound debridement and SSG Control arm: conventional tangential burn wound debridement using a Goulian knife and SSG Co‐intervention: general anaesthesia, antiseptic povidone–iodine operative site preparation, sterile draping
Outcomes	Primary review outcomes: time to complete healing, postoperative infection Secondary review outcomes: operative time, scar outcome
Notes	The primary outcome of the study was dermal preservation as measured by a 2‐millimetre punch biopsy taken pre‐ and postoperatively. All wounds had healed by the end of the study period. There was a key difference between the study protocol and report in the power calculation and primary outcome. We sought further information from the authors who clarified that the study was powered to detect a difference in the primary outcome of dermal preservation, despite this being different from the protocol (see 'Risk of bias' table).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The randomisation process was clearly described with no group imbalances.
Allocation concealment (selection bias)	Low risk	Allocation was concealed until enrolment. Allocation was revealed in the operating theatre immediately prior to intervention delivery.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Surgeons were not blinded, which is unavoidable in this study.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Participants and other health professionals were blinded. There were no deviations from the assigned intervention. There was appropriate blinding, no attrition for outcomes of interest, and no cross‐over.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There was no attrition.
Selective reporting (reporting bias)	High risk	There are some discrepancies between the published article and the protocol registered in the Australian and New Zealand Clinical Trials Registry (ACTRN12610000956000). The protocol states that the primary outcome would be the modified VSS at 3 months, with secondary outcomes as the % graft take, microbiology from swabs after debridement, duration of surgery, and histological assessment of the remaining dermis after debridement. However, the published report states that dermal preservation was the primary outcome. It is not clear why this change was made. Furthermore, the protocol states that the target sample size was 150, although the published report states 60 without any figures to support this power calculation. It is possible that recruitment was terminated before sufficient individuals were recruited to power the original planned primary outcome (modified VSS), so the authors selected a different outcome for the published report (dermal preservation), perhaps because it was the only comparison to yield a statistically significant difference between groups. Overall, the risk of bias is high.
Other bias	Unclear risk	We are told that “time to healing after‐skin grafting and burn was also determined independently by an experienced senior burn nurse as days to complete epithelialisation”, but it is unclear whether grafts/wounds were examined daily or at wider intervals (e.g. 2 or 3 days), or if this was consistent for all participants. It is not usual clinical practice to inspect burn wounds/grafts daily, and so it is likely that there was a greater time interval between wound inspections. Also, the ability of a nurse to determine re‐epithelialisation, the intra‐ and inter‐rate agreement and threshold used was not stated. Overall, there are concerns about inconsistency in the measurement of the outcome, putting this trial at unclear risk of bias for outcome measurement.
Overall risk of bias	High risk	Overall, there are some concerns that render the overall assessment of bias as high.

SSG: split‐thickness skin grafting
TBSA: total body surface area
VSS: Vancouver Scar Scale