Wittenberg 1999.

Study characteristics
Methods	Study design: prospective single‐blinded randomised, intra‐individual (split‐scar) controlled trial conducted in the USA. Scar segment as unit of randomisation. Duration of the study: 1 December 1996 to 31 May 1997 Follow‐up time: 40 weeks (intermediate‐term follow‐up) Protocol was published before recruitment of participants: approved by the institutional review board of the Henry Ford Health System, Detroit, Michigan Details of trial registration: not reported Funding sources: funding was provided through the Clarence S. Livingood Fund under the direction of Edward A. Krull, MD, and by a small projects fund, both at Henry Ford Hospital, Detroit, Michigan
Participants	Number of participants assigned: 20 participants (each participant's scar was divided in 3 segments ‐ 60 segments) Segment 1 (FLPDL ‐ 585‐nm Flashlamp‐Pumped Pulsed‐dye Laser): a total of 20 segments treated Segment 2 (SGS ‐ silicone gel sheeting): a total of 20 segments treated Segment 3 (control): a total of 20 segments not treated Loss: 1 One participant dropped out of the study and another 1 stopped using SGS (silicone gel sheeting). Number of participants assessed: 19 participants (with a total of 57 segments enrolled). Segment 1: a total of 19 segments treated Segment 2: a total of 18 segments treated Segment 3: a total of 19 segments not treated Inclusion criteria: participants older than 18 years with uniform, linear hypertrophic scars; non‐pregnant; skin type I ‐ VI. Exclusion criteria: treatment of the scar within the preceding 2 months; keloidal scarring; scars less than 8 cm long. Age: 49; 24 to 81 (mean; range) Gender: male: 5; 25 (number; %) female: 15; 75 (number; %) Duration of scar (months): 32; 4 to 240 (mean; range) Skin phototypes: Skin type II: 10; 50 (number; %) Skin type V: 2; 10 (number; %) Skin type VI: 8; 40 (number; %) Scar location: hip: 1; neck: 1; knee: 1; shoulder: 1; back: 1; thigh: 2; abdomen: 4; chest: 9 Previous treatment: silicone gel sheeting: 2; intralesional corticosteroid: 3; none: 15
Interventions	Segment 1: 1/3 of each participant’s scar was treated with an FLPDL (model SPTL‐1; Candela Laser Corp, Wayland, Mass) at a wavelength of 585‐nm, a pulsed duration of 450 microseconds, fluence per pulse between 6.5 and 8.0 J/cm², and a spot size of 5 mm with a 10% to 20% overlap. The fluence used on each participant was determined by the threshold dose—the lowest dose that produced nonblanchable purpura filling the entire spot size. Participants were given the option of applying a topical anaesthetic agent (2.5% lidocaine and 2.5% prilocaine cream) 2 hours before laser therapy. After laser treatments, the site was iced for 15 to 20 minutes. Participants (n = 19) received 4 laser treatments at 8‐week intervals. Segment 2: participants were instructed to wear SGS (Cica Care; Smith and Nephew, Largo, Fla) on the designated site (corresponding to 1/3 of their scar) for at least 12 continuous hours per day and to wash the SGS and test site with soap and water before and after treatment. The SGS was held in place using a dressing retention sheet (Hypafix tape; Smith and Nephew). Participants (n = 18) were treated for 24 weeks with SGS. Segment 3: 1/3 of the scar was randomised to control and left untreated for the study duration.
Outcomes	Primary outcomes Treatment‐related adverse effects: One participant dropped out of the study at week 24 because of pain during laser treatments. One participant did not use SGS because of skin irritation. Secondary outcomes Changes in burning pruritus and pain (not related to treatment): at each visit, participants rated their pain, burning, and pruritus based on a quartile scale from 1 (absent or minimal) to 4 (severe). Change in scar erythema: blood flow (erythema) was evaluated using a laser doppler. Measurements were taken at 0, 8, 16, 24, and 40 weeks. Change in scar elasticity: elasticity was evaluated with the handheld elastometer at 8, 16, 24, and 40 weeks. Change in scar volume: polydimethyl vinyl siloxane material was used to make negative impressions of the scars. Then, by a specific process final scar volumes were obtained. Change in scar histological analysis: in participants who consented, punch biopsy samples were taken from the treated and control sections of each scar and from healthy skin 3 cm from the scars. Each biopsy sample was fixed in formaldehyde, embedded in paraffin, and stained with hematoxylin‐eosin and Giemsa. A masked observer compared the degree of fibrosis in the biopsy samples at weeks 0 and 40.
Notes	The initial goal for this study was to enrol 30 participants, allowing a minimal difference between any 2 group means of 0.61 SD to be detected with 90% power. These calculations were based on paired t test analysis and assumed 2‐sided testing with α = .05. Twenty participants were enrolled in the study, which reduced the power to detect minimal differences to 74%. Funding was provided through the Clarence S. Livingood Fund under the direction of Edward. A. Krul, MD, and by a small projects fund, both at Henry Ford Hospital, Detroit, Mich. Smith and Nephew, Largo, Fla, for supplying silicone gel sheeting, and Hypafix tape; Heraeus Kulzer Dental, South Bend, Ind, for supplying dental impression material.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "using a computer‐generated randomization list"
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "Only the patient and the individual performing the treatments knew the treatment assignments"
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The measurements were taken by a masked observer at constant sites on the skin during each visit"
Incomplete outcome data (attrition bias) All outcomes	Low risk	One participant dropped out of the study (because of pain during laser treatments) and another one did not use the silicone gel sheeting until the end of the trial (because of skin irritation).
Selective reporting (reporting bias)	Low risk	No trial protocol is available, but all parameters listed in the methods section to assess changes in the scars were described.
Other bias	Low risk	No other sources of bias were found.

AFL: ablative fractional laser;CO₂: carbon dioxide; 5‐FU: Fluorouracil; HDTA: high‐density treatment arm; He‐Ne: helium‐neon; LDPI: Laser Doppler perfusion imager; LDTA: low‐density treatment arm MSS:Manchester Scar Scale; NdYAG:neodymium‐doped yttrium aluminium garnet; OFDI:optical frequency domain imaging; PDL: Pulsed‐Dye Laser; POSAS: Patient and Observer Scar Assessment Scale; RCT: randomised controlled trial; SD: standard deviation; SPSS: statistical package for social studies; TAC: triamcinolone acetonide; VBS: Vancouver Burn Scar Assessment Scale; VSS: Vancouver Scar Score.