Verhaeghe 2013.

Study characteristics
Methods	Study design: prospective, split lesion design, single‐blinded, intraindividual randomised controlled trial conducted in Belgium. Scar segment as unit of randomisation. Duration of the study: January 2010 to July 2011 Follow‐up time: 6 months from the beginning of treatment (intermediate‐term follow‐up). Protocol was published before recruitment of participants: the local ethical committee approved the study (registration number B 67020097471) Details of trial registration: registered on clinicaltrials.gov (NCT01056211) Funding sources: this work was supported by a grant of the Klinisch Onderzoeksfonds (Clinical Research Fund) of Ghent University Hospital
Participants	Number of participants assigned: 22 participants with a total of 22 scars (1 scar per person) Segment 1 (NAFL‐ Nonablative fractional laser): total of 22 segments treated Segment 2 (control): total of 22 segments not treated Loss: 4 participants Number of participants assessed: 18 participants (total of 36 segments) Segment 1 (NAFL‐ Nonablative fractional laser): 18 segments Segment 2 (control): 18 segment Inclusion criteria: aged 18 and older; skin type I–IV; a hypertrophic scar with 2 areas of similar sizes and appearance, within the same anatomic region; willingness and ability to adhere to the requirements of the protocol. Exclusion criteria: history of keloid formation. history of adverse outcome related to NAFL therapy; pregnancy; lactation; oral retinoid drugs within the past 6 months. Age: 41; 17 to 62 (mean; range) Race: Asian: 2; Caucasian: 16 Gender: male 3; 16.7 (number; %); female 15; 83.3 (number; %) Skin phototypes: Skin type I: 2; 11.1 (number; %) Skin type II: 7; 38.9 (number; %) Skin type III: 6; 33.3 (number; %) Skin type IV: 3; 16.7 (number; %) Duration of scar (months): 38; 1 to 223 (mean; range) Scar Location: lower limb: 2; head and neck: 3; joints: 4; upper limb: 4; trunk: 5 Scar cause: surgery: 15; trauma: 3 Previous Treatment: silicone dressing: 8; local corticosteroids: 5; moisturiser: 10; pressure therapy: 1; pulsed dye laser treatment: 1; none: 1
Interventions	Segment 1: 1 scar area was treated with 1540‐nm non‐ablative fractional laser therapy (Starlux 300 with Lux 1540‐nm fractional hand piece, Palomar Medical Technologies, Burlington, MA). A 10‐mm hand piece (100 microbeams/cm²) was used, with a 15‐ms pulse duration, and energies from 45 to 85 mJ/microbeam (mB) in 3 to 4 passes. The energy level was 45 to 55 mJ/mB during the 1st treatment and was increased 5 to 10 mJ/mB every subsequent session depending on side effects. The maximum delivered energy was 85 mJ/mB. Passes were applied by covering the entire treatment area 3 to 4 successive times. Participants were treated without topical or infiltrative anaesthesia. A single health professional performed all treatments. Segment 2: a scar area, similar to scar area treated with NAFL (in a same anatomic region), received no treatment (control group).
Outcomes	Primary outcomes: Severity of scar: Health Professional Global Assessment: measured for the treated and untreated control side on a visual analogue scale (VAS) ranging from 0 to 100 mm (0 = normal skin and 100 = worst possible scar) at baseline and 1 and 3 months after the last treatment. The health professional performing the evaluations was blinded to the treatment. Patient global assessment (PGA): measured for the treated and untreated control side on a VAS ranging from 0 to 100 mm (0 = as normal skin and 100 = worst possible scar) at baseline and 1 and 3 months after the last treatment. Patient and Observer Scar Assessment Scale (POSAS): scar assessment scale validated for the evaluation of burn scars and linear surgical scars. The combination of a participant and an observer scale allows for a more‐complete evaluation of the scar. The observer scale contains 6 parameters: vascularisation, pigmentation, thickness, relief, pliability, and surface area. The participant scale also contains 6 items: pain, itching, colour, stiffness, thickness, and relief. Each of the 6 items has a 10‐step score, with 10 indicating the worst imaginable scar or sensation. The total score of both scales is calculated by adding the scores of each of the 6 items (range 6 to 60). The lowest score (6) reflects normal skin, and the highest score (60) reflects the worst imaginable scar. Treatment‐related adverse effects: participants were interviewed 4 days after the treatment, over the telephone, and the treatment‐related adverse effects were registered on a standard form. Long‐term treatment‐related adverse effects were registered 3 months after the last treatment. Treatment‐related pain (patient self‐assessment): the participant assessed treatment‐related pain after each treatment on a VAS from 0 to 100 mm (0 = no pain and 100 = worst possible pain). Secondary outcomes: Skin reflectance measurements: quantified skin redness and pigmentation before and 1 and 3 months after the last treatment (DSM II Color meter, Cortex Technology, Hadsund, Denmark). The erythema (E) and melanin (M) index was measured 3 times for the treated side, the untreated control side, and the healthy surrounding skin at each time point. An average of the E and M index of the treated and untreated control area was calculated and subtracted from the average of the E and M index of the surrounding normal skin to adjust for fluctuations in the vascular bed and pigmentation (ΔE and ΔM). The location of the measurement was marked on plastic templates at baseline, and measurements were taken at 1‐ and 3‐month follow‐up at the same location. Digital photographs were taken for documentation in JPEG format using a digital camera.
Notes	The authors have indicated no significant interest with commercial supporters.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The randomization procedure involved computer‐generated randomization lists in which allocation was indicated. The random allocation sequence was created using a digital randomization program."
Allocation concealment (selection bias)	Low risk	Quote: "The allocated treatment was concealed from the assessor throughout the study and revealed only to the treating physician"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "Participant and treating physician were not blinded because this was not feasible from a practical perspective"
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The health professional performing the evaluations was blinded to the treatment. Quote: "The physician performing the evaluations was blinded to the treatment". "Blinded on‐site response evaluations were performed 1 and 3 months after the final treatment."
Incomplete outcome data (attrition bias) All outcomes	Low risk	One participant withdrew during the study for personal, non‐treatment‐related reasons and did not complete allocated treatments. Two participants were lost to follow‐up for personal non‐treatment‐related reasons, and 1 participant was excluded from statistical analysis because the blinded assessor judged that both parts of the scar were not identical at baseline. Eighteen participants were included in the statistical analyses.
Selective reporting (reporting bias)	Low risk	Protocol: NCT01056211. All parameters listed to assess changes in the scars were described in the overall result, such as health professional global assessment, patient global assessment (PGA), Patient and Observer Scar Assessment Scale (POSAS) and Skin reflectance measurements.
Other bias	Low risk	No other sources of bias were found.