Gunay 2018.

Study characteristics
Methods	2‐arm parallel RCT
Participants	Number recruited: 120 in total, 60 per group Sex (M:F): 37:83 Age (years): group 1: 15.7, group 2: 16.2 (SD not provided) Inclusion criteria: patients had undergone fixed appliance therapy, 60 with premolar extractions 60 without extractions; no missing mandibular incisors; no restorations on the mandibular incisors that might affect retainer bonding on lingual enamel surfaces; no morphologic crown anomalies Exclusion criteria: patients with rotations Setting: university hospital, Samsun, Turkey
Interventions	Two different techniques for bonding two types of lingual wire retainers Group 1: ⌀ 0.0175 inch 6‐strand stainless steel wire (Ortho Technology, Lutz, USA) was used, the lingual retainers were fabricated on plaster models, and a silicon transfer key was used.  Group 2: ⌀ 0.0195 inch dead‐soft coaxial wire (Respond; Ormco, Orange, USA) was used, and the lingual retainers were made in the participant's mandibular arch without a study model.
Outcomes	Stability assessed using LII, intercanine width and arch length (mm) (data not available) Survival: failure rate as a percentage Time points: 3, 6, 9 and 12  months
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	The study authors state “the patients were divided into 2 groups randomly with equal numbers of extraction and nonextraction patients”.
Allocation concealment (selection bias)	Unclear risk	No comment was made in the paper.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No comment was made in the paper.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No comment was made in the paper, but it appears from the failure data that there were no dropouts.
Selective reporting (reporting bias)	High risk	They measured arch length but did not report the data.
Other bias	Low risk	No areas of risk of bias were identified.