Pandis 2013.

Study characteristics
Methods	2‐arm parallel RCT
Participants	Number recruited: 220 participants (110 in each group) Median age (IQR)(years): 16.0 (2) Sex (M:F): 60:160 Inclusion criteria: no active caries, restorations, fractures on the mandibular anterior teeth or periodontal disease; adequate oral hygiene Exclusion criteria: unwilling to be assigned to any of the approaches or had any abnormal oral or medical condition contraindicating fixed retention Setting: private practice at Bern, Switzerland
Interventions	2 types of adhesives and composites Group 1: chemical group; maximum Cure 2‐part liquid adhesive (Reliance Orthodontic Products, Itasca, Ill) and Excel 2‐part paste (Reliance Orthodontic Products) Group 2: photo polymerisation group; light‐cured liquid (Assure; Reliance Orthodontic Products) and paste in 2 layers (Flow‐Tain; Reliance Orthodontic Products) All participants received a soft‐bonded lingual retainer of 0.022 inch (Tru‐Chrome multi‐stranded wire; Rocky Mountain Orthodontics, Denver, Colo) that was fabricated intraorally.
Outcomes	Survival: failure of the lingual retainer, failure risk (hazard ratio), adhesive remnant index (ARI) scores Time points: 3, 6, 12, 18, 24 months
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Well‐reported randomisation process, using a computer random number generator (Stata software, Corp, College Station, Tex) in random permuted blocks of 20 participants, ensuring equal distribution in the 2 groups
Allocation concealment (selection bias)	Low risk	Central allocation by the practice manager responsible for opening the next envelope in sequence
Blinding of outcome assessment (detection bias) All outcomes	Low risk	The outcome assessment was blinded. "Blinding was applicable for outcome assessment only."
Incomplete outcome data (attrition bias) All outcomes	Low risk	Trial authors imputed missing data using appropiate methods. Trial authors carried out imputations under the missing at random assumption.
Selective reporting (reporting bias)	Unclear risk	The study protocol is not available. Insufficient information to permit judgement of ‘low risk’ or ‘high risk’
Other bias	Low risk	The study appears to be free of other sources of bias.