Data sourcesCochrane Oral Health Group's Trials Register, Cochrane Central Register of Controlled Trials (CENTRAL), Medline, Embase, US National Institutes of Health Trials Registry and The World Health Organization (WHO) Clinical Trials Registry Platform, abstracts from the British Orthodontic Conference, the European Orthodontic Conference and the International Association for Dental Research (IADR) from 2011 to 2015 and the bibliographies of identified studies.Study selectionRandomised controlled trials (RCTs) involving children and adults who had had retainers fitted or adjunctive procedures undertaken to prevent relapse following orthodontic treatment with braces were considered.Data extraction and synthesisTwo reviewers independently selected studies, abstracted data and assessed study quality. For continuous data mean differences (MD) with 95% confidence intervals (CI) were calculated with ratios (RR) and 95% CI for dichotomous outcomes.ResultsFifteen studies involving a total of 1722 patients were included. Seven studies were considered to be at high risk of bias, four at low risk and four at unclear risk. For removable retainers versus fixed retainers (three studies) there was low quality evidence that thermoplastic removable retainers provided slightly poorer stability in the lower arch than multistrand fixed retainers: MD (Little's Irregularity Index, 0 mm is stable) 0.6 mm (95% CI 0.17 to 1.03) and of less gingival bleeding with removable retainers: RR 0.53 (95%CI; 0.31 to 0.88). Patients found fixed retainers more acceptable to wear, with a mean difference on a visual analogue scale (VAS; 0 to 100; 100 being very satisfied) of -12.84 (95% CI -7.09 to -18.60).For different types of fixed retainers (four studies) data from three studies (228 patients) comparing polyethylene ribbon bonded retainer versus multistrand retainer were pooled showing no evidence of a difference in failure rates. RR = 1.10 (95%CI; 0.77 to 1.57).Pooled data from two trials (174 patients) comparing the same types of upper fixed retainers, showed a similar finding: RR =1.25 (95%CI; 0.87 to 1.78).For different types of removable retainers (eight studies) one study at low risk of bias comparing upper and lower part-time thermoplastic versus full-time thermoplastic retainers showed no evidence of a difference in relapse (graded moderate quality evidence). Another study, comparing part-time and full-time wear of lower Hawley retainers, found no evidence of any difference in relapse (low quality evidence). Two studies at high risk of bias suggested that stability was better in the lower arch for thermoplastic retainers versus Hawley, and for thermoplastic full-time versus Begg (fulltime) (both low quality evidence). In one study, participants wearing Hawley retainers reported more embarrassment more often than participants wearing thermoplastic retainers: RR 2.42 (95% CI 1.30 to 4.49; one trial, 348 participants, high risk of bias, low quality evidence). They also found Hawley retainers harder to wear. There was conflicting evidence about survival rates of Hawley and thermoplastic retainers.For combination of upper thermoplastic and lower bonded versus upper thermoplastic with lower adjunctive procedures versus positioner (one study) there was no evidence of a difference in relapse between the combination of an upper thermoplastic and lower canine to canine bonded retainer and the combination of an upper thermoplastic retainer and lower interproximal stripping, without a lower retainer. Both these approaches are better than using a positioner as a retainer.ConclusionsWe did not find any evidence that wearing thermoplastic retainers fulltime provides greater stability than wearing them part-time, but this was assessed in only a small number of participants. Overall, there is insufficient high quality evidence to make recommendations on retention procedures for stabilising tooth position after treatment with orthodontic braces. Further high quality RCTs are needed.