The aim of the study was to assess treatment effects and potential side effects of different archwires used on patients receiving orthodontic therapy. Electronic and manual unrestricted searches were conducted in 19 databases including MEDLINE, Cochrane Library, and Google Scholar until April 2012 to identify randomized controlled trials (RCTs) and quasi-RCTs. After duplicate study selection, data extraction, risk of bias assessment with the Cochrane risk of bias tool, and narrative analysis, mean differences (MDs) with confidence intervals (CIs) of similar studies were pooled using a random-effects model and evaluated with GRADE. A total of 16 RCTs were included assessing different archwire characteristics on 1108 patients. Regarding initial archwires, meta-analysis of two trials found slightly greater irregularity correction with an austenitic-active nickel-titanium (NiTi) compared with an martensitic-stabilized NiTi archwire (corresponding to MD: 1.11 mm, 95% CI: -0.38 to 2.61). Regarding archwire sequences, meta-analysis of two trials found it took patient treated with a sequence of martensitic-active copper-nickel-titanium (CuNiTi) slightly longer to reach the working archwire (MD: 0.54 months, 95% CI: -0.87 to 1.95) compared with a martensitic-stabilized NiTi sequence. However, patients treated with a sequence of martensitic-active CuNiTi archwires reported general greater pain intensity on the Likert scale 4 h and 1 day after placement of each archwire, compared with a martensitic-stabilized NiTi sequence. Although confidence in effect estimates ranged from moderate to high, meta-analyses could be performed only for limited comparisons, while inconsistency might pose a threat to some of them. At this point, there is insufficient data to make recommendations about the majority of initial archwires or for a specific archwire sequence.