Extraction vs nonextraction of premolars for orthodontic treatment: a scoping review examining the extent, range, and characteristics of the literature

Introduction: The debate about whether malocclusion can or should be treated with or without extraction of premolars continues. This scoping review quantifies the literature, summarizes the outcomes researched and methods, and proposes a way to reduce uncertainty in this area.

Methods: Electronic and gray literature searches were undertaken without language restriction, but non-English language titles and abstracts were not translated. A minimum of 2 people independently screened the titles and abstracts.

Results: Searches identified 9010 articles, of which 3851 were duplicates; 5159 were screened, and 4617 were excluded (1092 laboratory or animal studies, 1219 case reports or series, 2306 with no information). By consensus, 399 articles contained information concerning differences between orthodontic patients treated with or without premolar extractions (143 were unclear). The majority (n = 372) reported outcomes in 8 areas. Fifty-seven were review articles (32 systematic reviews and 25 nonsystematic reviews or opinions). The most common research design in the remainder was a cohort (n = 280, 82% of 342 articles reporting primary data), of which a very large majority were considered retrospective (n = 249, 89% of articles reported for subjects over ≥2 time points). Only 28 (8% of articles reporting primary data) were judged to involve prospective data collection (4 randomized controlled trials [RCTs], 23 cohorts, 1 unclear design). Excluding reviews and unclear articles, 99% (332 out of 336) were considered observational research and only 1% were interventional.

Conclusions: There was limited low-quality evidence that extracting premolars in orthodontic patients have a possible negative effect in 2 outcome areas and a positive effect in 1 outcome area. Most study reports were of low methodological quality, and further reviews are unlikely to provide new information. Investigators should concentrate on collecting primary data of outcomes important to patients. A protocol has been made available to help reduce methodological differences, assist future meta-analyses and increase the generalizability of findings: https://doi.org/10.17605/OSF.IO/CQ49Y.