Randomized controlled trials in oral and maxillofacial surgery

This chapter reviews the published literature for randomized controlled trials in oral and maxillofacial surgery. Two-hundred-thirty-two studies were included and the content is divided into eight categories: general medicine and oral surgery, outpatient anesthesia, dentoalveolar surgery, orthognathic surgery, facial trauma, pathology, reconstruction, and temporomandibular disorders. Key findings are summarized in a table format in each section, and the study quality was assessed in several ways: sample size and number of enrolling institutions, risk of bias in four different domains, and whether the study used a standard diagram to illustrate study enrollment and data collection. Most investigations were single-center studies (94%) with a median sample size of 46 subjects. Bias was a significant concern since the vast majority of studies had either a moderate risk (66%) or high risk (28%) of bias based on Cochrane assessment tools. The content and quality analyses are then expounded upon by discussing gaps and common pitfalls in the current body of literature. The goal of this discussion is to give concrete recommendations on both design and content to today's aspiring surgeon-scientist to carry out the studies needed in this field. Major recommendations include a sample size greater than 70 subjects, multicenter collaboration, and inclusion of fundamental details of methodology: randomization techniques, baseline differences among study arms, subject adherence to assigned treatment, blinding status of evaluators, and completion of patient follow-up. The principle of clinical equipoise should guide surgeon-scientists to find the balance between intellectual curiosity and therapeutic obligations. Lastly, both the physiology and logistics of a given disease process or intervention may influence how easily it is studied using a randomized controlled trial; other study designs may still be indicated for outcomes that are rare, take a very long time to assess, or in diseases with a known best simple treatment.