The etiologic relevance of craniofacial structure to obstructive sleep apnea syndrome (OSAS) is controversial yet the premise of a causal association serves to justify many treatments. A qualitative and quantitative analysis of the literature was performed to examine the foundation for any relationship between craniofacial structure and OSAS. A MEDLINE search and investigation of the published and unpublished literature on diagnostic imaging and OSAS was toxonomically arranged. Each sample study was evaluated by using the following criteria: (a) appropriate control group, (b) "blinding" of evaluators, (c) reliability measured, (d) random assignment of treatment, and (e) "success" was defined adequately in efficacy studies. Morphologic variables were combined among studies and compared with controls drawn from either the same patient pool as the OSAS group, or matched for gender, age, and body mass index. Analysis revealed 32 review articles, 16 case reports, and 95 sample studies. Only seven sample studies drew a control group from the same patient pool, whereas five used matched controls. Only one of these studies satisfied all the qualitative criteria. Of the treatment efficacy studies, 10 defined outcome adequately. However, none of these met all the qualitative criteria. The most consistent, strong effect sizes with the highest potential diagnostic accuracies were for mandibular plane to hyoid, mandibular plane angle, and mandibular body length. Only mandibular body length demonstrated a clinically significant association with and diagnostic accuracy for OSAS. However, since this variable's controls were selected from the literature, possible explanations for a positive association include methodologic differences between studies, varying magnification factors, and morphologic differences.