We read with much interest “Morphometric analysis of cervical vertebrae in relation to mandibular growth” (Gray S, Bennani H, Kieser JA, Farella M. Am J Orthod Dentofacial Orthop 2016;149:92-8). One stated aim of this study was “to test the validity of the CVM method for determining the mandibular growth peak.” When performing their mixed model analysis, however, it seems that the authors changed their minds and decided to test the validity of the CVM method to determine mandibular length. As a matter of fact, in the mixed model analysis, they used mandibular length, not mandibular growth peak, as the response variable. Therefore, we do not agree with the interpretation of the results of the mixed model analysis that erroneously led the authors to conclude that “chronologic age is a better predictor of the mandibular growth peak than CVM stage” (final sentence of the Discussion). If we look closer at the results of the mixed model analysis reported in the Table, it appears that in the Conclusions the authors did not report that the sex-specific assessment of CVM stage is a significant predictor of mandibular length. In our opinion, the results of this study showed that both chronologic age and sex-specific assessment of the CVM stage can be used to predict mandibular length accurately.
Moreover, in the Conclusions the authors stated that “neither morphometric templates nor subjectively assessed CVM stages could accurately identify the mandibular growth peak.” It should be emphasized that these findings were based solely on either visual analysis or descriptive statistics, whereas inferential statistics are lacking.