Reply: Risk of Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, and Executive Function Impairment in Metopic Craniosynostosis

We appreciate the letter from Tio et al. regarding our article.1 They wrote that an association between increased age at surgery and worse executive function is at odds with existing research, citing a systematic review by Osborn et al.2 claiming no significant difference in neurocognition in operatively versus nonoperatively managed metopic synostosis. Our study does not include any nonoperatively managed patients and is outside the scope of the present work. However, in their own article, Osborn et al. wrote, "of particular note, is the fact that children who are treated conservatively appear to experience more difficulties than both their healthy peers and the children with [metopic synostosis] who receive surgery."2 Evaluations of neurocognitive outcomes in conservatively managed metopic synostosis utilize neurocognitive testing at a young age with limited predictive accuracy (mean age at testing, <5 years), and they often do not include a morphologic severity assessment.2 Our decision to include a higher threshold for testing was intended to avoid the worse predictive value of cognitive and behavioral testing below 6 years of age.3 Further, a lack of morphologic severity data begs the question of whether the non-surgically treated patients are not operated on for a reason. Perhaps these patients have very mild forms of synostosis or ridging. Tio et al. also remarked that we failed to provide an etiology for the outcome differences. As we have previously published, focal areas of growth restrictions in the frontal lobe may account for worse neurocognitive sequelae.4 Regardless, we believe that clinical decision-making should be based on clinical outcomes. The lack of a definitive etiology for the trends does not invalidate our results. Finally, Tio et al. raised methodologic concerns. Patient families opted in from 2 large craniofacial centers, and all patients that met inclusion criteria were contacted and invited to participate in the study. Potential selection bias is a characteristic of any retrospective study. While this is an acknowledged limitation of our work, it does not reflect a lack of transparency. Tio et al. claimed that we only evaluated patient age at surgery as less than 6 months or greater than 6 months. However, we also analyzed the impact of each additional month of surgery on behavioral outcomes. As our article states, increasing age at surgery by month was associated with reaching potentially clinically significant levels on the executive function subscale of the Conners Rating Scales, 3rd edition (Short Form), and subscales of the Behavior Rating Inventory of Executive Function, 2nd edition (Behavioral Regulation Index, Cognitive Regulation Index, and Global Executive Composite). Evaluating age in both ways revealed consistent significant trends. Notably, the systematic review by Mandela et al.,5 which they cite, compared age as a grouped variable rather than a continuous variable in every single study. They also incorrectly stated that we did not control for socioeconomic factors while controlling for race, sex, parental education, and household income. Next, they criticized that radiographic data exist for only two-thirds of patients. As mentioned earlier, our study demonstrated more due diligence to assess severity than nearly every study they cited. Regarding multiple statistical tests, significance across related subtests suggests a clinical outcome finding rather than a statistical aberration. Even with its limitations, our study is among the most comprehensive in the literature to date. DISCLOSURE Dr. Alperovich is a consultant for Johnson & Johnson and LifeNet Health. The other authors have no conflicts of interest relevant to this article to disclose.