75 Microstructures | Trigonocephaly were investigated. If this transformation did not improve the normal distribution, the untransformed data were used. FA and MD values in each of the chosen tracts were investigated by box plots and scatterplots. Subsequent linear regression models were used to further examine the FA values and MD values, with patient/control, sex, and age as independent variables. Sex and age have previously been shown to affect FA and MD values.18, 23-28 13-Coefficients were calculated (stats package) for each regression. Affected tracts would be further investigated by assessing the diffusivity values AD and RD. The FMA was assessed as a control tract to further validate our results; we compared the FA FMI/FMA ratio between patients and controls and the MD FMI/FMA ratio between patients and controls using a t test. The Bonferroni correction was conducted, and a P-value < 0.0025 (P-value = 0.05/20) was considered statistically significant. The effect of age on FA and MD between patients with trigonocephaly and controls was investigated using effect plots. As an additional analysis, we created subgroups based on age categories, to further correct for any effect due to age. Because of the low number of patients and controls, this is enclosed in Supplemental Table 1. RESULTS Characteristics Fourty-six patients with trigonocephaly with a median age of 0.49 years (IQR: 038) and twenty-one control subjects with a median age of 1.44 years (IQR: 0.98) were included in this study as presented in Table 1. Table 1. Descriptives Trigonocephaly Controls n MRI 46 21 f : m 12:34 14:07 median age (IQR) 0.49 (0.38) 1.44(0.98) Fractional anisotropy By linear regression, we found no significant effect of trig-onocephaly or gender on FA values of the investigated white matter tracts (Table 2). The effect of age was significant for all tracts, including the control tract FMA (P < 0.0025). The FA FMI/FMA ratio between patients and controls did not show a significant difference (P = 0.79) as well. We visualized the effect of age on increase of the FA value per tract in patients compared to controls, using scatterplots and effect plots (Figure 2) to correct for the age difference between the patient and control group, which confirmed again the nonsignificant difference between patients with trigonocephaly and controls. 5
RkJQdWJsaXNoZXIy MTk4NDMw