Page 91 - The autoimmune hypothesis of narcolepsy and its unexplored clinical features M.S. Schinkelshoek

Figure 5.3. Mixed model estimations of the averaged subject in four groups based on gender and medication type. Body mass index (BMI) at baseline is set at 28.0 kg/m2, the grand mean of our study population. Values for different groups at Follow-up duration = 0 depicts the average baseline BMI value of all patients in that particular group. SXB = sodium oxybate.
BMI change in patients is not correlated with either modafinil or SXB dose
A higher dose of either modafinil or SXB was not correlated with BMI change in our patient cohort. Spearman’s rho was 0.188 in modafinil-treated patients (p = 0.402) and 0.153 in SXB-treated patients (p = 0.294).
BMI decrease in SXB group not explained by methylphenidate or dexamphetamine use
Given the fact that stimulant use is significantly higher in the patient group in which SXB treatment was initiated, a mixed model was fitted on the dataset in which patients using methylphenidate or dexamphetamine were excluded (n = 24 in SXB group, n = 2 in modafinil group; data not shown). Also in this fitted model there was a significant main effect of SXB on BMI (F(1,40.073) = 16.003, p < 0.001). Additionally, we fitted a mixed model in which dexamphetamine or methylphenidate use was added both as a main effect and as an interaction effect with time (Supplementary Table 5.1). The main effect of dexamphetamine
Medication and BMI change
89
5