Page 93 - The autoimmune hypothesis of narcolepsy and its unexplored clinical features M.S. Schinkelshoek

2014). Participants in this study lost on average 5.2 kg in three months of SXB treatment, which supports our results (Husain et al., 2009). If weight loss is mediated through this pathway, gender differences in fat metabolism (Williams et al., 2004) could account for the different trends in BMI change between men and women in our study.
Another hypothesis is that BMI decrease results from the effect of SXB on ghrelin and leptin secretion. A comparison between NT1 patients and healthy controls (Donjacour et al., 2013) did not show any differences in ghrelin and leptin secretion after 3 months of SXB treatment. It was speculated there that the weight loss may also be due to an decrease of food intake and an increase of physical activity leading to a negative energy balance secondary to the sleep- promoting effects of SXB.
In addition to these hypotheses, we propose that the fact that patients who start using SXB are required to cease using alcohol might play an additional role in BMI loss in this patient group. Even though the exact relation between alcohol consumption and weight gain is complex, it can be said that alcohol consumption leads to weight gain (Suter et al., 2005).
The effects of SXB on BMI were compared with modafinil, a commonly prescribed therapy for narcolepsy. The decision to treat an individual with either SXB or modafinil is not always clear-cut. SXB is more often prescribed when cataplexy and disturbed nocturnal sleep are the most invalidating symptoms (Bosch et al., 2012; Boscolo-Berto et al., 2012) while modafinil is prescribed in those suffering most from excessive daytime sleepiness (Guilleminault and Cao, 2011). If cataplexy would have an effect on weight it could therefore lead to a selection bias in our study. There is no data on the relation between cataplexy and BMI. We have also no reason to expect other confounding parameters to be present than the ones we accounted for in our analysis. Earlier studies on the effect of modafinil in NT1 showed no significant BMI changes in those treated with it (Moldofsky et al., 2000; US Modafinil in Narcolepsy Multicenter Study Group, 1998). These studies had a shorter follow up time compared to our study.
Our study has a few limitations. Firstly, due to the retrospective nature we were not able to randomize patients and we were dependent on the methods and data collection which was chosen for the individual. The retrospective design
Medication and BMI change
91
5