Comments on the Letter to the Editor by Ferri et al. on Autonomic Response to Periodic Leg Movements in Sleep in Narcolepsy-Cataplexy

Thank you for your comment on our recent study. In this work, we documented that periodic leg movements during sleep (PLMS) of subjects with narcolepsy with cataplexy (NC) were associated with a lower amplitude of tachycardia and bradycardia than PLMS of normal control subjects.¹ We suggested that this attenuation could be explained by an impairment in both sympathetic and parasympathetic activities due to the hypocretin-deficient condition. In a letter to the editor, Ferri et al.² addressed some considerations that, in their opinion, confound the results and conclusions of our results. First, they pinpointed that the subjects we used as normal controls had an abnormally high PLMS index. As stated in our paper, we selected control subjects with high levels of PLMS in order to have enough PLMS to be further analyzed and compared with a carefully age- and gender-matched population without any disease or medication interference, to avoid any effect related to these variables. However, to verify if a high PLMS index could influence the amplitude of heart rate changes, we checked for a relationship between the magnitude of the cardiac activation associated with PLMS and PLMS index in both patients and controls but could not find any correlation.

These authors also argued that a high PLMS index is likely a precursor of RLS and that our control subjects could eventually develop RLS symptoms. As noted by these authors, a recent study found an association between the RLS gene variant (BDBD9) and subjects with PLMS without RLS, but importantly such an association was absent in subjects with RLS without PLMS.³ In our opinion, this supports the hypothesis that PLMS and RLS are actually at least partially distinct phenomena, especially in the elderly, but does not necessarily support the idea that our control subjects with high PLMS indexes will develop RLS in the future. We agree that large longitudinal studies including the selection of genetic polymorphisms (such as BDBD9 and Meis 1 gene variants) are needed to better answer this question.

Ferri et al. also claimed that PLMS are associated with sympathetic activation, and probably cardiovascular consequences, irrespective of the presence or absence of RLS. However, to our knowledge, cardiovascular changes associated with PLMS have only been studied in patients with RLS, with periodic leg movement disorder (PLMD) and with REM behavior disorder. In contrast, we clearly documented that our control subjects did not present any complaint of daytime sleepiness, insomnia, or RLS symptoms, which excludes a PLMD or RLS diagnosis even in the presence of high PLMS indices. The authors also argued that the difference we observed in the amplitude of heart rate changes between narcoleptic patients and control subjects could be caused not only by a reduction in narcoleptics patients, but also by higher values in our controls. As our NC patients also had a high PLMS index, it would be hard to understand why a high PLMS index in controls could have this different impact. Furthermore, we have to emphasize that heart rate changes associated with PLMS observed in our control subjects are about of the same amplitude as heart rate changes already reported in RLS or PLMD,^4–6 strongly arguing that the decreased activation observed in the group of patients is indeed related to the narcolepsy/hypocretin deficiency. Controlled studies are however needed to better compare the amplitude of heart rate changes between RLS patients and healthy normal subjects with low or high PLMS index.

Ferri et al. also suggested that heightened metrics of sympathetic outflow in RLS/PLMS may be due to increased CSF hypocretin-1 levels previously described in RLS.⁷ This is a good point. Nonetheless, we mentioned that despite high PLMS indexes none of our controls presented a RLS diagnosis. Moreover, we note that the findings of increased CSF hypocretin-1 in RLS were only found in the evening, not replicated in other settings, and could be due to increased motor activity and wakefulness associated with RLS. For this reason, we felt this explanation to be too speculative.

Finally, these authors recommended caution before concluding on an increased risk of cardiovascular morbidity in narcolepsy. We fully concur as to date no study has shown an increased risk of cardiovascular diseases in narcolepsy with or without PLMS. In fact, our paper only stated that several previous studies have shown an association between a decrease in HR variability and increased risk of mortality due to cardiovascular diseases^8,9 and that patients affected with NC often present with obesity, type 2 diabetes, and metabolic syndrome,^10,11 conditions also known to be associated with higher risk of cardiovascular diseases. Patients with NC are often treated with psychostimulants like methylphenidate for years; these medications are known for their impacts on the autonomic nervous system and cardiovascular function.¹² Altogether, we believe that patients with NC are at especially high risk for cardiovascular diseases, and that autonomic activation with PLM, a frequent co-occurrence in patients with narcolepsy, could be an additional burden. This reasonable hypothesis expends on a parallel suggestion that has been previously made for RLS patients with PLMs. These hypotheses do indeed still need to be tested.

DISCLOSURE STATEMENT

Dr. Dauvilliers has consulted for UCB Pharma, Cephalon, and Bioprojet. Dr. Montplaisir has received research support from Boehringer-Ingelheim, Sanofi-Aventis, and Merck; served as an advisor for Boehringer-Ingelheim, Merck, and Servier; and participated in speaking engagements for Valeant Pharmaceutical, Boehringer-Ingelheim, Sanofi-Aventis, and GlaxoSmithKline. Ms. Pennestri has indicated no financial conflicts of interest.