Heart Rate Changes Accompanying Periodic Leg Movements during Sleep in Narcoleptic Patients

The neural, pharmacological, and genetic substrates governing interactions between periodic leg movements of sleep (PLMS) and the autonomic nervous system have garnered the attention of many investigators given converging lines of evidence that implicate PLMS in cardiovascular morbidity.¹ We therefore read with interest the manuscript by Dauvilliers et al.² describing reductions in the amplitudes of tachycardia and bradycardia that accompany PLMS in narcolepsy with cataplexy (NC). Taken together with other animal and human studies the authors proffer the most parsimonious explanation for this blunted autonomic variability to be reductions in hypocretin that are unique to NC. That being said, there are several considerations that confound the results and conclusions, and that limit interpretations regarding pathophysiology. Primary amongst these, and one that plagues all like investigations of PLMS correlates, is the composition of the population from which “control” PLMS are selected. Advanced quantitative analyses^3,4 and dissection of the genetic contributions to PLMS⁵ establish that there are unique footprints, thresholds (e.g., PLMS > 5/hour), and molecules that characterize PLMS in individuals who share an underlying diathesis to restless legs syndrome (RLS). While there are no population-based or longitudinal data that inform us as to what proportion of individuals with PLMS fulfill, or eventually develop, a portion, or all of the diagnostic criteria for RLS, it is becoming quite clear that PLMS are an endophenotype of RLS⁶ whose ubiquity—by default unavoidable in those individuals comprising the control PLMS group selected by Dauvilliers et al.²—reflects the commonality of genes that also confer risk for RLS. Asymptomatic PLMS, for example, are often a precursor to RLS, are more common in relatives of RLS patients, and are more prevalent and abundant in ethnic groups that have the highest frequencies of RLS/PLMS risk alleles (e.g., North Americans of European vs. African descent).⁵ The BTBD9 and Meis1 gene variants conferring risk for RLS, in fact, are more robust predictors of PLMS in individuals who experience symptoms atypical of RLS, or none at all.⁷ Irrespective of the absence or presence of RLS symptoms, PLMS share heightened measures of sympathetic outflow and, presumably, cardiovascular consequences.^8,9 The 14 subjects comprising the control groups of PLMS in the Dauvilliers et al.² study exhibited an average PLMS index of 43.6. This value far exceeds the age-matched normative values of ∼10/hour reported by the same group,¹⁰ the 5/hr threshold at which the frequency of the RLS-associated BTBD9 genetic variant significantly exceeds that of the population,⁷ and the threshold of 15/hr reflecting current ICSD recommendations for a diagnosis of PLM disorder (PLMD).¹¹ Thus, it is unclear whether the dampened amplitudes of heart rate changes observed by Dauvilliers et al.² are due as much to genuine reductions in NC as opposed to heightened values observed by the authors themselves,¹² and others in previous studies of PLMD,⁹ and PLMS in the setting of RLS.^8,13 Moreover, it is interesting to speculate that heightened metrics of sympathetic outflow in RLS/PLMS may be due to increased hypocretin that has been observed in some¹⁴ but not all¹⁵ assessments of RLS. We therefore recommend caution least the cart gets out in front of the horse. Whether the reductions in HRV described by Dauvilliers et al.² in NC are indeed genuine needs to be replicated cognizant of the limitations that we have highlighted, before one dare speculate about a pathophysiological basis, and an as yet unproven increased risk of cardiovascular morbidity in narcolepsy. Future analyses of PLMS correlates beg for some consensus on what characteristics be relied upon going forward in deriving proper “control” PLMS—e.g., age-adjusted quantitative norms,¹⁰ periodicity metrics,³ presence/ absence of symptoms or comorbidities, or stratification/selection by genetic variants known to confer risk for RLS/PLMS.

DISCLOSURE STATEMENT

Dr. Ferri has consulted for Merck, Sanofi-Aventis, Sapio Life, and ATES Medica Device. Dr. Rye has been advisor and consultant to UCB and External Data Monitoring Committee member for Merck. Dr. Zucconi has indicated no financial conflicts of interest.