Circadian Biomarkers in Asymptomatic Offspring of Patients with Late-onset Alzheimer?s Disease and their Relationship with Cognitive Performance

CAROLINA ABULAFIA; BÁRBARA DUARTE-ABRITTA; STELLA M. SÁNCHEZ; MIRTA F. VILLARREAL; MARÍA S. LADRÓN DE GUEVARA; GUSTAVO SEVLEVER; LETICIA FIORENTINI; SALVADOR M. GUINJOAN; DANIEL E. VIGO

Punta del Este

Congreso; XVII Congreso Internacional de Medicina del Sueño - FLASS; 2018

Introduction: Early neuropathological changes characteristicof late-onset Alzheimer?s disease (LOAD) impact structuresthat regulate circadian rhythms and particularly sleep. Indeed,sleep pattern is emerging as a potential biomarker, mechanisticpathway and treatment target in LOAD. We hypothesizedthat circadian rhythm anomalies would already be present inasymptomatic, middle-aged offspring of patients with LOAD(O-LOAD) prior to cognitive decline. Methods: We tested 35subjects with at least one parent with LOAD (O-LOAD) and 31healthy individuals without family history of Alzheimer?s disease(control subjects, CS) with a series of cognitive tests, as well asactigraphy measures of sleep?wake rhythm, cardiac autonomicfunction via heart rate variability (HRV), and bodily temperature.Results: O-LOAD displayed subtle yet signifcant defcits inverbal episodic memory (RAVLT learning 48.32 ± 1.59 vs. 44.12± 1.21, p = 0.005; delayed recall 10.55 ± 0.38 vs. 8.68 ± 0.52, p= 0.005) and language (Vocabulary 50.5 ± 1.06 vs. 45.06 ± 1.48,p= 0.004) compared to CS. O-LOAD showed a more extendedsleep duration (439.26 min ± 9.41 vs. 473.66 min ± 10.57, p =.018) and reduced sleep effciency (97.07 % ± .41 vs. 95.75 % ±.48, p = .042). No signifcant differences were found for bodytemperature or HRV variables. Correlations between increasedsleep duration and poorer cognition were found in CS but notin O-LOAD. Improved cognitive performance was associatedto indicators of greater sympathetic activity. Discussion: Ourresults support the hypothesis that sleep pattern disturbances arealready present very early on in relatively young asymptomaticsubjects. The unexpected reduced cognitive results found inO-LOAD suggest that cognitive decline could start earlier thananticipated in the form of subtle cognitive changes within theclinically normal range. It is widely accepted that sleep patterndisturbances would result in cognitive alterations. Taken theseinformation together with the correlations between sleepduration and cognition present in CS but absent in O-LOADsuggest some impairment in the mechanisms underlying thesleep-cognitive relationship. Sleep pattern deserves further studyas a potential biomarker in LOAD, even in healthy middle-agedindividuals.