An observational study in Obstructive Sleep Apnea and Cognitive Impairment, sponsored by University of Split, School of Medicine. Not yet recruiting. Open to participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-10-27.
Sponsored by University of Split, School of Medicine · Observational
This cross-sectional study examines whether EEG markers derived from full-night polysomnography (sleep spindle density, K-complex density, and REM EEG slowing) are associated with cognitive performance in adults with obstructive sleep apnea, and whether they explain more variance in cognitive test scores than classical OSA metrics (AHI, T90, minimal SpO2).
BACKGROUND Obstructive sleep apnea (OSA) is a common chronic disorder and a risk factor for metabolic and cardiovascular disease. A subset of patients with OSA also exhibits cognitive impairment. Classical clinical metrics of OSA, such as the apnea-hypopnea index (AHI), time spent with oxygen saturation below 90% (T90) and minimal oxygen saturation (MinSpO2), show weak and inconsistent associations with cognitive outcomes. Sleep EEG markers derived from polysomnography (PSG), including sleep spindles, K-complexes and REM EEG slowing, have been proposed as more sensitive biomarkers of cognitive vulnerability. Sleep spindles have been linked to cognition in OSA, whereas the association of K-complexes and REM EEG slowing with cognitive performance in OSA remains understudied.
OBJECTIVES The primary objective is to examine whether sleep spindle density, K-complex density and REM EEG slowing are associated with cognitive test performance in adults with OSA, after adjustment for age, sex, body mass index, education and sleepiness. The secondary objective is to compare how strongly these EEG markers and the classical OSA metrics (AHI, T90, MinSpO2) are associated with performance on the same cognitive tests.
DESIGN This is a prospective, cross-sectional observational study conducted at the Sleep Medicine Center Split, Croatia. Consecutive adult patients referred for diagnostic full-night PSG will be invited to participate. After written informed consent, participants complete a standardized cognitive test battery before PSG, in a fixed order: Montreal Cognitive Assessment (MoCA), digit span test, Trail Making Test (TMT), Psychomotor Vigilance Test (PVT), Stroop test, Cognitive Flexibility Inventory (CFI) and Digit Symbol Substitution Test (DSST). Sleepiness is assessed with the Stanford Sleepiness Scale. Full-night PSG is recorded according to routine clinical procedures. Participants are included in the analysis if PSG confirms OSA (AHI ≥5 events/hour) and total sleep time exceeds 4 hours. Demographic, anthropometric and clinical data are obtained by direct measurement and from medical documentation. No intervention is applied.
EEG MEASURES From the PSG recordings, sleep spindle density (events/min of NREM sleep), K-complex density (events/min of NREM sleep) and REM EEG slowing (ratio of slow [delta, theta] to fast [alpha, beta, gamma] EEG bands) will be extracted and quantified with appropriate software. Classical OSA metrics (AHI, T90, MinSpO2) are taken from the same PSG.
OUTCOMES Primary outcomes are the scores of the individual cognitive tests (MoCA, digit span, PVT, TMT, Stroop, CFI, DSST), each tested for association with the three EEG markers. Secondary outcomes include the association of AHI, T90 and MinSpO2 with the same test scores, a comparison of the explanatory power of classical OSA metrics and EEG markers, and an exploratory analysis of the combination of all three EEG markers.
STATISTICAL ANALYSIS Data distribution will be assessed with the Shapiro-Wilk test and visual inspection of histograms and Q-Q plots. Associations will be examined with Pearson or Spearman correlation, as appropriate, and with multiple linear regression adjusted for age, sex, body mass index, education and sleepiness. Models based on classical OSA metrics and on EEG markers will be compared using regression coefficients, standardized beta coefficients, model fit indices and explained variance (R²). Results will be reported with 95% confidence intervals.
SAMPLE SIZE Sample size for the primary analysis was calculated for a correlation coefficient of 0.30 (alpha 0.05, power 80%), giving 85 participants. For the secondary analysis (multiple regression, 6 predictors, f2 = 0.15, alpha 0.05, power 80%), the minimum sample size is 97, which is the target sample size of the study.
The study will be reported according to the STROBE guidelines.
2,267 studies on the registry are indexed under Sleep Apnea, Obstructive; 468 are open to participants now.
This study's planned enrollment of 100 is close to the median of 105 across 667 observational studies indexed under Sleep Apnea, Obstructive.
Browse Sleep Apnea, Obstructive studies →University of Split, School of Medicine is the lead sponsor of 51 studies on the registry; 17 are open to participants now.
Counted across the registry records on this site, refreshed daily.
Adult patients (18 years or older) with obstructive sleep apnea (apnea-hypopnea index ≥5 events/hour) referred for diagnostic full-night polysomnography at the Sleep Medicine Center Split, Croatia. Participants are recruited consecutively and complete a cognitive test battery before polysomnography. Patients with neurological disease, other sleep disorders, history of head trauma or head surgery, or chronic disease that may substantially affect cognition or sleep architecture are not included.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Exclusion Criteria:
Routine full-night diagnostic polysomnography (PSG) performed at the Sleep Medicine Center Split according to standard clinical procedures; no change to the clinical procedure is made for the study. From the recorded EEG, sleep spindle density (events/min of NREM sleep), K-complex density (events/min of NREM sleep) and REM EEG slowing (ratio of slow \[delta, theta\] to fast \[alpha, beta, gamma\] band power) are extracted and quantified with appropriate software. Classical OSA metrics (AHI, T90, minimal SpO2) are obtained from the same recording. This is an assessment applied to all participants, not a treatment. Standardized cognitive assessment administered once, before PSG, by trained assessors using fixed instructions and a fixed test order: Montreal Cognitive Assessment (MoCA), digit span test, Trail Making Test (TMT), Psychomotor Vigilance Test (PVT), Stroop test, Cognitive Flexibility Inventory (CFI) and Digit Symbol Substitution Test (DSST).
Correlation between sleep spindle density and Montreal Cognitive Assessment (MoCA) total score
The correlation between sleep spindle density, quantified from full-night polysomnography (PSG), and global cognitive performance assessed using the Montreal Cognitive Assessment (MoCA; total score, 0-30 points, with higher scores indicating better cognitive performance). The correlation will be assessed using Pearson's or Spearman's correlation coefficient, as appropriate. The correlation coefficient (r) and its 95% confidence interval will be reported. An adjusted multiple linear regression analysis will additionally assess the relationship between sleep spindle density and MoCA total score, controlling for age, sex, body mass index, education, and daytime sleepiness.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
Correlation between sleep spindle density and Digit Span Test forward score
The correlation between sleep spindle density, quantified from full-night polysomnography (PSG), and Digit Span Test forward performance, measured as the maximum span length (number of digits correctly recalled). Higher scores indicate better performance. Pearson's or Spearman's correlation coefficient, as appropriate, and its 95% confidence interval will be reported. Multiple linear regression adjusted for age, sex, body mass index, education, and daytime sleepiness will additionally be performed.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
Correlation between sleep spindle density and Psychomotor Vigilance Test (PVT) mean reaction time
The correlation between sleep spindle density, quantified from full-night polysomnography (PSG), and mean reaction time on the Psychomotor Vigilance Test (PVT), measured in milliseconds (ms). Shorter reaction times indicate better sustained attention. Pearson's or Spearman's correlation coefficient, as appropriate, and its 95% confidence interval will be reported. Multiple linear regression adjusted for age, sex, body mass index, education, and daytime sleepiness will additionally be performed.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
Correlation between sleep spindle density and Trail Making Test Part A (TMT-A) completion time
The correlation between sleep spindle density, quantified from full-night polysomnography (PSG), and TMT-A completion time, measured in seconds. Shorter completion times indicate better processing speed. Pearson's or Spearman's correlation coefficient, as appropriate, and its 95% confidence interval will be reported. Multiple linear regression adjusted for age, sex, body mass index, education, and daytime sleepiness will additionally be performed.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
Correlation between sleep spindle density and Stroop Test interference-condition completion time
The correlation between sleep spindle density, quantified from full-night polysomnography (PSG), and completion time for the interference condition of the Stroop Test, measured in seconds. Shorter completion times indicate better performance. Pearson's or Spearman's correlation coefficient, as appropriate, and its 95% confidence interval will be reported. Multiple linear regression adjusted for age, sex, body mass index, education, and daytime sleepiness will additionally be performed.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
Correlation between sleep spindle density and Digit Symbol Substitution Test (DSST) score
The correlation between sleep spindle density, quantified from full-night polysomnography (PSG), and the Digit Symbol Substitution Test (DSST) score, measured as the number of correctly completed symbol substitutions according to the test scoring rules. Higher scores indicate better processing speed. Pearson's or Spearman's correlation coefficient, as appropriate, and its 95% confidence interval will be reported. Multiple linear regression adjusted for age, sex, body mass index, education, and daytime sleepiness will additionally be performed.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
Variance in cognitive test scores explained by a multivariable model including all three EEG markers
Exploratory multiple linear regression models will assess the combined contribution of sleep spindle density, K-complex density, and REM EEG slowing to cognitive test performance. Separate models will be fitted for each prespecified cognitive outcome: MoCA total score, Digit Span forward and backward scores, PVT mean reaction time, TMT-A and TMT-B completion times, TMT B-A difference score, prespecified Stroop Test measures, and DSST score. Models will be adjusted for age, sex, body mass index, education, and daytime sleepiness. The combined contribution of the EEG markers will be summarized using the coefficient of determination (R²), adjusted R², regression coefficients, and their 95% confidence intervals, as appropriate.
Time frame: Single assessment at baseline (at study enrollment), on the day of the overnight PSG.
No study locations are listed for this record.
Plan to share: No — There is no plan to make individual participant data available to other researchers. Participant-level data will be kept at the study site to protect participant privacy; aggregate results will be published.
No publications or documents are linked to this record.
Start is estimated for Nov 2026; primary completion for Jun 2028, 19 months later.
How these figures are worked out, and which studies are left out of them.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
This study is not yet recruiting, as verified in Oct 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.
University of Split, School of Medicine