Between those treated with CPAP and those who withdrew treatment, the difference in N3 sleep represented 15.7% of variance beyond a base model that included age and sex alone.
In some of the first evidence of its kind, results from a randomized crossover study showed that 3-night withdrawal of continuous airway pressure (CPAP) in patients with obstructive sleep apnea (OSA) has an effect on neurodegenerative and amyloid plasma biomarkers, most notably N3 sleep.
Presented at 2022 SLEEP Annual Meeting, June 4-8, in Charlotte, North Carolina, the analysis included 30 individuals (mean age, 52 years) who were randomly assigned to either CPAP treatment or 3-night CPAP withdrawal and had plasma samples collected every morning and night. The study used paired t-tests to compare measures across sleep conditions while hierarchical linear regression with difference in the overnight change of each biomarker between conditions were set as dependent variables. Age and sex were also used as covariates.
Lead investigator Korey Kam, PhD, assistant professor, Icahn School of Medicine at Mount Sinai, and colleagues noted that they expected sleep disruption to follow the withdrawal of CPAP. Compared with the treated group, those off the treatment showed significant differences in several measures, including percentage of N3 sleep (off: 6.1% [95% CI, 3.7-8.5]; on: 15.1% [95% CI, 10.6-19.6]; P <.001) and percentage of rapid eye movement (REM; off: 11.8% [95% CI, 8.8-14.7]; on: 20.6% [95% CI, 18.3-22.9]; P <.001).
The untreated group continued to show significant differences in other measures such as percentage of apnea hypopnea index-4 (AHI4; off: 63 per hr [95% CI, 54-72]; on: 3 per hr [95% CI, 2-4]; P <.001), oxygen desaturation (Sp02) below 90% (off: 20 min [95% CI, 14-26]; on: 1 min [0-3]; P <.001) and Sp02 min (off: 77% [95% CI, 74-80]; on: 88% [95% CI, 86-90]; P <.001). Patients who withdrew CPAP treatment demonstrated overnight increases in neurofilament light (NfL) and decreases in amyloid-ß40 (Aß) relative to those who stayed on treatment.
Between the 2 groups, there were no change observed with tau or Aß42. Additionally, the difference in percentage of N3 sleep among those on and off CPAP significantly explained an additional 15.7% of the variance beyond a base model including age and sex alone. No other difference in sleep architecture or apnea severity/hypoxemic burden metric significantly contributed to the variance in overnight change in Aß40 between conditions. Notably, there were no significant predictors for overnight change in NfL.
Sleep-breathing disorders such as OSA have previously been shown to be associated with a higher risk of progression to Alzheimer disease (AD). One notable 2021 study by Diaz-Roman et al provided evidence that OSA could be related to the pathophysiological mechanisms involved in neurodegeneration among patients with mild cognitive impairment.2 In that analysis, unadjusted correlation analyses showed that a higher AHI was related to higher phosphorylated tau and total tau (P = .001 after Bonferroni correction). Importantly, these associations were observed even after adjusting for potential confounders such as age, sex, body mass index, sleep medication, smoking, hypertension, and heart disease.