TABLE VIII.B. 5.

PAP to improve cognitive measures

Study	Year	LOE	Study design	Study groups	Clinical endpoints	Conclusion
Cognitively intact (i.e., inclusion criteria did not require mild cognitive impairment or dementia)
Kushida et al. (APPLES trial)1272	2012	1b	Randomized, double-blind, controlled trial comparing CPAP to sham-CPAP (placebo), multicenter trial	1098 participants enrolled across five sites, with assessments at baseline, 2 and 6 months.	Neurocognitive domains: 1) Executive and frontal lobe function (Sustained Working Memory Test-Overall Mid-Day Score); 2) Attention and psychomotor function (Pathfinder Number Test-Total Time); 3) Learning and memory (Buschke Selective Reminding Test-Sum Recall). Sleepiness: 1) Epworth Sleepiness Scale; 2) Maintenance of Wakefulness Test (MWT)	Neurocognitive domains: transient improvement in executive and frontal lobe function only at 2 months, but no sustained benefit. Sleepiness: CPAP led to improvements in subjective and objective sleepiness, especially in those with an AHI>30 events/h.
Pan et al.1270	2015	1a	Meta-analysis of randomized controlled trials (duration 1–24 weeks)	PubMed, CINAHL, Medline, PsycInfo, EMBASE, Cochrane Library, CNKI, WanFang, VIP, and CBMdisc from June 1971 to July 2014. 13 studies were identified involving 1744 participants. 1098 were from APPLES trial. Did not include trials of cognitively impaired patients.	Cognitive domains: 1) attention; 2) vigilance; 3) processing speed; 4) working memory; 5) memory; 6) verbal fluency; 7) visuoconstructive skills	CPAP led to a small but statistically significant improvement in vigilance only.
Dalmases et al.1281	2015	2b	Open label randomized pilot study of CPAP vs. conservative care for 3 months	33 cognitively intact participants age >65 years with severe OSA enrolled at a single site, parallel arm study.	Cognitive domains: 1) episodic memory; 2) short-term memory; 3) executive function-working memory, speed, visuomotor speed, mental flexibility, and verbal fluency. MRI scans at baseline and 3 months also	CPAP led to statistically significant improvements in episodic, short-term memory, speed of mental processing, and mental flexibility. MRI scan showed reduced cortical thinning and increased right middle frontal gyrus connectivity in the CPAP group. No change in Epworth Sleepiness Scale.
McMillan et al.1282	2015	1b	Open label randomized study of CPAP + best supportive care (BSC) vs. BSC alone	278 participants age > 65 years with an oxygen desaturation index >7.5 events/h + Epworth Sleepiness Scale ≥ 9 for a 12-month period.	Cognitive measures included the Mini-Mental Status Exam, Trail Making Test Part B, Digit Symbol Substitution Test, and reaction time test	CPAP did not lead to any cognitive changes. Of note, CPAP objective adherence (defined as ≥4 h of use/night) was 35% and median usage was 2:22 (h:min)/night at 12 months.
Bubu et al.832	2020	2a	Systematic review	PubMed/Medline, Embase, Psych INFO, and Cochrane library for clinical trials published prior to May 1, 2019 related to OSA and cognitive function.	Included research in cognitively intact and cognitively impaired populations	CPAP treatment may offer cognitive benefits for middle-aged and older adults, including those with Alzheimer’s disease.
Martinez-Garcia et al.1283	2015	2b	Open label randomized study of CPAP vs. no CPAP for 3 months	224 older adults age ≥70 years, with severe OSA (AHI ≥ 30 events/h) enrolled at 12 sites; excluded participants with severe impairment of cognitive performance and ESS ≥ 18.	Cognitive domains: executive function, visual attention, speed of processing/mental flexibility, and working memory. Other domains: subjective sleepiness, anxiety, depression, and sleep quality	CPAP treatment led to statistically significant improvements in depression, anxiety, and sleepiness (small, moderate, and large effect sizes, respectively). Statistically significant improvements for working memory (digit symbol test and Trail Making A) that were small and moderate, respectively. No effect on other domains.
Patil et al.39	2019	1a	AASM systematic review and meta-analysis	9 RCTs of variable patient populations with mild-mod OSA; control groups were sham CPAP, oral placebo, or conservative treatment. Duration of RCTs was at least 1 month (1–12 months).	Neurocognitive function across domains of executive function, processing speed, attention and vigilance, memory, and intelligence	Meta-analyses showed no clinically significant difference between CPAP and control groups in any of the neurocognitive domains.
Wang et al.1274	2020	1a	Meta-analysis of randomized controlled trials (duration 1–24 weeks)	PubMed, EMBASE, and Cochrane Library were systematically searched for RCTs from database inception to October 24, 2019. 14 studies were identified, involving 1926 participants; 843 were from APPLES trial. Did not include trials of cognitively impaired patients except for one study of Parkinson’s disease patients and one of stroke patients.	Cognitive domains: 1) attention and speed of information processing; 2) executive function; and 3) memory. Also reviewed individual cognitive scales when available, such as Trail Making A/B and subjective sleepiness	CPAP led to a small but statistically significant improvement in attention and vigilance only in severe OSA. No difference in subjective sleepiness. Of note, did not include APPLES trial data and large heterogeneity existed for sleepiness data analyses.
Cognitively impaired (i.e., mild cognitive impairment or Alzheimer’s disease)
Ancoli-Israel et al.1275	2008	1b	Randomized, controlled trial comparing CPAP to sham-CPAP (placebo)	52 participants with mild to moderate Alzheimer’s disease and OSA. Assessed at baseline and 3 weeks, then the placebo sham-CPAP arm crossed over to receive CPAP.	Fourteen cognitive measures assessing multiple domains including attention, vigilance, psychomotor speed, memory, and executive function. Also derived a composite neurocognitive score from these 14 items	No significant difference in cognitive measures for a priori analyses, possibly due to sample size. When an additional analysis was conducted including the cross-over data (pre-/post-paired comparison), statistically significant benefits noted on the composite score, Hopkins Verbal Learning Test-Revised (HVLT-R), and Trail Making test.
Richards et al.1271	2019	2b	Quasi-experimental pilot study comparing CPAP adherent and non-adherent participants	54 participants with mild cognitive impairment. Assessed at one year follow-up.	Memory (Hopkins Verbal Learning Test-Revised), psy- chomotor/cognitive processing speed (Digit Symbol subtest from the Wechsler Adult Intelligence Scale Substitution Test). Progression measures were the Everyday Cognition, Alzheimer’s Disease Cooperative Study–Clinical Global Impression of Change Scale, and Clinical Dementia Rating Scale	At 1-year: statistically significant improvements in psychomotor/cognitive processing speed in the CPAP adherent patients; small to moderate but not statistically significant improvements noted in memory, attention, and everyday function.
Perez-Cabezas et al.1255	2020	2a	Systematic review of randomized controlled trials and cohort studies for CPAP treatment of OSA in Alzheimer’s disease (AD)	Review of publications in PubMed, PEDro, SCOPUS, SPORTDiscus, CINHAL and Web of Science up to July 2019 using PRISMA framework	Identified five studies involving CPAP treatment of patients with AD and OSA: three clinical trials and two pilot studies. Treatment duration ranged from 3 weeks to 3 months	Noted some evidence to support attenuation of cognitive decline in AD patients, but there were several methodological limitations and a paucity of studies.