Brian James Murray

MRI-visible brain perivascular spaces (PVSs) represent an imaging feature of cerebral small vessel disease that is associated with hypertension, diabetes, and poor sleep quality and correlated with cognitive impairment and dementia. Although research shows that sleep apnea and sleep disruption are associated with PVS volumes, few studies have investigated whether sleep apnea treatment can modify this association. We aimed to test the hypothesis that continuous positive airway pressure (CPAP) therapy can change the trajectory of PVS progression in adults with sleep apnea. In this longitudinal observational treatment study, we recruited participants from sleep clinics at the Royal Infirmary of Edinburgh and Sunnybrook Health Sciences Centre with moderate-to-severe hypoxemic sleep apnea (apnea hypopnea index ≥15 and oxygen desaturation index ≥10). Before and after a minimum 4 months of CPAP, we obtained pulse oximetry (WatchPAT) and performed brain MRI, with quantification of PVS volumes in the basal ganglia (BG) and centrum semiovale (CSO). We used linear mixed-effects models to test for associations between CPAP usage, changes in PVS volumes, and baseline sleep apnea severity. Among 64 participants (mean age 54.7 years old, 31% female) who were followed for more than a median (interquartile range) of 140 (124-163) days, there were 2 subgroups: 37 participants who used CPAP optimally (≥4 hours a night ≥70% of nights) and 27 participants who did not. Optimal CPAP usage modified the association between visit and BG PVS volumes (interaction estimate = -0.007, SE = 0.002, p = 0.007). Participants with the poor CPAP adherence experienced an increase in BG PVS (estimate = +0.005, SE = 0.001, p = 0.002). Participants with consistent CPAP adherence experienced stabilization of BG PVS volumes (estimate = -0.001, SE = 0.001, p = 0.42). This benefit was more pronounced with increased hypoxemia (mean interaction estimate = +0.004, SE = 0.001, p = 0.0009) and increased AHI (interaction estimate = -0.002, SE = 0.00009, p = 0.008). Optimal CPAP use may moderate the effects of sleep apnea on trajectories of MRI-visible PVS volumes. Although the observational nonrandomized nature of this study is a limitation, these data suggest that CPAP may be an effective means of slowing PVS dysfunction progression, a key correlate of aging and dementia, in adults with sleep apnea. NCT03410095. This study provides Class IV evidence that, in patients with moderate-to-severe sleep apnea, optimal adherence to CPAP treatment stabilizes the progression of BG PVS volume.

Sleep is thought to play a critical role in the retention of memory for past experiences (episodic memory), reducing the rate of forgetting compared with wakefulness. Yet it remains unclear whether and how sleep actively transforms the way we remember multidimensional real-world experiences, and how such memory transformation unfolds over the days, months and years that follow. In an exception to the law of forgetting, we show that sleep actively and selectively improves the accuracy of memory for a one-time, real-world experience (an art tour)—specifically boosting memory for the order of tour items (sequential associations) versus perceptual details from the tour (featural associations). This above-baseline boost in sequence memory was not evident after a matched period of wakefulness. Moreover, the preferential retention of sequence relative to featural memory observed after a night’s sleep grew over time up to 1 year post-encoding. Finally, overnight polysomnography showed that sleep-related memory enhancement was associated with the duration and neurophysiological hallmarks of slow-wave sleep previously linked to sequential neural replay, particularly spindle–slow wave coupling. These results suggest that sleep serves a crucial and selective role in enhancing sequential organization in our memory for past events at the expense of perceptual details, linking sleep-related neural mechanisms to the days-to-years-long transformation of memory for complex real-life experiences.

Study Objectives: Periodic limb movements (PLMs) of sleep, which may be linked to increased vascular events via nighttime sympathetic overactivity, have shown associations with cerebral small vessel disease (CSVD) in small studies. This study examined the relationship between PLMs and CSVD in a larger cohort, accounting for comorbidities. Methods: Patients with first-ever stroke or transient ischemic attack (TIA) were retrospectively analyzed. Polysomnography assessed the PLM index (PLMI) and PLM arousal index (PLMAI). CSVD was measured using radiographic markers, including the Fazekas score, total age-related white matter changes (ARWMC) score, microbleed counts, and lacunar infarcts. Multivariable regression models analyzed the association between an elevated PLMI and PLMAI with CSVD markers. Results: This study included 86 patients (mean age 62.2 ± 14.3 years, 66.3% male, mean BMI 28.1 ± 5.7), 36 with a PLMI ≥ 5 (41.9%) and 11 with a PLMAI ≥ 5 (12.8%). Regression analyses showed that PLMI ≥ 5 and PLMAI ≥ 5 both predicted increased Fazekas and ARWMC total scores after adjusting for age, sex, BMI, and other comorbidities. Sensitivity analyses using age- and sex-specific PLMI cutoffs also showed that a PLMI exceeding the upper limit of normal predicted an increased Fazekas score and approached significance for the ARWMC score. PLMI was not significantly associated with cerebral microbleeds or lacunar infarcts. Conclusions: In patients with first-ever minor stroke and TIA, a significant association was observed between PLMI and PLMAI with white matter hyperintensities (WMHs) after adjusting for confounders. Future studies may help determine the directionality of this association and whether PLMs independently predict CSVD. Clinical Trial Information: Name: Sleep Disorders Managed and Assessed Rapidly in Transient Ischemic Attack (TIA) and In Early Stroke (SMARTIES). URL: https://clinicaltrials.gov/study/NCT01528462 ClinicalTrials.gov ID: NCT01528462. Name: SLEep APnea Screening Using Mobile Ambulatory Recorders After TIA/Stroke (SLEAP SMART). URL: https://clinicaltrials.gov/study/NCT02454023. ClinicalTrials.gov ID: NCT02454023.

Sleep is thought to play a critical role in the retention of episodic memories. Yet it remains unclear whether and how sleep actively transforms memory for specific experiences. More generally, little is known about sleep’s effects on memory for multidimensional real-world experiences, both overnight and in the days to months that follow. In an exception to the law of forgetting, we showed that sleep actively and selectively improves retrieval of a one-time real-world experience (a controlled but immersive art tour) – specifically boosting memory for the order of tour items (sequential associations), but not perceptual details from the tour (featural associations). This above-baseline increase in sequence memory was not evident after a matched period of wakefulness. Moreover, the sleep-induced advantage of sequence over featural memory grew over time up to one-year post-encoding. Finally, overnight polysomnography showed that sleep-related memory enhancement was associated with the duration and neurophysiological hallmarks of slow-wave sleep previously linked to neural replay, particularly spindle-slow wave coupling. These results suggest that sleep serves a crucial and selective role in enhancing sequential organization in episodic memory at the expense of specific details, linking sleep-related neural mechanisms to the transformation and enhancement of memory for complex real-life experiences. Significance Statement Sleep affects the retention of episodic memories. Yet, it remains unclear whether sleep active transforms how we remember past experiences, overnight and beyond. We investigated memory for different dimensions underlying a dynamic real-world event – sequential associations versus atemporal featural associations – before and after sleep or wakefulness, and serially up to a year later. Sleep actively and selectively enhanced sequence memory, with this preferential sequence retention growing with time. Overnight memory enhancement is associated with the duration and neurophysiological hallmarks of slow-wave sleep previously linked to sequential neural replay, particularly spindle-slow wave coupling. Our findings support an active role for sleep in transforming different aspects of real-world memory, with sequence structure coming to dominate long-term memory for dynamic real-world experiences.

Background The DOC screen was developed to identify Depression, Obstructive sleep apnea, and Cognitive impairment (“DOC” comorbidities) after stroke. Each component has its own score, but additional information may be gained from the time to complete the screen. Cognitive screening completion time is rarely used as an outcome measure. We assessed the added value of using DOC screen completion time as a predictor of impairment on detailed cognitive assessments. Methods Consecutive English-speaking new referrals to the stroke prevention clinic were consented to participate in detailed neuropsychological testing (n=437). DOC screen scores and times were compared to cognitive test scores using multiple linear regression and receiver operating characteristic (ROC) analysis. All linear regression analyses controlled for age, sex, years of education, and functional outcome as assessed by the modified Rankin score. Results Average completion time for the DOC screen was 3.8 ± 1.3 minutes. After accounting for age, sex and cognitive screen score, completion time was a significant independent predictor, of speed of processing (p = .002, 95% CI: -0.016 to -0.004), verbal fluency (p < .001, CI: -0.012 to -0.006) and executive (p = .004, CI: -0.006 to -0.001), but not memory, function. Completion time above 5.5 minutes (332.5 seconds) was associated with a high likelihood of impairment on gold standard executive (likelihood ratios 3.9-5.2) and speed of processing (likelihood ratio = 5.2) tasks. Conclusions DOC screen completion time is easy to collect in routine care and is independently associated with speed of processing, language and executive dysfunctions after stroke. People who take more than 5.5 minutes to complete the DOC screen are likely to have deficits in executive functioning and speed of processing. These domains can be challenging to screen for in stroke survivors, and this measure provides a simple, clinically feasible method to screen for these under-appreciated concerns. Clinical Trials Registration Identifier: NCT02363114 Clinical Trials URL: https://clinicaltrials.gov/ct2/show/NCT02363114

Research imaging in healthy and clinical youth populations yields incidental findings that require a management strategy. Our primary objective was to document the frequency and nature of incidental findings within a research group integrating multiple imaging modalities. A second objective was to describe the evolution of an approach to handling incidental findings. A case example was included to display the intricacies of some of these scenarios. Youth, ages 13–20 years, with bipolar disorder, familial risk for bipolar disorder, or healthy controls, obtained one or a combination of neuroimaging, cardio-thoracic imaging, retinal imaging, and carotid imaging. All images were systematically reviewed for incidental findings. Overall, of 223 participants (n = 102 healthy controls), 59% (n = 131) had a brain magnetic resonance imaging (MRI) incidental finding and 27% (n = 60) had at least one incidental brain finding requiring non-urgent follow-up. In addition, of 109 participants with chest/cardiac MRI and carotid ultrasound, 3% (n = 3) had chest findings, 2% (n = 2) had cardiac findings, and 1% (n = 1) had a carotid finding. Of 165 youth with retinal imaging, 1% (n = 2) had incidental findings. While the vast majority of these incidental findings were of a non-serious, non-urgent nature, there were noteworthy exceptions. Imaging research groups need a system that emphasizes the value of clinical review of research images and one that is collaborative and responsive in order to inform follow-up plans. Rating systems that have been developed and used in neuroimaging for the classification of incidental findings can be adapted for use in areas other than the brain. Regardless of severity, incidental findings may raise anxiety in youth participants and their parents. The optimal threshold is one that balances transparency with utility.