INTRODUCTION

Idiopathic normal pressure hydrocephalus (iNPH) is a neurological condition typical of the elderly, characterized by enlargement of the cerebral ventricular system. Clinically, it presents with locomotor, cognitive and urinary disturbances [1]. iNPH often mimics other neurodegenerative diseases, leading to misdiagnosis and treatment delays [2]. Several pathogenic mechanisms, primarily involving cerebrospinal fluid (CSF) circulation and the glymphatic system, have been proposed [3], yet its pathogenesis remains controversial [4].

The role of sleep is crucial in many physiological processes, including cognitive functions [5], and sleep-related disorders are common in many neurological diseases [6]. Sleep disorders, potentially causing malfunctioning of the glymphatic system, could result in the toxic accumulation of misfolded proteins and contribute to the pathogenesis of neurodegenerative diseases [3]. Indeed, this is the hypothesized pathogenetic mechanism through which obstructive sleep apnoea (OSA) may increase the risk of Alzheimer's disease (AD) [7, 8].

However, the clinical relevance of sleep disturbances in patients with iNPH remains unclear. The few available studies have primarily focused on the prevalence of sleep-related breathing disorders, specifically OSA [9-11], but evidence regarding the clinical impact of OSA in iNPH patients is lacking.

The aim of our study was to investigate the clinical correlates of sleep profiles and OSA in iNPH patients, focusing on clinical and neuropsychological features as well as neurodegeneration biomarkers in CSF.

METHODS

RESULTS

Fifty-one patients were initially screened, of whom 12 were excluded due to prior shunt surgery and one unable to comply with study procedures. Therefore, a total of 38 patients were included in the analysis, with 23 being male (61%). Nineteen patients (50%) were diagnosed with OSA, with 12 (63.1%) classified as moderate (7; 36.8%) or severe (5; 26.3%). No significant differences were found amongst the considered variables when comparing OSA− and mild OSA+ versus moderate–severe OSA+ patients.

The time (47.8 s, SD 21.6, vs. 34.1s, SD 12.2; p = 0.02) and number of steps (61.2, SD 20.4, vs. 48.8, SD 15.7; p = 0.04) required for the 10-m walk test were significantly higher in OSA+ iNPH patients. Similarly, scores on the ‘gait’ subitem of the iNPHRS were lower (32.5, SD 15.7, vs. 44.9, SD 20.2; p = 0.04) in OSA+ iNPH patients. OSA+ iNPH patients were more frequently categorized as having the locomotor phenotype compared to the disequilibrium phenotype (p = 0.02) and all OSA+ iNPH patients with moderate–severe OSA exhibited the locomotor phenotype (p = 0.008).

Regarding cardiovascular risk factors, only the presence of cardiovascular disease was more prevalent in the apnoeic group (p = 0.04). No significant differences in radiological features were observed between the two groups. Table 1 summarizes the demographic and clinical characteristics of the cohort.

OSA+ iNPH patients exhibited higher levels of CSF t-Tau (232, IQR 57, vs. 157, IQR 86; p = 0.02) and p-Tau (31.2, IQR 10.9, vs. 23, IQR 12.1; p = 0.03) with no significant difference in Aβ42 levels (445, IQR 366, vs. 615, IQR 276; p = 0.68). Eighteen patients (47%) had Aβ42 levels below the cutoff for AD (≤599 pg/mL) with no difference observed between OSA+ and OSA− patients, and only one patient had CSF biomarker profile consistent with AD pathology.

No significant differences were found in MMSE (25.7, IQR 2.6, vs. 26.7, IQR 3.6; p = 0.26) and MoCA (18.7, SD 3.2, vs. 20.3, SD 3.9; p = 0.17) corrected scores, or in the distribution of MCI, between OSA+ and OSA− iNPH patients. However, OSA+ iNPH patients demonstrated lower corrected scores in the Rey Auditory Verbal Learning Test immediate recall (29.8, SD 4.5, vs. 34.7, SD 8.1; p = 0.03) and the Rey–Osterrieth Complex Figure copy (26.8, SD 7.3, vs. 32.8, SD 3.7; p = 0.01).

Except for reduced REM sleep in OSA+ iNPH patients (10.7%, IQR 10.6, vs. 12.5%, IQR 5.9; p = 0.03), no other statistical differences emerged from polysomnographic sleep parameters or sleep self-reported questionnaires.

Tables 2 and 3 summarize comparisons between OSA+ and OSA− iNPH patients regarding CSF neurodegeneration biomarkers, neuroradiological features and polysomnographic sleep parameters.

A correlation was found between SpO₂ mean and time (Spearman's ρ = −0.42, p < 0.01) and step (Spearman's ρ = −0.35, p < 0.01) required for the 10-m walk, in addition to obesity (Spearman's ρ = −0.41, p = 0.02).

DISCUSSION

In our extensive cohort of non-shunted iNPH patients, a high prevalence of OSA was observed, frequently manifesting in a moderate to severe form. This prevalence exceeds that observed in healthy age-matched individuals from the general population [15-17], although it is lower than reported in previous studies. Indeed, recent investigations into sleep breathing disorders in iNPH, including a smaller sample size and patients with and without shunt, reported prevalence rates of 90.3% (80.6% moderate to severe) [10] and 96% (82% severe) [11] respectively. Earlier studies also indicated high rates of respiratory disturbance index greater than 10 in 65% of patients and over 12 in 47% [11]. Despite these findings, previous research has not thoroughly explored potential correlations between OSA and clinical outcomes [9, 10]. Addressing this gap in the literature, our study demonstrates that OSA in iNPH patients is associated with poorer motor and cognitive performance.

It was found that OSA+ iNPH patients exhibited diminished motor performances as measured by the 10-m walk test and iNPHRS, whereas the UPDRS part III, which did not differentiate significantly between the groups, appeared less sensitive to the type of motor impairment present in these patients. Interestingly, our study linked OSA with the clinical phenotype of iNPH predominantly characterized by parkinsonism. In Parkinson's disease (PD) the Timed Up and Go test has been shown to more accurately reflect overall functional status compared to UPDRS [18]. In PD, the presence of OSA may worsen motor symptoms and increase disability risk [19]. Our findings suggest a similar impact of OSA on iNPH patients. However, the mechanisms underlying the potential deterioration of motor function due to OSA remain incompletely understood.

Obstructive sleep apnoea induces intermittent hypoxia, alters intrathoracic and intracranial pressure, triggers sympathetic activation and disrupts sleep architecture, all of which are closely interrelated. Sleep fragmentation and hypoxaemia have been hypothesized as potential causes of worsened motor symptoms in PD patients with OSA [18]. Our data did not reveal differences in sleep macrostructure (e.g., sleep efficiency, time awake during bedtime) between OSA+ and OSA− iNPH patients, but a correlation was found between the hypoxic burden, expressed as mean oxygen saturation, and poorer motor performance, specifically in the time and steps needed for a 10-m walk. Intermittent hypoxia resulting from OSA may lead to structural damage and dysfunction within the central nervous system, involving areas crucial for motor functions (i.e., bilateral anterior cingulate cortex, superior frontal gyrus and the cerebellum) [20]. Additionally, hypoxaemia may particularly impact dopaminergic neurons in the substantia nigra, which are susceptible to ischaemic-anoxic insults [21], as well as the excitability and vulnerability of noradrenergic neurons in the locus coeruleus [22].

It was observed that OSA+ iNPH patients had poorer performances in verbal learning and memory [23] (as evidenced by the Rey Auditory Verbal Learning Test immediate recall) and in visuospatial constructional ability and visual memory [24] (as evidenced by the Rey–Osterrieth Complex Figure copy). OSA is a well-documented risk factor for cognitive impairment [25] with significant deficits reported across various cognitive domains such as executive function, attention, long-term verbal/visual memory, visuospatial/constructional ability and information processing [26, 27]. Moreover, structural damage and central nervous system dysfunction in OSA patients, predominantly affecting brain areas involved in integrative cognitive and emotional processes, are associated with neurocognitive deficits [8]. Several deficits have been shown to be improved following continuous positive airway pressure treatment [28, 29]. Cognitive impairment in OSA patients may involve sleep disruption, slow-wave sleep disturbances, hypoxaemia and increased intrathoracic and intracranial pressures due to apnoea, potentially triggering neuronal damage [30]. These mechanical changes related to OSA may impair CSF drainage through the venous system and disrupt glymphatic system function [31, 32]. Moreover, OSA reduces REM sleep [33], thereby compromising glymphatic system efficiency. Our findings align with these observations, as reduced REM sleep in OSA+ iNPH patients was observed. Altered CSF dynamics my lead to the accumulation of metabolites potentially toxic to neurons [34, 35].

The relationship between OSA and CSF neurodegeneration biomarkers remains poorly understood. Studies examining t-Tau and p-Tau levels in the CSF of OSA patients have yielded conflicting results [34-39]. Our study revealed higher levels of t-Tau and p-Tau in OSA+ compared to OSA− iNPH patients, with Aβ42 levels showing no statistically significant difference [40]. The elevated t-Tau levels detected in OSA+ iNPH patients might be explained by reduced exchanges between CSF and interstitial space [41]. Moreover, the soluble nature of t-Tau and its diffusion mechanism akin to prions may render it more detectable in CSF [42, 43]. The concurrent elevation of p-Tau in OSA+ iNPH patients may be induced by OSA-related hypoxia, facilitating tau hyperphosphorylation [30, 44, 45] under conditions where the substrate is more available. The mechanisms underlying these CSF findings remain only partially understood at present and warrant further, more comprehensive investigation. Furthermore, elevated levels of CSF t-Tau, indicative of neuronal damage, may reflect heightened vascular burden associated with OSA [46]. OSA is a well-known independent vascular risk factor [47] and it has been linked to brain white matter hyperintensities [48], analogous to findings in iNPH [49]. Whilst the association between OSA and other vascular risk factors in iNPH could theoretically exacerbate overall vascular alterations, our findings suggest that OSA-related vascular burden, potentially disrupting CSF influx through the brain parenchyma, did not differ significantly in OSA+ iNPH patients based on the Fazekas scale assessment.

Our study has several limitations, including its small sample size and single-centre design. Moreover, our findings do not establish a causal link or determine the direction of the pathophysiological processes linking OSA and the complex disease iNPH, necessitating further investigation. Longitudinal studies focusing on the potential impact of standard OSA treatment with continuous positive airway pressure therapy on the clinical manifestations of iNPH are warranted.

In conclusion, our study documents a high prevalence of mostly moderate–severe OSA amongst iNPH patients compared to age-matched individuals in the general population. OSA in iNPH patients correlates with poorer motor and cognitive performance, particularly affecting memory and visuospatial functions. The underlying mechanisms of motor and cognitive impairment in OSA patients remain speculative, potentially involving neural damage due to hypoxaemia and/or altered glymphatic/liquor drainage efficiency, which may be related to increased intrathoracic and intracranial pressures and changes in sleep macrostructure. Further studies involving larger patient cohorts and longitudinal evaluations are essential to elucidate the significance of OSA and the potential therapeutic impact of OSA treatment on the pathogenesis and severity of clinical manifestations in iNPH.

AUTHOR CONTRIBUTIONS

Regalbuto Simone: Conceptualization; investigation; writing—original draft; methodology; validation; visualization; data curation; resources. Zangaglia Roberta: Investigation; methodology; conceptualization; writing—review and editing; resources. Valentino Francesca: Investigation; resources; writing—review and editing. Todisco Massimiliano: Investigation; resources; writing—review and editing. Pacchetti Claudio: Conceptualization; investigation; resources; writing—review and editing. Cotta Ramusino Matteo: Writing—review and editing. Mazzacane Federico: Formal analysis; visualization; methodology. Picascia Marta: Investigation. Arceri Sebastiano: Investigation. Malomo Gaetano: Investigation. Capriglia Elena: Investigation. Spelta Laura: Investigation. Rubino Annalisa: Investigation. Pisani Antonio: Writing—review and editing. Terzaghi Michele: Conceptualization; investigation; writing—review and editing; methodology; validation; supervision; resources.

FUNDING INFORMATION

This research received no funding.

CONFLICT OF INTEREST STATEMENT

The authors report no relevant disclosures relating to the topic in this paper.