Mechanisms and Risk Factors of Cognitive Impairment in COPD

Abstract

Topic Importance

Cognitive impairment (CI) is an underrecognized extrapulmonary complication of COPD. Emerging evidence suggests that CI in COPD is both highly prevalent and clinically significant, with detrimental effects on disease management, medication adherence, exacerbation risk, and quality of life. The underlying biological mechanisms contributing to CI in COPD remain poorly defined.

Review Findings

Several plausible mechanisms have been implicated. For example, chronic hypoxemia has been associated with cerebral metabolic dysfunction and structural brain changes. Sleep-disordered breathing related to COPD can cause intermittent hypoxemia and sleep fragmentation, leading to glymphatic dysfunction, oxidative stress, and frontal lobe vulnerability. Oxidative stress, both systemic and cerebral, is exacerbated by smoking and inflammation, promoting endothelial dysfunction and DNA damage. Vascular dysfunction, including cerebral small-vessel disease and microbleeds, is associated with COPD and could be a co-occurrence because of shared risk factors, and correlates with cognitive decline. Finally, chronic hypercapnia contributes to CI through inflammatory, neurochemical, and cerebrovascular alterations, effects compounded when combined with hypoxemia.

Summary

Individuals with COPD that develop CI may reflect distinct biologically defined subtypes, rather than a uniform comorbidity. Current cognitive screening tools may underestimate CI in COPD, underscoring the need for validated, COPD-specific assessments. Although long-term oxygen therapy and treatment of sleep-disordered breathing show promise, their impact on cognition remains untested in trials with cognitive outcomes. Emerging tools (eg, multiomic profiling, tissue-based studies) may help identify high-risk subgroups and guide precision interventions. Given cost and eligibility constraints, precision risk stratification may improve feasibility of future trials.

COPD is a chronic respiratory condition characterized by progressive and largely irreversible airflow limitation because of damage in the airways and alveoli.¹ This damage most commonly results, in susceptible individuals, from prolonged exposure to harmful particles or gases (eg, those found in tobacco smoke). Although traditionally considered a disease confined to the lungs, growing evidence highlights a broad range of extrapulmonary manifestations.² The presence of systemic inflammation, chronic hypoxemia, oxidative stress, and shared risk factors (eg, age, cigarette smoke exposure) supports the growing view that COPD is a multisystem disease affecting almost every organ in the human body, including the cardiovascular, endocrine, hematologic, musculoskeletal, neuropsychiatric, and gastrointestinal systems.³ Notably, cognitive impairment (CI) has emerged as a clinically significant yet underrecognized consequence of COPD. CI in individuals with COPD is linked to a diverse array of pathophysiological mechanisms and is increasingly thought to be a significant contributor to poor clinical outcomes, through complicating disease management, reducing quality of life, and adversely impacting long-term prognosis. Nevertheless, CI remains frequently overlooked in this demographic, underscoring the urgent need for better understanding of altered biological mechanisms, greater awareness, early detection, and integration of comprehensive COPD care strategies.

CI has been reported in up to 77% of individuals with COPD and associated hypoxemia.⁴ Numerous cognitive domains have been shown to be especially vulnerable. These domains include memory, attention, executive function, processing speed, and visuospatial skills, with memory and attention consistently identified as the most adversely affected.⁵ Lung imaging studies have highlighted the association between CI and COPD-related radiographic abnormalities (eg, emphysema severity measured by lung density).⁶ In parallel, high-resolution MRI studies have shed light on structural brain alterations in individuals with COPD that may contribute to CI. For instance, reduced gray matter volumes were observed in regions involved in dyspnea, emotional processing, fear, and pain responses, notably in the anterior cingulate cortex, hippocampus, amygdala, and posterior cingulate cortex. Degeneration of anterior cingulate cortex gray matter correlated with disease duration and disease-specific fears.⁷ This may point toward a vicious cycle of structural brain changes and behavioral reinforcement, which could lead to a spiral of worsening cognitive decline in affected and at-risk individuals. Although progress has been made, further research is needed to better understand the associations between COPD and CI, particularly the underlying pathophysiological mechanisms (Table 1). Although cerebral hypoxia, inflammation, and vascular dysfunction have been proposed, current evidence remains conflicting regarding the relative contributions of each factor and their interactions with disease severity.⁵ This review aims to further explore the complex relationship between COPD and CI by examining proposed pathophysiological mechanisms, including chronic hypoxemia, sleep-disordered breathing (SDB), oxidative stress, vascular dysfunction, and hypercapnia. We then discuss the potential for therapeutic interventions and propose avenues for future research.

Table 1.

Mechanistic Pathways of CI in COPD

Pathways	Summary of Evidence and Implications
Chronic hypoxemia	COPD and severe hypoxemia with 5-fold increase in risk of CI; cerebral hypoxia can be seen with normal arterial oxygen concentration
Sleep-disordered breathing	Overlap syndrome with OSA exacerbates nocturnal hypoxemia and sleep fragmentation; linked to frontal lobe dysfunction and impaired glymphatic clearance
Oxidative stress and inflammation	Systemic redox imbalance contributes to endothelial dysfunction, neuroinflammation, and potential DNA repair deficits
Cerebrovascular dysfunction	COPD independently predicts cerebral microbleeds and small-vessel disease, correlating with cognitive decline
Hypercapnia	Chronic CO2 retention disrupts neurochemical balance and cerebral autoregulation; additive deleterious effects with concurrent hypoxemia

CI = cognitive impairment.

Prospective investigations should aim to better understand the mechanisms contributing to cognitive dysfunction beyond existing associations with broad biologic pathways (Fig 1). There is also a need to develop specific cognitive screening tools and biomarkers for use in routine respiratory care. Longitudinal studies are needed to understand how early CI in individuals with COPD develops and affects long-term outcomes. Finally, large-scale randomized trials will be required to confirm whether existing interventions (eg, pulmonary rehabilitation, long-term oxygen therapy) and potentially new or repurposed therapeutic agents can reduce, improve, or preserve cerebral function in individuals with COPD at the highest risk of CI. Recognizing CI in routine clinical practice may enhance treatment adherence, improve patient outcomes, and elevate quality of life in patients with COPD.

Figure 1.

Mechanisms of cognitive impairment in COPD. ROS = Reactive oxygen species.

Literature Search

We began our literature review process on February 20, 2025. We searched PubMed and Google Scholar databases for studies published in any language from database inception to February 20, 2025, using a combination of key words, including chronic obstructive pulmonary disease, COPD, cognitive impairment, memory, executive function, attention, MRI, brain structure, gray matter, hypoxemia, and vascular dysfunction, either separately or in combination. Snowball sampling was used to identify additional relevant articles from references of selected studies. Articles focusing on pathophysiological mechanisms, neuroimaging findings, cognitive domains affected, and intervention outcomes related to COPD and CI were retained for review.

Evidence Review

Mechanisms of CI in COPD

Chronic Hypoxemia

The brain is among the top oxygen-consuming organs in humans. Although only 15% of cardiac output is directed to the brain, it consumes up to 20% of total body oxygen (130 μmol oxygen per 100 g of brain tissue).⁸ The high oxygen requirements of the brain thus make it highly susceptible to scenarios where oxygen delivery is reduced. The lungs are often implicated in hypoxemia, accounting for 4 of 5 canonical mechanisms of hypoxemia (ventilation/perfusion mismatches, diffusion, restriction, and hypoventilation). COPD can manifest hypoxemia through several of the canonical mechanisms. For example, airway obstruction and collapse as a result of airway wall structural abnormalities related to COPD pathogenesis can result in hypoventilation. Alternatively, vascular pruning associated with COPD and emphysema can increase dead space ventilation, and airway inflammation can lead to shunting.⁹ Despite clear potential mechanisms of hypoxemia related to COPD, the prevalence of hypoxemia in COPD is not well defined. Clinical trials for COPD have shown only a small proportion, approximately 5% to 10%, develop severe hypoxemia (oxygen saturation < 88%).¹⁰ The rates of moderate hypoxemia (oxygen saturation ≥ 88% but < 93% or < 94%) are even less clear. However, the association of severe hypoxemia and CI in COPD has been well established, with individuals with COPD and severe hypoxemia at over a 5-fold increased risk of having CI.¹¹

Given the high metabolic and oxygen demands of the brain, severe hypoxemia is an obvious risk factor for brain dysfunction in COPD. However, studies have also shown that individuals with COPD but normal arterial oxygen concentration still show signs of cerebral hypoxia. For example, a pilot study using magnetic resonance spectroscopy demonstrated reduced levels of oxygen-dependent metabolites, including N-acetyl aspartate, creatine, and choline,¹² in individuals with COPD and impaired cognition. Although clinical trials of oxygen supplementation in COPD have focused on reducing mortality in individuals with severe hypoxemia, additional studies are needed to establish the role of moderate and subclinical hypoxemia in the development of CI in COPD and potentially expand the outcomes for oxygen supplementation beyond mortality to include measures of cognition.

SDB and Nocturnal Hypoxemia

Individuals with COPD frequently experience SDB and episodic oxygen desaturation during sleep. Nocturnal hypoxemia is common in COPD even when daytime oxygenation is adequate, often worsening during rapid eye movement sleep.¹³ Nocturnal breathing disturbances are compounded in the overlap syndrome (coexistent COPD and OSA), where individuals endure more severe oxygen desaturation and hypoventilation at night than those with COPD or OSA alone.¹⁴ Studies consistently show worse cognitive performance among patients with overlap syndrome. For instance, although mild CI has been reported in 36% of individuals with moderate-to-severe COPD vs 12% in control patients, individuals with COPD and severe OSA have nearly twice the prevalence of mild CI compared with patients with COPD without significant nocturnal breathing disturbances.^15,16 Similarly, a study focused on vascular pathology found that patients with overlap syndrome exhibited lower global cognitive function and processing speeds relative to individuals with isolated COPD or OSA.¹⁷

Although individuals with chronic hypoxemia, either nocturnal or awake, likely have faster cognitive decline, recurrent drops in oxygen saturation during sleep (as occur in OSA or COPD with nocturnal hypoxemia) can trigger damaging cycles of cerebral hypoxia and reoxygenation. These cycles induce oxidative stress and inflammation in neural tissue, leading to neuronal cell dysfunction and apoptosis.¹⁸ Notably, the degree of oxygen desaturation was more predictive of neurocognitive outcomes rather than the frequency of apneas. Large cohort studies of individuals with OSA indicate that chronic nocturnal hypoxemia preferentially affects frontal lobe networks, contributing to executive dysfunction, a key domain in the assessment of CI.¹⁹ In addition to direct cerebral damage, the systemic effects of intermittent hypoxia include multiple pathways (eg, oxidative stress, sympathetic activation, inflammation). Together, these can exacerbate neuronal injury via disruption of the cerebral microvasculature and integrity of the blood-brain barrier.

Individuals with overlap syndrome often present with fragmented sleep. Sleep fragmentation reduces deep slow-wave and rapid eye movement sleep, impairing daytime attention and psychomotor function.²⁰ Although intermittent hypoxemia predominantly affects memory, recurrent awakenings chiefly impact attention and cognitive speed. Fragmented sleep also disrupts the glymphatic system, which is a brain mechanism essential for clearing neurotoxic metabolites, including beta-amyloid, the main component of amyloid plaques found in individuals with Alzheimer disease.²¹ Glymphatic clearance primarily occurs during deep sleep. Thus, frequent arousals may inhibit this neuroprotective process, potentially accelerating CI.²² Studies show impaired glymphatic function correlating with worse cognitive outcomes in OSA, which CPAP therapy can partially reverse.²³ Chronic SDB and subsequent impaired glymphatic clearance may therefore cumulatively raise the risk of dementia. For example, higher dementia rates have been observed in older individuals with untreated nocturnal hypoxemia and apnea and accelerated cognitive decline has been linked to SDB severity.^14,24

The convergence of intermittent hypoxemia and sleep fragmentation in individuals with COPD provides a plausible pathophysiological basis for the development of CI in a subset of the overall COPD and CI population. The nocturnal insults can damage hippocampal and cortical neurons, impair synaptic plasticity, and promote small-vessel cerebrovascular disease. These changes are reflected in difficulties with memory, executive function, and attention commonly observed in COPD.²⁵ Accordingly, there is modest evidence that addressing nocturnal breathing disturbances can benefit cognition. In a large cohort, long-term nocturnal oxygen therapy was associated with a marked reduction in CI risk, underscoring the potential protective effect of correcting nocturnal hypoxemia.¹¹ Treatment of coexistent OSA is similarly important, with data suggesting that CPAP therapy can lead to improvements in cognitive function in individuals with COPD, particularly in domains of attention and vigilance, when significant OSA is present.¹⁴ Although more interventional studies are needed, these findings collectively reinforce that nocturnal hypoxemia and fragmented sleep in COPD can lead to compounding cerebral injury, beyond the effects of each individual insult.

Oxidative Stress and Inflammation

Oxidative stress is 1 of the pathophysiological drivers of tissue injury in COPD. Oxidative stress is triggered by exposure to cigarette smoke, which both depletes the supply of endogenous lung antioxidants and generates additional reactive oxygen species that are then left unmitigated, further propagating injury.²⁶ In addition, the redox imbalance in COPD has also been shown to extend outside of the lung. For example, a meta-analysis has shown significantly reduced circulating levels of glutathione in individuals with COPD compared with control individuals. Similarly, carbonyl stress from protein and lipid oxidation and hypochlorous acid generation from myeloperoxidase activation trigger autoimmune responses which further amplify the vascular injury.²⁷

Oxidative stress and associated mitochondrial dysfunction have similarly been implicated in the pathogenesis of CI, independent of COPD and other systemic diseases. However, oxidative stress in COPD is thought to set off a cascade of chronic systemic inflammation that promotes vascular endothelial dysfunction, impairs nitric oxide signaling, and disrupts the vascular tone and repair mechanisms.²⁶ Beyond the vascular consequences, one of the primary mechanisms of injury mediated by oxidative stress is DNA damage. Although brains from both individuals with and without CI show signs of DNA damage, DNA repair mechanisms appear to be impaired in individuals with CI.²⁸ Taken together, individuals with COPD and evidence of impaired DNA repair mechanisms may be a potential endotype at higher risk of developing CI.

Skeletal muscle weakness is also a well-recognized extrapulmonary manifestation of COPD and frequently co-occurs with CI.²⁹ Both conditions appear to share systemic contributors (eg, chronic inactivity, low-grade inflammation) which drive muscle wasting and are linked to worse cognitive performance. Additionally, COPD-related hypoxemia can trigger oxidative stress and inflammation that damage both peripheral muscles and brain tissue, compounding functional decline. Patients with pronounced muscle weakness often exhibit greater executive dysfunction and slower psychomotor speed, reflecting a distinct cognitive-metabolic impairment phenotype associated with hypoxia.³⁰ Thus, the shared oxidative and inflammatory milieu may jointly worsen muscle and cognitive health in COPD.

Vascular Dysfunction and Small Vessel Disease

Vascular dysfunction in COPD, manifesting as intraparenchymal vascular pruning, has been associated with accelerated emphysema progression and lung function decline.³¹ However, the effects of COPD on the systemic vasculature beyond the lung are increasingly being recognized. Vascular dysfunction also appears to be a critical pathway contributing to CI in COPD. There are likely multiple and overlapping pathways through which vascular dysfunction contributes to CI, including systemic inflammation and oxidative stress. However, cerebral microvascular pathology is a vascular-specific mechanism that can lead to CI in COPD.

Several analyses using data from the Rotterdam study demonstrated a high prevalence of cerebral microbleeds (CMBs) in patients with COPD, particularly in deep and infratentorial regions of the brain.32, 33, 34 In multivariable-adjusted analyses, COPD was an independent predictor of CMB development over time, even in individuals without CMBs at baseline. The risk of CMBs was also associated with worse airflow limitation and frequent exacerbations. Importantly, CMBs were associated with cognitive decline. However, as a marker of cerebral small-vessel disease, the authors suggest that COPD-associated CMBs could potentially be a part of broader atherosclerotic or hypertensive vasculopathy.³⁵ Thus, despite the independent association with COPD, individuals with CMBs and other evidence of cerebral small vessel disease may represent a vasculopathy subtype. In other words, the current evidence does not necessarily parse the direction of association of COPD and CMBs as a cause or potential co-occurring result of vascular susceptibility. Further investigation is needed to further define the risks and consequences of vascular dysfunction in COPD.

Hypercapnia and Neurochemical Alterations

Chronic CO₂ retention (hypercapnia) is another important pathophysiological factor linking COPD to cognitive decline. Many individuals with advanced COPD develop chronic hypercapnic respiratory failure. Estimates vary, but several studies have shown that 25% to 50% of individuals with Global Initiative for Chronic Obstructive Lung Disease stage 3 and 4 exhibit chronic hypercapnia (ie, Paco₂ ≥ 45 mm Hg).³⁶ Furthermore, the degree of CO₂ retention correlates closely with neuropsychological impairment.³⁷ Elevated Paco₂ levels have been associated with slower reaction times, memory deficits, reduced information processing speed, and impaired attention in individuals with COPD.³⁸ In addition, although acute CO₂ retention causes confusion and narcosis, chronic effects are more subtle. Acute hypercapnia leads to cerebral vasodilation and acutely increases cerebral blood flow, but chronic hypercapnia may disrupt normal autoregulation and raise intracranial pressure.³⁹ Aside from vascular changes, hypercapnia has been shown to increase cerebral inflammatory cytokines. In a large animal model study with goats exposed to 6% CO₂ for 30 days, IL-1β increased in the prefrontal cortex and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-type glutamate receptor expression in the hippocampus and insular cortex decreased in the first 24 hours. Although IL-1β subsequently normalized, reductions in α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-type glutamate receptor persisted.⁴⁰ These changes led to a > 50% drop in cognitive performance. Chronic hypercapnia can also trigger respiratory acidosis, leading to activation of inflammatory cascades. For example, in response to respiratory acidosis, NLRP3 inflammasome activation in microglia increases IL-1β, promoting neuronal apoptosis.⁴¹ In parallel, excess CO₂ can also lead to oxidative and nitrosative stress, which generates toxic intermediates (eg, peroxynitrite) that have been shown to damage cerebral vasculature and neurons, providing a clear pathway for impaired cognition in COPD.^42,43 Finally, experimental data indicate that prolonged hypercapnia can reduce cerebral neurotransmitters critical for learning and memory, including glutamate and acetylcholine.⁴⁴ Thus, despite compensation, chronic CO₂ retention can demonstrably drive cumulative neuronal injury.

Although chronic hypercapnia alone may account for a subtype of CI in COPD, several studies have shown that individuals with both hypercapnia and hypoxemia may be the most susceptible. For instance, rodent models show worsened memory and hippocampal injury with combined hypercapnia and hypoxemia.⁴⁵ In humans, individuals with COPD display reduced cerebral perfusion, especially in frontal and parietal regions, but those with chronic hypoxemia and hypercapnia show the greatest perfusion deficits and cognitive decline.⁴⁶ In vivo studies suggest that although baseline cerebrovascular reactivity to CO₂ can remain intact in patients with chronica hypercapnia, the overall burden of hypercapnia may lead to brain ischemic stress when combined with hypoxemia.⁴⁷ Therefore, concurrent hypercapnia and hypoxemia may be 1 of the highest risk subtypes of COPD for the development of CI.

Future Directions

Establishing Subtypes of COPD at the Highest Risk of CI

The diagnosis and severity assessment of COPD are dependent on spirometric measurement of FEV₁ and FVC. However, the severity of impairment in spirometry is not highly correlated with symptom severity or susceptibility to extrapulmonary manifestations of COPD. In addition, individuals with similar decrements in FEV₁ demonstrate substantial variability in the degree of radiographic emphysema.⁴⁸ Furthermore, there may be an association of lung-specific outcomes with systemic comorbidities—in this case, nonemphysematous COPD and type 2 diabetes mellitus. Thus, CI may represent 1 such related comorbidity that develops in a subgroup of individuals with COPD. However, the endotype (ie, biologically defined subtype) of COPD most at risk for CI remains undefined. Given the multiple putative mechanisms for CI related to COPD, there may indeed be multiple endotypes of COPD that ultimately lead to CI. However, further biological characterization and more precise subtype definitions are needed beyond the existing associations of individuals who fall under the broad umbrellas of COPD and CI.

Genetics and genomics therefore offer an attractive set of analyses to further define potential endotypes of COPD and related CI. For example, we and others have leveraged large repositories of biological data from individuals with COPD and other chronic diseases to use multiple omics technologies to establish subtypes of disease that may be amenable to precision interventions or potentially treatable traits.49, 50, 51 In parallel, there has been significant growth in available tools to study tissues with limited accessibility in clinical and research contexts, including both lung and brain tissue.^52,53 Combining different but nonetheless cutting-edge analytical methods may reveal important insights into identifying individuals with COPD at elevated risk of developing CI.

Clinical Implications of CI in COPD

The epidemiology of CI and COPD has come into focus over several decades of investigation, and future studies will serve to advance our understanding of the specific pathobiology and potentially identify individuals at highest risk. However, given the aging population and growing proportion of older individuals with COPD, clinical trials of existing interventions targeted toward individuals with COPD and prevalent CI could be implemented in the short run. Specifically, the 2 large clinical trials for oxygen supplementation were only designed to test for reduction in mortality.^54,55 Given the large role that chronic and intermittent hypoxemia may play in the development of CI in individuals with COPD, a future clinical trial of oxygen supplementation that includes cognitive function as the primary outcome could have an immeasurably large and immediate impact. However, because the proportion of individuals with COPD and hypoxemia is relatively small, precision definitions of individuals at highest risk may be required because of the potentially high costs of running a large-scale clinical trial and the significant financial burden of expanded oxygen supplementation on the already-strained health care system.

Aside from oxygen supplementation, there may be limited existing pharmacologic or physiological interventions for CI in COPD. On the other hand, behavioral interventions have been shown to have appreciable improvements in quality of life in individuals with CI, independent of COPD status.⁵⁶ Thus, integration of behavioral interventions into comprehensive COPD treatment plans may improve the care and quality of life for individuals with COPD and CI.

Current cognitive screening tools such as the Mini-Mental State Examination may underestimate impairment in this population.²⁵ Designed primarily to screen for dementia, the Mini-Mental State Examination often fails to capture the often-subtle cognitive deficits common in COPD.⁵⁷ For example, many individuals with mild CI or executive function loss can score in the normal Mini-Mental State Examination range despite significant neuropsychological deficits, resulting in the detection of fewer cases of CI in COPD compared with more sensitive instruments.⁵⁸ Consequently, experts recommend that clinicians use more sensitive screening tools (eg, Montreal Cognitive Assessment, multidomain cognitive testing) for patients with COPD, rather than relying solely on the Mini-Mental State Examination, to avoid missing subtle yet clinically important cognitive deficits.⁵⁹ Addressing these gaps in knowledge is essential for improving patient care and guiding the development of precision preventive and therapeutic strategies.

Summary

The burden of CI in individuals with COPD is high and is suspected to increase as the population ages. However, as smoking declines and environmental exposures rise, the landscape of COPD is shifting, making it difficult to predict how the disease will evolve in the coming years. This uncertainty also extends to the burden of associated CI, which may follow new patterns of risk. There are several mechanisms that may lead to CI in individuals with COPD, both independently and in overlap. Chronic hypoxemia is perhaps the most obvious; however, intermittent hypoxemia from SDB, oxidative stress from systemic inflammation and vascular dysfunction, and concomitant hypercapnia are also important pathways that contribute to cerebral dysfunction in COPD. Future studies should address 2 important issues: identification of endotypes of COPD with the highest risk for CI and development of new or existing interventions for CI. The application of omics technologies is one avenue ripe for breakthroughs to advance our understanding of CI in COPD.

Funding/Support

A. J. G. is supported by NIH K08HL168205.

Financial/Nonfinancial Disclosures

The authors have reported to CHEST Pulmonary the following: A. J. G. reports consulting or advisory fees from Grifols and TDA Research, Inc. None declared (S. P., M. W., I. A.).