Poor lung gas exchange may help explain lingering brain symptoms in Long COVID

By Priyanjana Pramanik, MSc.Reviewed by Susha Cheriyedath, M.Sc.Nov 25 2025

New imaging reveals how subtle lung changes may influence sleep and brain activity years after COVID, offering clues to why symptoms persist even when routine tests look normal.

Study: Long COVID: lung pathophysiology and its relationship with cognitive dysfunction. Image Credit: designium / Shutterstock

In a recent study published in the journal Scientific Reports, researchers examined whether lung abnormalities are associated with cognitive symptoms and brain scans in people with long coronavirus disease 2019 (COVID-19) symptoms, at an average of 32±5 months (range=24–43 months) after infection. Lower lung gas-exchange efficiency within the Long COVID-19 group was associated with worse sleep symptoms, poorer executive functioning, and higher cerebral perfusion, suggesting a potential lung–brain pathway contributing to persistent symptoms.

Persistent Respiratory and Cognitive Symptoms in Long COVID

Millions worldwide experience persistent symptoms, breathlessness, fatigue, exercise intolerance, and “brain fog” beyond three months after acute infection. Despite common respiratory complaints, conventional pulmonary tests often appear normal. Hyperpolarized xenon-129 (129Xe) MRI can sensitively measure regional ventilation and gas exchange, offering diagnostic value. Cognitive complaints are also prevalent, and prior MRI studies demonstrate gray matter changes, diffusion abnormalities, and altered perfusion. Yet the connection between lung physiology and brain alterations remains insufficiently understood.

Recruitment Criteria for Lung, Brain, and Cognitive Assessment

Adults aged 18 or older were recruited from a post-COVID-19 clinic if they experienced persistent dyspnea or fatigue. Eligibility required a negative PCR test and either previous hospitalization, abnormal pulmonary tests, or chest CT abnormalities. People with active respiratory infections or other cardiopulmonary conditions were excluded. A subset who completed 129Xe MRI, brain MRI, and cognitive testing during the same session formed the core sample.

Multimodal Testing: Symptoms, Cognition, Lung Function, and Brain Imaging

Participants completed questionnaires assessing cognitive complaints, mood, fatigue, dyspnea, pain, and sleep. Objective cognition was evaluated using the NIH Toolbox Cognition Battery with embedded validity indicators. Pulmonary testing followed standard guidelines. Brain MRI included structural imaging, diffusion tensor imaging, and arterial spin labeling to quantify volumetrics, white matter hyperintensities, diffusion metrics, and cerebral perfusion. 129Xe MRI assessed ventilation defects and gas-exchange efficiency; healthy controls completed only the lung-imaging component. Age-adjusted partial correlations linked lung metrics to symptoms, cognition, and MRI measures.

Symptom Profiling Shows Persistent Complaints but Intact Objective Cognition

The final sample included 12 Long COVID-19 participants and 10 matched controls, nearly three years post-infection. Long COVID-19 participants frequently reported fatigue, dyspnea, headaches, and substantial cognitive concerns. Yet objective cognitive performance remained normal for all but one participant, who showed domain-specific deficits in executive function, processing speed, and language. Pulmonary function tests were normal for all but one individual, and no group differences emerged in 129Xe gas-exchange values.

Lung Gas Exchange Correlates With Sleep Problems and Brain Perfusion

Even without group-level lung abnormalities, within-group associations were notable. Lower 129Xe gas-exchange ratios correlated with worse sleep disturbance, higher cerebral perfusion, and poorer executive functioning. Cerebral perfusion itself was strongly tied to sleep problems. Gas-exchange efficiency showed positive correlations with executive function and a trend with global cognitive scores; these persisted after adjusting for anxiety and depression. No associations emerged with subjective cognitive complaints, brain volumes, diffusion metrics, or white matter abnormalities after age adjustment.

Interpreting the Lung–Brain Connection in Long COVID

Findings suggest that impaired lung gas exchange may be associated with altered cerebral perfusion and sleep disruption, forming a potential physiological pathway that influences Long COVID-19 symptoms. The mismatch between subjective cognitive difficulties and normal objective cognition aligns with prior research and may reflect fluctuating deficits, heightened symptom awareness, or psychological factors.

Study Limitations and Future Directions for Long COVID Research

Strengths include a multimodal design integrating lung imaging, brain MRI, cognitive assessment, and symptom profiling. However, the small sample, partial imaging datasets, and lack of cognitive testing in controls limit generalizability. Age effects and broad composite scoring may obscure subtle deficits. Overall, the results highlight the interconnected nature of gas exchange, cerebral perfusion, and sleep disturbance in Long COVID-19, underscoring the need for larger, controlled studies with comprehensive brain and cognitive assessments to clarify mechanisms and inform targeted interventions.