Beyond AHI: What New Research Says About Sleep Apnea and Brain Health

By Dr. Chelsie Rohrscheib, Ph.D. — Last reviewed: May 15, 2026 · 5 min read

Key Takeaways

The apnea–hypopnea index (AHI) — the standard metric for diagnosing sleep apnea — measures how often breathing pauses occur, not how severe each pause is.

A new 7-Tesla MRI study found that T90 (the time spent with oxygen saturation below 90%) was significantly associated with brain injury in older adults, while AHI showed no meaningful relationship.

This points to a broader shift in sleep medicine toward physiologically meaningful metrics like hypoxic burden, which may better reflect cardiovascular and neurological risk.

AHI remains important, but a single-night, single-number diagnosis may miss meaningful physiological variability — making multi-night, oxygen-aware assessment more relevant.

Obstructive sleep apnea (OSA) is typically defined by a single number: the apnea–hypopnea index (AHI). This metric reflects how many times per hour a person’s breathing stops or slows during sleep, and it has long served as the foundation for diagnosis, severity classification, and treatment decisions [1]. While AHI has provided a useful framework for standardizing care, it was never designed to fully capture the biological complexity of sleep apnea. Increasingly, research suggests that it may not tell the whole story.

A recent study published in Sleep offers an important perspective on this issue by examining how sleep apnea affects the brain [2]. The researchers focused on the locus coeruleus, a small structure in the brainstem that plays a critical role in attention, memory, and wakefulness [3]. This region is also one of the earliest sites affected in Alzheimer’s disease, making it a particularly important target for understanding how sleep apnea may contribute to long-term neurological risk [4].

What the 7-Tesla MRI Study Found

Using ultra-high-resolution 7-Tesla MRI [5], the investigators assessed the structural integrity of the locus coeruleus in older adults and compared it to several commonly used sleep metrics. What they found was both surprising and informative. Traditional measures such as AHI, total sleep time, and sleep efficiency showed little to no relationship with brain structure. In contrast, a measure known as T90 — the cumulative time a person spends with oxygen saturation below 90% during sleep [6] — was significantly associated with reduced integrity of the locus coeruleus. In other words, it was not how often breathing events occurred that mattered most, but how much physiological stress those events placed on the brain.

This distinction is clinically meaningful. AHI treats all respiratory events equally, regardless of their duration or severity. However, two individuals with the same AHI can have very different oxygen profiles [7]. One may experience brief, mild disruptions with minimal desaturation, while another may spend prolonged periods in low oxygen states. From a physiological standpoint, these are fundamentally different conditions, and this study suggests that the brain may respond accordingly.

The findings are further strengthened by the study’s analytical approach. After accounting for age, sex, and body mass index, T90 explained a significant proportion of the variability in brain structure, while AHI did not meaningfully contribute. This supports the idea that oxygen deprivation may be a more relevant driver of neural injury than event frequency alone.

Important Caveats

At the same time, the study should be interpreted within its context. The sample size was relatively small, and the analysis was cross-sectional, meaning it cannot establish causality. It remains unclear whether hypoxia directly leads to structural changes in the brain or whether other underlying factors contribute to both. Additionally, the study focused on older adults without cognitive impairment, and the findings may not generalize to other populations. Sleep apnea is also a multifaceted condition involving not only hypoxia but also sleep fragmentation and changes in ventilatory control, which were not fully explored in this analysis [8].

A Broader Shift Toward Hypoxic Burden

Even with these limitations, the study aligns with a broader shift in sleep medicine toward more physiologically meaningful metrics. Rather than relying solely on event counts, there is increasing recognition that measures such as T90 and hypoxic burden may better reflect the true impact of the disease, particularly in relation to cardiovascular and neurological outcomes [9].

This raises an important question for clinical practice: if the most relevant signals are not fully captured by traditional metrics, are we measuring sleep apnea in the most effective way? Current diagnostic approaches are often based on a single night of testing, summarized into a small number of indices. However, sleep apnea is not a static condition. Oxygen levels, breathing patterns, and sleep architecture can vary significantly from night to night, influenced by factors such as body position, sleep stage, and environmental conditions [10]. A single snapshot may not adequately represent the patient’s typical physiological burden.

What Better Measurement Looks Like

As the field continues to evolve, there is growing interest in approaches that allow for more comprehensive and longitudinal assessment. Home sleep testing technologies, particularly those capable of capturing detailed physiological signals across multiple nights, offer a way to better understand how sleep apnea behaves in real-world settings. This type of monitoring may be especially valuable as clinicians begin to consider metrics beyond AHI when evaluating disease severity and treatment response.

Wesper Lab is designed with this shift in mind. By enabling multi-night assessment of respiratory patterns and oxygenation, it provides a view of sleep apnea physiology that single-night studies cannot. This can help clinicians identify patients with high hypoxic burden, supporting more individualized management strategies over time.

A More Complete Picture

The study from Kam and colleagues does not suggest that AHI should be abandoned. Rather, it highlights that AHI alone may not be sufficient. As our understanding of sleep apnea deepens, so too must our approach to measuring it. Sleep apnea is not defined solely by how often breathing stops. It is defined by how the body responds, and how much stress those disruptions place on critical systems, including the brain. Moving forward, aligning our metrics with these underlying biological processes may be key to improving both diagnosis and long-term outcomes.

Frequently Asked Questions

What is the AHI in sleep apnea?

The apnea–hypopnea index (AHI) is the number of times per hour a person’s breathing pauses or significantly slows during sleep. It is the primary metric used to diagnose sleep apnea and classify its severity, but it does not capture how severe each individual breathing event is.

What is T90 and why does it matter?

T90 is the cumulative time during sleep that a person’s oxygen saturation falls below 90%. It reflects the total burden of low oxygen — or nocturnal hypoxia — that the body experiences overnight. Research increasingly shows T90 may correlate with cardiovascular and brain-health risks more strongly than AHI.

Can sleep apnea damage the brain?

Untreated obstructive sleep apnea is associated with increased risk of cognitive impairment and neurological injury. A 2026 study using 7-Tesla MRI found that older adults with higher nocturnal hypoxia had reduced structural integrity of the locus coeruleus, a brain region involved in attention, memory, and wakefulness.

Is the AHI being replaced?

Not entirely. The AHI remains a clinically important and widely used metric. But increasingly, researchers and clinicians are evaluating additional physiological measures — such as hypoxic burden and T90 — alongside the AHI to better capture the real impact of sleep apnea on the body.

Why does multi-night testing matter for sleep apnea?

Sleep apnea severity varies night to night based on body position, alcohol intake, sleep stage, and other factors. A single-night study may not capture a patient’s typical physiological burden. Multi-night home sleep testing can reveal patterns that single-night studies miss.

What is the locus coeruleus?

The locus coeruleus is a small structure in the brainstem that regulates attention, memory, and wakefulness. It is also one of the earliest brain regions affected in Alzheimer’s disease, making it a key area of interest in understanding how sleep apnea may contribute to long-term neurological risk.

References

Berry RB, Quan SF, Abreu AR, et al. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications. Version 3. Darien (IL): American Academy of Sleep Medicine; 2023.
Kam K, Gaggi NL, Parekh A, et al. High obstructive sleep apnea hypoxic burden associates with reduced locus coeruleus structural integrity on 7T MRI in older adults. Sleep. 2026 Mar 11;49(3):zsaf398. doi: 10.1093/sleep/zsaf398.
Osorio-Forero A, Cherrad N, Banterle L, Fernandez LMJ, Lüthi A. When the Locus Coeruleus Speaks Up in Sleep: Recent Insights, Emerging Perspectives. Int J Mol Sci. 2022 Apr 30;23(9):5028.
Chen Y, Chen T, Hou R. Locus coeruleus in the pathogenesis of Alzheimer’s disease: A systematic review. Alzheimers Dement (N Y). 2022 Mar 7;8(1):e12257.
de Vries E, Hagbohm C, Ouellette R, Granberg T. Clinical 7 Tesla magnetic resonance imaging: Impact and patient value in neurological disorders. J Intern Med. 2025 Mar;297(3):244-261.
Wang L, Wei DH, Zhang J, Cao J. Time Under 90% Oxygen Saturation and Systemic Hypertension in Patients with Obstructive Sleep Apnea Syndrome. Nat Sci Sleep. 2022 Nov 30;14:2123-2132.
Parekh A. Hypoxic burden — definitions, pathophysiological concepts, methods of evaluation, and clinical relevance. Curr Opin Pulm Med. 2024 Nov 1;30(6):600-606.
Eckert DJ, Malhotra A. Pathophysiology of adult obstructive sleep apnea. Proc Am Thorac Soc. 2008 Feb 15;5(2):144-53.
Azarbarzin A, Sands SA, Stone KL, et al. The hypoxic burden of sleep apnoea predicts cardiovascular disease-related mortality. Eur Heart J. 2019 Apr 7;40(14):1149-1157.
Dzierzewski JM, Dautovich ND, Rybarczyk B, Taylor SA. Night-to-night fluctuations in sleep apnea severity: diagnostic and treatment implications. J Clin Sleep Med. 2020 Apr 15;16(4):539-544.

Author: Dr. Chelsie Rohrscheib, Ph.D., Wesper.

Back to Articles and Case Studies