Individuals diagnosed with obstructive sleep apnea (OSA) have higher volumes of white-matter hyperintensities (WMHs), according to a new analysis of data from the SHIP-Trend-0 cohort in Western Pomerania, Germany, which is part of the Study of Health In Pomerania. The association was true for individual measures of OSA, including apnea-hypopnea index (AHI) and oxygen desaturation index (ODI).
WMHs are often seen on MRI in older people and in patients with stroke or dementia, and they may be an indicator of cerebral small-vessel disease. They are linked to greater risk of abnormal gait, worsening balance, depression, cognitive decline, dementia, stroke, and death. Suggested mechanisms for harms from WMHs include ischemia, hypoxia, hypoperfusion, inflammation, and demyelination.
WMHs have been linked to vascular risk factors like smoking, diabetes, and hypertension. Brain pathology studies have found loss of myelin, axonal loss, and scarring close to WMHs.
Although a few studies have looked for associations between WMHs and OSA, they have yielded inconsistent results. The new work employed highly standardized data collection and more complete covariate adjustment. The results, published in JAMA Network Open, suggest a novel, and potentially treatable, pathological WMH mechanism, according to the authors.
“This is an important study. It has strong methodology. The automated analysis of WMH in a large population-based cohort helps to eliminate several biases that can occur in this type of assessment. The data analysis was massive, with adequate control of all potential confounders and testing for interactions. This generated robust results,” said Diego Z. Carvalho, MD, who was asked to comment on the findings. Dr. Carvalho is an assistant professor of neurology at the Center for Sleep Medicine at the Mayo Clinic, Rochester, Minn.
Worse apnea, worse hyperintensity
“The association varies according to the degree of apnea severity, so mild OSA is probably not associated with increased WMH, while severe OSA is mostly likely driving most of the associations,” said Dr. Carvalho.
If a causal mechanism were to be proven, it would “bring a stronger call for treatment of severe OSA patients, particularly those with increased risk for small-vessel disease, [such as] patients with metabolic syndrome. Likewise, patients with severe OSA would be the best candidates for therapeutic trials with [continuous positive airway pressure] with or without possible adjunctive neuroprotective treatment for halting or slowing down WMH progression,” said Dr. Carvalho.
Stuart McCarter, MD, who is an instructor of neurology at the Center for Sleep Medicine at the Mayo Clinic, Rochester, Minn., also found the results interesting but pointed out that much more work needs to be done. “While they found a relationship between OSA as well as OSA severity and WMH despite adjusting for other known confounders, it is unlikely that it is as simple as OSA is the main causal factor for WMH, given the complex relationship between OSA, hypertension, and metabolic syndrome. However, this data does highlight the importance of considering OSA in addition to other more traditional risk factors when considering modifiable risk factors for brain aging,” said Dr. McCarter. The study cohort was mostly of White European ancestry, so more work also needs to be done in other racial groups.
The study underlines the importance of screening among individuals with cognitive impairment. “If OSA represents a modifiable risk factor for WMH and associated cognitive decline, then it would represent one of the few potentially treatable etiologies, or at least contributors of cognitive impairment,” said Dr. McCarter.
The SHIP-Trend-0 cohort is drawn from adults in Western Pomerania. The researchers analyzed data from 529 patients who had WMH and for whom intracranial volume data were available. Each member of the cohort also underwent polysomnography.
Based on AHI criteria, 24% of the overall sample had mild OSA, 10% had moderate OSA, and 6% had severe OSA.
After adjustment for sex, age, intracranial volume, and body weight, WMH volume was associated with AHI (beta = 0.024; P < .001) and ODI (beta = 0.033; P < .001). WMH counts were also associated with AHI (beta = 0.008; P = .01) and ODI (beta = 0.011; P = .02).
The effect size increased with greater OSA severity, as measured by AHI for both WMH volume (beta = 0.312, 0.480, and 1.255 for mild, moderate, and severe OSA, respectively) and WMH count (beta = 0.129, 0.107, and 0.419). The ODI regression models showed similar associations for WMH volume (beta = 0.426, 1.030, and 1.130) and WMH count (beta = 0.141, 0.315, and 0.538).
Dr. Carvalho and Dr. McCarter disclosed no relevant financial relationships.
A version of this article first appeared on Medscape.com.