Jim Kling

December 10, 2024

07

Added to Email Alert

LOS ANGELES — Individuals with both uncontrolled epilepsy and sleep apnea are at significantly higher risk for mortality than those with epilepsy alone, according to a new analysis of over 2 million patient-years drawn from the Komodo Health Claims Database.

“A 10-year-old with uncontrolled central sleep apnea is about 200 times more likely to die than a general population 10-year-old. That’s comparable to a 10-year-old with [uncontrolled sleep apnea and] congestive heart failure. Noncentral sleep apnea is comparable to being paralyzed. It’s a huge risk factor,” said poster presenter Dan Lloyd, advanced analytics lead at UCB, which sponsored the research.

The ordering of sleep apnea tests for patients with epilepsy is widely variable, according to Stefanie Dedeurwaerdere, PhD, who is the innovation and value creation lead at UCB. “Some doctors do that as a general practice, and some don’t. There’s no coherency in the way these studies are requested for epilepsy patients. We want to create some awareness around this topic,” she said, and added that treatment of sleep apnea may improve epileptic seizures.

The findings were presented on December 7 at the American Epilepsy Society (AES) 78th Annual Meeting 2024.

The study included mortality rates between January 2018 and December 2022, with a total of 2,355,410 patient-years and 968,993 patients, with an age distribution of 19.1% age 1 to less than 18 years, 23.7% age 18-35 years, and 57.2% age 36 years or older. Sleep apnea prevalences were 0.7% for central sleep apnea (CSA), 14.0% for obstructive sleep apnea (OSA), and 85.3% with no sleep apnea.

Among those aged 1-18 years, the standardized mortality ratio (SMR) for those with uncontrolled epilepsy was 27.7. For those with comorbid OSA, the SMR was 74.2, and for comorbid CSA, the SMR was 135.9. The association was less pronounced in older groups, dipping to 7.0, 11.3, and 19.5 in those aged 18-35 years, and 3.3, 3.1, and 2.8 among those aged 36 years or older.

Among the 1-18 age group, SMRs for other comorbidities included 132.3 for congestive heart failure, 74.9 for hemiplegia/paraplegia, 55.3 for cerebrovascular disease, and 44.6 for chronic pulmonary disease.

Asked for comment, Gordon Buchanan, MD, PhD, welcomed the new work. “The results did not surprise me. I study sleep, epilepsy, and [sudden unexplained death in epilepsy (SUDEP)] in particular…and every time I speak on these topics, someone asks me about risk of SUDEP in patients with sleep apnea. It’s great to finally have some data,” said Buchanan. He is a professor of neurology at the University of Iowa, Iowa City, Iowa.

The authors found that patients undergoing continuous positive airway pressure (CPAP)/bi-level positive airway pressure therapy had a higher mortality risk than those not undergoing CPAP therapy but cautioned that uncontrolled confounders may be contributing to the effect.

Buchanan wondered if treatment with CPAP would be associated with a decreased mortality risk. “I think that would be interesting, but I know that, especially in children, it can be difficult to get them to remain compliant with CPAP. I think that would be interesting to know, if pushing harder to get the kids to comply with CPAP would reduce mortality,” he said.

The specific finding of heightened mortality associated with CSA is interesting, according to Buchanan. “We think of seizures propagating through the brain, maybe through direct synaptic connections or through spreading depolarization. So I think it would make sense that it would hit central regions that would then lead to sleep apnea,” he said.

The relationship between OSA and epilepsy is likely complex. Epilepsy medications and special diets may influence body composition, which could in turn affect the risk for OSA, as could medications associated with psychiatric comorbidities, according to Buchanan.

The study is retrospective and based on claims data. It does not prove causation, and claims data do not fully capture mortality, which may lead to conservative SMR estimates. The researchers did not control for socioeconomic status, treatment status, and other comorbidities or conditions.

Lloyd and Dedeurwaerdere are employees of UCB, which sponsored the study. Buchanan had no relevant financial disclosures.

0