Scientists analyze the EEG characteristics of patients with narcolepsy (with and without cataplexy) and idiopathic hypersomnia compared to controls.
Alyssa Cairns, PhD, a research scientist at SleepMed Inc, talks about her latest sleep research and how there’s a crucial need for developing a better way of identifying patients with narcolepsy that would hasten the diagnostic process.
Sleep Review (SR): What prompted you to evaluate the microarchitecture of the nighttime polysomnograms of different patient groups, including narcolepsy patients with and without cataplexy, patients with hypersomnia, and patients without any sleep disorders?
Alyssa Cairns (AC), PhD: Essentially we know that there’s a historic delay in the diagnosis of narcolepsy of about 10 years. This means that from symptom onset to the time the person is typically diagnosed and hopefully treated for narcolepsy is about 10 years. It can be sometimes a little sooner and sometimes a lot later, and it can be 20 to 30 years by the time they’re diagnosed.
The reason is because narcolepsy is a pretty heterogeneous brain disorder. So some patients have cataplexy and, in combination with excessive daytime sleepiness, that’s what we know as narcolepsy with cataplexy. We have another group of patients with narcolepsy that don’t have that muscle characteristic, and we call that narcolepsy without cataplexy. Those people tend to have a particularly long diagnostic delay, longer than those with narcolepsy with cataplexy, because their symptoms may not be as obvious. Without the obvious muscle atonia, physicians and other medical practitioners might suspect other underlying causes for their sleepiness like a major depressive disorder.
We wanted to explore if there was a better way of identifying these patients with narcolepsy with and without cataplexy that would help us hasten the diagnostic process so we don’t have to wait until either muscle atonia develops or they have such severe presentation of narcolepsy that they end up in a car accident or develop some other neurological condition or reduced quality of life to the point they have to be on disability. We tried to explore some other ways to identify narcolepsy via the nocturnal polysomnogram, which often occurs prior to being evaluated for hypersomnia, for example when getting evaluated for sleep apnea. We also wanted to compare the sleeping brain’s characteristics of patients with something called “idiopathic hypersomnia” compared to narcolepsy and matched controls. Idiopathic hypersomnia occurs when a person is inexplicably sleepy, but fails to meet the criteria for narcolepsy. Controls were clinic-based patients that were not sleepy and who had fairly “normal” EEG [electroencephalogram] characteristics.
The patients with idiopathic hypersomnia are like the group of patients with narcolepsy without cataplexy, but the caveat is that patients with hypersomnia basically have a default diagnosis of such because their MSLT [multiple sleep latency test]—the current diagnostic procedure for narcolepsy—failed to show they had met the criteria for narcolepsy.
Again, this study was conducted really to understand the nocturnal microarchitecture of the different groups of people with different hypersomnias versus matched controls, which were clinic-based controls who’d have fairly “normal” EEGs characteristics. The controls consisted of people who weren’t sleepy and were confirmed to not have anything that would disrupt their sleep architecture.
We looked at the EEG characteristics of narcolepsy patients with and without cataplexy and also the EEG characteristics of that intermediate group with hypersomnia. What we also know that’s very important to understand is a lot of patients with hypersomnia end up getting diagnosed eventually with narcolepsy and vice versa. If you test a patient who has narcolepsy without cataplexy and then a patient with hypersomnia later on with a MSLT, their diagnosis would theoretically flip flop about half of the time.
We’re looking for better ways to identify these individuals by using the advanced signal processing measure, which we hope is more empirical than the MSLT, which we know as a tool for diagnosing patients without cataplexy is fairly imperfect. Hence, there’s this diagnostic delay often because the patients are diagnosed first with hypersomnia because their MSLT failed to confirm narcolepsy. Our main objective was to better understand the underlying EEG data of these different groups of people to see if we can identify unique fingerprints that would help us then apply that algorithm, later on, to get these patients diagnosed quicker.
SR: Tell us more about the 50 cases from each diagnostic group that you examined. What really stood out to you as you were examining these different groups?
AC: This is the first time a study like this had been done. There really wasn’t any data out there on comparing each of these groups together. What we found was that there was significant heterogeneity in the groups, especially in patients with narcolepsy with cataplexy. Some patients had really disrupted sleep architecture, which is consistent with what we know of narcolepsy with cataplexy. We found some patients that fit well within that model but then some patients had fairly normal sleep architecture, which was kind of fascinating in itself. In other words, narcolepsy with cataplexy doesn’t necessarily need the MSLT to diagnose them; this is a physician-diagnosed phenomenon.
“There was significant heterogeneity in the groups, especially in patients with narcolepsy with cataplexy.”
It was interesting to find that so many patients had remarkably normal architecture, and then we had some patients with really disrupted architecture. We also saw quite a bit of heterogeneity in the narcolepsy without cataplexy group, with some patients having normal sleep architectures and others having a sleep architecture that resembled the ones of narcolepsy patients who had cataplexy. That reinforces our understanding that perhaps some of these diagnostic buckets we use might be arbitrary and there is some diagnostic overlap between the groups.
In a perfect world, the next time we do a study like this we definitely need to have a larger sample size in order to see if there is a better way to compartmentalize the fingerprints into a more meaningful explanation for why certain people have certain fingerprints.
SR: What are the key takeaways from your study that you think are important to how we understand the problem of delayed narcolepsy diagnoses?
AC: Essentially the inability to maintain sleep and to maintain wakefulness is probably one of the biggest things that contribute to their overwhelming sleepiness during the day. Because if you’re not able to have continuous sleep and if you’re not able to transition through the sleep stages normally, then you’re not going to feel restored during the day and you’re going to have lapses into sleep, which is what we’re finding with this data. What we found with patients who have hypersomnia is they didn’t seem to have as profound inability to maintain sleep as the narcolepsy groups, but instead were more likely to have nonrestorative sleep because the background of their sleep was composed of faster wave activity. So that is potentially an explanatory reason for why they feel so sleepy during the day.
I think the upshot is we need more data with larger sample sizes and we need to have a better understanding of how we should categorize patients based on the various presentations of sleepiness they exhibit. Ultimately there needs to be more data that will help us to understand nocturnal polysomnograms to see if we can help hasten the diagnostic process for these patients.
SleepMed acquires Novasom’s activity
SleepMed, one of the largest operators of sleep labs in the United States, has acquired Novasom’s assets, a leading provider of Home Sleep Testing (HST). Each service will continue to operate and the teams will work together to deliver the best care experience. Please follow this link to read the press release
Drowsy Drivers Cause More Accidents Than You Think!
Detecting drowsiness following a vehicle crash is difficult, which makes drowsy driving an underreported traffic safety issue. New research from the AAA Foundation for Traffic Safety provides an unprecedented analysis of in-vehicle dashcam video from more than 700 crashes—and the study found the percentage of crashes involving drowsiness to be nearly 8 times higher than federal estimates indicate