"Do I Have Narcolepsy?" Lawrence Martin

Lawrence Martin, M.D., FACP, FCCP

Associate Clinical Professor of Medicine
Case Western Reserve University School of Medicine, Cleveland
Board Certified in Pulmonary and Sleep Medicine

Sleep Home Page / Internet Sleep Links / Books about sleep disorders (with links to Amazon.com) / Drugs for Sleep & Awake / Index of Dr. Martin's web sites

Here's a little quiz. One or more statements may be correct.
People with narcolepsy often:

a) fall asleep during the day at inappropriate times

b) have episodes of muscle weakness brought on by emotions such as laughing

c) have nightmares just when they start to fall asleep.

d) have feelings of paralysis during sleep

Most people associate narcolepsy with "sleep attacks" or "falling asleep at inappropriate times." But that is just one aspect of narcolepsy. There can be associated nightmares at the start of sleep and episodes of Sleep paralysis. Even more common are sudden episodes of muscle weakness while awake, called 'cataplexy', which may herald a sleep attack. The answers are all correct.

Conversely, just because someone falls asleep at inappropriate times (answer a) doesn't mean he or she has narcolepsy. The most common causes of excessive daytime sleepiness, in fact, do not include narcolepsy. Far more common are simple sleep deprivation and obstructive sleep apnea.

Here are complaints from five patients. Which one likely has narcolepsy?

  1. Mary T., 31, housewife: "Doctor, I get so tired during the day I can't do the house work. I just have to sit down and then I fall alseep."
  2. John R., 38, truck driver. "I get so tired on the road I have to pull over and go to sleep. If I don't I'm afraid I'll crash the rig."
  3. Billy W., 44, advertising executive. "I need to drink 3 cups of strong coffee every morning before our sales meeting - It's all I can do to fight falling asleep - and I'm the team leader."
  4. Melinda W., 21, graduate student. "[My date] was telling us something about his job...and next thing I know he is standing over me, saying 'Melinda, wake up, are you all right?' "
  5. Reginald M., retired, age 70. "I've always fallen asleep easily during the day. Now my wife says its getting worse and asked me to see you."

The question is a bit sneaky, because there is simply no way to answer it without some testing in a sleep lab. Which is my point. Falling asleep during the day, no matter how inappropriate it may seem, is not equal to narcolepsy. No more than, say, all chest pain is heart attack or all cough and fever is pneumonia. There are many reasons people fall asleep when they don't want to.

Most of the time - including 4 of the 5 patients just presented - the problem is not narcolepsy, but something else. Melinda has narcolepsy.  (Diagnoses for patients without epilepsy is provided at the end.)

* * *

Melinda was only 21 at the time of her first visit, one of my younger patients (I don't see pediatric patients). At five feet three and 150 lbs., she was a bit on the heavy side but attractive.  She did not come across as sleepy or depressed, had no facial tics and, most important, made solid eye contact.

Like any field of medicine, certain sleep conditions cluster around certain age groups. The most common sleep disorder (after insomnia and snoring) is sleep apnea. Sleep apnea is usually found in men and women past 40; they are typically overweight and have heavy snoring. The younger and thinner the patient, the less likely will a sleep disorder relate to sleep apnea.

Her chief complaint was: "I fall asleep all the time." Melinda was a graduate student in psychology, unmarried. She told me she fell asleep in class on more than one occasion, that she fell asleep once driving and hit a parked car, fortunately without any personal injury. The incident which prompted the visit was her falling asleep on a date.  Here is her tale. The questions are mine.

"We were out at Nightown [a popular Cleveland restaurant], with another couple. Sitting at a table, waiting for our food."

"What time was it?"

"About 8 o'clock."

"And you were not tired going into the restaurant?"

"No, not at all. I had gotten a good night's sleep the night before."


"It's kind of a noisy place, and everyone around us was talking and laughing. Howard [her date] was telling us something about his job, about his boss actually, and it was funny. So we were laughing. Next thing I know Carol [woman sitting next to her] is shaking me and Howard is standing over me, saying 'Melinda, wake up, are you all right?' I never felt more embarrassed in my life."

"What did you do?"

"I excused myself, went to the bathroom and washed my face."

"Then what?"

"I went back to the table, told them I was fine, and we finished supper."

"How'd the rest of the evening go?"

"Fine. No more episodes."

"This happened when?"

"A week ago today."

"Did Howard or the other couple say what they saw? I mean, did they describe how you looked before falling asleep?"

"Not then, not in the restaurant. Later, when we were alone, Howard did say I looked kind of funny for a moment, then just nodded off. He thought I might be having a stroke! [Here, Melinda gave a short, nervous laugh.] I assured him I was OK, but then decided I should see someone about this."

"Any you were referred...by?"

"Oh, a friend gave me your name. Said you treated her father for 'sleep apnea'.

More history revealed that she had had the problem for about two and a half years. I asked why it took so long to see someone professionally.

"I don't know. At first I thought it was just because I didn't get enough sleep in college."

Which is true enough. College students get on average one to two hours of sleep below their daily need. They are notorious for falling asleep in class, or wherever. So in a college environment, Melinda didn't think too much about her "sleep attacks" as she called them.

"Can you describe another one?"

"She thought for a moment." Once, in a statistics class, I remember taking notes. It was a pretty boring class and the last of the day. I was leaving town right after class and was kind of anxious to get out. I remember stretching in my seat just before the hour was to end, and next thing I knew my head was on my shoulders and people were walking past my seat. A couple of classmates reached over and sort of gently woke me up. Seeing someone sleeping in this class is no big deal of course, except I don't remember being tired. I do remember thinking very clearly beforehand about my plans to leave town, so it seemed strange to just, like you know, fall asleep."

I asked her about sleep paralysis. This is a frightening symptom considered to be an abnormal episode of awakening from REM sleep. You awaken and find yourself paralyzed: unable to move any part of your body for up to several minutes (though it feels like an eternity). Sleep paralysis usually occurs upon falling asleep or just before waking up. My patient thinks she had one such episode, around the age of 16, but couldn't be sure. She does not recall any "hypnagogic hallucinations," the dream-like auditory or visual hallucinations that sometimes occur while dozing or falling asleep.

We talked about her general health, which was otherwise excellent. I asked her if she is still dating Howard. Yes, she said, and volunteered they were all but living together.

Her only medication was birth control pills and occasional Tylenol. Overall, there didn't seem to be any psychiatric or medical problems. A physically healthy, bright young woman with a very specific complaint. "Sleep attacks." We counted 8-10 such attacks in the previous 18 months, with more than half unrelated (as best she could tell) to lack of sleep the night before. Midway through her history she asked me the inevitable question.

"Do you think I have narcolepsy?"

"Yes, it's a good possibility, from your description. We'll need to schedule a couple of sleep studies to find out. What do you know about it?"

"I've read a little bit on the internet. It's not that common."

"Right, maybe 1 in 2000 people have it. Your description fits the condition, but it's really diagnosed by some sophisticated testing.  You need two different sleep tests. First, an overnight sleep study, followed in the morning by a daytime study to see how quickly you fall asleep during nap periods.  You come to the sleep lab (in the hospital) about 7 p.m. and leave the next day around 4 p.m." I spent another few minutes explaining the tests she needed (see table).



Test called "Polysomnogram" or PSG Test called "Multiple Sleep Latency Test"
Length: bedtime to wake up time Length:  4 or 5 separate naps beginning the morning after the PSG, usually spread out from 9 am to 3 or 5 pm
Used for diagnosing:  several sleep problems, mainly sleep apnea Used for diagnosing: mainly narcolepsy

Given her schedule and the sleep lab's backlog, it would take several weeks before we could get the studies done. In the meantime I asked her to keep a detailed sleep diary, recording both night time sleep (time to bed, time awake) and any daytime naps, intended or not.

I also asked her to see a neurologist as I am a sleep specialist with background in pulmonary diseases. Narcolepsy -- if that was the diagnosis -- is a brain disorder, and I didn't want to diagnose and treat it without a neurology specialist involved.

* * *

Like almost everything else about sleep medicine, the reliable diagnosis of narcolepsy is relatively new, dating from the 1970s. The First International Symposium on Narcolepsy was held in 1975, and only then was a clear consensus arrived at about definition and diagnosis. As part of the definition, it was concluded that narcoleptics should manifest early REM sleep during a controlled study.

Stanford University has been a pioneer in narcolepsy research. We now know narcolepsy is a frequent disorder. It is the second leading cause of excessive daytime sleepiness diagnosed by sleep centers after obstructive sleep apnea. The actual prevalence varies, from about 0.2 to 1.6 per thousand in European countries, Japan and the United States; this is about the same frequency as Multiple Sclerosis. There are several organizations that offer aid and information about narcolepsy.

REM (rapid eye movement) sleep, is where we do most of our dreaming. If you never enter REM sleep you will likely not dream very much. REM sleep occurs periodically throughout the night, typically first appearing 60 to 90 minutes after the onset of sleep. It then reappears every 90 minutes or so, and by the time normal sleep is over the person will have spent about 20-25% of sleep time in REM sleep.

Normally, REM is associated with loss of muscle tone and an 'active brain'. It is where dreaming usually occurs. Strange things can also happen in REM, including nightmares, 'fighting' movements and other abnormal behavior, and sleep paralysis. Fuseli's famous painting The Nightmare (1781) is widely viewed as a depiction of sleep paralysis occurring during REM.

Narcoleptics typically have early onset REM after going to sleep (i.e, well before 60 minutes) and -- this is most important -- can enter into REM sleep without warning DURING THE DAY. The 'sleep attacks' are in fact intrusions of REM sleep during the day, while doing ordinary activities.

But narcolepsy is not just sleep attacks. Another feature, present in about 75% of narcoleptics, is a strange affliction called catalplexy. Cataplexy is a sudden loss of muscle tone, often in a limb or one side of the face, that heralds the sleep attack. Typically, it could be a facial drooping, which is what I suspect Melinda's boyfriend saw in her at the restaurant. During cataplexy the person is awake, so it is not a "sleep attack."


A sudden muscular weakness that can afflict some narcoleptics. It is triggered by emotions, typically laughter or surprise. During attacks people are not unconscious: hearing and awareness remain intact. Cataplexy can manifest as:

  • Arm weakness
  • Sagging jaw
  • Drooping head
  • Slumping of the shoulders
  • Slurred speech
  • Generalized weakness
  • Knee buckling

Cataplexy of Narcolepsy -- Videos from You Tube:
Child with cataplexy
Adolescent with cataplexy
Woman with cataplexy

Generally, narcoleptics sleep fairly well. That is, night time sleep is usually not disturbed, and they wake up refreshed. But daytime sleepiness or cataplexy can come any time. Such 'sleep attacks' are often brought out by emotional outbursts, typically laughing. Why or how laughter triggers a narcoleptic episode is unknown. Melinda's restaurant attack likely was triggered by laughter.

Even with a history as good as Melinda's, you need both sleep studies (PSG and MSLT) for reliable diagnosis. The overnight PSG is to check for any type of traditional sleep disturbance, and to make sure there has been sufficient sleep before the daytime study begins. (Unless the subject sleeps well the night before, the daytime nap study is of no value. No point in seeing how long it takes someone to nap if they are already sleep deprived.)

For the nighttime PSG lights went out at 10:35 and Melinda was awakened at 6:15 am. During this almost 8-hour period the lab recorded EEG, oxygen saturation, body movements and air flow into and out of her throat. After awakening Melinda took a shower, went to breakfast and made some phone calls -- all while still in the hospital -- then returned to the lab at 8:30 am for the next phase, the Multiple Sleep Latency Test (MSLT).

Basically, MSLT measures how quickly the subject naps during the day, when lights are turned out. For the MSLT she was hooked up to some of the wires again and told what to expect.

(A related test, less frequently performed than MSLT, is the Maintenance of Wakefulness Test (MWT). This is a series of nap studies during the day in which the subject is asked to stay awake and not fall asleep. How well they can perform that task relates to their degree of daytime sleepiness. MWT is complementary to the MSLT, but it is usually not necessary when the latter gives diagnostic information.)

Here is what we tell patients having the MSLT: "We're going to ask you to take 4 or 5 naps: 9 am, 11 am 1 pm and 3 pm and, if necessary, 5 pm (5th nap needed if diagnosis is equivocal after just 4). Each nap will last about a half hour or less. After you fall asleep we'll wake you up. We are checking for two things, how long it takes you to fall asleep and what type of sleep you enter."

Doing the MSLT at the right time is every bit as important as how it's done. You don't want to test anyone sleep deprived, or anyone taking drugs that can interfere with sleep onset. Some labs actually check the patient's urine for drugs to make sure there is no interfering medication.  

The drug list is long. Sedatives, hypnotics, antihistamines, stimulants (including caffeine) and alcohol -- all should be withdrawn at least two weeks before the study (or else be prepared to interpret the study in light of possible drug effects).

The MSLT gives two types of information:

1. How long it takes the subject to enter into sleep.

From the time you lay down in bed, with lights out, until the first stage of sleep (determined by the EEG), is called "sleep latency." After a good night's sleep ending, say, at 7 a.m., how long before you would fall asleep around 9 in the morning (quiet room, lights out)? Normally, sleep latency is not less than 10 minutes. Since each nap trial lasts between 20-30 minutes, normal is simply any time above 10 minutes (including no sleep at all). Sleep latency between 5 and 10 minutes is considered mildly abnormal; less than 5 minutes is definitely abnormal. In practice, an average sleep latency of the four naps is taken, and the results can be viewed as follows:

Average time to fall asleep during an MSLT nap period.

>10 minutes (or no sleep) - normal

5-10 minutes - mildly to moderately abnormal. May require re-testing or treatment if the person has severe daytime sleepiness symptoms.

<5 minutes - definitely abnormal, and indicating excessive hypersomnia. Need to look for the cause (if not already evident from the overnight sleep study)

Before you decide to test yourself at home (Hey, let's do an MSLT!!) realize that sleep is defined ONLY by the EEG or brain wave test. You can't tell if someone is in the early or first stage of sleep by looking at them (at least, not reliably). While the patient is under visual observation via a video monitor, "going to sleep" is determined only by the brain wave, and the criteria are strict. The same is true for the second aspect of the test - entrance into REM sleep.

2. Whether or not the subject entered into REM sleep

Falling asleep early in one of the daytime naps only diagnoses excessive sleepiness, not narcolepsy. For narcolepsy to be confirmed, the patient should have early onset of REM sleep. Specifically, he or she should enter into REM sleep ("latency to REM") within 5 minutes after first falling asleep, and this should occur in at least two of the four/five naps.

Normally, you would never enter REM within 5 minutes of falling asleep after a good night's sleep. Remember, normal REM first occurs about 60 to 90 minutes after sleep onset, and then periodically thereafter. After a good night's sleep, REM occurrence in a brief nap period indicates a state of abnormal brain activity. Even one early-REM nap is abnormal, but to be sure of the diagnosis the standard has been accepted as two or more early REM periods out of four or five naps.

* * *

Melinda's neurologist agreed with my assessment and the intended sleep studies. He only wanted one other study, a brain CT scan, to make sure there was no lurking tumor. He arranged for the test and sent me the report (normal).

About a month after her visit she had the overnight polysomnogram, which was followed the next day by the MSLT. The polysomnogram showed no significant abnormalities.  A mountain of computerized data could be summarized as follows:

She fell asleep within 5 minutes after lights out (normal sleep latency) and entered into her first REM period 67 minutes later (normal). She had no breathing problems during sleep and maintained good heart and lung function all night. Her sleep efficiency -- the percent of time in bed actually spent sleeping -- was normal at 92%.

The normal night time study made her an excellent candidate for the following day's MSLT. Here are the MSLT results:

Patient: Melinda W., Age 21

Thursday, October 19

Nap 1 - lights out 9:24 a.m., sleep onset 7 minutes later, slept total of 27 minutes before awakened, no REM

Nap 2 - Lights out 11:23: a.m., sleep onset 3.5 minutes later, REM onset at 6.5 minutes (after 10 minutes of non-REM sleep). She woke up spontaneously after 16.5 minutes of sleep.

(She went to lunch at noon in the hospital cafeteria, returned around 1:15 p.m.)

Nap 3 - Lights out 1:45 p.m., sleep onset 2 minutes later, REM onset at 3.5 minutes (5.5 minutes after sleep onset). Technician awakened her after 29 minutes of total sleep.

Nap 4 - Lights out 3:15 p.m., sleep onset 3.5 minutes, REM onset 8 minutes (11.5 minutes after sleep onset). Technician awakened her after 25 minutes of total

Nap 5 - Not needed, since she had 2 early onset REM periods after 4 naps

Sleep Averages:
Sleep latency: 4 minutes -- excessively fast
REM latency: 6 minutes in 3 naps -- consistent with the diagnosis of narcolepsy.

Using sophisticated tools such as the polysomnogram and MSLT, it has become much easier to diagnose narcolepsy than a generation ago. But diagnosis would be to little avail if there was no treatment. The therapeutic aspect has greatly brightened in recent years.

I saw Melinda a week after the tests were completed. I felt secure about the diagnosis and had my prescription pad ready. But I knew accurate diagnosis and even 'the right drug' were not enough for a life long affliction. It's not like treating a simple infection, but more like treating diabetes -- the patient must accept ups and downs of therapy, be prepared to try new medications or different doses of the same drug, and "live with the affliction." I had no doubt Melinda was up to the task, but was a little taken aback by her first question of me after I explained the findings.

"Why me?"

What she meant (I thought) was 'Is it genetic, did I inherit this disease, did I catch it in childhood?' and so forth. I explained what we know about narcolepsy (not genetic, though it does run in some families, for unclear reasons), and switched over to the positive side of things.

"It's treatable."

Before 1998 various stimulants were all that were available, amphetamines being the most notorious. All had fair to poor results with high risk of side effects. In December 1998 the FDA approved the first new drug specifically for narcolepsy, Provigil (modafanil), which is now a mainstay of therapy. Provigil stimulates the brain to stay awake and works a large percentage of time to ameliorate the daytime sleep attacks of narcolepsy. The usual dose is 200 mg in the morning, one hour before or after breakfast. The most common side effects are headache, anxiety and insomnia.

A newer, but much less used therapy, for narcolepsy with cataplexy is gamma hydroxybutyrate, or GHB. The trade name is Xyrem.  Though Xyrem (sodium oxybate) is approved by the FDA for narcolepsy with cataplexy, it has a rather notorious history -- as the "date rape drug". When mixed with flunitrazepam, GHB induces a state of euphoria and the recipient (usually a woman) is prone to submit to an unwanted advance and then not remember anything about the affair.

FDA approval of GHB requires that the drug be tightly controlled.  It is dispensed from only a single source in the U.S. and it's use is closely monitored.  Through biochemical pathways too complicated to worry with here, GHB reduces cataplexy that often heralds or accompanies daytime sleep attacks.

GHB was not available when Melinda came for her follow up, and so I prescribed only Provigil, at 200 mg/day. A month later she reported feeling about the same, but admitted there had been no sleep attacks in that period. A year later she had what she thought was "maybe one attack, I'm not sure." I decided it was time to repeat her MSLT. Here are results of the second study.

Patient: Melinda W.

Age 22

Tuesday, November 6

Nap 1 - lights out 9:12 a.m., sleep onset 9 minutes later, slept total of 24 minutes before awakened, no REM

Nap 2 - Lights out 11:18: a.m., sleep onset 7 minutes later, no REM. Slept total of 18 minutes before awakened.

Nap 3 - Lights out 1:22 p.m., sleep onset 6.5 minutes later, REM onset at 9 minutes (13.5 minutes after sleep onset). Awakened her after 28 minutes of total sleep.

Nap 4 - Lights out 3:04 p.m.. Did not sleep.

Nap 5 - Lights out 4:45 p.m.  Sleep onset 12 minutes later, no REM.  Awakened after another 15 minutes.

Sleep Averages:
Sleep latency: 8.6 minutes over 4 naps -- mildly fast
REM latency: 9 minutes in one nap

Improvement in the MSLT confirmed her clinical improvement, and I attributed it to Provigil. Absent any further sleep attacks (or if they are very infrequent), she probably will not need GHB.

Melinda's case is actually unusual for being:  straightforward, diagnosed early, and relatively benign. It is estimated that the large majority of narcoleptic patients in the U.S. are still undiagnosed. Often years go by before the right tests are done and the diagnosis made. In one study there was an average of 14 years before onset of first symptoms and diagnosis! One result of delay in diagnosis, in all too many cases, is psychosocial impairment and disability. As far as I could tell Melinda seemed well adjusted, without these problems. She never asked, and I never suggested, psychological help for coping with her affliction. Nor did she suffer physical harm from cataplexy. Narcoleptics are at great risk of injury from cataplexic attacks (e.g., falling and head injury).

Will she be on Provigil forever? I don't know. I have actually turned her care over to the neurologist consulted, and I no longer follow her. But there is exciting research that promises new drug therapy. A defective gene for narcolepsy has been found in mice and dogs, and probably also exists in humans. The gene is called hypocretin receptor 2, which encodes a protein that sits on nerve cells. Almost all patients with narcolepsy and cataplexy have a deficiency of brain hypocretin (another name for this protein is orexin). Work is underway to develop hypocretin (orexin) replacement therapy for this group of patients.

Diagnoses for the 5 patients:

  1. Mary T. - Overwork:  Two young children, up all night, exhausted by morning.
  2. John R. - Sleep apnea and obesity:  Frequent apnea episodes during the night leads to not getting enough deep sleep and feeling tired the next day
  3. Billy W. - Over medication for depression
  4. Melinda W. - Narcolepsy
  5. Reginald - Sedentary life style, lack of stimulation.

Sleep Home Page/ Internet Sleep Links/ Books about sleep disorders (with links to Amazon.com) / Index of Dr. Martin's web sites
Forward any comments to:

Copyright © 2009, Lawrence Martin, M.D.