11/20/2015

1

PSG/MSLT Testing in Children: From Protocols to Interpretation

Sponsored byThe Warren Alpert Medical School

of Brown University

Kiran Maski, MD

Dept of Neurology Center for Pediatric Sleep Medicine

Boston Children’s Hospital

Site PI for sodium oxybate trial in children with

narcolepsy sponsored by Jazz Pharmaceuticals, Inc.

Indications for MSLT

Highlight findings on PSG/MSLT commonly found

among children with narcolepsy

Review PSG/MSLT protocols for children

Review MSLT normative values for children

Discuss reliability of MSLT findings for narcolepsy

with and without cataplexy

CNS Hypersomnia*

Chronic

(≥3 month)

Primary

Narcolepsy

Type 1

(with cataplexy)

Type 2

(no cataplexy)

Idiopathic

Hypersomnia

Secondary

Medical

condition

Acquired Genetic/

Metabolic

Substance/

Medication

Psychiatric

Disorder

Recurrent or

Periodic

(once per 12-18

mo)

Kleine Levin

syndrome

Menstrual

Related KLS

A. Persistent lapses of daytime sleep occurring for at least 3 months

B. No symptoms of cataplexy

C. No increased REM pressure <2 Sleep Onset REM Periods on MSLT

Or if there is SOREMP on PSG, no SOREMP on MSLT

D. Hypersomnia Mean Sleep Latency ≤8 minutes OR

Actigraphy shows total sleep time is ≥660 minutes averaged over at least 7 days

OR 24 hour PSG shows >660 minutes of sleep

Other causes of hypersomnia (medical, medications)

ICSD version 3, 2014.

Chronic neurologic disorder characterized by the

following:◦ Excessive daytime sleepiness

Worse when sedentary

Automatic behavior

+/- irresistible sleep attacks

◦ Cataplexy

sudden loss of muscle tone triggered by strong emotions

◦ Hypnagogic and hypnopompic hallucinations

◦ Sleep paralysis

Median time to diagnosis has been reported

as 10.5 years!

11/20/2015

2

Narcolepsy Type I (Narcolepsy with Cataplexy)

•EDS for at least 3 months.

•At least one of the following:

– Cataplexy and a positive MSLT*

– Low CSF hypocretin-1 concentrations (≤110 pg/ml or <1/3 of normal)

Narcolepsy Type 2 (Narcolepsy without Cataplexy)

•EDS for at least 3 months.

•Positive MSLT*

* Positive MSLT : mean sleep latency of ≤8 minutes and≥2 SOREMP’s.**

** A SOREMP on the preceding nocturnal polysomnogram may replace one of the SOREMPs on the MSLT.

Disrupted nighttime sleep

-30-95% of patients

-Inability to stay asleep with frequent wakings, excessive shifts to N1or wake from deeper sleep stages

(Roth T et al. JCSM 2013)

Vivid dreams/Nightmares◦ -83% of patients report confusing dreams with reality vs

15% controls (p=10-10)

(Wamsley E et al. SLEEP 2014)

REM behavior disorder◦ Clinically reported frequency of 45-61% vs. 1.6% in

general population

◦ REM without atonia detected on PSG in 36-43% among

patients with narcolepsy ◦ Cipolli C et al. Sleep Med 2011; Franceschini C et al. Sleep Med 2011;

Dauvilliers Y et al. Sleep Med 2013.

Periodic limb movements◦ Index >5/hour (25% of patients)

◦ Index >15/hour (10% of patients)Sasai-Sakuma et al. PLOS One 2015

OSA

RBDDisorder

Obstructive Sleep Apnea◦ 25% of narcolepsy patients in one adult study had

AHI >10/hour

10/33 patients were diagnosed with OSA resulting in

delayed diagnosis of mean 6 years!

(Sansa G et al. 2010)

High rate of obesity at onset of disease symptoms

may contribute to OSA co-morbidity

Overnight PSG features in narcolepsy

Frequent periodic

limb movement

s

Frequent sleep stage transitions and high arousal index

May have co-morbid

SDB

Sleep Onset REM Period

(<15 minutes)

Nocturnal Sleep onset REM period has 48% sensitivity and 97% specificity in children with symptoms of narcolepsy with cataplexy

Reiter J et al. SLEEP 2015

11/20/2015

3

This is a series of four to five “opportunities” for the

patient to sleep at a 2 hour interval each nap. The first

nap should start 1.5 to 3 hours after awakening the

patient from the PSG.

Measure of tendency to fall asleep in absence of

externalizing factors

Sleep latency is an objective measure of sleepiness

Results will be influenced by sleep disturbance, insufficient

amount of sleep, OSA, medications, circadian factors

It is strongly recommended that adequate sleep be

documented by sleep log and whenever possible

actigraphy for 1-2 weeks prior to PSG/MSLT

Participants are encouraged to sleep as much as

possible during the week and during the night prior to

the MSLT

PSG recommended immediately prior to MSLT

◦ Rule out underlying sleep disorder

◦ Ensure adequate sleep duration

◦ AASM guidelines state <6 hours of sleep can reduce

next day sleep latency timeLittner MR et al. SLEEP 2005;

ICSD 3rd ed

MSL and actigraphy measured sleep: r=0.43, p=0.002

MSL and self-report measured sleep: r=-0.05, p=0.7

MSL and sleep log measured sleep: r=0.001, p=0.99

Bradshaw DA et al. JCSM 2007

• Wake promoting agents, Stimulants, stimulant-like

medications, and REM suppressing medications

should be ideally stopped 2 weeks before the

MSLT.

-clonidine, stimulants, SSRI, SNRIs

(including buproprion), antipsychotics

• Use of the patient’s other usual medications (anti-

hypertensives, insulin, etc.) should be thoughtfully

planned by the sleep clinician before MSLT testing

Littner et al. SLEEP 2005; Rye DB Depress Anx 1998

Drug screening may be indicated to ensure that

sleepiness on the MSLT is not pharmacologically not

induced

◦ Generally performed on day of MSLT

◦ Timing and circumstances may be modified by

clinician

◦ Smoking stopped 30 min prior to each nap opportunity

11/20/2015

4

214 MSLTs conducted 1998-2013, during which

screening for drugs of abuse (cocaine, PCP,

marijuana, opiates)◦ All negative

189 of these patients had GC/MS:◦ 5% positive for SSRI

◦ 4% positive for OTC (i.e. diphenhydramine, ibuprofen,

pseudoephedrine)

◦ 31% positive for caffeine

1 case of caffeine positive patient with MSLT 12.2 min, 2

SOREMP who met definition of narcolepsy when screen

negative 4 weeks laterKatz E, Maski KM, Jenkins A. JCSM 2014

The morning after the polysomnogram, the

technician is to remove the

following:

– Leg leads

– Respiratory belts

– Intercostal EMGs

– Thermocouple

– Oximeter

– Watch/Alarm clock

Littner et al. SLEEP 2005

The patient is to change into their daytime clothes.◦ Hoodie to cover leads specified

It is important that the technician NOT wake the patient out of the last REM period of the night because the patient may experience a REM rebound on Nap #1.

Patients should be asked if they need to go to thebathroom prior to the start of each nap opportunity.

Sleep technologists who perform the MSLT should be experienced in conducting the test.

The montage should entail the following:

• 4 EEG leads (2 Central: C3,4 and 2 Occipital: O1,O2)

• 2 Referential leads (M1 and M2)

• 2 EOG leads (ROC/M1, LOC/M2)*

• 3 EMG leads (mentalis and 2 submentalis)

• EKG lead

* RUE, RLE (Right Upper Eye and Right Lower

Eye) could be used to pick up more questionable Rapid

Eye Movements

• Sleep rooms should be quiet and dark during testing.

• Room temperature should be set based on patient’s

comfort level.

• Vigorous physical activity should be avoided during the

day and any stimulating activities by the patient should

end at least 15 minutes prior to each nap opportunity.

• Patients must abstain from any caffeinated beverages

and avoid unusual exposures to bright sunlight.

• A light breakfast is recommended at least 1 hour

prior to the first nap.

• A light lunch is recommended immediately after the

termination of the second nap.

11/20/2015

5

The Patient calibrations should be performed prior

to commencing each nap.◦ Eye open 30 seconds

◦ Close both eyes 30 seconds

◦ Look all directions

◦ Blink eyes 5 times

◦ Grit teeth

Impedances should always be checked prior to

each nap as well.

Instruct the patient to “lie quietly and do whatever comes naturally.”

• Turn off lights and start the test.

• If the patient does not sleep, you are to run the nap for 20 minutes or 40 epochs.

If during this 20-minute time frame the patient falls asleep, you are to run the nap for 15 minutes or 30 epochs from the point where the patient fell asleep.

◦ 15 seconds of cumulative sleep marks sleep onset

• If the patient gets into REM during this 15-minute timeframe, you are to end the nap at the end of the 15thminute of sleep (clock time after sleep onset).• On the Technician Report Form, you are to compute theSleep and REM Latencies (if applicable). If the patientdoes not sleep, you are to put as a Sleep Latency time of20 minutes.• To compute Sleep Latency, you would calculate “SleepOnset Epoch – Lights Out Epoch” then divide by 2.

• To compute REM Latency you would calculate “REMOnset Epoch – Sleep Onset Epoch ” then divide by 2.

• At the end of each nap, have the patient complete the Post-Nap Questionnaire

Disconnect the jack box and inform the patient that he must stay out of the bed and remain awake until the next nap commences.

• To calculate for the Mean Sleep Latency, you would add up all of the Sleep Latencies then divide by the number

of naps.

Did you sleep? Y/N

How long did you sleep?

Did you dream? Y/N

How long did it take you to fall asleep?

11/20/2015

6

• Lights Out at 9:00 am

• If no sleep, you would end the nap at: (1)_____

• Sleep onset occurred at 9:06 am

• Sleep latency is: (2)_____ minutes

• Lights On at: (3)_____

• Lights Out at 9:00 am

• If no sleep, you would end the nap at: 9:20 am

• If Sleep onset occurred at 9:06 am

• Sleep latency is: 6 minutes

• Lights On at: 9:21 am

• Lights Out at: (1)_____ am

• Sleep onset occurred at 11:03 am

• Sleep latency was: (2)_____ minutes

• Wake observed at 11:04 am

• Lights On at: (3)_____ am

• Lights Out at: 11:00 am

• Sleep onset occurred at 11:03 am

• Sleep latency was: 3 minutes

• Lights On at: 11: 18 am

• Lights Out at: (1)_____ pm

• Sleep onset occurred at 1:00 pm

• Sleep latency was: (2)_____ minutes

• REM onset occurred at 1:02 pm

• REM latency was: (3)_____ minutes

• Lights On at: (4)_____ pm

• Lights Out at: 1:00 pm

• Sleep onset occurred at 1:00 pm

• Sleep latency was: 0 minutes

• REM onset occurred at 1:02 pm

• REM latency was: 2 minutes

• Lights On at: 1:15 pm

11/20/2015

7

• Lights Out at: (1)_____ pm

• Sleep was not observed

• Sleep latency would be: (2)_____ minutes

• Lights On at: (3)_____ pm

• Lights Out at: 3:00 pm

• Sleep was not observed

• Sleep latency would be: 20 minutes

• Lights On at: 3:20 pm

• Lights Out at epoch 350

• Sleep onset occurred at epoch 355

• Sleep latency was: (1)_____ minutes

• REM onset occurred at epoch 365

• REM latency was: (2)_____ minutes

• Lights On at epoch: (3)_____

• Lights Out at epoch 350

• Sleep onset occurred at epoch 355

• Sleep latency was: 2.5 minutes

• REM onset occurred at epoch 365

• REM latency was: 5 minutes

• Lights On at epoch: 385

(1) Sleep Latency is computed using the formula:

Sleep Onset Epoch (355) - Lights Out Epoch (350) ÷ 2

(to convert epoch to minutes)

(2) REM Latency is computed using the formula:

REM Onset Epoch (365) – Sleep Onset Epoch (355) ÷ 2

(to convert epoch to minutes)

(3) Lights On will also be after 30 epochs or 15 minutes

from sleep onset.

Example

Nap #1: 6.0 minutes

Nap #2: 3.0 minutes

Nap #3: 0.0 minutes

Nap #4: 20.0 minutes

Nap #5: 2.5 minutes

31.5 minutes ÷ 5 = 6.3 minutes

11/20/2015

8

Normative Data for MSLTs

Normative values not available in children <6

years

Pre-pubertal data: 18 children ages 8-12 years

with 2 consecutive PSG done in home and MSLT

in lab PSG◦ MSLT (30 min naps; 4 nap protocol at 10 am, 12 pm, 2

pm, 4 pm)

MSL 10 min or less on at least 2 naps

Daily MSL <20 min

Palm L et al. SLEEP 1989

-PSG: SOL 8.2-13.5 minutes; REM latency 118-159 min; TST

540-581 min

-MSLT:

- 5 participants did not fall asleep at all

-13 participants fell asleep at least once (avg MSL 1

trial=16.2 min; minimum MSL 9 min); Most after lunch

-2 of 18 participants fell asleep in all 5 naps (average

MSL=22.8 min)

Pre-puberty, children have less tendency to fall

asleep during daytime

-Daily MSL of 20 min or less on 30 min nap

protocol is rarely found

-No SOREMPs

Carskadon MA et al. SLEEP 1980

Carskadon MA. 1982. The second decade. Guilleminault C, ed. Sleeping and

Waking Disorders: Indications and Techniques. Menlo Park, CA: Addison-Wesley.

Pp. 99–125.

11/20/2015

9

Multiple SOREMPs in 16% of 10th grade adolescents

(notably in morning naps) due to delayed sleep phase

syndrome and insufficient sleep

Carskadon MA SLEEP 1998

5% of pediatric patients with OSA had 2 SOREMPs on

MSLT Chervin RD and Aldrich M. AJRCCM 2000

◦ Test-rest reliability in 14 healthy normal subjects

between 2 MSLT sessions 4-14 months apart was

highly reliable (r=0.97, p<0.001). Zwyghuizen A, Roth T et al. Sleep. 1998

◦ Test-retest reliability of MSLT among patients with

narcolepsy type 1 was excellent (r=0.81, p<0.01) but

re-testing 5 days apart. Folkerts M, Roth T et al. Biol Psych 1996

Retrospective database study evaluating MSLT test-retest reliability among patients with IH (n=13), narcolepsy without cataplexy (n=7)and physiologic hypersomnia (n=16)

Trotti LM et al. J. Clinical Sleep Med. 2013.

-On repeat testing, only 47% of patients retained their original diagnosis (5 of 7 narcolepsy patients kept diagnosis) -42% (n=15) crossed 8 minute MSLT threshold and 31% (n=11) had >2 SOREMS at repeat testing Trotti LM, et al. 2013

Is drug screening necessary for protocol?◦ Order as needed

◦ Counsel on caffeine avoidance 2 days prior to testing

Different protocols for pre-pubertal children?◦ 30 minutes naps, 4 nap protocols

Poor reliability of MSLT also seen in children?◦ Effects of sleep restriction more evident in post-

pubertal children

Should we be requiring actigraphy prior to testing in this population