Pediatric Pulmonology 49:252–256 (2014)

Prevalence of Pediatric Periodic Leg Movements ofSleep After Initiation of PAP Therapy

Vidya Pai, MD,1,2 Umakanth Khatwa, MD,2,3 Sriram Ramgopal, MD,1 Kanwaljit Singh, MD, MPH,4

Rebecca Fitzgerald, RPSGT,2 and Sanjeev V. Kothare, MD1,2*

Abstract. Introduction: Positive-airway-pressure (PAP) is a treatment for obstructive-sleep-apnea

(OSA). In adults, initiation of PAP-therapy may unmask periodic-limbmovements-of-sleep (PLMS).

Wepresent a series of children inwhomPLMSwasaggravatedor induced following initiation ofPAP

therapy. Methods: Retrospective analysis was performed on children who had PAP-studies

performed over a 3-year period. Children with OSA without significant PLMS on baseline studies

and who then showed a periodic-limb-movement-index (PLMI) (>5/hr) during titration with PAP-

therapy were identified. Charts were reviewed for demographics, OSA severity based on apnea

hypopnea index (AHI), PLMI, PLM-arousal-index (PLMAI) and pressures titrated. Results: Two

hundred fourteen PAP-titration studies (151 on continuous positive airway pressure [CPAP] and 63

on bilevel positive airway pressure [BiPAP]) were done. Eleven (10 on CPAP, 1 on BiPAP) met study

criteria. Eight patients were boys. The median age was 12 years (range 6 months–18 years). On

baseline studies,medianAHIwas 5/hr (range 1.5–32/hr),medianPLMIwas 1/hr (range 0–4/hr) and

median PLMAI 0.3/Hr (range 0–2/hr). On titration studies, pressures ranged from 4 to 14 cm of

water, median AHI was 0.5/hr (range 0–1), median PLMI was 12/hr (range 5–55/hr) and median

PLMAI 9/hr (range 0–25). PLMS were seen predominantly during N1, N2 sleep and on PAP of

>7 cm of water. One patient had resolution of PLMS at 1-year follow-up. Conclusion: PLMS were

seen in 5.1% of children after initiation of PAP. OSA may mask PLMS, which appears as breathing

improves. Alternatively, PAP may unmask or induce PLMS. PLMS may potentially be a cause of

treatment failure in childrenon recent-onset PAP therapy. PediatrPulmonol. 2014; 49:252–256.

� 2013 Wiley Periodicals, Inc.

Key words: polysomnography; PLMS; PAP-therapy; children.

Funding source: None reported.

INTRODUCTION

Periodic limb movements of sleep (PLMS) are a well-characterized phenomenon characterized by stereotypicalmovements of the lower extremities. PLMS were initiallydescribed as “nocturnalmyoclonus” andwere thought to bea type of sleep-related epilepsy.While PLMShas long been

known to occur in children, though less frequently than inadults, thereportsof thisconditionoccurring in thepediatricpopulation have only been published in the last twodecades.1 The prevalence of PLMD, according to a recentretrospective review, is estimated to be 14% in children.2

The presence of PLMS in relation to sleep-disorderedbreathing, such as obstructive or central sleep apnea, has

1Department of Neurology, Boston Children’s Hospital, Harvard Medical

School, Boston, Massachusetts.

2Center for Pediatric Sleep Disorders, Boston Children’s Hospital, Harvard

Medical School, Boston, Massachusetts.

3Division of Respiratory Diseases, Department of Medicine, Center for

Pediatric Sleep Disorders, Boston Children’s Hospital, Harvard Medical

School, Boston, Massachusetts.

4Lurie Center, Massachusetts General Hospital for Children, Boston,

Massachusetts.

Disclosures: Vidya Pai, Sriram Ramgopal, Umakanth Khatwa, Kanwaljit

Singh, and Rebecca Fitzgerald have nothing to disclose.

Sanjeev Kothare is funded by the following grants: National Institutes of

Health: 1 RC1 HL099749-01 (R21), and RFA-HL-09-001, and Investigator

Initiated Grant from Eisai Pharma, Inc, to assess safety and efficacy of

rufinamide in children, and the Harvard Catalyst to assess cardiorespiratory

abnormalities during seizures in children.

�Correspondence to: Sanjeev V. Kothare, MD, Fegan 9, 300 Longwood

Avenue, Children’s Hospital Boston, Boston, MA 02115.

E-mail: sanjeevkothare@hotmail.com

Received 30 October 2012; Accepted 27 February 2013.

DOI 10.1002/ppul.22802

Published online 1 July 2013 in Wiley Online Library

(wileyonlinelibrary.com).

� 2013 Wiley Periodicals, Inc.

been noted by several investigators. In a large study with1,124 adults evaluated for sleep disordered breathing,PLMS were found to be significantly associated withobjective sleepiness.3 Curiously, in a few reported cases ofadults, PLMS were noted to occur following the initiationof positive airway pressure (PAP) therapy, including in apreviously published series in which 19/81 adults in whomPLMS increased following the initiation of CPAP.4

To date, this phenomenon has not been reported inchildren. In this report, we document a series of childrendiagnosed for obstructive sleep apnea (OSA) and whodeveloped PLMS on the follow-up study during titrationwith PAP therapy.

METHODS

Patient Selection

We reviewed charts of patients with sleep studies bothprior to and following initiation of PAP therapy. Patientswho met all of the following criteria were included in thestudy population:

(1) Presence of mild to severe obstructive sleep apnea on baseline

PSG with an apnea hypopnea index (AHI) greater than 1.5/hr.5

(2) Improvement in OSA following CPAP titration as reflected by

an overall AHI <1.5/hr and less than the baseline AHI.

(3) Periodic limb movement index (PLMI) of <5/hr on baseline

(pre-treatment) studies and >5/hr on subsequent PAP titration

studies.

A total of 214 PAP titration studies were retrospectivelyreviewed between the months of January 2008 toDecember 2011. Prior to data review, approval wassought and obtained by the Institutional Review Board.Informed consent was not deemed necessary for thisretrospective study.

Sleep Study, Sleep Parameters andPolysomnographic Diagnosis

All patients in the study population attended anovernight PSG. Sleep studies were carried out usingdigital PSG equipment (Sleep Scan Vision 2 polysom-nography, Biologic SystemCorporation, Mundellein, IL).All studies were performed using simultaneous pressuretransducer and thermistor for nasal flow, piezoelectricchest and abdominal belts, end tidal (ET) CO2 monitor-ing, pulse oximetry, snore microphone, two leg channels,chin EMG, extended EEG montage (standard EEG withadditional temporal coverage), EKG, and simultaneousvideo monitoring.6

Detailed sleep parameters were assessed, includingtotal sleep time (TST), sleep and rapid eye movement(REM) latency, sleep efficiency, REM and slow wavesleep (SWS) percentages, arousal index, periodic limbmovement index, periodic limb movement arousals,

oxygen saturation nadir, apnea-hypopnea index (AHI)and peak ETCO2.

6

The PSG diagnosis of the patients was based on thefollowing definitions: obstructive apnea was defined ascessation of airflow for two respiratory cycles, with�90%reduction in airflow for �90% of the event with ongoingrespiratory effort; hypopnea as decrease in amplitude ofairflow signal by>50% lasting two respiratory cycles andaccompanied by oxygen desaturation of �3% or anarousal. Central events were scored when there was norespiratory effort and the event lasted 20 s or lasted twobreaths and had an arousal, awakening or at least 3%oxygen desaturation. Periodic Limb Movements of Sleep(PLMS) were defined as repetitive jerking of the legsduring sleep �5 times per hour, 0.5–5 s in duration, inclusters of four or more separated by 5–90 s.7,8 Legmovements that occurred as part of general bodymovements in association with EEG arousals followingobvious apneas or hypopneas were not scored as PLMS.9

The PAP titration was begun at 4 cm of water and wasadjusted in increments of 1 cm of water until apneas,hypopneas and snoring were eliminated and isolatedelectroencephalographic arousals were minimized.

Data Analysis

AHI, arousal index, PLM index, O2 saturation wereanalyzed. To adjust for the within-subject correlation onaccount of the pre- and post-PAP/PSG nature of the data, apaired t-test was used for analysis. Scatter plots and boxplots were also constructed and superimposed using aSASmacro program. SAS v. 9.3 (SAS, Inc, Cary, NC)wasused for statistical analysis.

RESULTS

Study Population

Fourteen patients who had PLMI > 5/hr were identi-fied, of which 11 (10 on CPAP and 1 on BiPAP) had dataavailable from both a pre- and post-PAP therapy PSG. Themedian age was 12 years (range 6 months to 18 years).Eight patients in the series were boys. The median ferritinlevel among the five patients with available data was 47.2(range 28.9–344.8; Table 1).

Pre-PAP Studies

From the baseline studies, the median AHI was foundto be 5/hr (range 1.5–32/hr), the median PLMI was 1/hr(range 0–4/hr) and the median PLMAI was 0.3/hr (range0–2/hr).

Post-PAP Studies

All of the children in the series had PAP pressuresranging from 4 to 14 cm of water (cw). After PAP

PLMS in Children on PAP-Therapy 253

Pediatric Pulmonology

titration, the median AHI was reduced to 0.5/hr (range 0–0.5/hr; P ¼ 0.0276), the median PLMI increased to 12/hr(range 5–55/hr; P ¼ 0.004) and the median PLMAI 9/hr(range 0–25; P ¼ 0.0256; Fig. 1).

PLMS Characteristics

PLMSwere seen in 5.1%of children following initiationof PAP therapy (Fig. 1). PLMS were seen predominantlyduring NREM N1 and N2 sleep and were observed toappear in the second half of PAP titration as breathing andobstruction improved (Fig. 2). Additionally, PLMS weremainly seen on PAP pressures >7 cm of water, althoughtherapeutic pressures ranged from4 to 14 cmofwater. Onepatient had a follow-up PAP therapy 1 year later, whodemonstrated complete resolution of PLMS.

DISCUSSION

Summary of Most Important Results

We investigated the emergence of PLMS after initiationof PAP therapy in children with diagnosed OSA. Wefound that 5.1% of children were seen to have PLMS oninitiation of PAP therapy. PLMS occurred more frequent-ly in the latter half of the night and occurred despite anadequate titration of PAP therapy. This finding has notbeen previously reported in a pediatric population.

Pathophysiology of PLMS

Dopaminergic dysfunction is thought to mediatePLMS. Activation of the sympathetic nervous systemfrom the stress of breathing against positive pressure ordisruption of sleepmay lead to limbmovements.10,11 Thismay occur despite the fact that OSA itself causesdysfunction of autonomic function in children whichmay be relieved by treatment.12 Evidence for this theory isderived from observations from adults with chroniccardiovascular conditions and congestive heart failure.13

Additional evidence for this theory is derived fromfindings that periodic leg movements during sleep areassociated with rise in heart rate which are greater thanthose associated with leg movements during wakeful-ness.14 Younger individuals may demonstrate a moredramatic swing in autonomic response associated withPLMS, potentially predisposing them to even greater risksof cardiovascular and neurovascular complications. Thelack of reversal of the metabolic complications associatedwith OSA may be related to the underlying persistence ofsympathetic over-activity induced by the PLMS.

PLMS Following PAP Therapy

Avariety of theories have been put forth to explain theonset of PLMS following onset of PAP therapy. PAP may

TABLE 1—Comparison of Polysomnographic Variables in the Case Series Both Pre- and Post-PAP Therapy

Parameter

Pre-PAP Post-PAP

PRange Median Range Median

AHI 1.5–3.2 2.5 0–0.5 0.5 0.0276

PLMI 0–4 1 5–55 12 0.004

AI 6–26 15 0–14 5 0.12

PLMAI 0–1.2 0.3 0–25 9 0.025

O2 Sat % 78–99 87 88–99 97 <0.001

TST 148–476 433 339–477 409

REM latency 29–244 96 59–387 114.5

SWS % 18–50% 33% 25–65% 32%

REM % 6–44% 20% 3–42% 22%

Peak ET CO2 37–60 50 39–52 47

Data expressed as ranges and median values. The increase in PLMI post PAP therapy is statistically significant P < 0.05. AHI, Apnea Hypopnea

Index; PLMI, Periodic Limb Movement Index; AI, Arousal Index; PLMAI, Periodic Limb Movement Arousal Index; TST, total sleep time; SWS

%, percentage of slow wave sleep; REM %, percentage of REM sleep; ET CO2, end tidal carbon dioxide.

Fig. 1. Superimposed scatter and box plots of periodic limbmovement index (PLMI) with positive airway pressure (PAP)status. The X-axis shows increasing PAP pressures on studynight. The Y-axis is the corresponding increase in PLMI withincreasing PAP pressures. Prior to initiation of CPAP, themedian PLMI was 1. Following optimization of PAP therapy,PLMI rose in this series to a median of 12/hr.

254 Pai et al.

Pediatric Pulmonology

induce PLMS, or alternatively, severe sleep fragmentationmay mask the presence of PLMS, which then becomesapparent when the obstruction is adequately controlled onPAP therapy.15 Another hypothesis suggests that residualupper airway resistance syndrome (UARS) due toincomplete treatment by PAP may trigger PLMS.16

PLMS has been thought to increase post-PAP therapy inmoderate to severe OSA due to an unmasking effect.Conversely, in mild OSA, PAP therapy may actuallyameliorate PLMS. This has given rise to the idea thatPLMSmay therefore have more than one etiology or maybe categorized as spontaneous or induced.17

LIMITATIONS

Our findings need to be interpreted in the appropriateclinical context. This was a retrospective series withlimited numbers. Long-term follow-up data was notavailable. As demonstrated in some adult series, testingover multiple nights may be necessary to truly identifyPLMS.18,19 Not all children in the series had ferritin levelsand follow-up data were limited. We did not assess

daytime sleepiness or fatigue in our patient population orassess symptoms of restless legs syndrome (RLS), sleep-fragmentation or family history of RLS or PLMD. Ideally,a comparison group of children with OSA treated withadenotonsillectomy would also have been included,whichwould help to differentiatewhether OSA resolutionunmasks PLMS, or whether PAP therapy alone inducesPLMS. This control group was not available. Appropriatescreening questionnaires need to be developed toprospectively assess these symptoms in the pediatricpopulation.20

CONCLUSION

Clinicians should be aware that aggravated or de-novoPLMS may be a possible cause for a child with OSA onPAP therapy to have continued sleep disruption, daytimefatigue and behavioral disturbances despite adequatecontrol of the underlying sleep apnea. In order to fullymanage and treat this condition, more data are needed tounderstand the frequency and severity of this issue, inaddition to identifying the causes of these conditions and

Fig. 2. Hypnogram showing that with increasing positive airway pressure (PAP, dashed line) ontitration night periodic limb movements were seen to progressively increase (solid line). Inaddition, legmovementswere seenmore inN1 andN2 sleep stages.W,wake; R, REMsleep; 1, N1sleep; 2, N2 Sleep; 3, N3 sleep.

PLMS in Children on PAP-Therapy 255

Pediatric Pulmonology

potential treatment approaches. It would also be ofinterest to identify children with PLMS prior to initiationof PAP therapy to investigate whether the conditionworsens following the initiation of therapy. Potentialavenues of research include the performance of prospec-tive studies utilizing multi-night PSG to assess extent ofPLMS on PAP therapy, assessment of night and daytimesymptoms as a result of the PLMS, and investigation ofcomorbidities, such as attention deficit hyperactivitydisorder both in patients and in family histories. It wouldalso be of value to study changes in PLMI according tochanges in PAP titration. Studies with ferritin levels andpossible trial with oral iron in children with low ferritinlevels should also be prospectively studied, along withinvestigations of the effect of habituation of PAP therapyon the PLMS.

REFERENCES

1. Walters AS, Picchietti DL, Ehrenberg BL, Wagner ML. Restless

legs syndrome in childhood and adolescence. Pediatr Neurol

1994;11:241–245.

2. Picchietti DL, Walters AS. Moderate to severe periodic limb

movement disorder in childhood and adolescence. Sleep 1999;22:

297–300.

3. Chervin RD. Periodic leg movements and sleepiness in patients

evaluated for sleep-disordered breathing. Am J Respir Crit Care

Med 2001;164:1454–1458.

4. Seo WH, Guilleminault C. Periodic leg movement, nasal CPAP,

and expiratory muscles. Chest 2012;142:111–118.

5. Witmans MB, Keens TG, Davidson Ward SL, Marcus CL.

Obstructive hypopneas in children and adolescents: normal values.

Am J Respir Crit Care Med 2003;168:1540.

6. Bokkala S, Napalinga K, Pinninti N, Carvalho KS, Valencia I,

Legido A, Kothare SV. Correlates of periodic limb movements

of sleep in the pediatric population. Pediatr Neurol 2008;39:

33–39.

7. International classification of sleep disorders. Diagnostic and

coding manual. Winchester, IL: American Academy of Sleep

Medicine; 2005.

8. Recording and scoring leg movements. The Atlas Task Force.

Sleep 1993;16:748–759.

9. Schenck CH, Mahowald MW. Motor dyscontrol in narcolepsy:

rapid-eye-movement (REM) sleep without atonia and REM sleep

behavior disorder. Ann Neurol 1992;32:3–10.

10. Ware JC, Blumoff R, Pittard JT. Peripheral vasoconstriction in

patients with sleep related periodic leg movements. Sleep 1988;

11:182–186.

11. McCall WV, Haponik E. Relationship of nasal continuous positive

airway pressure to periodic limb movement disorder in a patient

without sleep apnea. Chest 1993;103:1609–1611.

12. Muzumdar HV, Sin S, Nikova M, Gates G, Kim D, Arens R.

Changes in heart rate variability after adenotonsillectomy in

children with obstructive sleep apnea. Chest 2011;139:1050–1059.

13. Hornyak M, Trenkwalder C. Restless legs syndrome and periodic

limb movement disorder in the elderly. J Psychosom Res 2004;56:

543–548.

14. Winkelman JW. The evoked heart rate response to periodic leg

movements of sleep. Sleep 1999;22:575–580.

15. Fry JM, DiPhillipoMA, PressmanMR. Periodic legmovements in

sleep following treatment of obstructive sleep apnea with nasal

continuous positive airway pressure. Chest 1989;96:89–91.

16. Baran AS, Richert AC, Douglass AB, May W, Ansarin K. Change

in periodic limb movement index during treatment of obstructive

sleep apnea with continuous positive airway pressure. Sleep

2003;26:717–720.

17. Carelli G, Krieger J, Calvi-Gries F, Macher JP. Periodic limb

movements and obstructive sleep apneas before and after continuous

positive airway pressure treatment. J Sleep Res 1999;8:211–216.

18. Sforza E, Johannes M, Claudio B. The PAM-RL ambulatory

device for detection of periodic leg movements: a validation study.

Sleep Med 2005;6:407–413.

19. Morrish E, King MA, Pilsworth SN, Shneerson JM, Smith IE.

Periodic limb movement in a community population detected by a

new actigraphy technique. Sleep Med 2002;3:489–495.

20. Chervin RD, Hedger KM. Clinical prediction of periodic leg

movements during sleep in children. Sleep Med 2001;2:501–510.

256 Pai et al.

Pediatric Pulmonology