DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

httpwwwneurologyorgcontent6361065fullhtmllocated on the World Wide Web at

The online version of this article along with updated information and services is

0028-3878 Online ISSN 1526-632Xsince 1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN

reg is the official journal of the American Academy of Neurology Published continuouslyNeurology

Restless legs syndrome and pregnancyM Manconi MD V Govoni MD A De Vito MD NT Economou MD E Cesnik MD I Casetta MD

G Mollica MD L Ferini-Strambi MD and E Granieri MD

AbstractmdashObjective To perform a large and detailed epidemiologic study on restless legs syndrome (RLS) duringpregnancy and the puerperium Methods A structured clinical interview assessing symptoms since the beginning ofpregnancy was performed to a population of 642 pregnant women at the time of delivery and at follow-up evaluation (1 3and 6 months after delivery) Main hematologic tests were also evaluated A woman was considered affected if she met theInternational RLS Study Group criteria for RLS diagnosis Results Twenty-six percent of women were affected by RLSduring their pregnancy The disease was strongly related to the third trimester of pregnancy and tended to disappearreaching the time of delivery Affected women presented lower values of hemoglobin and mean corpuscular volumecompared with healthy subjects (both groups received the same supplemental iron and folate therapy) ConclusionsPregnancy is associated with transient restless legs syndrome

NEUROLOGY 2004631065ndash1069

The diagnostic criteria for restless legs syndrome(RLS) defined in 1995 by an International commit-tee and recently revised are the following 1) desireto move the limbs usually associated with paresthe-siasdysesthesias 2) motor restlessness 3) worsen-ing of symptoms at rest with at least partial andtemporary relief by activity and 4) worsening ofsymptoms in the evening or night12 Based on thepresence of associated conditions such as iron defi-ciency uremia peripheral neuropathy and preg-nancy RLS could be classified as idiopathic orsymptomatic34 The idiopathic RLS form may be in-herited with an autosomic dominant transmission5-7

The RLS prevalence ranges between 5 and 10 andincreases with age8 The pathogenesis of RLS is stillunknown but there is evidence that supports thehypothesis of a dopaminergic neurotransmissionweakness in these patients3910

The association between RLS and pregnancy wasnoted first in 1940 reported again in Ekbomrsquos workand confirmed later by a few epidemiologicinvestigations11-13 A recent cross-sectional surveyusing an autoadministered questionnaire on a widesample of Japanese pregnant women found an RLSprevalence rate of 19914 No epidemiologic studiesinvestigated the relationship between RLS and preg-nancy using the four standard diagnostic criteria

Methods Subjects and data collection Six hundred forty-twoconsecutive pregnant women admitted to the Department of Gy-necology and Obstetrics of Ferrara University (Italy) from Febru-ary 2000 to June 2002 were contacted Six hundred twenty-sixagreed to participate Women affected by pre-eclampsia or eclamp-sia severe ankle edema and diseases known to cause RLS wereexcluded No women were receiving drug therapy except for sup-

plemental iron and folate Six hundred six women were includedin the study the most frequent cause of exclusion was drug addic-tion (in six women) followed by diabetes mellitus (in five women)Two neurologists certified in sleep medicine interviewed thewomen within 2 days after parturition each Tuesday and FridayThe interview lasted ~20 minutes and consisted of a structuredquestionnaire that included the following information demo-graphic data personal and family medical history course of preg-nancy mother and newborn anthropometric data iron and folatetherapy sleep habits and presence of sleep disorders A detaileddescription of RLS symptoms if present during and before preg-nancy was also evaluated Main hematologic and urinary testswere analyzed within 24 hours before delivery (to not be influ-enced by the parturition hemorrhage) Laboratory data of 14women were not considered because they were outside the afore-mentioned period All the women found affected by RLS under-went a telephone follow-up evaluation performed by the sameinterviewer at the end of the first third and sixth month afterdelivery A woman was considered affected by RLS if she met allfour International RLS Study Group criteria1 A minimal fre-quency of occurrence of RLS symptoms that is not included in thestandard criteria was established and defined by the authors Awoman was considered affected by RLS if the frequency of symp-toms was at least three times in the same month or four times intwo consecutive months All women included in this study signedan informed consent form approved by the local ethical committee

Subgroup classification and statistical analysis According tothe aforementioned criteria women were classified in two groupshealthy group (HG) and total RLS group (tRLS) The tRLS groupwas subclassified into two subgroups the pre-existing RLS group(pRLS) which included subjects who already experienced RLSsymptoms in their medical history before the current pregnancybut during periods of no pregnancy and the new RLS group(nRLS) which included subjects who experienced RLS symptomsfor the first time during the current pregnancy (and possibly alsoduring previous ones)

Continuous variables were compared with the Studentrsquos t-testUnivariate analysis of categorical variables was performed usingthe 2 test Multivariate logistic regression was used to investi-gate the independent effect of each putative risk factor SPSSprogram (Chicago IL) for Windows 98 was used for statisticalanalysis

From the Section of Neurological Clinic (Drs Govoni De Vito Economou Cesnik Casetta and Granieri) and Section of Gynecology and Obstetrics (DrMollica) Ferrara University Italy Sleep Disorders Center Department of Neurology (Drs Manconi and Ferini-Strambi) Vita-Salute University IRCCSHS Raffaele Milan ItalySupported by the Italian Ministry of University and Research (MIUR) Preliminary results have been presented at the European Sleep Research Societymeeting (Reykjavik 2002) with a grant by the International Restless Legs Syndrome Study GroupReceived February 3 2004 Accepted in final form May 21 2004Address correspondence and reprint requests to Dr Mauro Manconi Sleep Disorders Center Department of Neurology Vita-Salute University IRCCS SanRaffaele Hospital Via Stamira drsquoAncona 20 20127 Milan Italy e-mail manconimaurohsrit

Copyright copy 2004 by AAN Enterprises Inc 1065

Results The population of 606 women had a mean ageof 318 47 years (mean SD) a mean body mass indexof 270 and a mean pregnancy duration of 38 weeks 594(n 360) were primiparae and 13 (n 8) had twinpregnancies More than 75 of women received iron andfolate supplementation during pregnancy Seventy-sevenpercent of the women reported that during pregnancysleep quality was reduced

One hundred sixty-one women reported the occurrenceof RLS symptoms (tRLS group) during pregnancy givingan RLS prevalence rate of 266 Among these 101women (167) never experienced RLS symptoms in theirlife and were classified as nRLS Of the tRLS (n 161) 60women (99) already experienced RLS symptoms beforepregnancy and were classified as pRLS When analyzingmore specifically the frequency of symptoms in our tRLSwe found that the prevalence decreased to 24 when con-sidering the presence of symptoms at least once a weekand to 15 if symptoms were present at least three timesper week (figure 1)

Comparing the data among the HG tRLS and nRLSgroups no significant differences were found in anthropo-metric measures of mothers and babies pregnancy dura-tion modality of delivery number of previous pregnanciesnumber of twin pregnancies motor habits (amount of restduring the day) time spent in bed and smoking The tRLSgroup was slightly but significantly older than the HGWomen affected by RLS received the same amount (meandaily intake and duration of therapy) of supplemental fo-late and iron during pregnancy as the HG Iron and folatewere usually taken together this therapy was usually in-troduced between the third and the eighth month of preg-nancy without differences among groups The hematologicstorage indicators revealed a mild but not significant de-crease in plasmatic iron values in the tRLS group (663 341 gdL) compared with the HG (745 462 gdL)However a significant difference for hemoglobin and meancorpuscular volume values was found between the tRLSgroup and the HG (hemoglobin 111 12 vs 115 12gdL p 0005 mean corpuscular volume 864 79 vs877 66 fl p 0005) When only the nRLS group wasconsidered a more important reduction of these valueswas observed (hemoglobin 109 12 gdL mean corpus-

cular volume 854 88 fl) No differences were found inurinary values between the HG and tRLS groups

The analysis of risk factors showed a significant associ-ation with mother weight 79 kg (odds ratio [OR] 1695 CI 11 to 24) and low hemoglobin levels (OR 1595 CI 103 to 23) in the tRLS group Only a low hemo-globin level was found to be significantly associated withthe risk of developing nRLS during pregnancy (OR 1695 CI 102 to 25)

Thirty-six percent of multiparous women in the tRLSgroup experienced the same symptomatology during atleast one of their previous pregnancies

The main RLS characteristics related to pregnancywere similar to those of the idiopathic form The anatomicdistribution of the symptomatology showed in 94 of thewomen a bilateral involvement of the limbs with a restric-tion to the area between the knee and ankle in 40 Ap-proximately one of four women had the occurrence of legsjerks during wakefulness before sleep

Twenty-nine percent of tRLS women reported a first-degree familiarity for these symptoms specifically 65 in thepRLS group 9 in the nRLS group and 2 in the HG

Sleep impact As shown in the table RLS symptomshad a significant impact on sleep Affected women reporteda reduced total sleep time and a longer sleep latency aswell as more frequent insomnia and excessive daytimesleepiness when compared with the HG No differenceswere found on nap habits Snoring and nocturnal eatingwere slightly more frequent in the tRLS group whereasbruxism leg cramps and other parasomnias were equallydistributed between RLS and non-RLS women Frequentlyreported causes of insomnia (especially during the thirdtrimester of pregnancy) were nicturia gastroesophagealreflux back pain forced body position in bed fetal move-ments and anxiety for the delivery

Course and follow-up evaluation To observe thecourse of RLS symptoms during pregnancy the RLS prev-alence rate was assessed for each month of pregnancy Asshown in figure 2 the number of women affected con-stantly increased throughout the pregnancy with a modaldistribution that peaked during the third trimester partic-ularly at the seventh and eighth months If we considerseparately the subgroups the peak of frequency for thepRLS group almost disappeared whereas the nRLS groupshowed the same trend as the tRLS group This suggeststhat in the tRLS group the maximum peak of prevalencein the third trimester is almost totally caused by the nRLSgroup The mean duration of symptoms during pregnancywas ~4 months for the tRLS group 3 months for the nRLSgroup and 6 months for the pRLS group The mean onsetof the RLS symptoms in the nRLS group was around thesixth month whereas most of the women in the pRLSgroup were already symptomatic since the initial monthsof pregnancy and often even before the pregnancy Sevenwomen affected by a pre-existing form of RLS (11 ofpRLS) reported an improvement of symptoms during preg-nancy 17 women (28) did not observe a change in theseverity of symptoms before and during pregnancywhereas the rest of the sample (61 n 36) had a signif-icant worsening of the symptoms after pregnancy beganparticularly during the third trimester

The RLS prevalence dramatically decreased around thetime of delivery 1 month after delivery the RLS preva-

Figure 1 Restless legs syndrome (RLS) prevalence rateduring pregnancy The rate changes based on the minimalthreshold of frequency of symptoms established to classifya woman as affected by RLS nRLS the new RLS grouppRLS the pre-existing RLS group

1066 NEUROLOGY 63 September (2 of 2) 2004

lence was 68 and it remained in the range between 5and 6 up to 6 months after delivery Most of the womenwho reported the persistence of RLS symptoms in the 6months after pregnancy belong to the pRLS group At thefirst follow-up evaluation after 1 month 13 of the nRLSgroup was symptomatic and the percentage decreased to5 (eight women) after 6 months (see figure 2) In themajority of women symptoms were stable without periodsof remission throughout the pregnancy and at thefollow-up evaluations symptoms were intermittent only ina few cases

Discussion Pregnancy may be considered a riskfactor for the occurrence or the worsening of RLSsymptomatology The RLS prevalence during preg-nancy of 266 that we found was higher than that

reported by most of the previous studies12-15 The ap-pearance or the worsening of RLS was closely relatedto the third trimester of pregnancy The tRLS groupsignificantly differed from the HG with a highermean age lower plasmatic storage iron indicatorshigher familiarity for RLS higher prevalence of in-somnia and higher association with snoring

The high percentage of incidence of symptoms inthe third trimester of pregnancy suggests that dur-ing this period one or more pregnancy-related factorscould facilitate the RLS appearance or its worseningWhat pregnancy-related factors could induce RLSsymptoms Our results suggest two hypotheses 1)ldquosomethingrdquo happens only in 26 of the women whobecome affected and not in the HG 2) ldquosomethingrdquohappens in almost all women but induces RLS onlyin women already predisposed to the disease In ouropinion the first hypothesis is weaker because if thispathogenetic factor would involve only one-fourth ofthe total population we would expect that in thesame way it would involve approximately one-fourthof the pRLS women But our results for the pRLSgroup do not confirm this in fact not one-fourth butthe majority of the pRLS group reported a significantincrease of the severity of symptoms during latepregnancy

Which pregnancy-related factors could reduce thesymptomatic RLS threshold Previous articles sup-port the hypothesis that RLS in pregnant womenmay be caused by an iron or folate deficiency121617

Iron and folate deficiency are well-known conditionscorrelated with RLS18 Plasmatic iron and folate lev-

Table Characteristics of pregnant women population subgroups

Total HG tRLS nRLS p Value

Age (y) 318 47 316 46 325 49 324 46 005

Mother weight (kg) 725 126 720 122 736 134 737 128 NS

Newborn weight (g) 31874 6561 31733 6725 32267 6085 32421 6425 NS

Delivery week 38 3 38 3 39 3 39 3 NS

Folate therapy (mo) 57 58 57 25 58 26 57 27 NS

Iron therapy (mo) 46 26 45 26 48 27 50 27 NS

Plasmatic iron (gdL) 720 430 745 462 663 341 663 340 NS

Hemoglobin (gdL) 113 12 115 12 111 12 109 12 0005

Mean corpuscular volume (fl) 873 70 877 66 864 79 854 88 0005

Total sleep time (min) 3918 1290 4045 1292 3568 1231 3459 1341 0005

Sleep latency (min) 243 347 232 338 274 371 295 408 005

Insomnia () 490 296 642 618 0005

EDS () 291 254 385 431 005

Snoring () 232 204 304 300 005

Leg cramps () 213 235 149 156 NS

Family history of RLS () 94 18 304 99 0005

Significance compared with HG

HG healthy group tRLS total restless legs syndrome group nRLS new restless legs syndrome group pRLS pre-existing rest-less legs syndrome group EDS excessive daytime sleepiness

Figure 2 Restless legs syndrome (RLS) course Percent-age of women affected by RLS from preconception to 6months after delivery tRLS total RLS group nRLS new RLS group pRLS pre-existing RLS group

September (2 of 2) 2004 NEUROLOGY 63 1067

els could easily decrease in pregnant women partic-ularly during the second half of pregnancy19 In ourstudy we evaluated a larger sample to assess thedose and duration of therapy and we found no differ-ences in folate and iron intake between RLS womenand the HG Our results of a significant difference inplasmatic iron storage indicators during pregnancybetween normal and affected groups support the hy-pothesis that a relative iron deficit could have anetiopathogenetic burden in this form of RLS Be-cause iron supplementation did not prevent RLSsymptoms in our study we have hypothesized thefollowing 1) women who became symptomatic dur-ing pregnancy could have had a lower iron storagealready before pregnancy and the supplementationtherapy was not sufficient to prevent the symp-toms17 2) a possible defective oral iron absorption inRLS patients could explain past failed attempts toimprove RLS by supplemental iron therapy20 and 3)there is evidence that RLS patients could have nor-mal serum iron storage indicators but low CSF con-centrations of ferritin and transferrin2122 Contraryto the iron theory is the fact that the majority ofwomen affected by RLS had remission of symptomsduring the ninth month of pregnancy andor arounddelivery even though the largest loss of blood andiron is at delivery (at least 3 months are needed torecover this loss)19

The possibility that hormonal changes duringpregnancy could reduce the symptomatic RLSthreshold has never been extensively consideredThe physiologic increase of estrogens progesteroneand prolactin levels accompanies the increase of RLSseverity until the third trimester peak is reached23

Dopamine is the strongest prolactin inhibitor Be-cause RLS has the same circadian rhythm of prolac-tin secretion some authors postulated that thenocturnal increase in prolactin might be associatedwith a reduction in dopaminergic activity inducingRLS symptoms24 Although these authors failed todemonstrate an alteration of prolactin secretion inidiopathic RLS no investigations are available forthe pregnant-related RLS form

Estrogens have been found to increase the run-ning wheel activity in rodents through an antido-paminergic effect mediated by the alpha estrogenreceptors of the medial preoptic area25-29

During late pregnancy a physiologic propriocep-tive hyperreflexia can also be observed This effectseems to be mediated by the progesterone that isknown to increase the neuronal excitability of sev-eral nervous centers30 it has been supposed that theperiodic limb movements and RLS could be furthermanifestations of this hyperexcitability31

The rapid and important RLS improvement afterdelivery gives more power to the hormonal ratherthan to the iron hypothesis

The forms of RLS secondary to radiculopathy sug-gest the possibility that during late pregnancy thegrowing volume of the fetus causes a mechanicalstrain effect on the nervous roots that resolves at

delivery32 However the similar newborn anthropo-metric values that we found in the tRLS group andHG do not confirm this hypothesis

Pregnancy is an important risk factor for RLSEven if the real cause of the association betweenRLS and pregnancy remains unclear this study isthe first to show a significant correlation betweenlow iron indicator values and the risk to developRLS Further investigations are needed to evaluatethe role of hormonal state and of personal geneticbackground predisposition in the pathogenesis ofthis reversible form of RLS

AcknowledgmentThe authors thank Dr Vincenza Castronovo for intellectual con-tribution and valuable suggestions The authors also thank DrSelena Harrison for the English language improvement

References1 Walters AS Toward a better definition of the restless legs syndrome

The International Restless Legs Syndrome Study Group Mov Disord199510634ndash642

2 Allen RP Picchietti D Hening WA et al Restless legs syndrome diag-nostic criteria special considerations and epidemiology A report fromthe restless legs syndrome diagnosis and epidemiology workshop at theNational Institutes of Health Sleep Med 20024101ndash119

3 Allen RP Earley CJ Restless legs syndrome a review of clinical andpathophysiologic features J Clin Neurophysiol 200118128ndash147

4 Stiansny K Oertel WH Trenkwalder C Clinical symptomatology andtreatment of restless legs syndrome and periodic limb movement disor-der Sleep Med Rev 20026253ndash265

5 Ondo WG Vuong KD Wang O Restless legs syndrome in monozygotictwins clinical correlates Neurology 200014551404ndash1406

6 Desautels A Turecki G Montplaisir J Sequeira A Verner A RouleauGA Identification of a major susceptibility locus for restless legs syn-drome on chromosome 12q Am J Hum Genet 2001691266ndash1270

7 Bonati MT Ferini-Strambi L Aridon P Oldani A Zucconi M Casari GAutosomal dominant restless legs syndrome maps on chromosome 14qBrain 20031261485ndash1492

8 Lavigne GJ Montplaisir JY Restless legs syndrome and sleep bruxismprevalence and association among Canadians Sleep 199417739ndash743

9 Ondo WG Vuong KD Jankovic J Exploring the relationship betweenParkinson disease and restless legs syndrome Arch Neurol 200259421ndash424

10 Michaud M Soucy JP Chabli A Lavigne G Montplaisir J SPECTimaging of striatal pre- and postsynaptic dopaminergic status in rest-less legs syndrome with periodic leg movements in sleep J Neurol2002249164ndash170

11 Mussio-Fournier JD Rawak F Familiaumlres auftreten von pruritus ur-tikaria und paraumlsthetischer hyperkinese der unteren extremitaumltenConfin Neurol 19403110ndash114

12 Ekbom KA Restless legs syndrome Acta Med Scand 1945158(suppl)4ndash122

13 Goodman JDS Brodie C Ayida GA Restless leg syndrome in preg-nancy BMJ 19882971101ndash1102

14 Suzuki K Ohida T Sone T et al The prevalence of restless legs syn-drome among pregnant women in Japan and relationship between rest-less legs syndrome and sleep problems Sleep 200326673ndash677

15 Jolivet B Parestheacutesies agitantes nocturnes des membres infeacuterieursimpatiences Theacutese de Paris 1953

16 Boetz MI Lambert B Folate deficiency and restless legs syndrome inpregnancy N Engl J Med 1977297670

17 Lee KA Zaffke ME Baratte-Beebe K Restless legs syndrome and sleepdisturbance during pregnancy the role of folate and iron J WomenrsquosHealth Gen Med 200110335ndash341

18 OrsquoKeeffe ST Gavin K Lavan JN Iron status and restless legs syn-drome in the elderly Age Ageing 199423200

19 Puolakka J Jaumlnne O Pakarinen A et al Serum ferritin as a measureof iron stores during and after normal pregnancy with and without ironsupplements Acta Obstet Gynecol Scand 198095(suppl)43ndash51

20 Davis BJ Rajput A Rajput ML Aul EA Eichhorn GR A randomizeddouble-blind placebo-controlled trial of iron in restless legs syndromeEur Neurol 20004370ndash75

21 Earley CJ Connor JR Beard JL Malecki EA Epstein DK Allen RPAbnormalities in CSF concentrations of ferritin and transferrin in rest-less legs syndrome Neurology 2000541698ndash1700

1068 NEUROLOGY 63 September (2 of 2) 2004

22 Sun ER Chen CA Ho G Earley CJ Allen RP Iron and restless legssyndrome Sleep 199821371

23 Coats P Walter E Youssefnejadian E Craft IL Variations in plasmasteroid and prostaglandin concentrations during human pregnancyActa Obstet Gynecol Scand 197756453ndash457

24 Wetter TC Collado-Seidel V Oertel H et al Endocrine rhythms inpatients with restless legs syndrome J Neurol 2002249146ndash151

25 Euvrard C Oberlander C Boissier JR Antidopaminergic effect of estro-gens at the striatal level J Pharmacol Exp Ther 1980214179ndash185

26 Ogawa S Chan J Gustafsson JA Korach KS Pfaff DW Estrogenincreases locomotor activity in mice through estrogen receptor alphaspecificity for the type of activity Endocrinology 2003144230ndash239

27 Morgan MA Pfaff DW Estrogenrsquos effects on activity anxiety and fearin two mouse strains Behav Brain Res 200213285ndash93

28 Becker JB Beer ME The influence of estrogen on nigrostriatal dopa-mine activity behavioral and neurochemical evidence for both pre- andpostsynaptic components Behav Brain Res 19861927ndash33

29 Bedard PJ Malouin F Dipaolo T Labrie F Estradiol TRH and striataldopaminergic mechanisms Prog Neuropsychopharmacol Biol Psychia-try 19826555ndash561

30 Roberts JM Pregnancy related hypertension In Creasy RK Resnik Ret al Maternal-Fetal Medicine Principle and Practice 2nd ed Phila-delphia WB Saunders 1989783ndash808

31 Santiago JR Nolledo MS Kinzler W Santiago TV Sleep and sleepdisorders in pregnancy Ann Intern Med 2001134396ndash408

32 Walters AS Wagner M Hening WA Periodic limb movements as theinitial manifestation of restless legs syndrome triggered by lumbosacralradiculopathy Sleep 199619825ndash826

CME TAKE ADVANTAGE OF NEUROLOGY CME ONLINEThe online version of Neurology includes a continuing medical education (CME) component that is only available tosubscribers The AAN designates this educational activity for up to 72 hours in Category 1 towards the AmericanMedical Association (AMA) Physicianrsquos Recognition Award (3 hours per completed online issue)

Using this system subscribers canbull Take the quizzes onlinebull Receive instant feedback on their selected answersbull Submit completed issue quizzes for CME credit and receive confirmation of credit immediately via emailbull Review all the credits theyrsquove earned via Neurology Onlinersquos individualized CME Summary Report

September (2 of 2) 2004 NEUROLOGY 63 1069

DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

ServicesUpdated Information amp

httpwwwneurologyorgcontent6361065fullhtmlincluding high resolution figures can be found at

References

httpwwwneurologyorgcontent6361065fullhtmlref-list-1at This article cites 28 articles 4 of which you can access for free

Citations

cleshttpwwwneurologyorgcontent6361065fullhtmlotherartiThis article has been cited by 9 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionrestless_legs_syndromeRestless legs syndrome

httpwwwneurologyorgcgicollectionendocrineEndocrine

httpwwwneurologyorgcgicollectionall_sleep_disordersAll Sleep Disordersfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Results The population of 606 women had a mean ageof 318 47 years (mean SD) a mean body mass indexof 270 and a mean pregnancy duration of 38 weeks 594(n 360) were primiparae and 13 (n 8) had twinpregnancies More than 75 of women received iron andfolate supplementation during pregnancy Seventy-sevenpercent of the women reported that during pregnancysleep quality was reduced

One hundred sixty-one women reported the occurrenceof RLS symptoms (tRLS group) during pregnancy givingan RLS prevalence rate of 266 Among these 101women (167) never experienced RLS symptoms in theirlife and were classified as nRLS Of the tRLS (n 161) 60women (99) already experienced RLS symptoms beforepregnancy and were classified as pRLS When analyzingmore specifically the frequency of symptoms in our tRLSwe found that the prevalence decreased to 24 when con-sidering the presence of symptoms at least once a weekand to 15 if symptoms were present at least three timesper week (figure 1)

Comparing the data among the HG tRLS and nRLSgroups no significant differences were found in anthropo-metric measures of mothers and babies pregnancy dura-tion modality of delivery number of previous pregnanciesnumber of twin pregnancies motor habits (amount of restduring the day) time spent in bed and smoking The tRLSgroup was slightly but significantly older than the HGWomen affected by RLS received the same amount (meandaily intake and duration of therapy) of supplemental fo-late and iron during pregnancy as the HG Iron and folatewere usually taken together this therapy was usually in-troduced between the third and the eighth month of preg-nancy without differences among groups The hematologicstorage indicators revealed a mild but not significant de-crease in plasmatic iron values in the tRLS group (663 341 gdL) compared with the HG (745 462 gdL)However a significant difference for hemoglobin and meancorpuscular volume values was found between the tRLSgroup and the HG (hemoglobin 111 12 vs 115 12gdL p 0005 mean corpuscular volume 864 79 vs877 66 fl p 0005) When only the nRLS group wasconsidered a more important reduction of these valueswas observed (hemoglobin 109 12 gdL mean corpus-

cular volume 854 88 fl) No differences were found inurinary values between the HG and tRLS groups

The analysis of risk factors showed a significant associ-ation with mother weight 79 kg (odds ratio [OR] 1695 CI 11 to 24) and low hemoglobin levels (OR 1595 CI 103 to 23) in the tRLS group Only a low hemo-globin level was found to be significantly associated withthe risk of developing nRLS during pregnancy (OR 1695 CI 102 to 25)

Thirty-six percent of multiparous women in the tRLSgroup experienced the same symptomatology during atleast one of their previous pregnancies

The main RLS characteristics related to pregnancywere similar to those of the idiopathic form The anatomicdistribution of the symptomatology showed in 94 of thewomen a bilateral involvement of the limbs with a restric-tion to the area between the knee and ankle in 40 Ap-proximately one of four women had the occurrence of legsjerks during wakefulness before sleep

Twenty-nine percent of tRLS women reported a first-degree familiarity for these symptoms specifically 65 in thepRLS group 9 in the nRLS group and 2 in the HG

Sleep impact As shown in the table RLS symptomshad a significant impact on sleep Affected women reporteda reduced total sleep time and a longer sleep latency aswell as more frequent insomnia and excessive daytimesleepiness when compared with the HG No differenceswere found on nap habits Snoring and nocturnal eatingwere slightly more frequent in the tRLS group whereasbruxism leg cramps and other parasomnias were equallydistributed between RLS and non-RLS women Frequentlyreported causes of insomnia (especially during the thirdtrimester of pregnancy) were nicturia gastroesophagealreflux back pain forced body position in bed fetal move-ments and anxiety for the delivery

Course and follow-up evaluation To observe thecourse of RLS symptoms during pregnancy the RLS prev-alence rate was assessed for each month of pregnancy Asshown in figure 2 the number of women affected con-stantly increased throughout the pregnancy with a modaldistribution that peaked during the third trimester partic-ularly at the seventh and eighth months If we considerseparately the subgroups the peak of frequency for thepRLS group almost disappeared whereas the nRLS groupshowed the same trend as the tRLS group This suggeststhat in the tRLS group the maximum peak of prevalencein the third trimester is almost totally caused by the nRLSgroup The mean duration of symptoms during pregnancywas ~4 months for the tRLS group 3 months for the nRLSgroup and 6 months for the pRLS group The mean onsetof the RLS symptoms in the nRLS group was around thesixth month whereas most of the women in the pRLSgroup were already symptomatic since the initial monthsof pregnancy and often even before the pregnancy Sevenwomen affected by a pre-existing form of RLS (11 ofpRLS) reported an improvement of symptoms during preg-nancy 17 women (28) did not observe a change in theseverity of symptoms before and during pregnancywhereas the rest of the sample (61 n 36) had a signif-icant worsening of the symptoms after pregnancy beganparticularly during the third trimester

The RLS prevalence dramatically decreased around thetime of delivery 1 month after delivery the RLS preva-

Figure 1 Restless legs syndrome (RLS) prevalence rateduring pregnancy The rate changes based on the minimalthreshold of frequency of symptoms established to classifya woman as affected by RLS nRLS the new RLS grouppRLS the pre-existing RLS group

1066 NEUROLOGY 63 September (2 of 2) 2004

lence was 68 and it remained in the range between 5and 6 up to 6 months after delivery Most of the womenwho reported the persistence of RLS symptoms in the 6months after pregnancy belong to the pRLS group At thefirst follow-up evaluation after 1 month 13 of the nRLSgroup was symptomatic and the percentage decreased to5 (eight women) after 6 months (see figure 2) In themajority of women symptoms were stable without periodsof remission throughout the pregnancy and at thefollow-up evaluations symptoms were intermittent only ina few cases

Discussion Pregnancy may be considered a riskfactor for the occurrence or the worsening of RLSsymptomatology The RLS prevalence during preg-nancy of 266 that we found was higher than that

reported by most of the previous studies12-15 The ap-pearance or the worsening of RLS was closely relatedto the third trimester of pregnancy The tRLS groupsignificantly differed from the HG with a highermean age lower plasmatic storage iron indicatorshigher familiarity for RLS higher prevalence of in-somnia and higher association with snoring

The high percentage of incidence of symptoms inthe third trimester of pregnancy suggests that dur-ing this period one or more pregnancy-related factorscould facilitate the RLS appearance or its worseningWhat pregnancy-related factors could induce RLSsymptoms Our results suggest two hypotheses 1)ldquosomethingrdquo happens only in 26 of the women whobecome affected and not in the HG 2) ldquosomethingrdquohappens in almost all women but induces RLS onlyin women already predisposed to the disease In ouropinion the first hypothesis is weaker because if thispathogenetic factor would involve only one-fourth ofthe total population we would expect that in thesame way it would involve approximately one-fourthof the pRLS women But our results for the pRLSgroup do not confirm this in fact not one-fourth butthe majority of the pRLS group reported a significantincrease of the severity of symptoms during latepregnancy

Which pregnancy-related factors could reduce thesymptomatic RLS threshold Previous articles sup-port the hypothesis that RLS in pregnant womenmay be caused by an iron or folate deficiency121617

Iron and folate deficiency are well-known conditionscorrelated with RLS18 Plasmatic iron and folate lev-

Table Characteristics of pregnant women population subgroups

Total HG tRLS nRLS p Value

Age (y) 318 47 316 46 325 49 324 46 005

Mother weight (kg) 725 126 720 122 736 134 737 128 NS

Newborn weight (g) 31874 6561 31733 6725 32267 6085 32421 6425 NS

Delivery week 38 3 38 3 39 3 39 3 NS

Folate therapy (mo) 57 58 57 25 58 26 57 27 NS

Iron therapy (mo) 46 26 45 26 48 27 50 27 NS

Plasmatic iron (gdL) 720 430 745 462 663 341 663 340 NS

Hemoglobin (gdL) 113 12 115 12 111 12 109 12 0005

Mean corpuscular volume (fl) 873 70 877 66 864 79 854 88 0005

Total sleep time (min) 3918 1290 4045 1292 3568 1231 3459 1341 0005

Sleep latency (min) 243 347 232 338 274 371 295 408 005

Insomnia () 490 296 642 618 0005

EDS () 291 254 385 431 005

Snoring () 232 204 304 300 005

Leg cramps () 213 235 149 156 NS

Family history of RLS () 94 18 304 99 0005

Significance compared with HG

HG healthy group tRLS total restless legs syndrome group nRLS new restless legs syndrome group pRLS pre-existing rest-less legs syndrome group EDS excessive daytime sleepiness

Figure 2 Restless legs syndrome (RLS) course Percent-age of women affected by RLS from preconception to 6months after delivery tRLS total RLS group nRLS new RLS group pRLS pre-existing RLS group

September (2 of 2) 2004 NEUROLOGY 63 1067

els could easily decrease in pregnant women partic-ularly during the second half of pregnancy19 In ourstudy we evaluated a larger sample to assess thedose and duration of therapy and we found no differ-ences in folate and iron intake between RLS womenand the HG Our results of a significant difference inplasmatic iron storage indicators during pregnancybetween normal and affected groups support the hy-pothesis that a relative iron deficit could have anetiopathogenetic burden in this form of RLS Be-cause iron supplementation did not prevent RLSsymptoms in our study we have hypothesized thefollowing 1) women who became symptomatic dur-ing pregnancy could have had a lower iron storagealready before pregnancy and the supplementationtherapy was not sufficient to prevent the symp-toms17 2) a possible defective oral iron absorption inRLS patients could explain past failed attempts toimprove RLS by supplemental iron therapy20 and 3)there is evidence that RLS patients could have nor-mal serum iron storage indicators but low CSF con-centrations of ferritin and transferrin2122 Contraryto the iron theory is the fact that the majority ofwomen affected by RLS had remission of symptomsduring the ninth month of pregnancy andor arounddelivery even though the largest loss of blood andiron is at delivery (at least 3 months are needed torecover this loss)19

The possibility that hormonal changes duringpregnancy could reduce the symptomatic RLSthreshold has never been extensively consideredThe physiologic increase of estrogens progesteroneand prolactin levels accompanies the increase of RLSseverity until the third trimester peak is reached23

Dopamine is the strongest prolactin inhibitor Be-cause RLS has the same circadian rhythm of prolac-tin secretion some authors postulated that thenocturnal increase in prolactin might be associatedwith a reduction in dopaminergic activity inducingRLS symptoms24 Although these authors failed todemonstrate an alteration of prolactin secretion inidiopathic RLS no investigations are available forthe pregnant-related RLS form

Estrogens have been found to increase the run-ning wheel activity in rodents through an antido-paminergic effect mediated by the alpha estrogenreceptors of the medial preoptic area25-29

During late pregnancy a physiologic propriocep-tive hyperreflexia can also be observed This effectseems to be mediated by the progesterone that isknown to increase the neuronal excitability of sev-eral nervous centers30 it has been supposed that theperiodic limb movements and RLS could be furthermanifestations of this hyperexcitability31

The rapid and important RLS improvement afterdelivery gives more power to the hormonal ratherthan to the iron hypothesis

The forms of RLS secondary to radiculopathy sug-gest the possibility that during late pregnancy thegrowing volume of the fetus causes a mechanicalstrain effect on the nervous roots that resolves at

delivery32 However the similar newborn anthropo-metric values that we found in the tRLS group andHG do not confirm this hypothesis

Pregnancy is an important risk factor for RLSEven if the real cause of the association betweenRLS and pregnancy remains unclear this study isthe first to show a significant correlation betweenlow iron indicator values and the risk to developRLS Further investigations are needed to evaluatethe role of hormonal state and of personal geneticbackground predisposition in the pathogenesis ofthis reversible form of RLS

AcknowledgmentThe authors thank Dr Vincenza Castronovo for intellectual con-tribution and valuable suggestions The authors also thank DrSelena Harrison for the English language improvement

References1 Walters AS Toward a better definition of the restless legs syndrome

The International Restless Legs Syndrome Study Group Mov Disord199510634ndash642

2 Allen RP Picchietti D Hening WA et al Restless legs syndrome diag-nostic criteria special considerations and epidemiology A report fromthe restless legs syndrome diagnosis and epidemiology workshop at theNational Institutes of Health Sleep Med 20024101ndash119

3 Allen RP Earley CJ Restless legs syndrome a review of clinical andpathophysiologic features J Clin Neurophysiol 200118128ndash147

4 Stiansny K Oertel WH Trenkwalder C Clinical symptomatology andtreatment of restless legs syndrome and periodic limb movement disor-der Sleep Med Rev 20026253ndash265

5 Ondo WG Vuong KD Wang O Restless legs syndrome in monozygotictwins clinical correlates Neurology 200014551404ndash1406

6 Desautels A Turecki G Montplaisir J Sequeira A Verner A RouleauGA Identification of a major susceptibility locus for restless legs syn-drome on chromosome 12q Am J Hum Genet 2001691266ndash1270

7 Bonati MT Ferini-Strambi L Aridon P Oldani A Zucconi M Casari GAutosomal dominant restless legs syndrome maps on chromosome 14qBrain 20031261485ndash1492

8 Lavigne GJ Montplaisir JY Restless legs syndrome and sleep bruxismprevalence and association among Canadians Sleep 199417739ndash743

9 Ondo WG Vuong KD Jankovic J Exploring the relationship betweenParkinson disease and restless legs syndrome Arch Neurol 200259421ndash424

10 Michaud M Soucy JP Chabli A Lavigne G Montplaisir J SPECTimaging of striatal pre- and postsynaptic dopaminergic status in rest-less legs syndrome with periodic leg movements in sleep J Neurol2002249164ndash170

11 Mussio-Fournier JD Rawak F Familiaumlres auftreten von pruritus ur-tikaria und paraumlsthetischer hyperkinese der unteren extremitaumltenConfin Neurol 19403110ndash114

12 Ekbom KA Restless legs syndrome Acta Med Scand 1945158(suppl)4ndash122

13 Goodman JDS Brodie C Ayida GA Restless leg syndrome in preg-nancy BMJ 19882971101ndash1102

14 Suzuki K Ohida T Sone T et al The prevalence of restless legs syn-drome among pregnant women in Japan and relationship between rest-less legs syndrome and sleep problems Sleep 200326673ndash677

15 Jolivet B Parestheacutesies agitantes nocturnes des membres infeacuterieursimpatiences Theacutese de Paris 1953

16 Boetz MI Lambert B Folate deficiency and restless legs syndrome inpregnancy N Engl J Med 1977297670

17 Lee KA Zaffke ME Baratte-Beebe K Restless legs syndrome and sleepdisturbance during pregnancy the role of folate and iron J WomenrsquosHealth Gen Med 200110335ndash341

18 OrsquoKeeffe ST Gavin K Lavan JN Iron status and restless legs syn-drome in the elderly Age Ageing 199423200

19 Puolakka J Jaumlnne O Pakarinen A et al Serum ferritin as a measureof iron stores during and after normal pregnancy with and without ironsupplements Acta Obstet Gynecol Scand 198095(suppl)43ndash51

20 Davis BJ Rajput A Rajput ML Aul EA Eichhorn GR A randomizeddouble-blind placebo-controlled trial of iron in restless legs syndromeEur Neurol 20004370ndash75

21 Earley CJ Connor JR Beard JL Malecki EA Epstein DK Allen RPAbnormalities in CSF concentrations of ferritin and transferrin in rest-less legs syndrome Neurology 2000541698ndash1700

1068 NEUROLOGY 63 September (2 of 2) 2004

22 Sun ER Chen CA Ho G Earley CJ Allen RP Iron and restless legssyndrome Sleep 199821371

23 Coats P Walter E Youssefnejadian E Craft IL Variations in plasmasteroid and prostaglandin concentrations during human pregnancyActa Obstet Gynecol Scand 197756453ndash457

24 Wetter TC Collado-Seidel V Oertel H et al Endocrine rhythms inpatients with restless legs syndrome J Neurol 2002249146ndash151

25 Euvrard C Oberlander C Boissier JR Antidopaminergic effect of estro-gens at the striatal level J Pharmacol Exp Ther 1980214179ndash185

26 Ogawa S Chan J Gustafsson JA Korach KS Pfaff DW Estrogenincreases locomotor activity in mice through estrogen receptor alphaspecificity for the type of activity Endocrinology 2003144230ndash239

27 Morgan MA Pfaff DW Estrogenrsquos effects on activity anxiety and fearin two mouse strains Behav Brain Res 200213285ndash93

28 Becker JB Beer ME The influence of estrogen on nigrostriatal dopa-mine activity behavioral and neurochemical evidence for both pre- andpostsynaptic components Behav Brain Res 19861927ndash33

29 Bedard PJ Malouin F Dipaolo T Labrie F Estradiol TRH and striataldopaminergic mechanisms Prog Neuropsychopharmacol Biol Psychia-try 19826555ndash561

30 Roberts JM Pregnancy related hypertension In Creasy RK Resnik Ret al Maternal-Fetal Medicine Principle and Practice 2nd ed Phila-delphia WB Saunders 1989783ndash808

31 Santiago JR Nolledo MS Kinzler W Santiago TV Sleep and sleepdisorders in pregnancy Ann Intern Med 2001134396ndash408

32 Walters AS Wagner M Hening WA Periodic limb movements as theinitial manifestation of restless legs syndrome triggered by lumbosacralradiculopathy Sleep 199619825ndash826

CME TAKE ADVANTAGE OF NEUROLOGY CME ONLINEThe online version of Neurology includes a continuing medical education (CME) component that is only available tosubscribers The AAN designates this educational activity for up to 72 hours in Category 1 towards the AmericanMedical Association (AMA) Physicianrsquos Recognition Award (3 hours per completed online issue)

Using this system subscribers canbull Take the quizzes onlinebull Receive instant feedback on their selected answersbull Submit completed issue quizzes for CME credit and receive confirmation of credit immediately via emailbull Review all the credits theyrsquove earned via Neurology Onlinersquos individualized CME Summary Report

September (2 of 2) 2004 NEUROLOGY 63 1069

DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

ServicesUpdated Information amp

httpwwwneurologyorgcontent6361065fullhtmlincluding high resolution figures can be found at

References

httpwwwneurologyorgcontent6361065fullhtmlref-list-1at This article cites 28 articles 4 of which you can access for free

Citations

cleshttpwwwneurologyorgcontent6361065fullhtmlotherartiThis article has been cited by 9 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionrestless_legs_syndromeRestless legs syndrome

httpwwwneurologyorgcgicollectionendocrineEndocrine

httpwwwneurologyorgcgicollectionall_sleep_disordersAll Sleep Disordersfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

lence was 68 and it remained in the range between 5and 6 up to 6 months after delivery Most of the womenwho reported the persistence of RLS symptoms in the 6months after pregnancy belong to the pRLS group At thefirst follow-up evaluation after 1 month 13 of the nRLSgroup was symptomatic and the percentage decreased to5 (eight women) after 6 months (see figure 2) In themajority of women symptoms were stable without periodsof remission throughout the pregnancy and at thefollow-up evaluations symptoms were intermittent only ina few cases

Discussion Pregnancy may be considered a riskfactor for the occurrence or the worsening of RLSsymptomatology The RLS prevalence during preg-nancy of 266 that we found was higher than that

reported by most of the previous studies12-15 The ap-pearance or the worsening of RLS was closely relatedto the third trimester of pregnancy The tRLS groupsignificantly differed from the HG with a highermean age lower plasmatic storage iron indicatorshigher familiarity for RLS higher prevalence of in-somnia and higher association with snoring

The high percentage of incidence of symptoms inthe third trimester of pregnancy suggests that dur-ing this period one or more pregnancy-related factorscould facilitate the RLS appearance or its worseningWhat pregnancy-related factors could induce RLSsymptoms Our results suggest two hypotheses 1)ldquosomethingrdquo happens only in 26 of the women whobecome affected and not in the HG 2) ldquosomethingrdquohappens in almost all women but induces RLS onlyin women already predisposed to the disease In ouropinion the first hypothesis is weaker because if thispathogenetic factor would involve only one-fourth ofthe total population we would expect that in thesame way it would involve approximately one-fourthof the pRLS women But our results for the pRLSgroup do not confirm this in fact not one-fourth butthe majority of the pRLS group reported a significantincrease of the severity of symptoms during latepregnancy

Which pregnancy-related factors could reduce thesymptomatic RLS threshold Previous articles sup-port the hypothesis that RLS in pregnant womenmay be caused by an iron or folate deficiency121617

Iron and folate deficiency are well-known conditionscorrelated with RLS18 Plasmatic iron and folate lev-

Table Characteristics of pregnant women population subgroups

Total HG tRLS nRLS p Value

Age (y) 318 47 316 46 325 49 324 46 005

Mother weight (kg) 725 126 720 122 736 134 737 128 NS

Newborn weight (g) 31874 6561 31733 6725 32267 6085 32421 6425 NS

Delivery week 38 3 38 3 39 3 39 3 NS

Folate therapy (mo) 57 58 57 25 58 26 57 27 NS

Iron therapy (mo) 46 26 45 26 48 27 50 27 NS

Plasmatic iron (gdL) 720 430 745 462 663 341 663 340 NS

Hemoglobin (gdL) 113 12 115 12 111 12 109 12 0005

Mean corpuscular volume (fl) 873 70 877 66 864 79 854 88 0005

Total sleep time (min) 3918 1290 4045 1292 3568 1231 3459 1341 0005

Sleep latency (min) 243 347 232 338 274 371 295 408 005

Insomnia () 490 296 642 618 0005

EDS () 291 254 385 431 005

Snoring () 232 204 304 300 005

Leg cramps () 213 235 149 156 NS

Family history of RLS () 94 18 304 99 0005

Significance compared with HG

HG healthy group tRLS total restless legs syndrome group nRLS new restless legs syndrome group pRLS pre-existing rest-less legs syndrome group EDS excessive daytime sleepiness

Figure 2 Restless legs syndrome (RLS) course Percent-age of women affected by RLS from preconception to 6months after delivery tRLS total RLS group nRLS new RLS group pRLS pre-existing RLS group

September (2 of 2) 2004 NEUROLOGY 63 1067

els could easily decrease in pregnant women partic-ularly during the second half of pregnancy19 In ourstudy we evaluated a larger sample to assess thedose and duration of therapy and we found no differ-ences in folate and iron intake between RLS womenand the HG Our results of a significant difference inplasmatic iron storage indicators during pregnancybetween normal and affected groups support the hy-pothesis that a relative iron deficit could have anetiopathogenetic burden in this form of RLS Be-cause iron supplementation did not prevent RLSsymptoms in our study we have hypothesized thefollowing 1) women who became symptomatic dur-ing pregnancy could have had a lower iron storagealready before pregnancy and the supplementationtherapy was not sufficient to prevent the symp-toms17 2) a possible defective oral iron absorption inRLS patients could explain past failed attempts toimprove RLS by supplemental iron therapy20 and 3)there is evidence that RLS patients could have nor-mal serum iron storage indicators but low CSF con-centrations of ferritin and transferrin2122 Contraryto the iron theory is the fact that the majority ofwomen affected by RLS had remission of symptomsduring the ninth month of pregnancy andor arounddelivery even though the largest loss of blood andiron is at delivery (at least 3 months are needed torecover this loss)19

The possibility that hormonal changes duringpregnancy could reduce the symptomatic RLSthreshold has never been extensively consideredThe physiologic increase of estrogens progesteroneand prolactin levels accompanies the increase of RLSseverity until the third trimester peak is reached23

Dopamine is the strongest prolactin inhibitor Be-cause RLS has the same circadian rhythm of prolac-tin secretion some authors postulated that thenocturnal increase in prolactin might be associatedwith a reduction in dopaminergic activity inducingRLS symptoms24 Although these authors failed todemonstrate an alteration of prolactin secretion inidiopathic RLS no investigations are available forthe pregnant-related RLS form

Estrogens have been found to increase the run-ning wheel activity in rodents through an antido-paminergic effect mediated by the alpha estrogenreceptors of the medial preoptic area25-29

During late pregnancy a physiologic propriocep-tive hyperreflexia can also be observed This effectseems to be mediated by the progesterone that isknown to increase the neuronal excitability of sev-eral nervous centers30 it has been supposed that theperiodic limb movements and RLS could be furthermanifestations of this hyperexcitability31

The rapid and important RLS improvement afterdelivery gives more power to the hormonal ratherthan to the iron hypothesis

The forms of RLS secondary to radiculopathy sug-gest the possibility that during late pregnancy thegrowing volume of the fetus causes a mechanicalstrain effect on the nervous roots that resolves at

delivery32 However the similar newborn anthropo-metric values that we found in the tRLS group andHG do not confirm this hypothesis

Pregnancy is an important risk factor for RLSEven if the real cause of the association betweenRLS and pregnancy remains unclear this study isthe first to show a significant correlation betweenlow iron indicator values and the risk to developRLS Further investigations are needed to evaluatethe role of hormonal state and of personal geneticbackground predisposition in the pathogenesis ofthis reversible form of RLS

AcknowledgmentThe authors thank Dr Vincenza Castronovo for intellectual con-tribution and valuable suggestions The authors also thank DrSelena Harrison for the English language improvement

References1 Walters AS Toward a better definition of the restless legs syndrome

The International Restless Legs Syndrome Study Group Mov Disord199510634ndash642

2 Allen RP Picchietti D Hening WA et al Restless legs syndrome diag-nostic criteria special considerations and epidemiology A report fromthe restless legs syndrome diagnosis and epidemiology workshop at theNational Institutes of Health Sleep Med 20024101ndash119

3 Allen RP Earley CJ Restless legs syndrome a review of clinical andpathophysiologic features J Clin Neurophysiol 200118128ndash147

4 Stiansny K Oertel WH Trenkwalder C Clinical symptomatology andtreatment of restless legs syndrome and periodic limb movement disor-der Sleep Med Rev 20026253ndash265

5 Ondo WG Vuong KD Wang O Restless legs syndrome in monozygotictwins clinical correlates Neurology 200014551404ndash1406

6 Desautels A Turecki G Montplaisir J Sequeira A Verner A RouleauGA Identification of a major susceptibility locus for restless legs syn-drome on chromosome 12q Am J Hum Genet 2001691266ndash1270

7 Bonati MT Ferini-Strambi L Aridon P Oldani A Zucconi M Casari GAutosomal dominant restless legs syndrome maps on chromosome 14qBrain 20031261485ndash1492

8 Lavigne GJ Montplaisir JY Restless legs syndrome and sleep bruxismprevalence and association among Canadians Sleep 199417739ndash743

9 Ondo WG Vuong KD Jankovic J Exploring the relationship betweenParkinson disease and restless legs syndrome Arch Neurol 200259421ndash424

10 Michaud M Soucy JP Chabli A Lavigne G Montplaisir J SPECTimaging of striatal pre- and postsynaptic dopaminergic status in rest-less legs syndrome with periodic leg movements in sleep J Neurol2002249164ndash170

11 Mussio-Fournier JD Rawak F Familiaumlres auftreten von pruritus ur-tikaria und paraumlsthetischer hyperkinese der unteren extremitaumltenConfin Neurol 19403110ndash114

12 Ekbom KA Restless legs syndrome Acta Med Scand 1945158(suppl)4ndash122

13 Goodman JDS Brodie C Ayida GA Restless leg syndrome in preg-nancy BMJ 19882971101ndash1102

14 Suzuki K Ohida T Sone T et al The prevalence of restless legs syn-drome among pregnant women in Japan and relationship between rest-less legs syndrome and sleep problems Sleep 200326673ndash677

15 Jolivet B Parestheacutesies agitantes nocturnes des membres infeacuterieursimpatiences Theacutese de Paris 1953

16 Boetz MI Lambert B Folate deficiency and restless legs syndrome inpregnancy N Engl J Med 1977297670

17 Lee KA Zaffke ME Baratte-Beebe K Restless legs syndrome and sleepdisturbance during pregnancy the role of folate and iron J WomenrsquosHealth Gen Med 200110335ndash341

18 OrsquoKeeffe ST Gavin K Lavan JN Iron status and restless legs syn-drome in the elderly Age Ageing 199423200

19 Puolakka J Jaumlnne O Pakarinen A et al Serum ferritin as a measureof iron stores during and after normal pregnancy with and without ironsupplements Acta Obstet Gynecol Scand 198095(suppl)43ndash51

20 Davis BJ Rajput A Rajput ML Aul EA Eichhorn GR A randomizeddouble-blind placebo-controlled trial of iron in restless legs syndromeEur Neurol 20004370ndash75

21 Earley CJ Connor JR Beard JL Malecki EA Epstein DK Allen RPAbnormalities in CSF concentrations of ferritin and transferrin in rest-less legs syndrome Neurology 2000541698ndash1700

1068 NEUROLOGY 63 September (2 of 2) 2004

22 Sun ER Chen CA Ho G Earley CJ Allen RP Iron and restless legssyndrome Sleep 199821371

23 Coats P Walter E Youssefnejadian E Craft IL Variations in plasmasteroid and prostaglandin concentrations during human pregnancyActa Obstet Gynecol Scand 197756453ndash457

24 Wetter TC Collado-Seidel V Oertel H et al Endocrine rhythms inpatients with restless legs syndrome J Neurol 2002249146ndash151

25 Euvrard C Oberlander C Boissier JR Antidopaminergic effect of estro-gens at the striatal level J Pharmacol Exp Ther 1980214179ndash185

26 Ogawa S Chan J Gustafsson JA Korach KS Pfaff DW Estrogenincreases locomotor activity in mice through estrogen receptor alphaspecificity for the type of activity Endocrinology 2003144230ndash239

27 Morgan MA Pfaff DW Estrogenrsquos effects on activity anxiety and fearin two mouse strains Behav Brain Res 200213285ndash93

28 Becker JB Beer ME The influence of estrogen on nigrostriatal dopa-mine activity behavioral and neurochemical evidence for both pre- andpostsynaptic components Behav Brain Res 19861927ndash33

29 Bedard PJ Malouin F Dipaolo T Labrie F Estradiol TRH and striataldopaminergic mechanisms Prog Neuropsychopharmacol Biol Psychia-try 19826555ndash561

30 Roberts JM Pregnancy related hypertension In Creasy RK Resnik Ret al Maternal-Fetal Medicine Principle and Practice 2nd ed Phila-delphia WB Saunders 1989783ndash808

31 Santiago JR Nolledo MS Kinzler W Santiago TV Sleep and sleepdisorders in pregnancy Ann Intern Med 2001134396ndash408

32 Walters AS Wagner M Hening WA Periodic limb movements as theinitial manifestation of restless legs syndrome triggered by lumbosacralradiculopathy Sleep 199619825ndash826

CME TAKE ADVANTAGE OF NEUROLOGY CME ONLINEThe online version of Neurology includes a continuing medical education (CME) component that is only available tosubscribers The AAN designates this educational activity for up to 72 hours in Category 1 towards the AmericanMedical Association (AMA) Physicianrsquos Recognition Award (3 hours per completed online issue)

Using this system subscribers canbull Take the quizzes onlinebull Receive instant feedback on their selected answersbull Submit completed issue quizzes for CME credit and receive confirmation of credit immediately via emailbull Review all the credits theyrsquove earned via Neurology Onlinersquos individualized CME Summary Report

September (2 of 2) 2004 NEUROLOGY 63 1069

DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

ServicesUpdated Information amp

httpwwwneurologyorgcontent6361065fullhtmlincluding high resolution figures can be found at

References

httpwwwneurologyorgcontent6361065fullhtmlref-list-1at This article cites 28 articles 4 of which you can access for free

Citations

cleshttpwwwneurologyorgcontent6361065fullhtmlotherartiThis article has been cited by 9 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionrestless_legs_syndromeRestless legs syndrome

httpwwwneurologyorgcgicollectionendocrineEndocrine

httpwwwneurologyorgcgicollectionall_sleep_disordersAll Sleep Disordersfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

els could easily decrease in pregnant women partic-ularly during the second half of pregnancy19 In ourstudy we evaluated a larger sample to assess thedose and duration of therapy and we found no differ-ences in folate and iron intake between RLS womenand the HG Our results of a significant difference inplasmatic iron storage indicators during pregnancybetween normal and affected groups support the hy-pothesis that a relative iron deficit could have anetiopathogenetic burden in this form of RLS Be-cause iron supplementation did not prevent RLSsymptoms in our study we have hypothesized thefollowing 1) women who became symptomatic dur-ing pregnancy could have had a lower iron storagealready before pregnancy and the supplementationtherapy was not sufficient to prevent the symp-toms17 2) a possible defective oral iron absorption inRLS patients could explain past failed attempts toimprove RLS by supplemental iron therapy20 and 3)there is evidence that RLS patients could have nor-mal serum iron storage indicators but low CSF con-centrations of ferritin and transferrin2122 Contraryto the iron theory is the fact that the majority ofwomen affected by RLS had remission of symptomsduring the ninth month of pregnancy andor arounddelivery even though the largest loss of blood andiron is at delivery (at least 3 months are needed torecover this loss)19

The possibility that hormonal changes duringpregnancy could reduce the symptomatic RLSthreshold has never been extensively consideredThe physiologic increase of estrogens progesteroneand prolactin levels accompanies the increase of RLSseverity until the third trimester peak is reached23

Dopamine is the strongest prolactin inhibitor Be-cause RLS has the same circadian rhythm of prolac-tin secretion some authors postulated that thenocturnal increase in prolactin might be associatedwith a reduction in dopaminergic activity inducingRLS symptoms24 Although these authors failed todemonstrate an alteration of prolactin secretion inidiopathic RLS no investigations are available forthe pregnant-related RLS form

Estrogens have been found to increase the run-ning wheel activity in rodents through an antido-paminergic effect mediated by the alpha estrogenreceptors of the medial preoptic area25-29

During late pregnancy a physiologic propriocep-tive hyperreflexia can also be observed This effectseems to be mediated by the progesterone that isknown to increase the neuronal excitability of sev-eral nervous centers30 it has been supposed that theperiodic limb movements and RLS could be furthermanifestations of this hyperexcitability31

The rapid and important RLS improvement afterdelivery gives more power to the hormonal ratherthan to the iron hypothesis

The forms of RLS secondary to radiculopathy sug-gest the possibility that during late pregnancy thegrowing volume of the fetus causes a mechanicalstrain effect on the nervous roots that resolves at

delivery32 However the similar newborn anthropo-metric values that we found in the tRLS group andHG do not confirm this hypothesis

Pregnancy is an important risk factor for RLSEven if the real cause of the association betweenRLS and pregnancy remains unclear this study isthe first to show a significant correlation betweenlow iron indicator values and the risk to developRLS Further investigations are needed to evaluatethe role of hormonal state and of personal geneticbackground predisposition in the pathogenesis ofthis reversible form of RLS

AcknowledgmentThe authors thank Dr Vincenza Castronovo for intellectual con-tribution and valuable suggestions The authors also thank DrSelena Harrison for the English language improvement

References1 Walters AS Toward a better definition of the restless legs syndrome

The International Restless Legs Syndrome Study Group Mov Disord199510634ndash642

2 Allen RP Picchietti D Hening WA et al Restless legs syndrome diag-nostic criteria special considerations and epidemiology A report fromthe restless legs syndrome diagnosis and epidemiology workshop at theNational Institutes of Health Sleep Med 20024101ndash119

3 Allen RP Earley CJ Restless legs syndrome a review of clinical andpathophysiologic features J Clin Neurophysiol 200118128ndash147

4 Stiansny K Oertel WH Trenkwalder C Clinical symptomatology andtreatment of restless legs syndrome and periodic limb movement disor-der Sleep Med Rev 20026253ndash265

5 Ondo WG Vuong KD Wang O Restless legs syndrome in monozygotictwins clinical correlates Neurology 200014551404ndash1406

6 Desautels A Turecki G Montplaisir J Sequeira A Verner A RouleauGA Identification of a major susceptibility locus for restless legs syn-drome on chromosome 12q Am J Hum Genet 2001691266ndash1270

7 Bonati MT Ferini-Strambi L Aridon P Oldani A Zucconi M Casari GAutosomal dominant restless legs syndrome maps on chromosome 14qBrain 20031261485ndash1492

8 Lavigne GJ Montplaisir JY Restless legs syndrome and sleep bruxismprevalence and association among Canadians Sleep 199417739ndash743

9 Ondo WG Vuong KD Jankovic J Exploring the relationship betweenParkinson disease and restless legs syndrome Arch Neurol 200259421ndash424

10 Michaud M Soucy JP Chabli A Lavigne G Montplaisir J SPECTimaging of striatal pre- and postsynaptic dopaminergic status in rest-less legs syndrome with periodic leg movements in sleep J Neurol2002249164ndash170

11 Mussio-Fournier JD Rawak F Familiaumlres auftreten von pruritus ur-tikaria und paraumlsthetischer hyperkinese der unteren extremitaumltenConfin Neurol 19403110ndash114

12 Ekbom KA Restless legs syndrome Acta Med Scand 1945158(suppl)4ndash122

13 Goodman JDS Brodie C Ayida GA Restless leg syndrome in preg-nancy BMJ 19882971101ndash1102

14 Suzuki K Ohida T Sone T et al The prevalence of restless legs syn-drome among pregnant women in Japan and relationship between rest-less legs syndrome and sleep problems Sleep 200326673ndash677

15 Jolivet B Parestheacutesies agitantes nocturnes des membres infeacuterieursimpatiences Theacutese de Paris 1953

16 Boetz MI Lambert B Folate deficiency and restless legs syndrome inpregnancy N Engl J Med 1977297670

17 Lee KA Zaffke ME Baratte-Beebe K Restless legs syndrome and sleepdisturbance during pregnancy the role of folate and iron J WomenrsquosHealth Gen Med 200110335ndash341

18 OrsquoKeeffe ST Gavin K Lavan JN Iron status and restless legs syn-drome in the elderly Age Ageing 199423200

19 Puolakka J Jaumlnne O Pakarinen A et al Serum ferritin as a measureof iron stores during and after normal pregnancy with and without ironsupplements Acta Obstet Gynecol Scand 198095(suppl)43ndash51

20 Davis BJ Rajput A Rajput ML Aul EA Eichhorn GR A randomizeddouble-blind placebo-controlled trial of iron in restless legs syndromeEur Neurol 20004370ndash75

21 Earley CJ Connor JR Beard JL Malecki EA Epstein DK Allen RPAbnormalities in CSF concentrations of ferritin and transferrin in rest-less legs syndrome Neurology 2000541698ndash1700

1068 NEUROLOGY 63 September (2 of 2) 2004

22 Sun ER Chen CA Ho G Earley CJ Allen RP Iron and restless legssyndrome Sleep 199821371

23 Coats P Walter E Youssefnejadian E Craft IL Variations in plasmasteroid and prostaglandin concentrations during human pregnancyActa Obstet Gynecol Scand 197756453ndash457

24 Wetter TC Collado-Seidel V Oertel H et al Endocrine rhythms inpatients with restless legs syndrome J Neurol 2002249146ndash151

25 Euvrard C Oberlander C Boissier JR Antidopaminergic effect of estro-gens at the striatal level J Pharmacol Exp Ther 1980214179ndash185

26 Ogawa S Chan J Gustafsson JA Korach KS Pfaff DW Estrogenincreases locomotor activity in mice through estrogen receptor alphaspecificity for the type of activity Endocrinology 2003144230ndash239

27 Morgan MA Pfaff DW Estrogenrsquos effects on activity anxiety and fearin two mouse strains Behav Brain Res 200213285ndash93

28 Becker JB Beer ME The influence of estrogen on nigrostriatal dopa-mine activity behavioral and neurochemical evidence for both pre- andpostsynaptic components Behav Brain Res 19861927ndash33

29 Bedard PJ Malouin F Dipaolo T Labrie F Estradiol TRH and striataldopaminergic mechanisms Prog Neuropsychopharmacol Biol Psychia-try 19826555ndash561

30 Roberts JM Pregnancy related hypertension In Creasy RK Resnik Ret al Maternal-Fetal Medicine Principle and Practice 2nd ed Phila-delphia WB Saunders 1989783ndash808

31 Santiago JR Nolledo MS Kinzler W Santiago TV Sleep and sleepdisorders in pregnancy Ann Intern Med 2001134396ndash408

32 Walters AS Wagner M Hening WA Periodic limb movements as theinitial manifestation of restless legs syndrome triggered by lumbosacralradiculopathy Sleep 199619825ndash826

CME TAKE ADVANTAGE OF NEUROLOGY CME ONLINEThe online version of Neurology includes a continuing medical education (CME) component that is only available tosubscribers The AAN designates this educational activity for up to 72 hours in Category 1 towards the AmericanMedical Association (AMA) Physicianrsquos Recognition Award (3 hours per completed online issue)

Using this system subscribers canbull Take the quizzes onlinebull Receive instant feedback on their selected answersbull Submit completed issue quizzes for CME credit and receive confirmation of credit immediately via emailbull Review all the credits theyrsquove earned via Neurology Onlinersquos individualized CME Summary Report

September (2 of 2) 2004 NEUROLOGY 63 1069

DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

ServicesUpdated Information amp

httpwwwneurologyorgcontent6361065fullhtmlincluding high resolution figures can be found at

References

httpwwwneurologyorgcontent6361065fullhtmlref-list-1at This article cites 28 articles 4 of which you can access for free

Citations

cleshttpwwwneurologyorgcontent6361065fullhtmlotherartiThis article has been cited by 9 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionrestless_legs_syndromeRestless legs syndrome

httpwwwneurologyorgcgicollectionendocrineEndocrine

httpwwwneurologyorgcgicollectionall_sleep_disordersAll Sleep Disordersfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

22 Sun ER Chen CA Ho G Earley CJ Allen RP Iron and restless legssyndrome Sleep 199821371

23 Coats P Walter E Youssefnejadian E Craft IL Variations in plasmasteroid and prostaglandin concentrations during human pregnancyActa Obstet Gynecol Scand 197756453ndash457

24 Wetter TC Collado-Seidel V Oertel H et al Endocrine rhythms inpatients with restless legs syndrome J Neurol 2002249146ndash151

25 Euvrard C Oberlander C Boissier JR Antidopaminergic effect of estro-gens at the striatal level J Pharmacol Exp Ther 1980214179ndash185

26 Ogawa S Chan J Gustafsson JA Korach KS Pfaff DW Estrogenincreases locomotor activity in mice through estrogen receptor alphaspecificity for the type of activity Endocrinology 2003144230ndash239

27 Morgan MA Pfaff DW Estrogenrsquos effects on activity anxiety and fearin two mouse strains Behav Brain Res 200213285ndash93

28 Becker JB Beer ME The influence of estrogen on nigrostriatal dopa-mine activity behavioral and neurochemical evidence for both pre- andpostsynaptic components Behav Brain Res 19861927ndash33

29 Bedard PJ Malouin F Dipaolo T Labrie F Estradiol TRH and striataldopaminergic mechanisms Prog Neuropsychopharmacol Biol Psychia-try 19826555ndash561

30 Roberts JM Pregnancy related hypertension In Creasy RK Resnik Ret al Maternal-Fetal Medicine Principle and Practice 2nd ed Phila-delphia WB Saunders 1989783ndash808

31 Santiago JR Nolledo MS Kinzler W Santiago TV Sleep and sleepdisorders in pregnancy Ann Intern Med 2001134396ndash408

32 Walters AS Wagner M Hening WA Periodic limb movements as theinitial manifestation of restless legs syndrome triggered by lumbosacralradiculopathy Sleep 199619825ndash826

CME TAKE ADVANTAGE OF NEUROLOGY CME ONLINEThe online version of Neurology includes a continuing medical education (CME) component that is only available tosubscribers The AAN designates this educational activity for up to 72 hours in Category 1 towards the AmericanMedical Association (AMA) Physicianrsquos Recognition Award (3 hours per completed online issue)

Using this system subscribers canbull Take the quizzes onlinebull Receive instant feedback on their selected answersbull Submit completed issue quizzes for CME credit and receive confirmation of credit immediately via emailbull Review all the credits theyrsquove earned via Neurology Onlinersquos individualized CME Summary Report

September (2 of 2) 2004 NEUROLOGY 63 1069

DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

ServicesUpdated Information amp

httpwwwneurologyorgcontent6361065fullhtmlincluding high resolution figures can be found at

References

httpwwwneurologyorgcontent6361065fullhtmlref-list-1at This article cites 28 articles 4 of which you can access for free

Citations

cleshttpwwwneurologyorgcontent6361065fullhtmlotherartiThis article has been cited by 9 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionrestless_legs_syndromeRestless legs syndrome

httpwwwneurologyorgcgicollectionendocrineEndocrine

httpwwwneurologyorgcgicollectionall_sleep_disordersAll Sleep Disordersfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

DOI 10121201WNL000013842783574A62004631065-1069 Neurology

M Manconi V Govoni A De Vito et al Restless legs syndrome and pregnancy

This information is current as of September 27 2004

ServicesUpdated Information amp

httpwwwneurologyorgcontent6361065fullhtmlincluding high resolution figures can be found at

References

httpwwwneurologyorgcontent6361065fullhtmlref-list-1at This article cites 28 articles 4 of which you can access for free

Citations

cleshttpwwwneurologyorgcontent6361065fullhtmlotherartiThis article has been cited by 9 HighWire-hosted articles

Subspecialty Collections

httpwwwneurologyorgcgicollectionrestless_legs_syndromeRestless legs syndrome

httpwwwneurologyorgcgicollectionendocrineEndocrine

httpwwwneurologyorgcgicollectionall_sleep_disordersAll Sleep Disordersfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpwwwneurologyorgmiscaboutxhtmlpermissionsor in its entirety can be found online atInformation about reproducing this article in parts (figurestables)

Reprints

httpwwwneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online