Effect of Levothyroxine on Prolonged Nocturnal Sleep Time and Excessive

Sleep Medicine 12 (2011) 578–583

Contents lists available at ScienceDirect

Sleep Medicine

journal homepage: www.elsevier.com/locate/sleep

Original Article Effect of levothyroxine on prolonged nocturnal sleep time and excessive daytime somnolence in patients with idiopathic hypersomnia ⇑ Hideto Shinno a, , Ichiro Ishikawa a, Mami Yamanaka a, Ai Usui a, Sonoko Danjo a, Yasushi Inami b, Jun Horiguchi b, Yu Nakamura a a Department of Neuropsychiatry, Kagawa University School of Medicine, 1750-1 Ikenobe, Miki, Kita, Kagawa 761-0793, Japan b Department of Psychiatry, Shimane University Faculty of Medicine, 89-1 Enya, Izumo, Shimane 693-8501, Japan article info abstract

Article history: Objective: This study aims to examine the effect of levothyroxine, a thyroid hormone, on a prolonged noc- Received 20 December 2010 turnal sleep and excessive daytime somnolence (EDS) in patients with idiopathic hypersomnia. Received in revised form 18 February 2011 Methods: In a prospective, open-label study, nine patients were enrolled. All subjects met criteria for idi- Accepted 22 February 2011 opathic hypersomnia with long sleep time defined by the International Classification of Sleep Disorders, Available online 12 May 2011 2nd edition (ICSD-2). Subjects with sleep apnea syndrome, obesity or hypothyroidism were excluded. Sleep architecture and subjective daytime somnolence were estimated by polysomnography (PSG) and Keywords: Epworth Sleepiness Scale (ESS), respectively. After baseline examinations, levothyroxine (25 lg/day) Excessive daytime somnolence was orally administered every day. Mean total sleep time, ESS score at baseline were compared with Levothyroxine Idiopathic hypersomnia those after treatment (2, 4 and 8 weeks). Idiopathic hypersomnia with long sleep Results: Mean age of participants was 23.8 ± 13.7 years old. At baseline, mean total sleep time (hours) and time ESS score were 12.9 ± 0.3 and 17.8 ± 1.4, respectively. Mean total sleep times after treatment were Epworth Sleepiness Scale 9.1 ± 0.7 and 8.5 ± 1.0 h at 4 and 8 treatment weeks, respectively. Mean ESS scores were 8.8 ± 2.3 and International Classification of Sleep 7.4 ± 2.8 at 4 and 8 treatment weeks, respectively. One patient dropped out at the 2nd week due to poor Disorders effect. No adverse effects were noted. Conclusions: After treatment with levothyroxine for over 4 weeks, prolonged sleep time and EDS were improved. Levothyroxine was effective for hypersomnia and well tolerated. Ó 2011 Elsevier B.V. All rights reserved.

1. Introduction multiple sleep latency test (MSLT) reveals reduced mean sleep latency and less than two sleep onset rapid eye movement (REM) Idiopathic hypersomnia was formerly characterized as pro- sleep periods (SOREMPs). In contrast to narcolepsy, idiopathic longed sleep episodes, excessive sleepiness, or excessively deep hypersomnia lacks specific clinical features such as cataplexy and sleep, which lasted for over 6 months. The International Classifica- characteristic polysomnographic features indicating alterations in tion of Sleep Disorders, 2nd edition (ICSD-2) has separated idio- rapid eye movement (REM) sleep. pathic hypersomnia into two entities [1]. The two conditions are Previous reports involving cerebrospinal fluid (CSF) analyses in referred to as idiopathic hypersomnia with long sleep time and idiopathic hypersomnia patients have revealed that cell counts, that without long sleep time. The former is characterized by exces- cytology, and proteins were not altered [2]. A decrease in dopa- sive daytime somnolence (EDS), prolonged nocturnal sleep and dif- mine and indoleacetic acid in the CSF was identified in patients ficulty in awakening, and is considered to be polysymptomatic, with hypersomnia including narcolepsy and idiopathic hypersom- primary, essential idiopathic hypersomnia. The latter is, on the nia [3]. Another study demonstrated a dysregulation of the other hand, remarkable only for EDS, and appears to be dopamine system in narcolepsy and of the norepinephrine system monosymptomatic. In both subtypes of idiopathic hypersomnia, in hypersomnia [4]. While there have been reports on the pathologies of idiopathic hypersomnia, its pathogenesis has not Abbreviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Scale; been sufficiently discussed, and a strategy for its treatment has ICSD-2, International Classification of Sleep Disorders, 2nd edition; MSLT, multiple not been established. While narcolepsy is treated with psychostim- sleep latency test; REM, rapid eye movement; SOREMP, sleep onset rapid eye ulants for excessive daytime sleepiness and antidepressants for movement sleep period; T3, triiodothyronine; T4, thyroxine; TSH, thyroid-stimulating hormone (thyrotropin). cataplexy and abnormal REM sleep [5], psychostimulants such as ⇑ Corresponding author. Tel.: +81 87 891 2165; fax: +81 87 891 2168. methylphenidate are not effective for excessive daytime sleepiness E-mail address: [email protected] (H. Shinno). in most patients with idiopathic hypersomnia [2]. Naps are of no

1389-9457/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.sleep.2011.02.004 H. Shinno et al. / Sleep Medicine 12 (2011) 578–583 579 use because they are lengthy and not refreshing. A strategy treat- 16:30. The patients entered their rooms at 18:00. Room lights were ing EDS in patients with idiopathic hypersomnia has not yet been put off when patients required. Our staff checked and recorded the established, and investigation to identify an appropriate strategy time. Nocturnal PSG was measured for 12 h (19:00 to 07:00). We for pharmacological intervention is necessary. performed overnight PSG by means of standard procedures that in- This study aims to investigate the effect of a thyroid hormone cluded recording a sleep electroencephalogram (C3-A2, C4-A1), on prolonged nocturnal sleep and excessive daytime somnolence bilateral eye movements, submental electromyography (EMG), an in patients with idiopathic hypersomnia. It is well known that pa- electrocardiogram, pulse oximetry, bilateral tibialis anterior EMG, tients with hypothyroidism usually exhibit daytime sleepiness. nasal air flow by a pressure sensor, as well as rib cage and abdom- Sleep apnea and its related arousal at night may cause reduction inal excursions. The sleep stage was scored according to standard in quality of nocturnal sleep and daytime somnolence. Therefore, criteria [8]. Sleep efficiency and the lengths of stages I, II, III, IV, patients with hypothyroidism or sleep apnea syndrome were ex- and REM were obtained independently. Periodic limb movements cluded. We previously reported two cases with latent hypothyroid- during sleep (PLMs) and the apnea-hypopnea index (AHI) were ism who presented prolonged daytime somnolence and EDS. They also estimated. were successfully treated with levothyroxine [6]. In the present To evaluate the sleep latency and sleep onset REM periods study, subjects with latent hypothyroidism were also excluded. (SOREMPs), the MSLT was performed following overnight PSG according to the standard guideline [9]. Sleep latency was calcu- 2. Methods lated from the results obtained by five sleep latency tests that were repeated at 2 h intervals (08:00, 10:00, 12:00, 14:00, and 16:00). A 2.1. Study design SOREMP was defined as the appearance of an epoch of REM sleep during the first 15 min of naps on the MLST. This study was a prospective, open-label study design to assess the therapeutic effect of levothyroxine. Data were collected be- 2.4.2. Evaluation of symptoms tween April 2008 and September 2010. Mean daily sleep time, daytime somnolence and symptom severity were evaluated at baseline and treated for 2, 4, and 2.2. Patients 8 weeks. Mean daily sleep time indicates nocturnal plus daytime sleep, and was calculated with sleep logs and interview. Values Nine patients with idiopathic hypersomnia with long sleep time were means of 7 days till each evaluation point. The subjective were enrolled in this study. The diagnosis of idiopathic hypersom- daytime somnolence was determined by the Epworth Sleepiness nia was made according to the criteria established by ICSD-2 [1]. Scale (ESS) [10]. Patients were eligible if (i) they were aged <60 years old; (ii) 2 their body mass index was <25 kg/m ; (iii) they had not been trea- 2.4.3. Laboratory data ted for hypersomnia and had not been medicated with psychotro- Blood samples were collected before breakfast. Serum free T3, pic agents such as psychostimulants, narcotics and free T4, and TSH were evaluated. antidepressants; (iv) no other drugs were prescribed during levothyroxine treatment; and (v) serum tri-iodothyronine (T ), thyrox- 3 2.4.4. Observation of adverse effect ine (T ) and thyrotropin (TSH) were within normal range. Patients 4 To examine whether adverse effects including symptoms due to were excluded for pregnancy or breast-feeding, for having contra- an alteration in thyroid function were present, we examined the indications to levothyroxine, for comorbidity with psychiatric dis- patients carefully at each visit, and laboratory data were also orders or medical illness. Psychiatrists diagnosed psychiatric examined. comorbidity using the Structured Clinical Interview for Diagnostic and Statistical Manual for Mental Disorders, 4th edition (DSM-IV) [7]. To exclude medical disorders, clinical interview and laboratory 2.5. Data analysis examinations were carried out. Patients were also excluded from this study if the baseline polysomnography demonstrated the exis- To assess changes in scores on the mean daily sleep time and tence of other sleep disorders or a high apnea-hypopnea index the ESS score, we used a Wilcoxon’s signed rank test. Calculation (AHI) (>10). was carried out with software PASW Statistics 18.0™. When the The local institutional research boards approved this study. All p value is less than 0.05, we considered the difference statistically patients gave informed consent according to institutional guide- significant. lines and the tenets of the Declaration of Helsinki. 3. Results 2.3. Treatment 3.1. The demography and baseline characteristics of subjects (Table 1) After the baseline examination, 25 lg/day of levothyroxine was administered in the morning. To examine whether adverse effects Nine patients were enrolled in this study (four males and five fe- including symptoms due to an alteration in thyroid function were males). All patients met criteria for idiopathic hypersomnia with present, we examined the patients carefully and laboratory data long sleep time. The mean age of diagnosis was 23.8 ± 13.7 years were also examined. old (14–59 years old). Prolonged nocturnal sleep and excessive daytime somnolence began in their teens, and the mean age of symp- 2.4. Measurements tom onset was 15.1 ± 1.1. The mean duration of hypersomnia was 8.1 ± 13.3 years (1.0–44.0 years). The mean body mass index was 2.4.1. Polysomnography and multiple sleep latency test 21.2 ± 2.5 kg/m2. The mean serum levels of fT3, fT4 and TSH were Each patient received a standardized evaluation including a 3.00 ± 0.42 pg/mL (normal range, 2.2–4.1 pg/mL), 1.17 ± 0.15 ng/ medical history, physical, and neurological examinations. At the mL (normal range, 0.88–1.81 ng/mL) and 1.71 ± 0.95 l-IU/mL (nor- baseline, polysomnography (PSG) was carried out following the mal range, 0.35–3.73 l-IU/mL), respectively. No subjects exhibited adaptation night. Electrodes for polysomnogram were attached at an altered thyroid function at baseline. 580 H. Shinno et al. / Sleep Medicine 12 (2011) 578–583

Table 1 The demography and baseline characteristics of the subjects.

Case number Mean ± SD 123456789 Gender Male Female Female Male Female Male Male Male Female Age of diagnosis (years) 22 14 59 21 17 25 17 22 17 23.8 ± 13.7 Age of symptom onset (years) 16 13 15 16 15 16 15 16 14 15.1 ± 1.1 Body mass index (kg/m2) 18.8 19.8 24.6 17.1 24.2 20.5 20.9 21.9 23.3 21.1 ± 2.5 Baseline daily sleep time Time in bed (hours) 14.6 16.0 15.4 14.2 16.3 15.1 15.7 14.0 15.1 15.1 ± 0.8 Total sleep time (hours) 13.0 13.3 12.8 12.6 13.2 12.8 13.1 12.4 12.6 12.9 ± 0.3 Nocturnal sleep time (hours) 10.8 11.1 11.0 11.1 10.8 10.9 11.0 10.6 10.7 10.9 ± 0.2 Daytime sleep time (hours) 2.2 2.2 1.8 1.5 2.4 1.9 2.1 1.8 1.9 2.0 ± 0.3 Epworth Sleepiness Scale score 18 19 17 16 19 20 18 17 16 17.8 ± 1.4 Thyroid function

Serum free T3 (pg/mL) 3.21 3.31 2.45 3.58 3.46 2.44 3.03 2.81 2.71 3.00 ± 0.42

Serum free T4 (ng/mL) 1.34 1.20 1.07 1.09 1.05 0.97 1.39 1.34 1.09 1.71 ± 0.15 Serum TSH (lIU/mL) 1.48 3.42 3.12 1.40 1.54 0.77 1.57 1.43 0.66 1.71 ± 0.95 Blood pressure Systolic/diastolic (mmHg) 127/83 115/68 137/91 114/74 110/58 90/68 101/45 128/70 114/72 115 ± 14/70 ± 13 Pulse rate (bpm) 73 72 87 74 66 79 62 79 63 72.8 ± 8.2

Daily sleep time means the time of nocturnal and daytime sleep, and was calculated with sleep logs and interview. The normal ranges of serum free T3,T4 and TSH were 2.2– 4.1 pg/mL, 0.88–1.81 ng/mL and 0.35–3.73 lIU/mL), respectively.

Table 2 Polysomnography and multiple sleep latency test.

Case number Mean ± S.D. 123456789 Nocturnal polysomnography Time in bed (min) 705 699 700 671 667 702 688 707 679 691 ± 15 Total sleep time(TST) (min) 669.0 662.7 656.7 628.3 620.3 657.0 643.0 656.7 636.0 648 ± 17 Sleep efﬁciency (%) 94.9 94.8 93.8 93.6 93.0 93.6 93.5 92.9 93.7 93.8 ± 0.7 Sleep latency (min) 8.7 8.3 9.7 8.3 8.7 9.3 9.7 11.3 10.7 9.4 ± 1.1 REM sleep latency (min) 71 73 89 68 92 94 86 68 71 79 ± 11 Stage I (%TST) 7.9 9.2 11.1 9.7 10.8 9.0 9.3 12.3 7.4 9.6 ± 1.6 Stage II (%TST) 50.5 52.1 51.3 48.4 48.9 47.8 46.1 44.7 50.5 48.9 ± 2.5 Stage III + IV (%TST) 24.8 19.6 19.4 24.3 20.5 21.0 24.7 25.4 23.1 22.5 ± 2.4 Stage REM (%TST) 16.8 19.1 18.2 17.6 19.8 22.3 19.9 16.6 19.0 18.8 ± 1.8 Apnea/hypopnea index(n/h) 2.2 3.5 8.2 1.9 3.2 4.1 2.7 2.8 2.3 3.4 ± 1.9 Periodic limb movement during sleep (n/h) 1.9 2.2 9.1 2.3 1.8 3.9 2.5 3.0 1.9 3.1 ± 2.3 Multiple sleep latency test Mean sleep latency(min) 6.3 6.3 6.7 7.3 6.0 5.7 6.7 7.3 7.0 6.6 ± 0.5 1st test 5.0 5.3 6.0 6.3 4.7 5.0 5.3 6.3 6.3 2nd test 6.0 6.3 5.7 7.3 6.0 6.0 6.7 7.7 7.0 3rd test 5.3 6.0 7.0 7.0 5.7 5.7 7.0 6.7 6.3 4th test 7.3 6.7 7.3 8.0 7.0 5.0 7.0 7.7 7.7 5th test 7.7 7.3 7.3 7.7 7.3 7.0 7.7 8.0 7.7

At baseline, polysomnography [8] and multiple sleep latency test [9] were carried out with standard guidelines.

Polysomnography was carried out at baseline. Their total sleep time at baseline was 12.9 ± 0.3 h. As defined by ICSD-2 criteria, all time and percentage of time spent in stages I, II, III + IV and REM patients had nocturnal sleep time more than 10 h. After treatment, are shown in Table 2. Mean REM sleep latency was 79 ± 11 min. the mean daily sleep time was 11.0 ± 1.4, 9.1 ± 0.7 and 8.5 ± 1.0 at Mean AHI was 3.4 ± 1.9. There were no subjects with AHI over 2nd, 4th and 8th week, respectively (Fig. 1). When compared with 10. MLST revealed that mean sleep latency was 6.6 ± 0.5 min, and baseline, reductions in nocturnal sleep after treatment were that all subjects had fewer than two sleep onset REM sleep periods statistically significant (p = 0.012 at 2nd week, p = 0.008 at 4th (SOREMPs; Table 2). week and p = 0.008 at 8th week). Proportion of patients whose 3.2. Outcome after levothyroxine treatment nocturnal sleep exceeded 10 h/day was 67%, 11% and 11% at 2nd, 4th and 8th week, respectively. The values include one dropped- Levothyroxine (25 lg/day) was administered orally once a day out patient. in the morning. Of nine subjects eight patients completed the 8 week study. One patient (case 4) dropped out, because treatment 3.2.2. The daytime somnolence with levothyroxine for 16 days failed to improve prolonged noctur- The mean ESS score was 17.8 ± 1.4 at baseline. After treatment, nal sleep and EDS. We analyzed daily sleep time and EDS of nine the mean ESS score was 12.8 ± 3.7, 8.8 ± 2.3 and 7.4 ± 2.8 at 2nd, cases including one dropped-out case. 4th and 8th week, respectively (Fig. 2). The values include one dropped-out patient. When compared with baseline, reductions 3.2.1. The daily sleep time in ESS score after treatment were statistically significant Time in bed, total sleep time, nocturnal sleep time and daytime (p = 0.017 at 2nd week, p = 0.008 at 4th week and p = 0.008 at sleep time at baseline were shown in Table 1. The mean daily sleep 8th week). H. Shinno et al. / Sleep Medicine 12 (2011) 578–583 581

mean daily sleep time (min) nocturnal sleep started when they were 13–16 years old and con- tinued for more than one year (1–44 years). Their PSG and MLST revealed a short sleep latency, a prolonged sleep time, a normal apnea/hypopnea index, and a normal REM sleep latency. Their labo-

ratory examinations revealed the normal serum levels of free T3 and free T4, and TSH. They met the ICSD-2 diagnostic criteria for idiopathic hypersomnia with long sleep time [1]. However, we adopted the sleep log and interview to calculate daily sleep time, and concluded that their daily and nocturnal sleep times were prolonged. Subjectively reported total sleep time may not necessarily be accurate to distinguish total sleep time from wake time in bed. We consider that actigraphy may be helpful to give a different account of total sleep time than a subjective report [11]. 0 2 4 8 The pathophysiology of idiopathic hypersomnia has not been treatment days (week) sufficiently understood, and the strategy for its treatment has not been established. In this study, the effect of levothyroxine (a thy- Fig. 1. Effect of levothyroxine on prolonged daily sleep time. Daily sleep time was calculated with each patient’s sleep log and interview, which include nocturnal and roid hormone) on prolonged nocturnal sleep and EDS was investi- daytime sleep. Mean daily sleep time at baseline was compared with that after gated. While mean sleep time and EDS began to reduce in the 2nd treatment for 2, 4, and 8 weeks. Mean daily sleep time shortened significantly after treatment week, nocturnal sleep times still exceeded 10 h for most treatment (p = 0.012 at 2nd week, p = 0.008 at 4th week and p = 0.08 at 8th week). of the patients. After treated for over 4 weeks, mean nocturnal While sleep time began to reduce at 2nd week, 6 of 9 patients still slept over 10 h/ sleep time was less than 10 h and EDS was also reduced in all sub- day. Values obtained from one dropped-out patient were indicated with dotted line. Data were analyzed by Wilcoxon’s signed-rank test. jects, which did not meet criteria for idiopathic hypersomnia. We demonstrated that treatment with 25 lg of levothyroxine for over 4 weeks improved prolonged nocturnal sleep and EDS, and levothyroxine was well tolerated. ESS score There have been several studies that investigated the association between hypothalamo–pituitary–thyroid (HPT) axis and alertness. Thyrotropin releasing hormone (TRH) has shown to be distributed widely in the CNS, and its receptors are reported to ex- ist in structures such as pituitary, cortex, brainstem, thalamus, hip- pocampus, amygdala, and spinal cord [12]. Besides its role in stimulating the release of thyroid stimulating hormone (TSH) and prolactin, TRH has been shown to exhibit various neuromodulating effects that are separate from its hormonal effects [13]. These effects include CNS stimulant and antidepressant effects and neuro- trophic effects. The clinical application of exogenous TRH, however, appears to be greatly limited because of a short biological half-life and limited access to the CNS. Therefore, biologically-stable TRH 02 4 8 analogs have been developed for possible clinical application. Sev- treatment days (week) eral reports have demonstrated the association between TRH and alertness in narcolepsy [14], while the association remains to be Fig. 2. Effect of levothyroxine on excessive daytime somnolence. Daytime somnolence was evaluated using Epworth Sleepiness Scale (ESS) [10]. The ESS scores at unclear in other diseases such as idiopathic hypersomnia and sleep baseline were compared with that after treatment for 2, 4, and 8 weeks. There were apnea syndrome. There have been reports that investigated the ef- significant reductions in the ESS scores after treatment (p = 0.017 at 2nd week, fect of TRH analogs in narcoleptic dogs. Acute and chronic oral p = 0.008 at 4th week and p = 0.08 at 8th week). Values obtained from one dropped- administration of CG-3703 (a TRH analog) was demonstrated to out patient were indicated with dotted line. Data were analyzed by Wilcoxon’s significantly reduce daytime sleep as well as cataplexy [14]. The ef- signed-rank test. fect of CG-3703 was also demonstrated to appear rapidly, while the effect of levothyroxine required about a month in our study. TRH and TRH analogs are also known to enhance dopaminergic transmission in the nucleus accumbens, which is important for locomo- 3.3. The adverse events tor activation and arousals [15]. It is possible that the effect of TRH on sleep and wakefulness may be mediated by enhancement of We observed the physical condition and examined adverse dopamine turnover, which is a common mechanism for most events at every visit. No subjects exhibited subjective and objective CNS stimulants [16]. TRH analogs could be beneficial for excessive adverse events. The baseline systolic/diastolic blood pressures and daytime sleepiness, but evidence has not accumulated in other pulse rates are indicated in Table 1. There were no differences in types of hypersomnia such as idiopathic hypersomnia. In addition, systolic (p = 0.285) and diastolic blood pressure (p = 0.188) as well previous studies demonstrated that oral administration of TRH as pulse rate (p = 0.439). analog did not cause significant changes in serum T3,T4 and TSH, and concluded that the effect may be independent of its effect on 4. Discussion the thyroid system [14,17]. On the other hand, it may be argued whether administration of levothyroxine induces suppression of All patients had complained of excessive daytime sleepiness TRH or TSH in subjects in the present study. Although data on thy- and prolonged nocturnal sleep (over 10 h). They had not felt roid function after treatment are available only in three subjects refreshed after naps. It had been difficult for them to wake up (data not shown) and are not sufficient for comparing with that in the morning. Excessive daytime sleepiness and prolonged at baseline, TSH as well as T3 and T4 after treatment were within 582 H. Shinno et al. / Sleep Medicine 12 (2011) 578–583 normal ranges in three patients. We consider two possible reasons. in somnological or hormonal properties between responders and Dose of levothyroxine was so low that it induced suppression of non-responders. TSH. Another is that patients with idiopathic hypersomnia may exhibit the alteration in HPT axis. Precise mechanism remains to be elucidated, and further investigations are necessary. Conflict of Interest Studies have also indicated the interaction between HPT axis and neuronal transmission systems that influence alertness. Cere- The ICMJE Uniform Disclosure Form for Potential Conflicts of brospinal fluid (CSF) homovanillic acid, a major metabolite of Interest associated with this article can be viewed by clicking on the following link: doi:10.1016/j.sleep.2011.02.004. dopamine, was negatively correlated with plasma TSH and T3 in healthy humans [18], which indicated the physiological signifi- cance of the interaction between dopamine and thyroid in central nervous system at the normal, euthyroid human. As clinical and Acknowledgement experimental data have suggested, thyroid hormone influences the central neurochemical systems, and may also affect the We certify that there was no financial support for this study. wake-promoting systems. The pathogenesis of idiopathic hypersomnia is unknown. In idiopathic hypersomnia the involvement References of alterations in HPT axis remains to be elucidated, while neuroen- docrinological investigations have been performed in narcolepsy [1] American Academy of Sleep Medicine. International classification of sleep [14]. disorders, 2nd edition: diagnostic and coding manual. Westchester, Illinois: American Academy of Sleep medicine, 2005. The histaminergic tuberomammillary nucleus of the hypothala- [2] Bassetti CL, Pelayo R, Guilleminault C. Idiopathic hypersomnia. In: Kryger M, mus [19] and the noradrenergic locus coeruleus of the pons [20] Roth T, Dement W, editors. Principles and practice of sleep medicine, 4th ed. are major wakefulness-promoting nuclei, and also play a role in Philadelphia: Elsevier Saunders; 2005. p. 791–800. [3] Montplaisir J, de Champlain J, Young SN, Missala K, Sourkes TL, Walsh J, et al. autonomic regulation. On the other hand, thyroid hormones have Narcolepsy and idiopathic hypersomnia: biogenic amines and related been reported to serve as co-transmitters with some neurotrans- compounds in CSF. Neurol 1982;32:1299–302. mitters such as histamine and noradrenalin [21]. Levothyroxine [4] Faull KF, Thiemann S, King RJ, Guilleminault C. Monoamine interactions in administration has been indicated to raise the histamine level of narcolepsy and hypersomnia: a preliminary report. Sleep 1986;9:246–9. [5] Mitler MM, Hajdukovic R, Erman M, Koziol JA. Narcolepsy. J Clin Neurophysiol the hypothalamus and cerebral cortex [22]. Previous reports have 1990;7:93–118. suggested that T3 is localized in brain nuclei receiving strong nor- [6] Shinno H, Inami Y, Inagaki T, Kawamukai T, Utani E, Nakamura Y, et al. adrenergic innervation [23] and concentrated in both noradrener- Successful treatment with levothyroxine for idiopathic hypersomnia patients with subclinical hypothyroidism. Gen Hosp Psychiatry 2009;31(2):190–3. gic centers and noradrenergic projection sites [21]. It is possible [7] American Psychiatric Association. Diagnostic and Statistical Manual of that the altered thyroid function or the serum TSH level may affect Mental Disorders. 4th ed. American Psychiatric Association. Washington, the arousal-promoting system. D.C.; 1994. [8] Rechtschaffen A, Kales A. A Manual of standardized terminology, techniques, To discriminate the influences by sleep-disordered breathing, and scoring system for sleep states of human subjects. Washington, DC: USPHS we excluded the subjects with hypothyroidism, sleep apnea syn- Publication No. 204. US Government printing Office; 1968. drome and obesity. The subjects with high BMI (over 25 kg/m2) [9] Carskadon MA, Dement WC, Milter MM, Roth T, Westbrook PR, Keenan S. Guidelines for the Multiple Sleep Latency Test (MSLT), a standard measure of or high AHI (over 10) were excluded in this study. Myxoedema is sleepiness. Sleep 1986;9:519–24. associated with sleep apnea. Several mechanisms have been pro- [10] Johns MW. A new method for measuring daytime sleepiness: the Epworth posed to explain the association between sleep apnea and hypo- sleepiness scale. Sleep 1991;14(6):540–5. [11] Kushida CA, Chang A, Gadkary C, Guilleminault C, Carrillo O, Dement WC. thyroidism. Patients with hypothyroidism commonly present Comparison of actigraphic, polysomnographic, and subjective assessment of with disordered breathing. Due to obstructive sleep apnea (OSA), sleep parameters in sleep-disordered patients. Sleep Med 2001;2(5): it is often difficult to maintain sleep continuity. Patients usually 389–96. complain of excessive daytime sleepiness, fatigue, snoring, de- [12] Winokur A, Manaker S, Kreider M. TRH and TRH receptors in the spinal cord. Ann New York Acad Sci 1989;553:314–24. creased libido, and impaired concentration. Several studies have [13] Kolesnick RN, Gershengorn MC. Thyrotropin-releasing hormone and the demonstrated that OSA is more prevalent in patients with hypo- pituitary: new insights into the mechanisms of stimulated secretion and thyroidism than control subjects [24–26]. Further examinations clinical usage. Am J Med 1985;79:729–39. [14] Riehl J, Honda K, Kwan M, Hong J, Mignot E, Nishino S. Chronic oral have focused on how strong hypothyroidism correlates to OSA. administration of CG-3703, a thyrotropin releasing hormone analog, Lin et al. also compared hypothyroidism patients with OSA and increases wake and decreases cataplexy in canine narcolepsy. without OSA, and suggested that age and body weight were related Neuropsychopharmacology 2000;23:34–45. [15] Momiyama T, Ishihara K, Kimura K, Todo N, Fujita Y, Serikawa T, et al. Long- to the development of OSA [24]. term antiepileptic effects of chronic intake of CNK-602A, a thyrotropin- To our knowledge, there have not been any studies that inves- releasing hormone analogue on spontaneously-epileptic rats. Epilepsia tigated the effect of thyroid hormone on symptoms presented in 1996;37(4):328–31. [16] Wisor JP, Nishino S, Sora I, Uhl GH, Mignot E, Edgar DM. Dopaminergic role in patients with idiopathic hypersomnia. The present study demon- stimulant-induced wakefulness. J Neurosci 2001;21(5):1787–94. strates the important and practical findings that had not been [17] Nishino S, Mignot E. Pharmacological aspects of human and canine narcolepsy. investigated in patients with idiopathic hypersomnia. Although a Prog Neurobiol 1997;52:27–78. [18] Strawn JR, Ekhator NN, D’Souza BB, Geracioti Jr TD. Pituitary–thyroid state lower dose of levothyroxine was prescribed and did not cause car- correlates with central dopaminergic and serotonergic activity in healthy diological or physiological alterations, it is necessary to determine humans. Neuropsychobiology 2004;49(2):84–7. carefully in every patient whether treatment with levothyroxine is [19] Steininger TL, Alam MN, Gong H, Szymusiak R, McGinty D. Sleep-waking appropriate. We fixed inclusion and exclusion criteria with careful discharge of neurons in the posterior lateral hypothalamus of the albino rat. Brain Res 1999;840:138–47. deliberations. Our study, however, has some limitations since it [20] Wu MF, Gulyani SA, Yau E, Mignot E, Phan B, Siegel JM. Locus coeruleus was an open-label trial. Further studies using a double-blind de- neurons: cessation of activity during cataplexy. Neuroscience sign or a crossover design with a larger sample size are recom- 1999;91(4):1389–99. [21] Dratman MB, Gordon JT. Thyroid hormones as neurotransmitters. Thyroid mended. Second, we have not investigated what proportions of 1996;6:639–47. the total patient with idiopathic hypersomnia respond to levothy- [22] Upadhyaya L, Agrawal JK. Effect of L-thyroxine and carbimazole on brain roxine administration because only nine subjects were enrolled in biogenic amines and amino acids in rats. Endocr Res 1993;19(2–3):87–99. [23] Dratman MB, Futaesaku Y, Crutschfield FL, et al. Iodine-125-labeled the present study. To realize the effect of levothyroxine in idio- triiodothyronine in rat brain: evidence for localization in discrete neural pathic hypersomnia, it may be helpful to elucidate the difference systems. Science 1982;215(4530):309–12. H. Shinno et al. / Sleep Medicine 12 (2011) 578–583 583

[24] Lin CC, Tsan KW, Chen PJ. The relationship between sleep apnea syndrome and [26] Misiolek M, Marek B, Namyslowski G, Scierski W, Zwirska-Korczala K, hypothyroidism. Chest 1992;102(6):1663–7. Kazmierczak-Zagorska Z, et al. Sleep apnea syndrome and snoring in [25] Jha A, Sharma SK, Tandon N, Lakshmy R, Kadhiravan T, Handa KK, et al. patients with hypothyroidism with relation to overweight. J Physiol Thyroxine replacement therapy reverses sleep-disordered breathing in Pharmacol 2007;58(S1):77–85. patients with primary hypothyroidism. Sleep Med 2006;7(1):55–61.