Effect of Growth Hormone Treatment on Sleep EEGs in Growth Hormone-Deficient Children

Transcription

1 Sleep 11(5): , Raven Press, Ltd., New York 1988 Association of Professional Sleep Societies Effect of Growth Hormone Treatment on Sleep EEGs in Growth Hormone-Deficient Children Richard Hk. Wu and Michael J. Thorpy Department of Pediatrics, Bronx Lebanon Hospital Center and the Sleep-Wake Disorders Center, Montefiore Hospital and Medical Center, Albert Einstein College of Medicine, Bronx, New York, U.S.A. Summary: Total sleep time, sleep stages 1-4, REM, REM latency, and sleep efficiency were analyzed in seven children with growth hormone deficiency (GHD) before and after growth hormone (GH) therapy. Before GH therapy, GHD children spent 19.5% oftheir total sleep time in REM sleep, 9.7% in stage 1,41.0% in stage 2, 10.0% in stage 3, and 19.7% in stage 4. GHD children had more stage 1 and 3 sleep and less REM as compared with age-matched normal children reported by Williams et a1. After GH therapy was initiated, six of the seven patients had decreases in the duration of stage 3 sleep, with an average decrease of 21.8 min. The difference between stage 3 sleep before and during GH treatment was significant, with a p value of < When the results were expressed as the percentage of the total sleep period, the difference was also significant, (10.0 ± 2.0 to 7.5 ± 3.1%, mean ± SD; p < 0.05). No other sleep parameters were significantly affected by GH therapy. The changes observed in stage 3 sleep, non-rem sleep, and the lack of any other changes in sleep before and after GH therapy have not been described before in GH-deficient children. They differ from studies in normal humans and animals which showed that REM sleep increased with administration of growth hormone. These differences suggest that GH deficiency is associated with a specific sleep EEG anomaly that is corrected in part by GH therapy. Key Words: Growth hormone deficiency-growth hormone therapy-sleep time-children. Growth hormone (GH) has an episodic, intradiem pattern of secretion related to the sleep-wake cycle (1-5). Growth hormone secretion is closely associated with slow wave sleep, stage 3, and stage 4. When sleep is acutely disrupted and sleep occurs during daytime hours, GH is secreted during the daytime sleep period rather than during the usual night hours (6,7). Data obtained in animals suggest that bovine GH Accepted for publication April Parts of this work were presented at the 7th International Congress of Endocrinology, Quebec City, Canada, July Address correspondence and reprint requests to Richard Hk. Wu at Department of Pediatrics, Bronx Lebanon Hospital Center, 1650 Grand Concourse, Bronx, NY

2 426 R. HK. WU AND M. 1. THORPY increases rapid eye movement (REM) sleep in rats and cats with intact hypothalamicpituitary function (8,9). Orr and coheagues anaiyzed sieep in chiidren with GH deficiency (GHD) who were not on treatment and found that children aged 6-8 years had more slow wave sleep than did children aged years (10). The present study assessed the acute effects of exogenous human GH on sleep stages, sleep efficiency, REM latency, and total sleep time in children with GHD. PATIENTS AND METHODS Seven prepubertal patients with GHD, diagnosed consecutively over a 4-year period, were studied at the Clinical Research and Sleep-Wake Disorder Centers at Montefiore Hospital and Medical Center. The study was approved by the Subcommittee on Human Research, and parental consent was obtained. Clinical data for the seven children are shown in Table 1. Five patients had idiopathic, isolated GHD. Patient B had septooptic dysplasia with unilateral microophthalmia and visual impairment. She had normal vision in the unaffected eye. Patient C had a craniopharyngioma and was studied 6 months after transcranial surgery and radiation therapy to the cranial vault. She had minimal visual impairment and took replacement hydrocortisone (10 mg/day) and levothyroxine (0.1 mg/day) orally. GHD was diagnosed on the basis of subnormal GH (:::;7.0 ng/ml) responses of two of the following tests: intravenous (Lv.) arginine-insulin (0.5 g/kg-o.1 V/kg), oral L-DOPA (250 mg) or intramuscular (i.m.) glucagon (0.1 mg/kg, maximum dose 5.0 mg). The sleep studies were conducted in the Sleep Laboratory as previously described (11). All patients had one adaptation night at the beginning of the study. Electroencephalogram (EEG), electrooculogram (EOG), and electromyogram (EM G) electrodes were placed on the patients prior to each night's study. Two pre-gh treatment polysomnographic studies were done before GH therapy was begun. The patients were then given human GH, provided by the National Hormone and Pituitary Program (NHPP). Lots A25, P8, Pll, and P25 were used. One international unit (IV) was given daily (i.m.) before bedtime, for 1-2 weeks before the polysomnographic studies were repeated (two successive nights). The GH dosage varied from 0.03 to 0.08 IV/kg because GH allocation by the NHPP was restricted and the maximum of 7 IU/week met the established guidelines (see Table 1). Reference materials for GH radioimmunoassay (RIA) were supplied by the NHPP. GH was assayed using the dextran-coated charcoal method (12). TABLE 1. Age Weight Pt. (yr, mo)/sex (kg) A B C D E F G 7,7/F ,3/F ,4/F ,0/M ,9/M ,8/F ,5/F 15.0 Clinical characteristics of children Height Dose-GHa Diagnosis (cm) (kg) (Medication) Isolated GHD Isolated GHD Septooptic dysfunction Craniopharyn (Cortisol, T4) Isolated GHD Isolated GHD Isolated GHD Isolated GHD GHD, growth hormone deficiency. All children were prepubertal. a Dose in kg/day for the acute treatment period, 1 IU/day. Subjects A-D were treated with NPA batches A25, P8, and PlI. Subjects E-G were treated with NHPP batch P25. Sleep, Vol. 11, No.5, 1988

3 EFFECT OF GROWTH HORMONE ON SLEEP 427 Sleep stages were scored using the method of Rechtschaffen and Kales (13). The records were scored anonymously, and the scorer had no knowledge of the patients' treatment status. The scores of the first night (adaptation night) were not used, and the scores of the second and third nights were averaged. The averages of the two pretreatment study night scores were compared with the averages of the posttreatment scores. Student's paired t tests and a two-tailed curve were used in the statistical analysis. In., analysis of the EEG, EOG, and EMG records, the following were considered: total sleep time (EEG-defined sleep period), sleep efficiency or time spent awake after sleep onset, REM latency (the time from sleep onset to first REM period), and time spent in each of the standard sleep stages (REM, 1, 2, 3, and 4). The data are presented as duration of time in each sleep stage as well as percentage of the total sleep period. RESULTS Total sleep time, REM sleep, and sleep stages 1-4 are shown in Table 2. The duration of stage 3 sleep (minutes) decreased significantly after GH therapy, p < Six of seven patients had decreases, and one (patient D) had no change. Together, the seven patients had a mean decline of 17 min. None of the other sleep parameters changed significantly. Table 3 shows the data as a percentage of the total sleep period. Data for normal age-matched males and females from the study of Williams and co-workers are included for comparison (14). Stage 3 sleep fell significantly after GH treatment, p < The same six patients who exhibited decreases in their duration of stage 3 sleep (minutes), also had decreases in the percentage of stage 3 sleep. Although stage 3 sleep decreased after GH, it did not return to the normal levels seen in the age-matched normal children. Our study patients had more stage 1, 3, 4, and sleep than did normal patients. Conversely, our patients had less REM and stage 2 sleep than did normal controls. Because the patients in this study served as their own controls, any differences between our patients and normal children should not affect the results. None of the children reported subjective changes in their sleep. DISCUSSION The effect of central nervous system (CNS) hormones and releasing factors on sleep have been studied with inconclusive results. In rats, thyrotropin-releasing hormone TABLE 2. Duration (min) of total sleep period, REM sleep and sleep stages 1-4 before/after GH Sleep stages Pt. Total sleep REM A 474/552 79/110 56/90 196/265 47/ B 527/557 84/92 67/47 246/260 53/36 74/110 C 524/ /61 196/168 59/35 99/83 D 674/ / /101 E 575/ / /29 65/125 F 933/ /177 34/69 275/ /191 G 878/946 72/69 79/ /345 70/49 283/259 Mean 655/ /82 262/263 66/49 137/135 ±SD /37 37/ /32 87/67 p value NS NS NS NS < NS GH, growth hormone. Sleep, Vol, 1l, No.5, 1988

4 428 R. HK. WU AND M. J. THORPY TABLE 3. Sleep parameters expressed as percentage of total sleep time beforelafter GH Pt. REM (%) A 16.7/20.0 B 16.6/18.6 C D E 26.9/25.0 F 27.6/19.2 G b Mean 19.5/17.9 ±SD 6.0/5.3 Normal c F 29.3 ± 4.8 M 27.3 ± / / / / / / / / ± ± 0.8 Sleep stages and / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / ± ± ± ± ± ± ± ± 3.5 GR, growth hormone. o p < 0,05. b Patient was 10 years 5 months, but there is no significant difference between the female groups aged 6-9 and years. C Normals for females and males aged 6-9 years. 14 (TRH) had no effect on REM sleep, whereas arginine vasotocin increased slow wave sleep (15,16). GH given to normal men had no effect on REM sleep at a dose of 2 IV/day, whereas 5 IV/day caused an increase in REM sleep (17). Studies with GHreleasing hormone (GH-RH) are of great interest although in one study it had no effect on sleep in normal men (18). To date, no data on the effect ofgh-rh in normal or GHD children have been reported. Our study demonstrates a significant change in a specific parameter of the sleep EEG in GHD children. The significance of these changes is unclear since little is known about the correlation of neurosecretory events and stages of sleep. Our data on sleep stages before GH therapy concur with those of Orr and colleagues for all stages of sleep in untreated GHD children (10). In addition, our data correlate well with the recent report by Taylor and Brook on untreated GHD children for sleep stages 2,3, and 4 (19). However, those investigators found less stage 1 (1 vs. 9.7%) and more REM ( vs. 19.5%). Our patients did not demonstrate any changes in REM after GH, unlike normal men who exhibiteq more REM when higher doses of GH (0.07 IV/kg) were used (17). Five of our patients received GH in the higher dosage range. The difference between our patients and normal men could be related to the impaired function of the hypothalamic-pituitary units in our patients. GH secretion during slow wave sleep has been described in detail by many investigators. Disruption of slow wave sleep can result in impaired GH secretion, leading some investigators to believe that slow wave sleep is the putative agent in GH release during sleep. Our data, however, demonstrate that exogenous GH may affect a specific phase of slow wave sleep, stage 3. If there is a causal relationship between slow wave sleep and GH release, our study may have demonstrated product feedback inhibition. Such a relationship has not been demonstrated in normal children. The theory of causality is weakened by the discovery of delta sleep-inducing peptide (DSIP) (20). When administered to dogs, DSIP caused increases in slow wave sleep (delta sleep) and GH release. Thus, neither GH nor slow wave sleep has any direct causal effect on each other but they may affect each other through another pathway. In summary, our study demonstrates that GH replacement may affect a specific Sleep. Vol. 11, No.5, 1988

5 EFFECT OF GROWTH HORMONE ON SLEEP 429 parameter of sleep EEG activity in GHD children. These preliminary findings warrant further investigation for confirmation and may lead to studies that will shed light on the relationship of neurosecretory events and sleep EEGs. ', Acknowledgment: We acknowledge the assistance of sleep technicians of the CRC and Sleep Wake Disorders Center; Nate Katz, Grace Amoye, and Bill Greenhut for laboratory analyses; Carmen Rosario for manuscript preparation; and Paul Saenger, M.D., for manuscript review. REFERENCES 1. Finkelstein JW, Roffwarg HP, Boyar RM, et ai, Age related change in the spontaneous secretion of growth hormone. J Clin Endocrinol Metab 1972;35: Finkelstein JW. "Early to bed" or sleep revisited. Am J Dis Child 1974;127: Sassin JF, Parker DC, Mace JW, et at. Human growth hormone release: relation to slow-wave sleep and sleep waking-cycles, Science 1969;165: Takahashi Y. Growth hormone secretion sleep: a review. In: Kawakami M, ed. Biological rhythms in neuroendocrine activity. Tokyo: Igak-Shoin, 1974: , Takahashi Y, Kipnis DM, Daughaday WH: "Growth hormone" secretion during sleep, J Clin Invest 1968;47: Sassin JF, Parker DC, Johnson LC, et ai. Effects of slow-wave sleep deprivation on human growth hormone release in sleep: preliminary study. Life Sci 1969;8: Guilhaume A, Benoit 0, Gourmelen M, et al. Relationship between sleep stage IV deficit and reversible hgh deficiency in psychosocial dwarfism. Pediatr Res 1982;16: Drucker-Colin RR, Spanis CW, Hunyadi J, et al. Growth hormone effects on sleep and wakefulness in the rat. Neuroendocrinology 1975;18: Stem WC, Jaldweic JE, Shabeshekowitz H, et al. Effects of growth hormone on sleep-wake patterns in cats. Horm Behav 1975;6: Orr WC, Vogel GW, Stahl ML, et ai. Sleep patterns in growth hormone deficient children and agedmatched controls: developmental consideration. Neuroendocrinology 1977;24: II. Wu RH, Boyar RM, Knight R, et al. Endocrine studies in a phenotypic girl with XY gonadal agenesis. J Clin Endocrinol Metab 1976;43: Albin J, Wu RH. Abnormal hypothalamic-pituitary function in polyostotic fibrous dysplasia. Clin Endoerinol (Oxford) 1981 ;14: Rechtschaffen A, Kales A, eds. A manual of standardized terminology, techniques and scoring systems for sleep stages of human subjects. Washington D.C.: US Government Printing Office, Williams RL, Karacan I, Hursch CJ. Encephalography of human sleep: clinical applications. New York: John Wiley, 1974: Pavel S, Petrescu M, Vicoleanu N. Evidence of central gonadotropin inhibiting activity of arginine vasotocin in the female mouse. Neuroendocrinology 1973;11: Mendelsohn WB, Gillin JC, Pisner G, et al. The effects of arginine vasotocin on sleep and behavior in the rat. Sleep Res 1980; Mendelsohn WB, Slater S, Gold P, et al. The effect of growth hormone administration on human sleep: a dose-response study. Bioi Psychiatry 1980;15: Garry P, Roussel R, Cohen B, et al. Diurnal administration of human growth hormone-releasing factor does not modify sleep and sleep-related growth hormone secretion in normal young men. Acta Endocrinol (Copenh) 1985;110: Taylor RJ, Brook cm. Sleep EEG in growth disorders. Arch Dis Child 1986;61: Takahashi Y, Kato N, Nakamura Y, et al. Delta sleep inducing peptide (DSIP), a preliminary report of effects of intraventricular infusion on sleep and GH secretion, and of plasma DSIP concentration in the dog and rat. In: Ito M, Tsukahara N, Kubota K, Yagi K, eds.lntegrative control functions of the brain, vol. 2. Amsterdam: Elsevier, 1980: Sleep, Vol. 11, No.5, 1988