Ul.ie

Sleep-Disordered Breathing Is Increased in ObeseAdolescents with Craniopharyngioma Comparedwith Obese Controls Clodagh S. O’Gorman, Judith Simoneau-Roy, Paul Pencharz, Jamie MacFarlane, Ian MacLusky, Indra Narang, Khosrow Adeli, Denis Daneman, and Jill Hamilton* Context: Retrospective studies suggest that adolescents with craniopharygnioma and hypotha-
lamic obesity have increased sleep-disordered breathing (SDB).
Objectives: The objectives of this study were to compare the prevalence of SDB in adolescents with
craniopharyngioma-related obesity compared with body mass index (BMI)-matched controls and
to explore possible relationships between SDB, insulin resistance, and adipocytokines.
Design: This was a cross-sectional study of obese craniopharyngioma and obese control adolescents.
Setting: Subjects were evaluated in the clinical investigation unit at the Hospital for Sick Children,
Toronto.
Patients: Fifteen patients with craniopharyngioma-related obesity and 15 BMI-matched controls
were recruited and tested.
Interventions: Each subject underwent fasting blood work, frequent sampled iv glucose tolerance
test, polysomnography, and abdominal magnetic resonance imaging with calculation of visceral
and sc adipose tissue.
Main Outcome Measures: Main measures included insulin sensitivity, sleep efficiency, and
fragmentation.
Results: Insulin sensitivity was lower in craniopharyngioma subjects compared with control subjects
(0.96 Ϯ 0.34 vs. 1.67 Ϯ 0.7, P ϭ 0.01). Sleep-onset latency (19.3 Ϯ 27.8 vs. 31.9 Ϯ 23.4, P ϭ 0.03) and
oxygen saturations (rapid eye movement sleep: 89.0 Ϯ 5.1 vs. 94.2 Ϯ 2.3, P Ͻ 0.001; non-rapid eye
movement sleep: 88.4 Ϯ 5.6 vs. 94.3 Ϯ 1.5, P Ͻ 0.001) were lower in craniopharyngioma. Obstructive
apnea-hypopnea index (OAHI) (7.5 Ϯ 9.0 vs. 1.5 Ϯ 1.5, P ϭ 0.03) was higher in craniopharyngioma.
Respiratory distress index and OAHI correlated negatively with adiponectin concentrations (r ϭ
Ϫ0.61, P ϭ 0.03, r ϭ Ϫ0.71, P ϭ 0.006, respectively) in craniopharyngioma. On multiple regression,
TNF-␣ and craniopharyngioma were independent positive predictors of sleep-onset latency and
adiponectin and craniopharyngioma were significant predictors (negative and positive, respec-
tively) of OAHI.
Conclusions: SDB is increased in adolescents with craniopharyngioma-related obesity compared
with BMI-matched controls. Routine polysomnography should be considered in obese patients
with craniopharyngioma and appropriate treatment initiated. (J Clin Endocrinol Metab 95:
2211–2218, 2010)
ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviations: BMI, Body mass index; CAI, central apnea index; CPAP, continuous positive airway pressure; CV, coefficient of variation; FSIGT, insulin-modified frequent sampling iv Copyright 2010 by The Endocrine Society glucose tolerance test; MRI, magnetic resonance imaging; NREM, non-rapid eye move- doi: 10.1210/jc.2009-2003 Received September 18, 2009. Accepted February 26, 2010.
ment; OAHI, obstructive apnea hypopnea index; OSA, obstructive sleep apnea; PSG, poly- somnography; RDI, respiratory disturbance index; REM, rapid eye movement; SaO2, oxygen * Author Affiliations are shown at the bottom of the next page.
saturations; SAT, sc adipose tissue volume; SDB, sleep-disordered breathing; SDS, SD score;Si, insulin sensitivity; SOL, sleep onset latency; TST, total sleep time; VAT, visceral adiposetissue volume; WC, waist circumference.
J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 Sleep-disorderedbreathing(SDB),specificallyobstruc- fallingasleep,nocturnalwaking,andextremesleepiness
tive sleep apnea (OSA) is common, occurring in during the afternoon. Furthermore, adolescents with cra- 1– 4% of healthy children (1). Estimates of the prevalence niopharyngioma are at increased risk for metabolic ab- of OSA in obese children vary from 13 to 60%, depending normalities, including dyslipidemia, increased insulin re- on the definitions of OSA and obesity (2). Evidence sug- sistance, and the metabolic syndrome (16, 17), which gests that insulin resistance is positively associated with could exacerbate or contribute to SDB. Given the hypo- SDB in adolescents (3). A pediatric study showed a positive thalamic disturbances in craniopharyngioma patients, it is correlation between fasting insulin and the respiratory dis- possible that they exhibit more SDB than otherwise turbance index (RDI), a measure of sleep fragmentation (4). Furthermore, a retrospective review suggested an as-sociation between lower total sleep time (TST) and insulin The aims of this study were to compare incidence of Sleep is complex and its physiological regulation is mul- SDB and explore the relationships between SDB, Si, and tifactorial. The cytokines, IL-6 and TNF␣, are involved in inflammatory cytokines in adolescents with cranio- physiological sleep regulation and correlate positively pharyngioma and hypothalamic obesity and age- and with the obstructive apnea-hypopnea index (OAHI; a BMI-matched controls. We hypothesized that patients measure of sleep fragmentation) in obese adults (6). Ad- with craniopharyngioma would have more SDB than ministration of exogenous IL-6 to adults results in in- creased sleepiness and fatigue (7). IL-6 and TNF-␣ areelevated in adults with either SDB or obesity (8). The adi-pocyte-derived proteins (adipokines) leptin and adiponec- tin play a role in energy homeostasis and insulin sensitivity(Si) (9). Leptin is secreted in proportion to adipose mass Our institutional ethics review board approved this pro- and is elevated in obesity. However, hyperleptinemia fails tocol. Inclusion criteria for craniopharyngioma subjects to suppress appetite or increase metabolic rates, presum- included craniopharyngioma diagnosed at least 1 yr be- ably due to central nervous system leptin resistance (9).
fore study participation and adequate replacement of all Leptin is increased in pediatric OSA, independent of the pituitary hormone deficiencies. Controls, recruited from a body mass index (BMI) (10). Adiponectin, an insulin sen- weight management clinic, were otherwise healthy and sitizer, is decreased in obesity (9). To date, adiponectin age, BMI, and gender matched to craniopharyngioma sub- levels have not been correlated with SDB independently of jects. Exclusion criteria included: use of medications that obesity (10, 11). Furthermore, sleep abnormalities coexist might alter lipid levels, Si, or adiposity; respiratory ab- frequently with cardiovascular disease, may activate path- normalities that might preclude interpretation of sleep ways that cause or aggravate cardiovascular damage, or may studies; or known sleep abnormalities. Craniopharyngi- cause resistance to conventional antiobesity therapies (12).
oma subjects took their usual hormone replacement med- A primary hypothalamic function is to regulate the ications before study participation.
sleep-wake cycle. Hypothalamic damage, complicated by At study initiation, each subject was admitted to the obesity, temperature, and sleep dysregulation and pitu- clinical research center for physical examination with an- itary hormone deficiency, occurs commonly after treat- thropometrics, performed by a single examiner. Height ment for pituitary or hypothalamic tumors, including cra- and weight were measured using a wall-mounted stadio- niopharyngioma (13). The literature evaluating SDB in meter and a calibrated scale, respectively. BMI was cal- this population is sparse. One questionnaire-based study culated as weight (kilograms)/height2 (square meters).
reported frequent sleep abnormalities in craniopharyngi- BMI SD scores (SDS) were calculated using a program from oma (14). A prospective study in pediatric craniopharyn- the Centers for Disease Control and Prevention web site gioma included three patients and identified increased (18). Waist circumference SDS was calculated using pub- SDB by questionnaire and polysomnography (PSG) (15).
Our clinical experience is that patients with craniophar- Fasting blood samples were taken for baseline labora- yngioma and hypothalamic obesity complain commonly tory evaluation (including glucose, insulin, leptin, adi- of specific disturbed sleeping patterns, including difficulty ponectin, IL-6, TNF␣). Insulin-modified frequent sam- Divisions of Endocrinology (C.S.O., J.S.-R., D.D., J.H.) and Gastroenterology, Hepatology, and Nutrition (P.P.), Pediatric Laboratory Medicine (K.A.), Paediatric Sleep (I.M., I.N.), Divisionof Respiratory Medicine, Department of Pediatrics (C.S.O., J.S.-R., P.P., I.N., K.A., I.M., D.D., J.H.), The Hospital for Sick Children (C.S.O., J.S.-R., P.P., I.M., I.N., K.A., D.D., J.H.), Physiologyand Experimental Medicine Program (C.S.O., J.S.-R., I.N., J.H.), The Hospital for Sick Children Research Institute, Departments of Nutritional Sciences (P.P.) and Psychiatry (J.M.), Universityof Toronto (C.S.O., J.S.-R., P.P., J.M., I.N., K.A., D.D., J.H.), Toronto, Ontario, Canada M5G 1X8; and Children’s Hospital of Eastern Ontario (I.M.), Ottawa, Ontario, Canada K1H 8LI.
J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 pling iv glucose tolerance test (FSIGT) was performed abdominal effort for 20 sec or longer or of the duration after an overnight fast (20 –22). Via an iv antecubital can- of two prior baseline breaths in which case the event nula, 0.3 g/kg of 25% dextrose, followed 20 min later by must be accompanied by one of the following: 1) 3% or 0.03 U/kg of regular insulin (Humulin R; Eli Lilly, India- greater drop in oxygen saturation; 2) an arousal; or napolis, IN), was injected. Thirty-four blood samples for glucose and insulin were taken over 3 h from the con-tralateral arm. The mathematical model of Bergman (min- Definitions
imal model) (23) using MINMOD computer software OSA severity was graded using current clinically ac- (MINMOD Millennium, Version 5.18, MINMOD, Inc., cepted criteria according to the OAHI, the number of ob- Pasadena, CA) and data from FSIGT was used to calculate structive apneas, and obstructive hypopneas per hour dur- Si (minutesϪ1 per milliunitϪ1 per milliliterϪ1).
ing sleep. OAHI less than 1.5 was normal. OSA was During a second admission, each subject had overnight defined as: mild, OAHI greater than 1.5–5; moderate, PSG for evaluation of sleep abnormalities, supervised by greater than 5–10; and severe, greater than 10. CAI was trained pediatric sleep technologists in the sleep medicine defined as the number of central apneas per hour during laboratory. Every PSG was performed in a quiet, darkened sleep; more than 1.0 per hour was abnormal. SOL was room, without the use of sleep-inducing medications, with defined as the length of time in minutes from lights out to one parent present, and was interpreted by a single pedi- the first epoch of sleep. TST was defined as the sum of all the minutes spent in all stages of sleep or the total time in All PSGs were conducted and scored according to the bed minus all time spent awake during the study. Sleep American Academy of Sleep Medicine manual for scoring efficiency was defined as TST divided by the time in bed sleep and associated events (24) with a computerized sys- multiplied by 100. RDI was defined as the total number of tem (XL-TEK, Oakville, Ontario, Canada). A standard apneas, hypopneas, and arousals per hour. The arousal overnight PSG included a 4-lead electroencephalogram index was defined as the number of arousals per hour.
(C3, C4, O1, and O2), two bilateral electrooculogram Sleep fragmentation is associated with a high arousal in- leads referenced to A1 or A2, and one submental and two dex, increase in wakefulness during the night, and an in- tibial electromyograms. Respiratory measurements in- crease in stage 1 sleep (typically 5–10%).
cluded chest wall and abdominal movement using induc- Visceral and sc adipose tissue volumes (VAT and SAT) tance pneumography; airflow using a nasal cannula con- were measured in each subject using abdominal magnetic nected to a nasal pressure airflow; oxygen saturation resonance imaging (MRI) (General Electric Twin Speed EXCITETM III 12.0 1.5 Tesla; GE Healthcare, Milwaukee 2); and transcutaneous carbon dioxide measure- ments. Video and audio recordings were obtained for each WI), using previously published methodology (26).
study. Sleep architecture was assessed by standard tech-niques (25). Information obtained from each PSG in- Biochemical analyses
cluded sleep onset latency (SOL) and rapid eye movement Insulin was measured by chemiluminescence using the sleep (REM)-onset latency, TST, sleep efficiency, time Immulite 2500 [Siemens, Melvern, PA; range of assay 15– spent in each sleep stage (minutes and percentage), and the 2165 pmol/liter, intra- and interassay coefficient of vari- number of arousals. Recorded respiratory data included ation (CV) Ͻ 7.6%]. Leptin and adiponectin were mea- counts and indices of the following events: obstructive sured by ELISA (Diagnostics Systems Laboratories Inc., apneas and hypopneas (OAHI), central apneas [central Webster, TX; range 0.1–50.0 ng/ml; interassay CV 1.5– apnea index (CAI)], and mixed apneas recorded in non- 6.2%; and Linco Research Inc., St. Charles, MO; range REM (NREM) sleep, REM sleep, and total sleep.
1–100 mg/liter; CV 2.4 – 8.4%, respectively). IL-6 and An obstructive apnea event was scored when airflow TNF␣ were measured using chemiluminescence (Siemens dropped at least 90% from baseline with chest and/or Immulite 1000; range: 2–1000 pg/ml; CV Ͻ 7.5%; and abdominal motion throughout the entire event, the dura- range 1.7–1000 pg/ml; CV Ͻ 6.5%, respectively).
tion of which was at least a minimum of two baselinebreaths. An obstructive hypopnea event was scored when Statistical analysis
airflow dropped at least 50% from baseline, the duration Statistical analysis was performed using SAS software of which was at least a minimum of two baseline breaths.
(SAS version 8.2, 1999; Cary, NC). Continuous variables The event must have been accompanied by one of the fol- were expressed as means Ϯ SDs. Categorical variables were lowing: 1) 3% or greater drop in SaO2; 2) an arousal; or expressed as frequencies. Comparisons between groups 3) an awakening (24). A central apnea was defined as a were performed using the Fisher’s exact test and ␹2 anal- cessation of airflow with an absence of respiratory and ysis for categorical parameters and Student’s t test for con- J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 tinuous variables, with Bonferroni correction for multiple were treated with GH if they were both GH deficient and had comparisons. Data distribution was first assessed using growth failure and with sex steroid replacement therapy if simple measures of central tendencies and completed by they had hypogonadism biochemically and clinically. Of performing Shapiro-Wilk’s test for normality. Log trans- seven of 15 patients GH untreated, five of seven had normal formations were performed on nonnormative variables growth without GH, and two of seven had completed their and then compared by Student’s t test. Pearson correlation growth and were postpubertal at the time of diagnosis with analyses were performed between measures of SDB and craniopharyngioma.Fivepatientsreceivedhormonereplace- measures of Si and cytokines. Multiple regression analysis ment therapy for hypogonadotrophic hypogonadism. No was performed using factors significant on univariate re- gression, with only three variables entered (due to the The duration of follow-up from diagnosis of cranio- pharyngioma to assessment was 4.9 Ϯ 3.0 yr and corre-lated positively with both BMI (r ϭ 0.58, P ϭ 0.02) and BMI SDS (r ϭ 0.64, P Ͻ 0.01). All craniopharyngiomasubjects had evidence of hypothalamic damage on neu- Of 43 children who had surgery for craniopharyngioma roimaging after tumor resection. Time between each of and were followed up at our institution, 22 developed the two visits was 4.6 months (range 0 –11) in the cra- postoperative obesity, defined as BMI greater than the niopharyngioma group and 3.0 months (range 0 – 8) in 95th percentile for age and gender (13). Sixteen (77.3%) consented to participate; one was subsequently excluded After their initial PSG, two craniopharyngioma sub- after starting metformin therapy. Therefore, we enrolled15 children aged 10 –21 yr. No subject had received ra- jects required initiation of nocturnal continuous posi- diotherapy. We also enrolled 15 age-, sex-, and BMI- tive airway pressure (CPAP) therapy. Their untreated matched children with exogenous obesity. However, cra- PSG results are reported here. No control required niopharyngioma children had higher waist circumference (WC) and WC-SDS than controls (Table 1).
Si from FSIGT data was significantly lower in cranio- Craniopharyngioma subjects had multiple pituitary pharyngioma subjects compared with controls (0.96 Ϯ hormonal insufficiencies but were receiving adequate re- 0.34 vs. 1.67 Ϯ 0.7, P ϭ 0.01) (Table 1). TNF␣ was sta- placement therapies including: 14 of 15 treated with des- tistically higher in craniopharyngioma compared with mopressin for diabetes insipidus; 14 of 15 treated with controls, with no difference in IL-6 (Table 1). On abdom- hydrocortisone (dose range 8 –12 mg/m2 ⅐ d); 14 of 15 inal MRI, there were no differences between groups in the 14.5 Ϯ 2.6 pmol/liter); and eight of 15 treated with sc GH trols 480.3 Ϯ 185.5, P ϭ 0.77), VAT (craniopharyngi- (all for more than 12 months). Craniopharyngioma subjects oma, 85.1 Ϯ 40.5; controls, 67.5 Ϯ 29.5, P ϭ 0.25) or TABLE 1. Comparison of patient demographics and biochemical measures between groups
Demographics
Craniopharyngioma
P value
Si (FSIGT) (ϫ10Ϫ4/minϪ1 ⅐ ␮IU ⅐ ml) Data given are means Ϯ SD, except where indicated.
J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 TABLE 2. Sleep measures
Craniopharyngioma
P value
VAT to SAT ratios (craniopharyngioma, 0.18 Ϯ 0.08; in controls. RDI correlated negatively with adiponectin in controls, 0.16 Ϯ 0.09, P ϭ 0.38).
craniopharyngioma subjects (r ϭ Ϫ0.61, P ϭ 0.03) (Fig.
1) but did not show significant correlations in controls. In Sleep indices (Table 2)
craniopharyngioma subjects, after controlling for WC and SOL was significantly lower in craniopharyngioma and WC-SDS, respectively, adiponectin correlated well with TST trended toward longer, suggesting that craniophar- each of OAHI (r ϭ Ϫ0.70, P ϭ 0.006; r ϭ Ϫ0.71, P ϭ yngioma subjects fall asleep quicker and remain asleep 0.007), CAI (r ϭ Ϫ0.92, P ϭ 0.06; r ϭ Ϫ0.85, P ϭ 0.07), longer. OAHI was significantly higher in craniopharyn- and RDI (r ϭ Ϫ0.076, P ϭ 0.008; r ϭ Ϫ0.74, P ϭ 0.009).
gioma subjects. All 13 obese controls had either normal or TNF␣ predicted negatively and craniopharyngioma mild OAHI (i.e. fewer than five episodes per hour) com- predicted positively SOL (t ϭ Ϫ0.32, P Ͻ 0.001; t ϭ 1.0, pared with seven of 13 craniopharyngioma subjects. Of P ϭ 0.02, respectively). Adiponectin predicted negatively the remaining six craniopharyngioma subjects, two had and craniopharyngioma predicted positively OAHI (t ϭ moderate and four had severe OAHI. There was no sig- Ϫ4.5, P Ͻ 0.001; t ϭ 4.4, P Ͻ 0.001, respectively). Si was nificant difference in RDI, percentage time in slow wave not a significant predictor on multiple regression.
sleep, or sleep efficiency index between groups. MeanSaO2 was lower in craniopharyngioma than control in both REM and NREM sleep. Minimum SaO2 was lower in craniopharyngioma than controls. Craniopharyngiomaand control subjects spent similar times in stage 1 sleep, Our data suggest that adolescents with craniopharyngio- but craniopharyngioma subjects had significantly shorter ma-related obesity have more SDB than obese, otherwise percentage of time in stage 2 sleep.
healthy, BMI-matched adolescents. This is evidenced by Whereas there were no differences found between sub- increased OAHI and a trend toward increased CAI in the jects with craniopharyngioma who were GH treated or craniopharyngioma group, suggesting more sleep frag- GH untreated, in OAHI (GH treated, 3.9 Ϯ 5.6; GH un- mentation and less sleep efficiency. They also exhibit de- treated, 13.3 Ϯ 11.0, P ϭ 0.17) or RDI (GH treated, creased SOL and a trend toward increased TST, suggesting 14.0 Ϯ 10.5; GH untreated, 24.8 Ϯ 16.9, P ϭ 0.22), withthe relatively small number of subjects, it was unlikelysuch a difference could be identified.
SOL correlated positively with Si (r ϭ 0.79, P ϭ 0.03) and leptin (r ϭ 0.63 P ϭ 0.02) in the craniopharyngiomagroup. SOL showed no significant correlations in the con-trol group. TST correlated positively with IL-6 (r ϭ 0.55,P Ͻ 0.01) in the control group. OAHI correlated nega-tively with adiponectin and WC in craniopharyngiomasubjects (r ϭ Ϫ0.7, P ϭ 0.01, and r ϭ 0.7, P ϭ 0.03, FIG. 1. A, Plot of adiponectin against OAHI. B, Plot of adiponectin
respectively) (Fig. 1) but showed no significant correlation J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 more sleepiness in craniopharyngioma subjects. In partic- An additional mechanism that may contribute to SDB in ular, obese craniopharyngioma subjects exhibit more ob- craniopharyngioma subjects is sleep disturbance and cen- structive episodes and lower SaO2 during both REM and tral apnea due to hypothalamic damage. Several hypotha- NREM sleep than BMI-matched controls. These data sug- lamic nuclei, including the suprachiasmatic nuclei (which gest that the mechanism of obesity, not just the degree of control the circadian drive to wakefulness) (33) and the obesity, may affect SDB. TNF␣ and diagnosis of cranio- ventrolateral preoptic nuclei (which control the homeo- pharyngioma are independent positive predictors of static drive to sleep) (34), are responsible for regulation of shorter SOL. Additionally, lower adiponectin and cranio- sleep. The circadian rhythm is the result of a complex pharyngioma group predict higher OAHI.
interaction between circadian and homeostatic processes To our knowledge, adiponectin levels have not previ- (34). Specifically important in SDB and hypothalamic ously been found to correlate with SDB independently of damage, three sets of hypothalamic nuclei are situated BMI or other measures of adiposity. Physiologically, adi- close to each other: those involved in sleep and wake reg- ponectin potentiates sensitivity to insulin peripherally. Re- ulation; those responsible for the release of pituitary-con- cently adiponectin receptors have been identified in the trolling releasing hormones; and those that produce and periventricular nuclei and area postrema in mice models, store hormones released by the posterior pituitary. Thus, in which their central effects include reduction of body fat, damage to one of these nuclei groups may coexist with glucose, and lipids (27). Thus, adiponectin may play a role in SDB. IL-6 did not emerge as a significant predictor of Our study did not attempt to determine the physiolog- SDB on multiple regression analyses. This is consistent ical reasons for SDB in craniopharyngioma. Furthermore, with recent data suggesting that obesity, not SDB, is the the craniopharyngioma subjects in this study all had hy- major risk factor for insulin resistance (28), with a recent pothalamic damage based on MRI. In subjects with cra- study that found no difference in IL-6 between groups with niopharyngioma but without hypothalamic obesity, the similar BMI but differed in the presence or absence of SDB results of sleep studies may be different. Correlation of (29). We found higher TNF␣ levels in patients with more specific areas of hypothalamic damage to sleep parameters SDB (as measured by lower sleep onset latency.) The in- would be interesting, but the neuroimaging of our cranio- terrelationship between TNF␣ and SDB is complex, and pharyngioma subjects are not sensitive enough to define the cause and direction of the association are unclear (11).
which specific hypothalamic areas and nuclei are affected.
Sleep deprivation alone results in increased TNF␣ and This problem has been acknowledged previously (15). In IL-6, whereas TNF␣ is itself a hypnotic (30). Both SDB and the future, however, with advancing neuroimaging tech- obesity are inflammatory states associated with increased niques, identification of these regions may be easier.
TNF␣ levels. One study suggests that TNF␣ levels are Pituitary hormone deficiencies also potentially in- increased in SDB and that the level of TNF␣ correlates creased SDB in craniopharyngioma in this study. Al- positively with both the degree of sleepiness and the se- though all craniopharyngioma patients had adequate hor- verity of hypoxia (31). Our study supports these findings mone replacement, there are possibly differences between because we identified a significant positive relationship prescribed hormone replacement therapies and hormone between TNF␣ and shorter SOL in this obese population.
levels altered by autoregulation and feedback mecha- Obesity contributes potentially to SDB via several po- nisms. Also, the close proximity of sleep nuclei and pitu- tential mechanisms. First, both obesity and SDB are sys- itary hormone nuclei implies that damage to both nuclei temic inflammatory disorders. Second, impaired Si is as- groups might occur simultaneously. Further studies sociated with sleep disturbances (3). Third, increased sc should evaluate this. We did not identify any association tissue in the neck increases the risk of obstructive sleep between GH deficiency and SDB. Previous studies have apneas (32). Fourth, increased WC in the craniopharyn- identified longer TST and less deep sleep in GH-deficient gioma group, with implied increased intraabdominal fat patients and decrease in TST after GH replacement (35).
accumulation, could also lead to SDB. However, adi- Furthermore, GH replacement in GH-deficient individu- ponectin, when controlled for WC, still correlated signif- als has been associated with improved qualify of life (36).
icantly with measures of SDB, suggesting that factors It is possible that we did not find any associations because other than visceral adiposity mediate alterations in SDB GH was adequately replaced in many of the craniophar- via Si. It is possible that in patients with craniopharyngi- yngioma subjects. For those not receiving GH therapy but oma, adiponectin is affected more by dysregulated insulin who may have been GH deficient, it may have been too secretion through, for example, increased vagal tone, than early in their course to pick up major differences in sleep it is affected by visceral adiposity. All four etiologies could related to GH deficiency, although there were too few contribute to SDB in patients with hypothalamic obesity.
subjects in our study to realistically examine this question.
J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 A study of sleep abnormalities in five pediatric patients tion and cardiovascular risk as well as impact on SDB (12).
with hypothalamic-pituitary obesity (three of five had cra- Future studies to include evaluation of early endothelial niopharyngioma) and five age- but not BMI-matched con- dysfunction in relation to other metabolic and sleep pa- trols described more subjective sleepiness, more inefficient rameters should address this question.
sleep (as measured by stage 3– 4 sleep), and lower meansleep length in patients with hypothalamic obesity (15).
Conclusion
This study did not measure Si or adipokines (15). A ret- This study demonstrates that SDB is common among rospective case series, in which subjects were not matched patients with hypothalamic dysregulation and obesity af- with controls, identified increased TST and decreased ter craniopharyngioma treatment and that SDB in this SaO2 in subjects with central nervous system tumors (37).
group is more common than in an age-, gender-, and BMI- The worst sleep abnormalities were identified in subjects matched population. Two of 15 subjects required inter- with pituitary or hypothalamic abnormalities (37).
vention with CPAP after the sleep assessment, indicating The type of central nervous system tumor might impact an unrecognized sleep disorder. We suggest that clinicians the severity of SDB. For example, a study using question- should consider performing routine PSGs in all patients naire assessments of sleep identified severe sleepiness in with craniopharyngioma and hypothalamic dysregula- 35% of 79 children with craniopharyngioma and 15% of tion. Further multicenter studies will be required to eval- 19 children with pilocytic astrocytomas (14). No objective uate longitudinally the potential effects of SDB treatment assessment of sleep abnormalities was performed, and on weight gain, and metabolic parameters.
there was no healthy control group. Another question-naire study, in an adult population, identified increasedsubjective sleepiness and increased snoring in patients with either craniopharyngioma or nonfunctioning macro-adenomas, compared with healthy controls (38). Sleep ab- Address all correspondence and requests for reprints to: Dr. Jill normalities were not assessed objectively.
Hamilton, Hospital for Sick Children, 555 University Avenue, To our knowledge, this is the largest study to date ad- Toronto, Ontario, Canada M5G1X8. E-mail: [email protected].
dressing SDB in a pediatric population with craniophar- This work was supported by a Hospital for Sick Children yngioma and hypothalamic dysregulation. Furthermore, Research Institute seed grant. J.H. is supported through a Novo our craniopharyngioma and control populations were Nordisk Young Investigator Career Development award. J.S.-R.
age, gender, and BMI matched, suggesting that differences was supported by a fellowship from the Canadian Pediatric En- between the two groups are due to craniopharyngioma and hypothalamic damage. Additionally, we included cra- Present address for C.S.O.: Department of Pediatrics, Graduate niopharyngioma and control subjects without prior Entry Medical School, University of Limerick, Limerick, Ireland.
screening with sleep questionnaires. Sleep questionnaires Present address for J.S.-R.: Department of Pediatrics, Uni- have not been found to accurately reflect SDB as detected versity of Sherbrooke, Sherbrooke, Que´bec, Canada.
by sleep studies assessments (39). However, measure- Present address for I.M.: Department of Pediatrics, Chil- dren’s Hospital of Eastern Ontario, Ottawa, Canada.
ments of neck circumference, which is a validated corre- Disclosure Summary: The authors have no disclosures late of SDB in adult populations (32), might have helped to distinguish OSA related to oropharyngeal sc fat depo-sition from central sleep apneas. We acknowledge that oursample size was relatively small. However, craniopharyn-gioma with hypothalamic dysregulation is a rare disease, and this is one of the largest studies of this population 1. Lumeng JC, Chervin RD 2008 Epidemiology of pediatric obstruc-
tive sleep apnea. Proc Am Thorac Soc 5:242–252 Longitudinal sleep and metabolic studies will be re- 2. Verhulst SL, Schrauwen N, Haentjens D, Rooman RP, Van Gaal L,
De Backer WA, Desager KN 2008 Sleep duration and metabolic
quired to establish the onset, etiology, and causation of dysregulation in overweight children and adolescents. Arch Dis SDB in craniopharyngioma. For example, sleep studies performed before surgery and the postoperative period of 3. Redline S, Storfer-Isser A, Rosen CL, Johnson NL, Kirchner HL,
Emancipator J, Kibler AM 2007 Association between metabolic syn-
rapid weight gain (13) might identify the natural course of drome and sleep-disordered breathing in adolescents. Am J Respir this disorder. Craniopharyngioma patients are at in- creased of cardiovascular disease (40), and the coexistence 4. de la Eva RC, Baur LA, Donaghue KC, Waters KA 2002 Metabolic
correlates with obstructive sleep apnea in obese subjects. J Pediatr of SDB may further compound this risk. Common path- ways of inflammation may increase endothelial dysfunc- 5. Flint J, Kothare SV, Zihlif M, Suarez E, Adams R, Legido A, De Luca
J Clin Endocrinol Metab, May 2010, 95(5):2211–2218 F 2007 Association between inadequate sleep and insulin resistance
derstanding of glucose tolerance. Minimal-model approach. Diabe- in obese children. J Pediatr 150:364 –369 6. Imagawa S, Yamaguchi Y, Ogawa K, Obara N, Suzuki N,
24. Iber CA-I, Chesson A, Quan S 2007 The American Academy of Sleep
Yamamoto M, Nagasawa T 2004 Interleukin-6 and tumor necrosis
Medicine. The AASM manual for the scoring of sleep and associated factor-␣ in patients with obstructive sleep apnea-hypopnea syn- events: rules, terminology and technical specifications. Winchester, 7. Mastorakos G, Chrousos GP, Weber JS 1993 Recombinant inter-
25. Rechtschaffen AK 1968 A manual of standardised terminology: tech-
leukin-6 activates the hypothalamic-pituitary-adrenal axis in hu- niques and scoring system for sleep stages of human subjects: UCLA mans. J Clin Endocrinol Metab 77:1690 –1694 Brain Information Service. Los Angeles: Brain Research Institute 8. Vgontzas AN, Papanicolaou DA, Bixler EO, Hopper K, Lotsikas A,
26. Caprio S, Hyman LD, Limb C, McCarthy S, Lange R, Sherwin RS,
Lin HM, Kales A, Chrousos GP 2000 Sleep apnea and daytime
Shulman G, Tamborlane WV 1995 Central adiposity and its metabolic
sleepiness and fatigue: relation to visceral obesity, insulin resistance, correlates in obese adolescent girls. Am J Physiol 269:E118–E126 and hypercytokinemia. J Clin Endocrinol Metab 85:1151–1158 27. Ahima RS 2006 Adipose tissue as an endocrine organ. Obesity (Sil-
9. Coll AP, Farooqi IS, O’Rahilly S 2007 The hormonal control of food
28. Tauman R, O’Brien LM, Ivanenko A, Gozal D 2005 Obesity rather
10. Tauman R, Serpero LD, Capdevila OS, O’Brien LM, Goldbart AD,
than severity of sleep-disordered breathing as the major determinant Kheirandish-Gozal L, Gozal D 2007 Adipokines in children with
of insulin resistance and altered lipidemia in snoring children.
sleep disordered breathing. Sleep 30:443– 449 11. Capdevila OS, Kheirandish-Gozal L, Dayyat E, Gozal D 2008 Pe-
29. Nakra N, Bhargava S, Dzuira J, Caprio S, Bazzy-Asaad A 2008 Sleep-
diatric obstructive sleep apnea: complications, management, and disordered breathing in children with metabolic syndrome: the role of long-term outcomes. Proc Am Thorac Soc 5:274 –282 leptin and sympathetic nervous system activity and the effect of con- 12. Shamsuzzaman AS, Gersh BJ, Somers VK 2003 Obstructive sleep
tinuous positive airway pressure. Pediatrics 122:e634–642 apnea: implications for cardiac and vascular disease. JAMA 290: 30. Mullington JM, Haack M, Toth M, Serrador JM, Meier-Ewert HK
2009 Cardiovascular, inflammatory, and metabolic consequences 13. Ahmet A, Blaser S, Stephens D, Guger S, Rutkas JT, Hamilton J
of sleep deprivation. Prog Cardiovasc Dis 51:294 –302 2006 Weight gain in craniopharyngioma—a model for hypotha- 31. Ryan S, Taylor C, McNicholas WT 2006 Predictors of elevated
lamic obesity. J Pediatr Endocrinol Metab 19:121–127 nuclear factor-␬B-dependent genes in obstructive sleep apnea syn- 14. Mu¨ller HL, Handwerker G, Wollny B, Faldum A, So¨rensen N 2002
drome. Am J Respir Crit Care Med 174:824 – 830 Melatonin secretion and increased daytime sleepiness in childhood cra- 32. Davies RJ, Ali NJ, Stradling JR 1992 Neck circumference and other
niopharyngioma patients. J Clin Endocrinol Metab 87:3993–3996 clinical features in the diagnosis of the obstructive sleep apnoea 15. Snow A, Gozal E, Malhotra A, Tiosano D, Perlman R, Vega C,
Shahar E, Gozal D, Hochberg Z, Pillar G 2002 Severe hypersom-
33. Mignot E, Taheri S, Nishino S 2002 Sleeping with the hypothala-
nolence after pituitary/hypothalamic surgery in adolescents: clinical mus: emerging therapeutic targets for sleep disorders. Nat Neurosci characteristics and potential mechanisms. Pediatrics 110:e74 16. Simoneau-Roy J, O’Gorman C, Pencharz P, Adeli K, Daneman D,
34. Saper CB, Chou TC, Scammell TE 2001 The sleep switch: hypo-
Hamilton J 2009 Insulin sensitivity and secretion in children and
thalamic control of sleep and wakefulness. Trends Neurosci 24: adolescents with hypothalamic obesity following treatment for cra- niopharyngioma. Clin Endocrinol (Oxf) 72:364 –370 35. Astrom C, Pedersen SA, Lindholm J 1990 The influence of growth
17. Srinivasan S, Ogle GD, Garnett SP, Briody JN, Lee JW, Cowell CT
hormone on sleep in adults with growth hormone deficiency. Clin 2004 Features of the metabolic syndrome after childhood cranio- pharyngioma. J Clin Endocrinol Metab 89:81– 86 36. Peker Y, Svensson J, Hedner J, Grote L, Johannsson G 2006 Sleep
18. Kuczmarski RJ, Ogden CL, Guo SS, Grummer-Strawn LM, Flegal
apnoea and quality of life in growth hormone (GH)-deficient adults KM, Mei Z, Wei R, Curtin LR, Roche AF, Johnson CL 2002 2000
before and after 6 months of GH replacement therapy. Clin Endo- CDC growth charts for the United States: methods and develop- 37. Rosen GM, Bendel AE, Neglia JP, Moertel CL, Mahowald M 2003
19. Ferna´ndez JR, Redden DT, Pietrobelli A, Allison DB 2004 Waist
Sleep in children with neoplasms of the central nervous system: case circumference percentiles in nationally representative samples of review of 14 children. Pediatrics 112:e46 –54 African-American, European-American, and Mexican-American 38. van der Klaauw AA, Biermasz NR, Pereira AM, van Kralingen KW,
children and adolescents. J Pediatr 145:439 – 444 Dekkers OM, Rabe KF, Smit JW, Romijn JA 2008 Patients cured
20. Finegood DT, Hramiak IM, Dupre J 1990 A modified protocol for
from craniopharyngioma or nonfunctioning pituitary macroad- estimation of insulin sensitivity with the minimal model of glucose enoma (NFMA) suffer similarly from increased daytime somnolence kinetics in patients with insulin-dependent diabetes. J Clin Endo- despite normal sleep patterns compared to healthy controls. Clin 21. Cutfield WS, Bergman RN, Menon RK, Sperling MA 1990 The
39. Muzumdar H, Arens R 2008 Diagnostic issues in pediatric obstruc-
modified minimal model: application to measurement of insulin sen- tive sleep apnea. Proc Am Thorac Soc 5:263–273 sitivity in children. J Clin Endocrinol Metab 70:1644 –1650 40. Pereira AM, Schmid EM, Schutte PJ, Voormolen JH, Biermasz NR,
22. Conwell LS, Trost SG, Brown WJ, Batch JA 2004 Indexes of insulin
van Thiel SW, Corssmit EP, Smit JW, Roelfsema F, Romijn JA 2005
resistance and secretion in obese children and adolescents: a vali- High prevalence of long-term cardiovascular, neurological and psy- dation study. Diabetes Care 27:314 –319 chosocial morbidity after treatment for craniopharyngioma. Clin 23. Bergman RN 1989 Lilly lecture 1989. Toward physiological un-

Source: http://www.ul.ie/4i/sites/default/files/SDB%20Obese%20Adolescents.pdf