Metabolic Disregulation in Obese Adolescents with Sleep-Disordered Breathing Before and After Weight Loss

Transcription

1 Metabolic Disregulation in Obese Adolescents with Sleep-Disordered Breathing Before and After Weight Loss K. Van Hoorenbeeck 1, H. Franckx 2, P. Debode 2, P. Aerts 1, J. Ramet 1,3, L.F. Van Gaal 1,4, K.N. Desager 3, W.A. De Backer 1,5 and S.L. Verhulst 1,3 Objective: Sleep-disordered breathing (SDB) is prevalent in obesity. Weight loss is one of the most effective treatment options. The aim was to assess the association of SDB and metabolic disruption before and after weight loss. Design and Methods: Obese adolescents were included when entering an in-patient weight loss program. Fasting blood analysis was performed at baseline and after 4-6 months. Sleep screening was done at baseline and at follow-up in case of baseline SDB. Results: 224 obese adolescents were included. Median age was 15.5 years ( ) and mean BMI z- score was About 30% had SDB at baseline (N ¼ 68). High-density lipoprotein (HDL)- cholesterol was associated with mean nocturnal oxygen saturation (<SaO2>) (partial r ¼ 0.21; P ¼ 0.002). Aspartate aminotransferase (ASAT) and alanine aminotransferase were related with oxygen desaturation index (partial r ¼ 0.15; P ¼ 0.03 and partial r ¼ 0.15; P ¼ 0.02), but this became insignificant after correction for sex. After weight loss, 24% had residual SDB. Linear regression showed an association between ASAT and <SaO2> (partial r ¼ 0.34; P ¼ 0.002). There were no significant correlations between improvements in laboratory measurements and sleep parameters. HDL-cholesterol improved in relation with the decrease in BMI z-score. Conclusion: SDB at baseline was associated with higher levels of liver enzymes and lower HDLcholesterol concentration. Improvements in sleep parameters were not associated with improvements in laboratory measurements. VC 2013 American Institute of Chemical Engineers AIChE J, 2013 (2013) 21, doi: /oby Introduction Sleep-disordered breathing (SDB) is highly prevalent in obese children and adolescents. Prevalence rates are as high as 13-59%, depending on the methodology used (1). The prevalence of SDB is about 2% in the general pediatric population and is most commonly treated with adenotonsillectomy (2). However, adenotonsillectomy for SDB is associated with a high recurrence risk in obese children because of postoperative weight gain (3). A recent multicenter study reported residual SDB in 88% of obese participants after surgery (4). We previously showed that weight loss for SDB in the obese population has a high success rate of 76% (5). Weight loss can therefore be considered as a first line treatment in this specific context, also because it improves the other obesity-related complications. SDB in obese children and adolescents is associated with components of the metabolic syndrome (6,7). This effect is even dose dependent (6). However, the reported studies were cross sectional and there were several reports that showed no effect of SDB on components of the metabolic syndrome (8,9). In this study, we focus on the effects of SDB on factors of metabolic disregulation, such as altered insulin sensitivity, dyslipidemia, and elevation of liver enzymes, in obese children and adolescents. Secondly, we evaluate whether these effects are modified by weight loss as a treatment of SDB. For this purpose we extended the study population of our previous report (5). Patients and Methods Obese adolescents were recruited consecutively while entering an inpatient weight loss treatment program between 2007 and They were admitted at the rehabilitation center Zeepreventorium, De Haan (Belgium). The weight reduction program was a multicomponent 1 Laboratory of Experimental Medicine and Pediatrics, University of Antwerp, Antwerp, Belgium. Correspondence: K. Van Hoorenbeeck (kim.vanhoorenbeeck@ua.ac.be) 2 Program, Zeepreventorium, De Haan, Belgium 3 Department of Pediatrics, Antwerp University Hospital, Edegem, Belgium 4 Department of Endocrinology, Diabetology and Metabolism, Antwerp University Hospital, Edegem, Belgium 5 Department of Respiratory Medicine, Antwerp University Hospital, Edegem, Belgium Disclosure: The authors have no conflicts of interest to disclose. Received: 26 August 2012 Accepted: 10 December 2012 Published online 14 February doi: /oby VOLUME 21 NUMBER 7 JULY

2 TABLE 1 Patient characteristics at baseline visit Variable ODI < 2 ODI 2 P value N Sex (% males within < subcategory) Age (years) 15.4 ( ) 15.9 ( ) 0.3 BMI z-score ODI (events/h) 0.64 ( ) 2.94 ( ) < AHI (events/h) 0.80 ( ) 2.21 ( ) < <SaO2> (%) 97 (95-98) 96 (84-98) < SaO2 nadir (%) 92 (71-98) 87 (70-93) < ODI, oxygen desaturation index; BMI, body mass index; h, hour; AHI, apnea-hypopnea index; <SaO2>, mean nocturnal oxygen saturation; SaO2 nadir, nocturnal oxygen saturation nadir. treatment with moderate dietary restrictions ( kcal/day), increased physical activity (minimum of 10 h/week), psychological individual and group support and medical supervision. No anti-obesity drugs were prescribed. Exclusion criteria were neuromuscular or known endocrine disease, genetic or craniofacial syndromes, acute illness, uncontrolled chronic disease, or the use of anti-inflammatory drugs at the moment of the study. Visits were organized at baseline and after 4-6 months of weight loss treatment. The Ethics committee of the University of Antwerp (Antwerp, Belgium) approved this study (A07-01) and an informed consent was obtained from both the patients and their parents. Anthropometry Skilled personnel measured height and weight at admission according to standard techniques. Body mass index (BMI) was calculated and expressed as BMI z-score using an electronic calculator based on the Flemish growth charts. This was repeated after 4-6 months of treatment and the relative decrease in BMI z-score was determined ((baseline BMI z-score BMI z-score after therapy) / baseline BMI z-score, in percent). Sleep assessment Sleep assessment was performed using a portable device (Apnea- Link, ResMed, Switzerland), at baseline and after 4 to 6 months of therapy when SDB was observed during the first investigation (10-12). This device has been validated recently in a population of adolescents between 9 and 18 years old (13). Respiratory airflow was measured by a nasal pressure cannula (detecting 10 to þ10 hpa) and oxygen saturation and pulse rate were obtained by using a pulse oxymeter and a pulse sensor (sampling rate of 1 Hz). Tracings were all manually reviewed and measurements associated with poor pulse tracing or aberrant respiratory signals were excluded from the analysis. Sleep study was repeated in case of less than 4 h of good quality signal. Total sleep time was established as recording time minus signal artefacts. Apnea was defined as the absence of airflow lasting at least two breaths. Hypopnea was a 50% decrease in amplitude of the airflow signal, lasting for 2 breaths and with a concurrent desaturation of 4%. The apnea-hypopnea index (AHI) was calculated as the sum of the recorded apneas and hypopneas divided by the total sleep time. As ApneaLink is a screening device it is not possible to make the difference between obstructive and central events, or to recognize arousals. All desaturations of 4% from the baseline oxygen saturation (SaO2) were quantified and the oxygen desaturation index (ODI) was calculated as the total number of desaturations divided by the total sleep time. The mean nocturnal SaO2 (<SaO2>) and SaO2 nadir were registered. SDB was diagnosed in the presence of an ODI of 2 (14). Children with residual SDB after weight loss were advised to undergo full polysomnography. Laboratory measurements A fasting venous blood sample was drawn in the morning at the moment of admission. This was repeated after 4-6 months of intervention. Glycemia, insulin level, alanine aminotransferase (ALAT) and aspartate aminotransferase (ASAT), total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides were determined. Glycemia was determined by an enzymatic UV-test using the hexokinase method (Beckman Coulter, Switzerland). Insulin levels were obtained by electrochemiluminescence (Roche Diagnostics, Switzerland). ASAT and ALAT were both the result of a kinetic UV-test (Beckman Coulter, Switzerland). Analysis of lipids was done by an enzymatic UV-test (Beckman Coulter, Switzerland). HOMA-index (homeostasis model assessment) was calculated as the product of glucose (mg/dl) and insulin (lu/ml), divided by 405. Statistical analysis Statistical analysis was performed using SPSS 20 (IBM, US). Normality was tested using the Shapiro-Wilk test. Normally distributed data are presented as mean 6 standard deviation; whereas skewed data are reported as median and range. Comparison of different groups was done using Fisher s exact, Student t-test and Mann-Whitney U-test, as appropriate. Comparison of variables before and after intervention was performed with the Wilcoxon signed ranks test. Spearman or Pearson correlation coefficients were determined, as appropriate. Linear regression was performed with acknowledgement of possible confounders. Conversion of skewed data was done to obtain normality. All P values are two tailed and statistical significance is set at less than Results Baseline characteristics We included 224 adolescents between 2007 and Median age was 15.5 years ( ). Mean baseline BMI z-score was One third of the participants (N ¼ 75) were boys. Median interval between the baseline and follow-up visit was 5.1 months ( ). Absolute decrease in BMI z-score was 0.81 ( ), which corresponds to a relative decrease of about 30% (2-115). Sixty-eight participants (30%) had SDB at the baseline visit and were asked for a repeat sleep study after 4-6 months of weight loss. Baseline patient characteristics are summarized in Table 1. Patients of both groups were comparable regarding age and BMI z-score. Significantly more boys were found in the SDB group, i.e., 52% versus 26%. Baseline associations between sleep screening and laboratory parameters Comparison of the laboratory measurements is presented in Table 2. Important differences between the two study groups exist of higher VOLUME 21 NUMBER 7 JULY

3 Metabolic Disruption and SDB in Obese Adolescents Van Hoorenbeeck et al. TABLE 2 Laboratory measurements at baseline visit Variable ODI a < 2 ODI 2 P value Glycemia (mg/dl) 83 (57-105) 85 (70-114) 0.2 Insulin (lu/ml) 14 (4-42) 14 (7-53) 0.5 HOMA b -index 3.09 ( ) 2.89 ( ) 0.6 Total cholesterol (mg/dl) 150 (66-360) 143 (85-247) 0.05 HDL c -cholesterol (mg/dl) 42 (21-78) 40 (28-60) 0.1 Triglycerides (mg/dl) 78 (37-206) 73 (41-344) 0.3 ASAT d (mg/dl) 25 (14-227) 29 (14-223) 0.02 ALAT e (mg/dl) 25 (8-512) 30 (12-99) 0.02 a ODI: oxygen desaturation index. b HOMA: homeostasis model assessment. c HDL: high density lipoprotein. d ASAT: aspartate aminotransferase. e ALAT: alanine aminotransferase. aminotransferases in the SDB-group compared to the patients with a normal study. There is a trend towards higher levels of HDL-cholesterol and total cholesterol in the children and adolescents without SDB. When further dividing the participants with SDB in those with mild versus severe SDB (ODI 5) no significant differences of the metabolic parameters were observed. Spearman correlation analysis showed significant correlations for HDL-cholesterol and ODI (r ¼ 0.15; P ¼ 0.03), <SaO2> (r ¼ 0.21; P ¼ 0.002), and SaO2 nadir (r ¼ 0.17; P ¼ 0.02). ASAT correlated with ODI (r ¼ 0.25; P ¼ <0.001) and <SaO2> (r ¼ 0.18; P ¼ 0.009). ALAT was associated with ODI (r ¼ 0.24; P ¼ <0.001) and <SaO2> (r ¼ 0.21; P ¼ 0.002). All parameters also correlated with BMI z-score and sex and HDL-cholesterol was significantly associated with age. Glycemia, fasting insulin and HOMA showed strong correlations with BMI z-score and age. No association was observed between these parameters and the assessed sleep indices. Linear regression was performed on data of all patients after transformation with correction for confounders. HDL-cholesterol, ASAT, and ALAT were normally distributed after log transformation, inversion, and inversion of the square root, respectively. HDL-cholesterol showed a linear association with <SaO2> after correction for age, sex, and BMI z-score (partial r ¼ 0.21; P ¼ 0.002). For ASAT and ALAT, a linear relation was found with ODI after correction for BMI z-score (respectively: partial r ¼ 0.15; P ¼ 0.03 and partial r ¼ 0.15; P ¼ 0.02). However, after correction for sex these relations became insignificant with a P-value of 0.1 and 0.08, respectively. Follow-up study Of the 68 patients with SDB at baseline 50 participated in the follow-up study. Twelve had residual SDB. Main reasons for study discontinuation were leaving the treatment program early or refusing a second sleep study. Characteristics of dropouts were comparable to those of the children and adolescents that continued participation (Table 3). TABLE 3 Comparison of dropouts Variable No dropout Dropout P value N BMI z-score ODI (events/h) 2.99 ( ) 2.82 ( ) 0.2 AHI (events/h) 2.11 ( ) 2.50 ( ) 0.8 <SaO2> (%) 96 (84-98) 97 (94-98) 0.3 SaO2 nadir (%) 87 (70-93) 89 (77-91) 0.2 Glycemia (mg/dl) 84 (71-105) 87 (70-114) 0.8 Insulin (lu/ml) 14 (7-53) 14 (8-28) 0.8 HOMA-index 2.91 ( ) 2.65 ( ) 0.9 Total cholesterol (mg/dl) 142 (85-208) 144 (95-247) 0.7 HDL-cholesterol (mg/dl) 40 (28-60) 42 (31-59) 0.3 Triglycerides (mg/dl) 73 (41-237) 73 (49-344) 0.9 ASAT (mg/dl) 27 (14-223) 30 (16-60) 0.7 ALAT j (mg/dl) 30 (12-88) 35 (15-99) 0.6 BMI, body mass index; ODI, oxygen desaturation index; h, hour; AHI, apnea-hypopnea index; <SaO2>, mean nocturnal oxygen saturation; SaO2 nadir, nocturnal oxygen saturation nadir; HOMA, homeostasis model assessment; HDL, high-density lipoprotein; ASAT, aspartate aminotransferase; ALAT, alanine aminotransferase. Comparison of the data before and after treatment was performed with the Wilcoxon signed ranks test. All parameters improved significantly after weight loss, both for patients with and without SDB at baseline. This is presented in Table 4. Only for glycemia no important differences were observed pre- versus post-treatment. Delta s (D s) were calculated for the sleep parameters and laboratory measurements. Patients were divided into two study groups depending on their control sleep study, which was normalized or still showed residual SDB. Characteristics of the patients in the follow-up study are presented in Table 5. Comparison of the post-treatment data showed no significant differences regarding demographics, anthropometry, and laboratory measurements. Patients with severe residual SDB (ODI 5) had no significant changes of the metabolic parameters compared to those with mild SDB. Spearman correlation analysis of the post-treatment parameters showed a significant association between ASAT and <SaO2>. To perform linear regression ASAT was transformed and normality was obtained with 1/LnASAT. The linear relation remained significant after correction for sex as a significant confounder (partial r ¼ 0.34; P ¼ 0.002). No significant correlations were observed between the improvements in laboratory measurements and improvements in sleep parameters when examining the delta s, except for a trend between DASAT and DODI (r ¼ 0.28; P ¼ 0.06) and DSaO2 nadir (r ¼ 0.28; P ¼ 0.06). HDL-cholesterol improved in relation with a lowering in BMI z-score. Discussion The association between the metabolic syndrome and SDB in obese children remains controversial. Both Verhulst et al. and Redline et al. previously showed an association between components of the metabolic syndrome varying with SDB severity independent of sex 1448 VOLUME 21 NUMBER 7 JULY

4 TABLE 4 Comparison of pre- and post-treatment data Total group of subjects Subjects with SDB Variable Pre Post P value Pre Post P value N BMI z-score < <0.001 ODI (events/h) 1.11 ( ) * 2.94 ( ) 0.89 ( ) <0.001 AHI (events/h) 1.13 ( ) * 2.21 ( ) 0.87 ( ) <0.001 <SaO2> (%) 97 (84-98) * 96 (84-98) 97 (95-98) <0.001 SaO2 nadir (%) 91 (70-98) * 87 (70-93) 91 (72-95) <0.001 Glycemia mg/dl 84 (57-114) 83 (69-100) (70-114) 84 (69-99) 0.6 Insulin (lu/ml) 14 (4-53) 9 (4-37) < (7-53) 10 (5-37) HOMA-index 3.01 ( ) 1.86 ( ) < ( ) 2.07 ( ) Total cholesterol (mg/dl) 147 (66-360) 137 (88-398) < (85-247) 130 (89-182) HDL-cholesterol (mg/dl) 42 (21-78) 43 (25-73) < (28-60) 42 (29-56) 0.02 Triglycerides (mg/dl) 75 (37-344) 70 (25-241) < (41-344) 64 (33-241) ASAT (mg/dl) 26 (14-227) 19 (11-84) < (14-223) 19 (11-32) <0.001 ALAT (mg/dl) 27 (8-512) 18 (6-171) < (12-99) 18 (10-42) <0.001 SDB, sleep-disordered breathing; BMI, body mass index; ODI, oxygen desaturation index; h, hour of sleep; AHI, apnea-hypopnea index; <SaO2>, mean nocturnal oxygen saturation; SaO2 nadir, oxygen saturation nadir; HOMA, homeostasis model assessment; HDL, high-density lipoprotein; ASAT, aspartate aminotransferase; ALAT, alanine aminotransferase. * Value not reported because a follow-up sleep study was only performed in patients with SDB at baseline. TABLE 5 Demographic, anthropometric, sleep and laboratory parameters at follow-up visit Variable Normalized Residual SDB P value N Sex (% male) BMI z-score Absolute decrease 0.85 ( ) 0.90 ( ) 0.9 BMI z-score Relative decrease 28 (2-76) 33 (2-71) 0.8 BMI z-score ODI (events/h) 0.64 ( ) 3.73 ( ) <0.001 AHI (events/h) 0.71 ( ) 1.43 ( ) 0.07 <SaO2> (%) 97 (95-98) 97 (95-98) 0.2 SaO2 nadir (%) 91 (74-95) 87 (72-92) Glycemia (mg/dl) 83 (72-98) 86 (78-93) 0.4 Insulin (lu/ml) 11 (5-37) 9 (6-11) 0.1 HOMA-index 2.39 ( ) 1.76 ( ) 0.2 Total cholesterol (mg/dl) 129 (94-176) 147 (89-176) 0.5 HDL-cholesterol (mg/dl) 42 (29-52) 41 (29-56) 0.5 Triglycerides (mg/dl) 64 (37-241) 73 (48-127) 0.2 ASAT (mg/dl) 19 (11-32) 19 (15-26) 0.8 ALAT (mg/dl) 18 (11-42) 20 (10-41) 0.8 SDB, sleep-disordered breathing; BMI, body mass index; ODI, oxygen desaturation index; h, hour of sleep; AHI, apnea-hypopnea index; <SaO2>, mean nocturnal oxygen saturation; SaO2 nadir, oxygen saturation nadir; HOMA, homeostasis model assessment; HDL, high-density lipoprotein; ASAT, aspartate aminotransferase; ALAT, alanine aminotransferase. and degree of obesity. However, no effect of treatment was addressed in these studies (6,7). Other studies found no effect of SDB on insulin levels in both obese and nonobese children (8,9). Tauman et al. stated that the effect of obesity on the development of metabolic disregulation was much stronger than the effect of disturbed nocturnal respiration (8). Although data are limited, literature is mostly consistent regarding findings on lipid levels as several studies reported on a clear link between HDL-cholesterol levels and severity of SDB (6,7,15,16). Unfortunately there exists no literature on the effect of weight loss as a treatment of SDB on lipid profiles. In this study we were unable to find a link between the glucose metabolism and homeostasis and SDB. Correlation analysis revealed a strong relation between these metabolic factors and BMI. We can therefore conclude that our data confirm the study of Tauman et al. in which these factors were mostly linked to the degree of obesity (8). However, we did not perform an oral glucose tolerance test and might have missed early changes in the glucose metabolism as previously reported (6). Our data on lipid profiles are consistent with previous reports regarding dislipidemia in children with SDB. Most studies report on a clear link between HDL-cholesterol levels and severity of SDB, whereas data on other lipids are more diverse (6,7,15,16). In our study the correlation with <SaO2> persisted after controlling for important confounders such as age, sex, and BMI z-score. Improvements in HDL-cholesterol concentration after treating SDB were not linked with improvements in sleep-parameters. The improvements were driven by a decrease in BMI z-score rather than an increase in <SaO2>. This unfortunately leads to the fact that the role of nocturnal hypoxemia in the pathogenesis of dislipidemia partly remains unresolved. VOLUME 21 NUMBER 7 JULY

5 Metabolic Disruption and SDB in Obese Adolescents Van Hoorenbeeck et al. As several studies report on a link between the presence of fatty liver disease (17,18), expressed by elevated liver enzymes, we chose to also consider ALAT and ASAT in this study. There was a clear linear association between these parameters and the ODI at baseline, independent of the BMI z-score. However, this relation was highly influenced by sex, as a correction for this confounder made the association insignificant. We were unable to find a link between improvements in sleep parameters or BMI z-score and improvements in liver enzymes after weight loss. However, ASAT at the follow-up study was significantly associated in linear regression with <SaO2> after treatment, even after correction for the usual confounders. The first drawback of this study is the fact that we chose to perform a sleep screening rather than a full polysomnography, which still remains the gold standard. Therefore, we were unable to differentiate between central and obstructive events and have no data on sleep architecture and arousals. The latter may have also led to an underestimation of the number of hypopneas. Because of the lack of information on sleep architecture total sleep time was defined as the total recording time minus the episodes of signal artefacts. We are aware that this is merely an estimation of sleep time. However, we feel that awakenings are usually accompanied by aberrant respiratory signals. After weight loss, in patients with residual SDB the median ODI was greater than the median AHI. This was because of spontaneous desaturations without concurring respiratory events in 20 out of 50 patients. We performed a correlation analysis for AHI and ODI at baseline and after the treatment and found a strong positive correlation between both parameters (respectively: r ¼ 0.56; P ¼ <0.001 and r ¼ 0.51; P ¼ <0.001). We chose to divide our patients into study groups based on the ODI, because intermittent hypoxia is most likely the cause of oxidative stress and inflammation leading to metabolic disregulation. We used a cutoff value of 4% for desaturations as the study was designed and inclusion of patients started before the publication of the pediatric AASM guidelines (19) that recommend a cutoff value of 3%. Because our patients have a more adult phenotype and because the first patients included were stratified based on the value of 4%, we chose not to change the methodology of the reported study. We did not assess the pubertal stages, which may be important in studying the insulin metabolism as puberty is a state of hyperinsulinism and were unable to correct for this possible confounder. Another limitation of the study is the rather short follow-up period of 6 months and the dropout rate. However, patients that dropped out of the study were comparable to those who chose to participate in the follow-up study. The effect of dropout is therefore limited. All patients with residual SDB after weight loss treatment were advised to undergo full polysomnography and work-up of SDB. However, they were not followed in this study protocol. We therefore have no data on further treatment options and results. This is subject of a follow-up study in our center. In conclusion, this study confirmed the independent effect of nocturnal hypoxia on HDL-cholesterol levels in obese children with SDB at baseline, even after correcting for BMI z-score. Improvements in HDL-cholesterol concentration after weight loss were driven by a decrease in BMI z-score. Liver enzymes were correlated with ODI at baseline and although this linear association remained after correcting for BMI z-score, it was highly dependent of sex. ASAT, at follow-up correlated with <SaO2>, independent of the degree of obesity. Improvements in liver enzymes were neither driven by a decrease in BMI z-score, nor by improvements in sleep parameters. Fasting glucose, insulin, and HOMA-indices were mostly influenced by BMI z- score and changes in BMI z-score. Acknowledgment The authors would like to thank Greet Stalpaert and the nursing staff of Zeepreventorium (De Haan) for performing the sleep assessments and Eddy Bassle for the laboratory assistance.o VC 2013 The Society References 1. Verhulst SL, Van Gaal L, De Backer W, Desager K. The prevalence, anatomical correlates and treatment of sleep-disordered breathing in obese children and adolescents. Sleep Med Rev 2008;12: Marcus CL. Sleep-disordered breathing in children. Am J Respir Crit Care Med 2001;164: Amin R, Anthony L, Somers V, et al. Growth velocity predicts recurrence of sleepdisordered breathing 1 year after adenotonsillectomy. Am J Respir Crit Care Med 2008;177: Bhattacharjee R, Kheirandish-Gozal L, Spruyt K, et al. Adenotonsillectomy outcomes in treatment of obstructive sleep apnea in children: a multicenter retrospective study. Am J Respir Crit Care Med 2010;182: Van Hoorenbeeck K, Franckx H, Debode P, et al. Weight loss and sleep-disordered breathing in childhood obesity: effects on inflammation and uric acid. 2012;20: Verhulst SL, Schrauwen N, Haentjens D, et al. Sleep-disordered breathing and the metabolic syndrome in overweight and obese children and adolescents. J Pediatr 2007;150: Redline S, Storfer-Isser A, Rosen CL, et al. Association between metabolic syndrome and sleep-disordered breathing in adolescents. AmJRespirCritCareMed2007;176: Tauman R, O Brien LM, Ivanenko, et al. rather than severity of sleep-disordered breathing as the major determinant of insulin resistance and altered lipidemia in snoring children. Pediatrics 2005;116:e66-e Kaditis AG, Alexopoulos EI, Damani E, et al. Obstructive sleep-disordered breathing and fasting insulin levels in nonobese children. Pediatr Pulmonol 2005;40: Chai-Coetzer CL, Antic NA, Rowland LS, et al. A simplified model of screening questionnaire and home monitoring for obstructive sleep apnoea in primary care. Thorax 2011;66: Ng SSS, Chan TO, To KW, et al. Validation of a portable recording device (Apnea- Link) for identifying patients with suspected obstructive sleep apnoea syndrome. Int Med J 2009;39: Erman MK, Stewart D, Einhorn D, Gordon N, Casal E. Validation of the Apnea- LinkTM for the screening of sleep apnea: a novel and simple single-channel recording device. J Clin Sleep Med 2007;3: Lesser DJ, Haddad GG, Bush RA, Pian MS. The utility of a portable recording device for screening of obstructive sleep apnea in obese adolescents. J Clin Sleep Med 2012;8: Verhulst SL, Schrauwen N, Haentjens D, Van Gaal, De Backer WA, Desager KN. Reference values for sleep-related respiratory variables in asymptomatic European children and adolescents. Pediatr Pulmonol 2007;42: Alexopoulos EI, Gletsou E, Kostadima E, et al. Effects of obstructive sleep apnea severity on serum lipid levels in Greek children with snoring. Sleep Breath 2011;15: Gozal D, Sans Capdevila O, Kheirandish-Gozal L. Metabolic alterations and systemic inflammation in obstructive sleep apnea among nonobese and obese prepubertal children. Am J Respir Crit Care Med 2008;177: Verhulst SL, Jacobs S, Aerts L, et al. Sleep-disordered breathing: a new risk factor of suspected fatty liver disease in overweight children and adolescents. Sleep Breath 2009;13: Kheirandish-Gozal L, Sans Capdevila O, Kheirandish E, Gozal D. Elevated serum aminotransferase levels in children at risk for obstructive sleep apnea. Chest 2008;133: Iber C, American Academy of Sleep Medicine. The AASM manual for the scoring of sleep and associated events: rules, terminology, and technical specifications. Westchester, IL: American Academy of Sleep Medicine; VOLUME 21 NUMBER 7 JULY