This article was downloaded by: [The University of British Columbia]On: 29 October 2014, At: 21:20Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Developmental NeuropsychologyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/hdvn20

Pediatric Obstructive Sleep Apnea inObese and Normal-Weight Children:Impact of Adenotonsillectomy onQuality-of-Life and BehaviorRon B. Mitchell a & Emily F. Boss ba Department of Otolaryngology–Head and Neck Surgery , CardinalGlennon Children's Medical Center, Saint Louis University School ofMedicine , St. Louis, Missourib Department of Otolaryngology–Head and Neck Surgery, Divisionof Pediatric Otolaryngology , Johns Hopkins Medical Institutions ,Baltimore, MarylandPublished online: 09 Sep 2009.

To cite this article: Ron B. Mitchell & Emily F. Boss (2009) Pediatric Obstructive Sleep Apnea inObese and Normal-Weight Children: Impact of Adenotonsillectomy on Quality-of-Life and Behavior,Developmental Neuropsychology, 34:5, 650-661, DOI: 10.1080/87565640903133657

To link to this article: http://dx.doi.org/10.1080/87565640903133657

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

Pediatric Obstructive Sleep Apnea in Obeseand Normal-Weight Children:

Impact of Adenotonsillectomy on Quality-of-Lifeand Behavior

Ron B. MitchellDepartment of Otolaryngology–Head and Neck Surgery

Cardinal Glennon Children’s Medical Center, Saint Louis University School of Medicine,St. Louis, Missouri

Emily F. BossDepartment of Otolaryngology–Head and Neck Surgery, Division of Pediatric

Otolaryngology, Johns Hopkins Medical Institutions, Baltimore, Maryland

Objectives: To evaluate the impact of adenotonsillectomy (T&A) on quality-of-life (QOL) and behav-ior in obese versus normal-weight children with Obstructive Sleep Apnea (OSA). Design: Prospec-tive, non-randomized, controlled study. Methods: Children with an apnea-hypopnea index (AHI) ≥2were studied. Polysomnography was performed before and after T&A. An age- and gender-specificbody mass index (BMI-for-age) percentile was determined preoperatively. Children who were obese(>95th percentile) were compared to normal-weight children (BMI-for-age > 5th–85th percentile).Caregivers completed the OSA-18 QOL survey and the Behavioral Assessment Survey for Children(BASC) before surgery and 3–6 months postoperatively. Pre- and postoperative scores were com-pared using paired t-tests, and the impact of covariants was analyzed using ANOVA. Results: Thestudy population consisted of 89 children, 40 of whom were obese (45%). Postoperative scores forAHI, OSA-18 total and domain scores, and BASC scales and composites were significantly lower(improved) compared to pre-operative values in all children (p < .001). All mean OSA-18 and BASCscores were higher (indicating worse quality-of-life and behavior) pre- and postoperatively in obesethan in normal-weight children. Postoperatively, the majority of OSA-18 total scores and domainscores were significantly higher in obese children. A comparison of the total OSA-18 scores betweenchildren with a postoperative AHI < 2 and AHI ≥ 2 in obese children and a similar comparison in nor-mal-weight children was not statistically significant. There was no significant difference for BASCscores pre- and postoperatively between obese and normal-weight children. The pre- and postopera-tive scores for the AHI had a poor correlation with the pre- and postoperative Behavioral SymptomsIndex (BSI) and total OSA-18 scores (r = .09), respectively. Conclusions: Following T&A all childrenhave improvements in AHI, QOL, and behavior. Obese children are more likely to have persistentOSA and poor QOL scores after T&A. Behavior improves postoperatively to a similar extent in allchildren regardless of obesity.

DEVELOPMENTAL NEUROPSYCHOLOGY, 34(5), 650–661Copyright © 2009 Taylor & Francis Group, LLCISSN: 8756-5641 print / 1532-6942 onlineDOI: 10.1080/87565640903133657

Correspondence should be addressed to Ron B. Mitchell, Department of Otolaryngology – Head and Neck Surgery,Cardinal Glennon Children’s Medical Center, Saint. Louis University School of Medicine, 1465 S. Grand, Suite 4740,St. Louis, MO 63104. E-mail: rmitch11@slu.edu

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

INTRODUCTION

In the past two decades, there has been increasing focus on the evaluation and treatment of ob-structive sleep apnea (OSA) in children. OSA in children has been associated with physicalsequelae such as systemic hypertension, pulmonary hypertension, enuresis, and growth impair-ment (Kwok, & Chan, 2008; Sans, Crabtree & Kheirandish-Gozal, et al., 2008; Somers, White, &Amin, 2008). OSA is also known to negatively impact the psychological well-being of children.Neurocognitive effects of OSA in children include hypersomnolence, hyperactivity, decreased at-tention, and poor school performance (Beebe, 2006; Chervin, Archbold, & Dillon, et al., 2002;Galland, Dawes, & Tripp, et al., 2006; Lewin, Rosen, & England, et al., 2002; Suratt, Barth, & Di-amond, et al., 2007).

More than one third of obese children have OSA (Shine, Coates, & Lannigan, 2005). The preva-lence of obesity in children in the United States has nearly tripled over the last three decades to in-clude 17% of the pediatric population (Ogden, Carroll, & Curtin, et al., 2006). This growing nationaland global obesity epidemic is likely to make the prevalence of OSA in children increase. Alongwith numerous negative physical effects, neurocognitive effects of obesity in children include in-creased depression, poorer quality-of-life (QOL), and behavioral problems (Lee, 2007; Rudnick &Mitchell, 2007; Datar & Sturm, 2006; Datar & Sturm, 2004; Goodman & Whitaker, 2002).

The outcome of adenotonsillectomy (T&A) in children with obesity continues to be elucidated.Children with obesity experience improvement in sleep parameters following surgery althoughmilder OSA frequently persists (Mitchell & Kelly, 2007). T&A in obese children tends to result inweight stability or even gain and this may result in worsening of OSA severity. Weight gain maybe due to a reduction in energy expenditure during sleep as well as improved growth hormone se-cretion (Shine et al., 2005; Bonuck, Freeman, & Henderson, 2008). Nonetheless, obese childrenwith OSA have been shown to have improved disease-specific QOL following T&A (Mitchell &Kelly, 2004). Likewise, outcomes of T&A for OSA in normal-weight children with behavioral orneurocognitive impairment are also encouraging. Treatment of mild OSA can even positively im-pact neurocognitive function in children with syndromes such as attention deficit hyperactivitydisorder (ADHD), autism, Prader-Willi, and epilepsy (Huang, Guilleminault, & Li, et al., 2007;Kaleyias, Cruz, & Goraya, et al., 2008; Malow, Marzec, & McGrew, 2006; Malow, Mcgrew& Harvey et al., 2006; O’Donoghue, Camferrman, & Kennedy, 2005; Brodsky, Moore, &Stanievich, 1987).

Although there are established associations between OSA, obesity, QOL, and behavior, the ef-fects of T&A on QOL and behavior in obese children remain loosely defined. Furthermore, it isunclear whether severity of OSA corresponds to the degree of QOL or behavioral impairment be-fore and after surgical therapy with T&A. Therefore, the objectives of this prospective, controlledstudy were to compare the following parameters in obese and normal-weight children with OSA:(1) QOL and behavioral deficits; (2) the impact of T&A for OSA on QOL and behavior; and (3)correlation of the severity of OSA with severity of QOL and behavioral impairment.

METHODS

Children referred to the pediatric otolaryngology service with a sleep disturbance were evaluatedfor inclusion in the study. Institutional review board approval was obtained and caregivers signed

PEDIATRIC OBSTRUCTIVE SLEEP APNEA 651

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

an informed consent document. Demographic information was collected for both caregiver andchild. All children considered for enrollment underwent polysomnography. Exclusion criteria in-cluded: (1) children less than 3 or more than 18 years of age; (2) children who previously had aT&A; (3) children with craniofacial syndromes, neuromuscular disease, or developmental delay;(4) children on psycho-stimulant therapy; and (5) children with an AHI < 2.

The age- and gender-corrected body mass index (BMI) was calculated for each child using es-tablished guidelines (Polhamus et al., 2003) and children were divided into 4 groups as follows:group 1, underweight (BMI less than or equal to the 5th percentile); group 2, normal (BMI greaterthan the 5th percentile but less than the 85th); group 3, at risk for obesity (BMI greater than the85th percentile but less than the 95th); group 4, obese (BMI greater than or equal to the 95th per-centile). Children in group 4 were considered obese and were compared to the normal weight chil-dren in group 2. Tonsils were graded according to the scheme proposed by Brodsky et al. (1987):(1) small tonsils confined to the tonsillar pillars; (2) tonsils that extend just outside the pillars; (3)tonsils that extend outside the pillars but do not meet in the midline; (4) large tonsils that meet inthe midline.

The following parameters were measured during polysomnography: four-channel electroen-cephalography with bilateral central and occipital leads; electrooculography to measure verticaland horizontal eye movements; electromyography with submental electrodes; electrocardiogra-phy; airflow recording through the nose and mouth by a nasal air pressure transducer with endtidal CO2; thoracic and abdominal effort by piezoelectric sensors; oxygen saturation throughpulse oximetry; and tracheal sound recording by using a microphone secured to the neck. Digitalvideotaping with sound recording was performed throughout the night. A sleep medicine physi-cian interpreted the polysomnography results. The AHI, defined as the average number of ob-structive apneas and hypopneas per hour of sleep, was used for diagnosis of OSA. Obstructiveapnea was defined as total absence of airflow through the mouth and nose with continued chestand abdominal movement for at least two respiratory cycles. Hypopnea was defined as nasal flowof greater than or equal to 50% in at least two breaths that is associated with a corresponding de-crease in oxygen saturation of greater than or equal to 3% or an arousal or awakening (AASMManual for the Scoring of Sleep and Associated Events, 2007). OSA was defined as an AHIgreater than or equal to 2 (Witmans, Keens, & Davidson, 2003; Marcus, Omlin, & Basinski,1992). Tonsillectomy was performed with a combination of sharp and blunt dissection utilizingprimarily a blade electrocautery. Adenoidectomy was performed using suction ablation andcautery.

Caregivers were asked to complete the OSA-18 Quality of Life and BASC surveys before and3–6 months after surgery (Franco, Rosenfeld, & Rao, 2000; Reynolds & Kamphaus, 1998). TheOSA-18 comprises 18 items in five domains of sleep disturbance, physical suffering, emotionaldistress, daytime problems and caregiver concerns. A point scale is used ranging from 1 (= noneof the time) to 7 (= all of the time) to grade the relative severity of the problem addressed in eachitem. The total score, the domain score, and the item score were recorded. Specific BASC instru-ments were used for the following age groups; 2.5–5 years; 6–11 years; and 12–18 years. OnlyBASC scales and composites that were applicable to all age groups were included in this study.The BASC behavioral scales used were atypicality, depression, hyperactivity, and somatization.BASC composite scales are made up of several behavioral scales that are weighted differently as afunction of age. This weighting reflects the relative contribution of individual scales to problembehaviors in different age groups. The composite scales were the behavioral symptoms index

652 MITCHELL AND BOSS

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

(BSI), externalizing problems, and internalizing problems. Responses to each question werescored as follows: never occurs = 0; sometimes occurs = 1; often occurs = 2; almost always occurs= 3. Raw scores on the BASC scales and composites were converted to T scores with a mean of 50and a standard deviation of 10. Children who scored from >60 were considered to have clinicallysignificant behavioral impairment.

Data were analyzed using SPSS 13.0 (SPSS Inc., Chicago, IL). Results were expressed as themean scores or the mean ± SD. Subtracting the mean postoperative score from the mean preopera-tive score derived the OSA-18 and BASC difference score. Because a decrease in mean score im-plies improvement, the mean change is a positive number. The 95% confidence intervals (95%CI) for difference scores were also calculated. Paired t tests were used to compare mean values ofAHI, OSA-18, and BASC scores before and after adenotonsillectomy. Analysis of variance(ANOVA) was used to establish the impact of covariants on surgical outcomes. Pearson correla-tion was used to compare the mean pre- and postoperative OSA-18, BASC, and AHI scores. Ap-value less than or equal to .05 was considered significant. OSA-18 total scores on the first sur-vey were classified as: mild (< 60); moderate (≥ 60 ≤ 80); or severe (> 80). Children were also di-vided into two groups based on pre-operative BASC scores: normal (T score less than 60); clini-cally significant (T score greater than or equal to 60). This was done to compare the number ofchildren with high OSA-18 and BASC scores before and after surgery. To calculate the samplesize for the present study, data from polysomnography for 26 children with OSA who underwentadenotonsillectomy and were studied by Suen, Arnold, and Brooks et al. (1995), were used as amodel. The mean preoperative AHI value for these children was 18.1 ± 11.3 and the postoperativevalue was 4.5 ± 9.4. Assuming a moderate correlation of 0.3 between pre- and postoperative val-ues, a population of 30 children in each group would be sufficient to detect a mean differencescore of 10. This indicates a clinically significant change in AHI with 80% power and α = .05 us-ing a 2-tailed t test. A sample size of 30 is also adequate to detect a 10-point difference in BASC Tscores and a 20-point difference in OSA-18 total scores with 80% power and α = .05. Because adifference of 10 in a BASC T score or a 20-point difference in OSA-18 total scores is clinicallysignificant (Franco et al., 2000; Reynolds & Kamphaus, 1998) the present study has sufficientpower to detect a significant change in AHI, behavior and quality-of-life for either obese or nor-mal-weight children after adenotonsillectomy.

RESULTS

Eighty-nine children were included in the study. Forty children were obese and 49 children werenormal-weight. The demographics and clinical characteristics are summarized in Table 1. Themean age was similar between the two groups as was the mean interval between pre- and postop-erative polysomnography and surgery. Hispanic or Latino children were equally represented inboth groups. Obese children were less likely to have marked enlargement of their tonsils.

Pre- and postoperative mean scores as well as difference scores with 95% confidence intervalsfor AHI, OSA-18, and BASC T-scores for obese and normal-weight children are presented in Ta-ble 2. Postoperative scores for AHI, all OSA domains and BASC scales and composites were sig-nificantly lower (improved) compared to preoperative values in obese and normal weight children(p < .001). Preoperatively, the mean AHI was 20.1 (range 3.1–67.7) for obese children and 13.2(range 3.0–88.0) for normal-weight children. Postoperatively, the mean AHI was 8.8 (range

PEDIATRIC OBSTRUCTIVE SLEEP APNEA 653

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

654

TABLE 1Demographics and Clinical Characteristics of the Study Population

Demographics and Clinical CharacteristicsObese Children

(N = 40)Normal Weight

Children (N = 49) p-value

Gender—Male 28 (70%) 26 (53%) NSMean age (range ) in yrs 9.1 (3.0–17.1) 6.7 (3.0–14.9) NSHispanic or Latino 19 (48%) 21 (43%) NSMean age- and gender-corrected BMI 28.1 14.1 <.001Children with tonsillar hypertrophy (Grades 3–4) 26 (65%) 47 (96%) <.001Months from Preoperative PSG to Tonsillectomy 2.6 2.3 NSMonths from Tonsillectomy to Postoperative PSG 5.4 5.1 NS

BMI = body mass index; NS, not significant; PSG = polysomnography. p > .05.

TABLE 2Polysomnography, OSA-18, and BASC Scores Pre- and Post-Adenotonsillectomy (T&A)

Domains and ParametersPre-SurgeryMean Score

Post-SurgeryMean Score

Mean Change(95% CI)

Normal Weight ChildrenApnea-Hypopnea Index (AHI) 13.2 1.3 11.9 (8.4–15.4)OSA-18 Total Score 71.2 32.2 39.0 (33.4–41.6)Sleep Disturbance 18.9 5.6 13.3 (11.6–15.0)Physical symptoms 14.8 7.8 7.0 (5.4–8.6)Emotional distress 10.4 6.7 3.7 (2.4–5.0)Daytime function 11.1 5.6 5.5 (3.3–6.7)Caregiver concerns 16.3 6.5 9.8 (8.0–11.6)BASC-BSI* 56.3 49.2 7.2 (2.9–11.5)Atypicality 53.6 48.5 5.1 (1.7–8.5)Depression 56.4 48.3 8.1 (3.0–13.2)Hyperactivity 57.0 49.1 7.9 (3.9–11.9)Somatization 60.3 51.1 9.2 (5.1–13.3)lparExternalizing Problems 53.9 47.4 6.5 (2.3–10.7)Internalizing Problems 58.4 49.3 9.1 (4.5–13.7)

Obese ChildrenAHI 20.1 8.8 11.3 (6.0–16.6)OSA-18 Total Score 86.0 42.9 43.2 (35.2–51.2)Sleep Disturbance 22.0 7.9 14.1 (13.5–15.7)Physical Symptoms 17.4 8.4 9.1 (6.8–11.4)Emotional Distress 12.4 9.3 3.1 (0.9–5.3)Daytime function 14.2 7.8 6.4 (4.8–8.0)Caregiver concerns 20.0 9.5 10.5 (8.1–12.9)BASC-BSI* 60.5 49.9 10.7 (6.3–15.1)Atypicality 58.4 51.5 6.9 (2.5–11.3)Depression 59.4 50.4 9.0 (4.2–13.8)Hyperactivity 57.8 52.3 5.2 (0.3–10.1)Somatization 64.8 54.1 10.7 (6.7–14.7)Externalizing Problems 55.9 48.4 7.5 (2.2–12.8)Internalizing Problems 60.1 51.3 8.8 (5.1–12.5)

*BSI = Behavior Symptoms Index; p-value < .001 2-tailed paired t-test for mean change in all pre- and post-operativescores for normal weight and obese children.

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

0–34.2) for obese and 1.3 (range 0–8.1) for normal-weight children (p < .01). The mean change inOSA-18 total score was 43.2 (CI 35.2–51.2) for obese and 39.0 (CI 33.4–41.6) for normal-weightchildren. The OSA-18 domain with the largest mean change score was sleep disturbance and thedomain with the smallest mean change score for both groups of children was emotional distress.The mean change in BSI was 10.7 (CI 6.3– 15.1) for obese children and 7.2 (2.9–11.5) for nor-mal-weight children. The BASC scale with the largest mean change score was somatization forboth groups of children. The domain with the smallest mean change score was hyperactivity forobese and atypicality for normal-weight children (Table 2).

A comparison of pre- and postoperative parameters for obese and normal-weight children ispresented in Table 3. All mean OSA-18 domains and total scores and BASC scales and compos-ites were higher (indicating worse quality-of-life and behavior) pre- and postoperatively in obesethan in normal weight children. Preoperatively, OSA-18 total scores and the domains of sleep dis-turbance, physical symptoms, daytime function, and caregiver concerns were significantly higherin obese compared to normal-weight children. There was no significant difference preoperativelyin the domain of emotional distress between the two groups. Postoperatively, OSA-18 total scoresand the domains of sleep disturbance, emotional distress, daytime function, and caregiver con-cerns were significantly higher in obese compared to normal-weight children. There was no sig-nificant difference postoperatively in the domain of physical symptoms between the two groups.Further analysis was done to look at whether obesity resulted in persistent disturbances in qual-ity-of-life independent of residual OSA. Both the obese and normal-weight children were dividedinto a “responder” (post T&A AHI <2) and “non-responder” (post T&A AHI ≥2) groups. Themean AHI for obese “responders” was 1.1 (0–1.9) and for “non-responders” it was 10.6 (2.1–34.2).The mean AHI for normal-weight “responders” was 1.4 (0–1.9) and for “non-responders” 4.4(2.4–8.1). The mean OSA-18 total score for obese “responders” was 38.1 (20–65) and for“non-responders” it was 44.4 (18–92). The mean OSA-18 total score for normal-weight “respond-ers” was 32.7 (18–77) and for “non-responders” 30.6 (20–49). A comparison of the total OSA-18scores between children with a postoperative AHI <2 and AHI ≥2 in obese children and a similarcomparison in normal-weight children was not statistically significant. There was no significantdifference for BASC scales and composite T scores pre- and postoperatively between obese andnormal-weight children.

The number of children with BSI scores that indicated clinically significant behavioralimpairment (T ≥ 60) before surgery was 21 (55%) for obese and 18 (39%) for normal-weightchildren. Postoperatively, 6 (17%) of obese and 9 (18%) of normal-weight children had sig-nificant behavior impairment. Similarly, the number of children with OSA-18 total scores thatindicated moderate to severe QOL impairment (OSA-18 total scores ≥ 60) before surgerywas 40 (100%) for obese and 35 (71%) for normal-weight children. Postoperatively, 11 (27%)of obese and 2 (4%) of normal-weight children had moderate to severe QOL impairment(Table 4).

The preoperative scores for BSI had a fair correlation with the preoperative total OSA-18scores (r = 0.26). The preoperative scores for AHI had a poor correlation with the preoperativeBSI and total OSA-18 scores (r = 0.11). The postoperative scores for BSI had a fair correlationwith the postoperative total OSA-18 scores (r = 0.26). The postoperative scores for the AHI had apoor correlation with the postoperative BSI and total OSA-18 scores (r = 0.09). Less than 10% ofthe variation in one score, either the pre- or postoperative AHI, could be explained by variations inthe pre- or postoperative BSI or total OSA-18 score, respectively.

PEDIATRIC OBSTRUCTIVE SLEEP APNEA 655

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

A repeated measures ANOVA showed that the covariates gender, age, ethnicity, and tonsillarsize did not contribute significantly to the observed changes in AHI, OSA-18, or BASC scores.BMI and preoperative AHI were correlated significantly with persistent OSA (p < .001).

DISCUSSION

This study demonstrates that children with OSA exhibit significant improvements in respiratorysleep parameters, QOL, and behavior following T&A. While obesity impacts the magnitude of

656 MITCHELL AND BOSS

TABLE 3OSA-18 and BASC Scores Pre- & Post-Adenotonsillectomy (T&A)

Domains and Parameters

Pre-T&A Mean Score(SD)

Normal Weight ChildrenPre-T&A Mean Score (SD)

Obese Children p-value

Quality of LifeOSA-18 Total Score 71.2 (19.5) 86.0 (15.5) <.0010Sleep Disturbance 32.2 (11.3) 42.9 (21.9) <.001Physical symptoms 14.8 (5.3) 17.4 (5.2) 0.02Emotional distress 10.4 (5.1) 12.4 (5.5) NSDaytime function 11.1 (4.6) 14.2 (3.5) <.001Caregiver concerns 16.3 (5.9) 20.0 (4.8) 0.002

BehaviorBASC-BSI* 56.3 (15.2) 60.5 (15.1) NSAtypicality 53.6 (9.9) 58.4 (13.4) NSDepression 56.4 (17.0) 59.4 (16.3) NSHyperactivity 57.0 (13.3) 57.8 (14.4) NSSomatization 60.3 (13.7) 64.8 (12.2) NSExternalizing Problems 53.9 (14.2) 55.9 (17.3) NSInternalizing Problems 58.4 (15.1) 60.1 (12.8) NSOSA-18 Total Score

Domains and ParametersPost-T&A Mean Score (SD)Normal Weight Children

Post-T&A Mean Score (SD)Obese Children p-value

Quality of LifeOSA-18 Total Score 32.3 (11.3) 42.9 (21.9) 0.008Sleep Disturbance 5.6 (2.1) 7.9 (3.8) 0.002Physical Symptoms 7.8 (3.0) 8.4 (5.0) NSEmotional Distress 6.7 (3.9) 9.3 (5.4) 0.01Daytime function 5.6 (2.6) 7.8 (4.4) 0.009Caregiver concerns 6.5 (2.9) 9.5 (5.9) 0.006

BehaviorBASC-BSI* 49.2 (8.2) 49.9 (10.9) NSAtypicality 48.5 (8.6) 51.5 (8.9) NSDepression 48.3 (9.0) 50.4 (11.7) NSHyperactivity 49.1 (8.9) 52.6 (7.6) NSSomatization 51.1 (10.9) 54.1 (10.9) NSExternalizing Problems 47.4 (7.6) 48.4 (9.2) NSInternalizing Problems 49.3 (10.1) 51.3 (12.3) NS

*BSI = Behavioral Symptoms Index; NS = not significant; SD = Standard DeviationDow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

change in postoperative respiratory sleep parameters and QOL, it does not appear to affect im-provement in behavioral impairment. Furthermore, no clear relationship is observed between theseverity of OSA, as measured by polysomnography, and the severity of QOL or behavioral im-pairment.

The relationship between OSA and neurocognition or behavior has been widely studied in re-cent years. Validated instruments have been used to methodologically identify impairments in be-havior and cognition in children who exhibit a broad range of obstructive sleep patterns (Lewin etal., 2002; Gottlieb, Chase, & Vezina, et al., 2004; Chervin et al., 2002). Behavioral problems asso-ciated with OSA have included inattention, hyperactivity, aggression, depression, and poor so-cialization. Neurocognitive impairment such as memory and intelligence deficits have been ob-served in children with all degrees of OSA. Even children with primary snoring have behavioralimpairment compared to controls (Gozal & Pope, 2001; Kennedy, Blunden, & Hirte, et al., 2004).There is also evidence that these neurocognitive impairments may persist into adolescence(Gottlieb et al., 2004; Gozal & Pope, 2001). These findings emphasize the importance of earlyrecognition and treatment of OSA in children. Cognitive and behavioral problems, as opposed tosevere physical sequelae, are more commonly recognized symptoms associated with OSA and aregenerally identified during office evaluation for sleep disturbance (Chervin, Ruzicka, & Giorani,et al., 2006). Following T&A, children with OSA have demonstrated improvement in behavioralparameters (Galland et al., 2006; Goldstein, Post, & Rosenfeld, et al., 2000; Mitchell & Kelly,2005), although postoperative cognitive improvement is less definitive (Chervin et al., 2006).

Neurocognitive and behavioral problems have also been associated with childhood obesity.Several studies have examined early academic performance as it relates to obesity in children. Anindependent association between behavioral problems and obesity has previously been estab-lished in a large nationally representative cohort of elementary school-aged children (Lumeng,Gannon, & Cabral, et al., 2003). In a separate study, girls who became obese during elementaryschool proved to demonstrate poorer test scores and teacher behavioral ratings when compared to

PEDIATRIC OBSTRUCTIVE SLEEP APNEA 657

TABLE 4Quality of Life and Behavior Scores Before and After T&A

Number of Children(%) Preoperatively

Number of Children(%) Postoperatively

Normal Weight Children (N = 49)OSA-18 OSA < 60 14 (29) 47 (96)

OSA > 60 < 80 20 (41) 2 (4)OSA > 80 15 (30) 0 (0)

BASC-BSI* < 60 31 (63) 40 (82)≤ 60 18 (37) 9 (18)

Obese Children (N = 40)OSA < 60 0 (0) 29 (73)

OSA-18 OSA > 60 < 80 13 (33) 7 (18)OSA > 80 27 (67) 4 (9)

BASC-BSI* < 60 18 (45) 33 (83)≥ 60 21 (55) 6 (17)

*BSI = Behavioral Symptoms Index

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

peers (Datar & Sturm, 2006). Furthermore, poor academic performance in obese school-agedchildren is also related to disparate socioeconomic variables, and may contribute to negative stig-matization of children from a young age (Datar, Sturm, & Magnabosco, 2004). Thus both OSAand obesity have separately been associated with behavioral and neurocognitive impairment inchildren. We, therefore, anticipated that obese children with OSA would demonstrate worsenedbehavioral impairment compared to normal-weight children. This would imply a cumulative ef-fect of obesity and OSA that was not seen in this study. Although children with obesity did dem-onstrate greater behavioral impairment both before and after surgery in all domains and indiceswhen compared to normal-weight children, these differences did not reach statistical significance.This may reflect the lack of a sufficiently large study population to detect this difference; it may bedue to BASC not being the best tool to evaluate behavioral problems in obese children or it maysimply be due to this difference being very small and clinically insignificant.

T&A is the primary treatment for pediatric OSA. It has been shown to lead to a significant im-provement or resolution of OSA in the majority of children on postoperative polysomnography(Shine et al., 2005). Furthermore, children experience marked symptomatic resolution and im-proved QOL following T&A for OSA. Similarly, behavioral problems significantly improve aftersurgery in children with OSA (Goldstein et al., 2000; Mitchell & Kelly, 2005), and this improve-ment is maintained over time (Mitchell & Kelly, 2006). The role of obesity in postoperative out-comes in children with OSA is less clear. McLaughlin, Varni, and Gozal (2004) reported that chil-dren with suspected OSA, regardless of the severity of the AHI or obesity, had more impairmentsin QOL than children without a sleep disturbance. Results of this study demonstrate that obesecompared to normal-weight children have less marked improvements in AHI and QOL, but simi-lar improvements in behavioral parameters. We expected obese children to have more behavioralproblems prior to T&A and demonstrate less improvement after T&A. Perhaps behavior prob-lems are much more a function of OSA than of obesity. Future studies are needed to evaluate be-havior and neurocognition in obese children with OSA, possibly with greater emphasis on differ-ent psychosocial aspects such as depression or academic achievement following surgery.

Interestingly, pre- and postoperative AHI did not correlate with total OSA-18 or BSI scores forthese children. We therefore conclude that children with mild, moderate, or severe OSA shouldexperience equivocal and significant improvement in QOL and behavior. These findings suggestthat any physical sleep disturbance can manifest in poor QOL and behavior, and that the degreesof hypoxemia, hypercapnia, or apneic events are more important at predicting physical sequelae.Perhaps a mild sleep disturbance over a long period of time is even more significant than severeOSA over a short period of time. The poor correlation between AHI with QOL and behavior alsounderscores the importance of recognition and treatment of SDB regardless of mild respiratorysleep parameters. This point is further emphasized by studies that show that primary snoring in theabsence of polysomnography-documented OSA is related to delayed neurocognitive develop-ment (Gozal & Pope, 2001; Kennedy et al., 2004).

There are numerous strengths to this study. First, the study evaluates data collected prospec-tively, and children with obesity were compared to controls. Next, children were evaluated usingobjective measures: polysomnography was used to define OSA, and a standardized measure ofweight (BMI-for-age) was used to define the study population. Outcomes were measured withvalidated instruments. In particular, the OSA-18 is a disease-specific QOL tool, and the BASC isan age-stratified validated behavioral tool. The study also benefited from a relatively large studypopulation.

658 MITCHELL AND BOSS

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

There were a number of limitations that warrant further discussion. First, the validated instru-ments used to measure QOL and behavioral impairment were completed by parents. Therefore itis possible that parental bias influenced the preoperative severity of disease or degree of postoper-ative improvement. Next, the BASC instrument is a general behavioral measure and is not specificfor OSA or for obesity. Furthermore, the BASC does not measure impairment in neurocognition,and aspects of neurocognition such as school performance may be influenced differently by T&Ain this population of children. Finally, postoperative results were obtained within 6 months of sur-gery. Long-term results would be more optimal, as it is uncertain whether the improvements notedwould be sustainable over time. Obese children who undergo T&A tend to remain overweight orbecome more obese (Shine et al., 2005). It is therefore possible that over a longer period of timeQOL and behavioral impairments would worsen due to the effects of weight gain. Despite theselimitations, this study clearly shows that obese children experience postoperative improvement inAHI, QOL, and behavior, but those behavioral improvements are not significantly impacted byobesity.

CONCLUSIONS

Children with OSA exhibit significant improvement in respiratory sleep parameters, disease-spe-cific QOL, and behavior following adenotonsillectomy. Following T&A, children with obesity aremore likely to have persistent OSA and poorer QOL than normal-weight children. Normal-weightand obese children with OSA display behavioral problems that improve following T&A, but thepresence of obesity does not impact this change. Severity of OSA does not correlate with degreeof QOL or behavioral impairment. These results suggest that although obesity may influence chil-dren with OSA to experience a poorer QOL than normal- weight children, it has no distinguish-able impact on behavior. Further investigation into the relationship between obesity and behavioris warranted.

REFERENCES

AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specification. (2007).Beebe, D. W. (2006). Neurobehavioral morbidity associated with disordered breathing during sleep in children: a compre-

hensive review. Sleep, 29(9), 1115–1134.Bonuck, K., Freeman, K., & Henderson, J. (2008). Growth and growth biomarker changes after adenotonsillectomy: Sys-

tematic review and meta-analysis. Archives of Disease in Childhood.Brodsky, L., Moore, L., & Stanievich, J. F. (1987). A comparison of tonsillar size and oropharyngeal dimensions in chil-

dren with obstructive adenotonsillar hypertrophy. International Journal of Pediatric Otorhinolaryngology, 13(2),149–156.

Chervin, R. D., Archbold, K. H., Dillon, J. E., Panahi, P., Pituch, K. J., Dahl, R. E., et al. (2002). Inattention, hyperactivity,and symptoms of sleep-disordered breathing. Pediatrics, 109(3), 449–456.

Chervin, R. D., Ruzicka, D. L., Giordani, B. J., Weatherly, R. A., Dillon, J. E., Hodges E. K., et al. (2006). Sleep-disorderedbreathing, behavior, and cognition in children before and after adenotonsillectomy. Pediatrics, 117(4), e769–e778.

Datar, A., & Sturm, R. (2004). Childhood overweight and parent- and teacher-reported behavior problems: Evidence froma prospective study of kindergartners. Archives of Pediatric Adolescent Medicine, 158(8), 804–810.

Datar, A., Sturm, R., & Magnabosco, J. L. (2004). Childhood overweight and academic performance: National study ofkindergartners and first-graders. Obesity Research, 12(1), 58–68.

PEDIATRIC OBSTRUCTIVE SLEEP APNEA 659

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

Datar, A., & Sturm, R. (2006). Childhood overweight and elementary school outcomes. International Journal of Obesity(Lond), 30(9), 1449–1460.

Franco, R. A., Jr., Rosenfeld, R. M., & Rao, M. (2000). First place—Resident clinical science award 1999.Quality of life for children with obstructive sleep apnea. Otolaryngology Head & Neck Surgery, 123(1 Pt1), 9–16.

Galland, B. C., Dawes, P. J., Tripp, E. G., Taylor, B. J. (2006). Changes in behavior and attentional capacity afteradenotonsillectomy. Pediatric Research, 59(5), 711–716.

Goldstein, N. A., Post, J. C., Rosenfeld, R. M., Campbell, T. F. (2000) Impact of tonsillectomy and adenoidectomy onchild behavior. Archives of Otolaryngology Head & Neck Surgery, 126(4), 494–498.

Goodman, E., & Whitaker, R. C. (2002). A prospective study of the role of depression in the development and persistenceof adolescent obesity. Pediatrics, 110(3), 497–504.

Gottlieb, D. J., Chase, C., Vezina, R. M., Heeren, T. C., Corwin, M. J., Auerbach, S. H., et al. (2004). Sleep-disorderedbreathing symptoms are associated with poorer cognitive function in 5-year-old children. Journal of Pediatrics, 145(4),458–464.

Gozal, D., & Pope, D. W., Jr. (2001). Snoring during early childhood and academic performance at ages thirteen to four-teen years. Pediatrics, 107(6), 1394–1399.

Huang, Y. S., Guilleminault, C., Li, H. Y., Yang, C.M., Wu, Y.Y., Chen, N.H., et al. (2007). Attention-deficit/hyperactiv-ity disorder with obstructive sleep apnea: A treatment outcome study. Sleep Medicine, 8(1), 18–30.

Kaleyias, J., Cruz, M., Goraya, J. S., Valencia, I., Khurana, D.S., Legido, A., et al. (2008). Spectrum of polysomnographicabnormalities in children with epilepsy. Pediatric Neurology, 39(3), 170–176.

Kennedy, J. D., Blunden, S., Hirte, C., Parsons, D.W., Martin, A.J., Crowe, E., et al. (2004). Reduced neurocognition inchildren who snore. Pediatric Pulmonology, 37(4), 330–337.

Kwok, K. L., Ng, D. K., & Chan, C. H. (2008). Cardiovascular changes in children with snoring and obstructive sleepapnoea. Annals of the Academy of Medicine, Singapore, 37(8), 715–721.

Lee, W. W. (2007). An overview of pediatric obesity. Pediatric Diabetes Suppl., 9, 76–87.Lewin, D. S., Rosen, R. C., England, S. J., Dahl, R. E. (2002). Preliminary evidence of behavioral and cognitive sequelae

of obstructive sleep apnea in children. Sleep Medicine, 3(1), 5–13.Lumeng, J. C., Gannon, K., Cabral, H. J., Frank, D. A., & Zuckerman, B. (2003). Association between clinically meaning-

ful behavior problems and overweight in children. Pediatrics, 112(5), 1138–1145.Malow, B. A., Marzec, M. L., McGrew, S. G., Wang, L., Henderson, L. M., & Stone, W. L. (2006). Characterizing sleep in

children with autism spectrum disorders: A multidimensional approach. Sleep, 29(12), 1563–1571.Malow, B. A., McGrew, S. G., Harvey, M., Henderson, L. M., & Stone, W.L. (2006). Impact of treating sleep apnea in a

child with autism spectrum disorder. Pediatric Neurology, 34(4), 325–328.Marcus, C. L., Omlin, K. J., Basinki, D. J., Bailey, S. L., Rachal, A. B., Von Pechmann, W. S., et al. (1992). Normal

polysomnographic values for children and adolescents. American Review of Respiratory Disease 146(5 Pt 1),1235–1239.

McLaughlin Crabtree, V., Varni, J. W., & Gozal, D. (2004). Health-related quality of life and depressive symptoms in chil-dren with suspected sleep-disordered breathing. Sleep, 27(6), 1131–1138.

Mitchell, R. B., & Kelly, J. (2004). Adenotonsillectomy for obstructive sleep apnea in obese children. OtolaryngologyHead & Neck Surgery, 131(1), 104–108.

Mitchell, R. B., & Kelly, J. (2005). Child behavior after adenotonsillectomy for obstructive sleep apnea syndrome. Laryn-goscope, 115(11), 2051–2055.

Mitchell, R. B., & Kelly, J. (2006). Long-term changes in behavior after adenotonsillectomy for obstructive sleep apneasyndrome in children. Otolaryngology Head & Neck Surgery, 134(3), 374–378.

Mitchell, R. B., & Kelly, J. (2007). Outcome of adenotonsillectomy for obstructive sleep apnea in obese and nor-mal-weight children. Otolaryngology Head & Neck Surgery, 137(1), 43–48.

O’Donoghue, F. J., Camfferman, D., Kennedy, J. D., Martin, A.J., Couper, T., Lack L.D. ,et al. (2005). Sleep-disorderedbreathing in Prader-Willi syndrome and its association with neurobehavioral abnormalities. Journal of Pediatrics,147(6), 823–829.

Ogden, C. L., Carroll, M. D., Curtin, L. R., McDowell, M. A., Tabak, C. J., & Flegal, K. M. (2006). Prevalence of overweightand obesity in the United States, 1999–2004. Journal of the American Medical Association, 295(13), 1549–1555.

Polhamus, B., Dalenius, K., Thompson, D., Scanlon, K., Borland, E., Smith, B., et al. (2003). Pediatric Nutrition Surveil-lance 2003 Report. Atlanta: U. S. Department of Health and Human Services, Centers for Disease Control and Preven-tion. Retrieved August 10, 2009, from http://www.cdc.gov/pedness/pdfs/PEDNESS_2002_Summary.pdf

660 MITCHELL AND BOSS

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4

Reynolds. C., & Kamphaus, R. (1998). Manual for behavioral assessment system for children. Circle Pines, MN: Ameri-can Guidance Service Inc.

Rudnick, E. F., & Mitchell, R. B. (2007). Behavior and obstructive sleep apnea in children: Is obesity a factor? Laryngo-scope, 117(8), 1463–1466.

Sans, C.O., Crabtree, V. M., & Kheirandish-Gozal, L. (2008). Increased morning brain natriuretic peptide levels in childrenwith nocturnal enuresis and sleep-disordered breathing: a community-based study. Pediatrics, 121(5), e1208–e1214.

Shine, N. P., Coates, H., & Lannigan, F. J. (2005). Obstructive sleep apnea, morbid obesity, and adenotonsillar surgery: Areview of the literature. International Journal of Pediatric Otorhinolaryngology, 69(11),1475–1482.

Somers, V. K., White, D. P., Amin, R., Abraham, W. T., Costa, F., & Culebras, A. (2008). Sleep apnea and cardiovasculardisease: An American Heart Association/American College of Cardiology Foundation Scientific Statement from theAmerican Heart Association Council for High Blood Pressure Research Professional Education Committee, Council onClinical Cardiology, Stroke Council, and Council on Cardiovascular Nursing. Journal of the American College of Car-diology, 52(8), 686–717.

Suen, J. S., Arnold, J. E., & Brooks, L. J. (1995). Adenotonsillectomy for treatment of obstructive sleep apnea in children.Archives of Otolaryngology Head & Neck Surgery, 121(5), 525–530.

Suratt, P. M., Barth, J. T., Diamond, R., D’Andrea, L., Nikova, M., Perriello,V. A., et al. (2007). Reduced time in bed andobstructive sleep-disordered breathing in children are associated with cognitive impairment. Pediatrics, 119(2),320–329.

Witmans, M. B., Keens, T. G., Davidson Ward, S. L., & Marcus, C. L. (2003). Obstructive hypopneas in children and ado-lescents: Normal values. American Journal of Respiratory Critical Care Medicine, 168(12),1540.

PEDIATRIC OBSTRUCTIVE SLEEP APNEA 661

Dow

nloa

ded

by [

The

Uni

vers

ity o

f B

ritis

h C

olum

bia]

at 2

1:20

29

Oct

ober

201

4