Relationship between Drug-Induced Sleep Endoscopy Findings,Tonsil Size,and Polysomnographic Outcomes of Adenotonsillectomy in Children

Abstract

Objective

(1) Determine the correlation of awake tonsil scores and preadenotonsillectomy (pre-AT) sleep endoscopy findings. (2) Assess the relationship between polysomnographic AT outcomes with awake tonsil scores and sleep endoscopy ratings of tonsil and adenoid obstruction.

Study Design

Retrospective case series with chart review.

Setting

Tertiary care children’s hospital.

Subjects and Methods

Children aged 1 to 18 years who underwent sleep endoscopy and AT from January 1, 2013, to August 30, 2016, were included. Pre-AT sleep endoscopy findings were scored with the Sleep Endoscopy Rating Scale. Awake tonsil scores and sleep endoscopy ratings were compared with Spearman correlation. Associations between changes in pre- and post-AT polysomnography parameters and (1) awake tonsil scoring and (2) sleep endoscopy scoring were assessed with 1-way analysis of variance and linear regression.

Results

Participants included 36 children (mean ± SD age, 6.8 ± 4.3 years; 68% male, 44% obese). Awake tonsil scores and sleep endoscopy ratings were strongly correlated (R = 0.58, P = .003). Awake tonsil scores were not associated with changes in any polysomnography parameters after AT (all P > .05), while sleep endoscopy ratings of adenotonsillar obstruction were significantly associated (all P < .05, R² = 0.16-0.35). Patients with minimal adenotonsillar obstruction during sleep endoscopy had less improvement than those with partial or complete obstruction (mean obstructive apnea-hypopnea index change: −8.2 ± 11.5 vs −15.9 ± 14.3, and −46.8 ± 31.3, respectively; P < .001).

Conclusions

In children at risk for AT failure, assessment of dynamic collapse with sleep endoscopy may better predict the outcome of AT than awake tonsil size assessment, thus helping to inform surgical expectations.

Keywords

obstructive sleep apnea pediatric drug-induced sleep endoscopy adenotonsillectomy polysomnography

Although adenotonsillectomy (AT) is the primary treatment for obstructive sleep apnea (OSA) in children, complete resolution of OSA after AT may be seen in only 25% to 75% of patients depending on the criteria used for diagnosing childhood OSA.^1-3 The influence of tonsil size on the outcome of AT has not been well studied. It seems intuitive to assume that small tonsils observed during a routine examination are unlikely to contribute significantly to OSA; therefore, AT is not likely to offer significant improvement. However, assessments of tonsil size during awake clinic examination have not correlated well with disease severity or outcome of AT.^4-7

Drug-induced sleep endoscopy (DISE) allows for the assessment of dynamic obstruction during sedated sleep. DISE therefore may be beneficial in the evaluation of children who have not yet undergone AT, by allowing a dynamic assessment of anatomic contributors to obstruction rather than a limited awake oropharyngeal examination in the clinic.^8,9 While a few studies have compared pre-AT DISE findings and awake tonsil scoring and generally found good correlation between these assessments,^8-10 there is a paucity of literature examining the relationship of either assessment with the outcome of AT.^7,11 This study had 2 objectives: first, to determine the correlation of tonsil size assessment and pre-AT DISE findings; second, to assess the relationship between polysomnography (PSG) outcomes of AT and (1) tonsil scores and (2) DISE ratings of tonsil and adenoid obstruction.

Materials and Methods

Study Population

This study is a retrospective case series with chart review of children aged 1 to 18 years who underwent DISE at the time of AT for OSA or sleep-disordered breathing from January 1, 2013, to August 30, 2016. Indications for DISE were based on risk factors for persistent post-AT OSA—specifically, obesity (body mass index >95 percentile for age), severe baseline OSA (obstructive apnea-hypopnea index [oAHI] >10), neuromuscular disorder, Down syndrome, or small (1+) tonsils. Only patients with pre- and post-AT PSG were included in the final analysis. Demographic data (age and sex), body mass index, clinic assessment of tonsil size, comorbidities (neuromuscular disorders, craniofacial anomalies, Down syndrome, asthma, prematurity, etc), and pre- and post-AT PSG data were extracted from the electronic medical record. Awake tonsil scores and DISE ratings made at the time of surgery were all assessed by the primary author (D.J.L.). This study was approved by the Oregon Health and Science University Institutional Review Board.

Tonsil Scoring

Tonsil size was assessed during an awake clinic examination with the Friedman scoring¹²: 1+, within the tonsillar fossa; 2+, beyond the tonsillar pillars but <50% toward midline; 3+, >50% toward midline; 4+, meeting at midline (“kissing tonsils”). With cooperative children, this scoring was made over 1 to 3 attempts with maximal oral opening and a relaxed tongue position in the floor of the mouth. With uncooperative children or where the tonsils could not be clearly visualized with mouth opening alone, ≥1 tongue blades were inserted into the oral cavity to depress the tongue and allow visualization of the oropharynx.

DISE Rating and Patterns of Obstruction

DISE was performed with a flexible 2.8-mm fiberoptic endoscope prior to AT. Patients breathed spontaneously and were sedated with a combination of dexmedetomidine (1- to 2-mcg/kg bolus over 5-10 minutes, followed by 0.5- to 1-mcg/kg/h infusion) and ketamine (1-mg/kg boluses up to 40 mg, as needed). Topical decongestant was not applied to the nasal cavities prior to endoscopic evaluation. After the optimal sedation level was achieved (defined as lack of response to transnasal insertion of the endoscope with active snoring), the DISE evaluation was completed over a period of 5 to 15 minutes depending on the ease of visualization due to severity of obstruction or presence of excess secretions. DISE findings were scored according to the previously published Sleep Endoscopy Rating Scale (SERS).¹³ The SERS is a validated tool for the systematic assessment of upper airway obstruction from the nose to the supraglottis in children undergoing DISE. Briefly, ratings were assessed on a 3-point scale (0 = no obstruction, 1 = partial obstruction, 2 = complete obstruction) at 6 anatomic levels of the upper airway: nasal airway, nasopharynx, velopharynx, oropharynx (including tonsils), base of tongue (including epiglottis), and arytenoids. Patterns of adenotonsillar obstruction were based on the combined obstructive ratings of the adenoids and tonsils: minimal, no significant obstruction of adenoids and tonsils (SERS ratings = 0 for both, defined as tonsils occupying <50% of oropharynx, adenoids not extending beyond eustachian tube opening); partial, complete obstruction at 1 site (SERS = 2) and no or partial obstruction at the other (SERS = 0 or 1); or complete, complete obstruction at both sites (SERS = 2 for both).

Polysomnography

All PSG was performed in an attended overnight sleep laboratory according to the guidelines of the American Academy of Sleep Medicine.¹⁴ This included electroencephalography, electro-oculography, electromyography of the submentalis and anterior tibialis, electrocardiography, nasal and oral airflow monitoring, thoracic and abdominal sensor belts, pulse oximetry, and audio/video recording. PSG parameters obtained from pre- and post-AT and included in the analysis were oAHI, rapid eye movement (REM) AHI, oxygen nadir, and oxygen desaturation index. OSA severity was defined as mild, moderate, or severe according to oAHI (1-4.9, 5-9.9, and ≥10, respectively). Residual OSA was defined as oAHI >1 on post-AT PSG.

Analysis

Cohort Characteristics

Patient characteristics were recorded for all patients who underwent DISE during the study period, regardless of PSG testing. To test for possible selection bias, patient characteristics of those included and excluded from the study were compared with 1-way analysis of variance for parametric continuous variables, with Kruskal-Wallis test for nonparametric continuous variables, and with either chi-square or Fisher’s exact test for categorical variables.

Comparison of Clinic Tonsil Size Assessment and DISE Findings

The correlation between awake tonsil score and SERS oropharyngeal/tonsil rating was quantified with a Spearman correlation coefficient. To test for differences in SERS ratings across awake tonsil scores, we used a Fisher’s exact test.

Response to AT Based on Awake Tonsil Score versus SERS Rating

Differences in mean pre- and post-AT PSG parameters were compared with a Student’s t test. To test for significant differences in the change from baseline for each parameter across categories of awake tonsil scoring or SERS rating of adenotonsillar obstruction, we used a 1-way analysis of variance. Univariate linear regression was used to quantify the strength of association between the change in each PSG parameter and (1) the awake tonsil score and (2) the SERS rating of combined adenotonsillar obstruction.

Results

A total of 88 children underwent DISE at the time of AT during the study period. Of these, 22 (25%) had no PSG testing at all; 22 (25%), only pre-AT PSG; 7 (8%), only post-AT PSG; and 36 (41%), pre- and postoperative PSG. Overall, there were no significant differences in patient characteristics based on pre- or post-AT PSG testing. However, children who underwent only pre-AT testing had fewer comorbidities than other subgroups, and those who had only post-AT PSG were more likely to be obese. The remainder of the analysis focused on patients who underwent pre- and postoperative PSG. Among these patients, age at the time of AT ranged from 1.5 to 16.2 years with a mean of 6.8 years, which was slightly younger than the other subgroups. There was a slightly higher proportion of boys (25 of 36, 69%), though these differences were not statistically significant ( Table 1 ). Nearly half were obese (16 of 36, 44%). The most common comorbidities were asthma (4 of 36, 11%) and Down syndrome (5 of 36, 14%). All patients were treated with AT, but some patients underwent concurrent procedures: tympanostomy tube placement (7), direct laryngoscopy and bronchoscopy (6), marsupialization of an ear dermoid cyst (1), interarytenoid injection laryngoplasty (1), and endoscopic repair of a type I laryngeal cleft (1). Residual post-AT OSA (oAHI >1) was present in 27 (75%) patients, and 18 (50%) patients had at least moderate residual OSA (oAHI ≥5). A total of 16 (44%) patients had minimal AT obstruction during DISE, while 7 (19%) had complete obstruction of the adenoids and tonsils. The most common sites of obstruction based on DISE findings were the oropharynx (including tonsils), velopharynx, and nasal airway ( Table 2 ).

Table 1.

Cohort Characteristics, Stratified by Polysomnography Testing.^a

Characteristic	Pre- and Post-AT PSG (n = 36)	No PSG (n = 22)	Pre-AT PSG Only (n = 22)	Post-AT PSG Only (n = 7)	P Value
Age at time of AT, y, mean ± SD	6.8 ± 4.3	8.1 ± 4.8	8.7 ± 4.6	9.4 ± 3.2	.33
Male	25 (69)	14 (64)	12 (55)	3 (43)	.59
BMI
Percentile, median [IQR]	92 [64, 100]	86 [48, 100]	86 [28, 100]	98 [85, 99]	.89
Obese	16 (44)	9 (41)	9 (41)	4 (57)	.88
Medical comorbidities^b
Neurologic condition	1 (3)	2 (9)	1 (5)	1 (14)	.46
Craniofacial anomaly	4 (11)	1 (5)	1 (5)	0 (0)	.79
Down syndrome	5 (14)	0 (0)	3 (14)	1 (14)	.24
Prematurity	3 (8)	3 (14)	0 (0)	0 (0)	.34
Asthma	4 (11)	2 (9)	1 (5)	2 (29)	.32
Other^c	9 (25)	6 (29)	2 (9)	2 (29)	.33
None	19 (53)	12 (55)	17 (77)	2 (29)	.10
Tonsil size (Friedman scale)^d
1+	3 (9)	0 (0)	5 (23)	1 (14)	.10
2+	6 (17)	5 (23)	7 (32)	2 (29)
3+	23 (66)	13 (59)	9 (41)	2 (29)
4+	3 (9)	4 (18)	1 (5)	2 (29)
Pattern of AT obstruction^e
Minimal	16 (44)	11 (50)	11 (50)	4 (57)	.41
Partial	13 (35)	11 (50)	8 (36)	2 (29)
Complete	7 (19)	0 (0)	3 (14)	1 (14)

Abbreviations: AT, adenotonsillectomy; BMI, body mass index; IQR, interquartile range; PSG, polysomnography.

Values are presented as n (%) unless noted otherwise.

Some patients had >1 comorbidity, so column percentages do not necessarily total 100%.

Includes Arnold-Chiari malformation, spina bifida, neurofibromatosis, developmental delay, and eosinophilic esophagitis.

Based on awake clinical examination.

Based on drug-induced sleep endoscopy findings.

Table 2.

Sleep Endoscopy Rating Scale Scores Among Patients with Pre- and Post-PSG (n = 36).^a

Anatomic Level	Score, Mean ± SD
Nasal airway	0.97 ± 0.64
Nasopharynx	0.84 ± 0.87
Velopharynx	1.27 ± 0.73
Oropharynx (including tonsils)	1.24 ± 0.76
Base of tongue (including epiglottis)	0.59 ± 0.76
Arytenoids	0.19 ± 0.46
Total score	5.17 ± 1.95

Abbreviation: PSG, polysomnography.

Scores based on 3-point rating scale: 0, no obstruction; 1, partial obstruction; 2, complete obstruction at each anatomic level.

Tonsil scores and SERS oropharyngeal ratings demonstrated a strong correlation (R = 0.58, P = .003). There was perfect agreement for small (1+) tonsils and for very large (4+) tonsils. However, there was a broad range of SERS scores among patients with tonsil scores of 2+ or 3+ ( Table 3 ). One patient did not have awake tonsils scores available due to lack of cooperation with clinic examination.

Table 3.

Frequency Distribution of Awake Tonsil Scores and SERS Oropharyngeal Rating.^a

	Awake Tonsil Score, n (%)^b
SERS Rating	1+	2+	3+	4+	Total
0	3 (100)	2 (33)	2 (9)	0 (0)	7 (20)
1	0 (0)	3 (50)	10 (43)	0 (0)	13 (37)
2	0 (0)	1 (17)	11 (48)	3 (100)	15 (43)
Total^c	3 (9)	6 (17)	23 (66)	3 (9)	35

Abbreviation: SERS, Sleep Endoscopy Rating Scale.

P = .005, Fisher’s exact test.

Column percentages.

Row percentages.

PSG parameters showed significant improvement after AT, with the exception of mean and maximal end tidal CO₂ ( Table 4 ). When stratified by the degree of AT obstruction during DISE, patients with minimal adenotonsillar obstruction had less post-AT improvement than those with partial or complete adenotonsillar obstruction (mean oAHI change: −8.2 ± 11.5 vs −15.9 ± 14.3 and −46.8 ± 31.3, respectively, P < .001; Table 5 ). In general, there was significant improvement in most PSG parameters (oAHI, REM AHI, O₂ nadir, and oxygen desaturation index) when more severe AT obstruction was observed preoperatively.

Table 4.

Polysomnography Parameters Before and After Adenotonsillectomy.^a

Parameter	Pre-AT	Post-AT	Change	P Value
AHI, events/h
Obstructive	23.2 ± 21.7	5.0 ± 4.1	−18.2 ± 21.6	<.0001
REM	42.7 ± 37.6	14.0 ± 11.2	−28.6 ± 39.3	.0003
O₂ nadir, %	81.9 ± 8.3	86.8 ± 6.1	4.9 ± 10.2	.007
ODI, events/h	18.6 ± 18.8	4.9 ± 4.8	−13.6 ± 18.2	.0002
ETCO₂, mm Hg
Mean	39.5 ± 9.6	41.2 ± 9.2	1.8 ± 10.9	.41
Maximal	51.2 ± 5.2	53.2 ± 14.6	2.0 ± 17.4	.56

Abbreviations: AHI, apnea-hypopnea index; ETCO₂, end-tidal carbon dioxide; O₂, oxygen; ODI, oxygen desaturation index; REM, rapid eye movement.

Data presented as mean ± SD.

Table 5.

Change in PSG Parameters after AT Based on Pattern of Obstruction.^a

	AT Obstruction
PSG Parameter	Minimal	Partial	Complete	P Value
AHI, events/h
Obstructive	−8.2 ± 11.5	−15.9 ± 14.3	−46.8 ± 31.3	.0005
REM	−10.2 ± 24.7	−28.0 ± 38.7	−72.9 ± 36.9	.002
O₂ nadir, %	1.8 ± 7.1	4.0 ± 11.7	13.6 ± 9.4	.03
ODI, events/h	−6.2 ± 10.9	−12.8 ± 14.3	−32.3 ± 26.6	.009

Abbreviations: AHI, apnea-hypopnea index; ETCO₂, end-tidal carbon dioxide; O₂, oxygen; ODI, oxygen desaturation index; PSG, polysomnography; REM, rapid eye movement.

Values are presented as change from baseline, mean ± SD.

There was no significant association between awake tonsil scores and any PSG parameter change after AT (all P > .05; Table 6 ). In contrast, SERS ratings of adenotonsillar obstruction were significantly associated with change in oAHI, REM AHI, oxygen nadir, and oxygen desaturation index (all P < .05). The ratings of AT obstruction accounted for approximately one-third of the variance of change in oAHI and REM oAHI (R² = 0.35 and 0.33, respectively).

Table 6.

Changes in PSG Parameters and Their Associations with Awake Tonsil Scores and SERS Ratings.^a

	Beta [95% CI]; R²
	oAHI	REM AHI	O₂ Nadir	ODI
Awake tonsil score	−8.9 [−20.4, 2.6]; 0.08	−22.6 [−45.9, 0.7]; 0.12	2.0 [−2.9, 6.8]; 0.02	−4.2 [−15.8, 7.4]; 0.02
SERS TA	−17.5 [−26.4, −8.5]; 0.35	−29.2 [−45.0, −13.4]; 0.33	5.3 [1.1, 9.5]; 0.16	−12.0 [−19.7, −4.3]; 0.26

Abbreviations: AHI, apnea-hypopnea index; O₂, oxygen; oAHI, obstructive apnea-hypopnea index; ODI, oxygen desaturation index; PSG, polysomnography; REM, rapid eye movement; SERS, Sleep Endoscopy Rating Scale; TA, combined tonsil and adenoid obstruction.

R² values indicate the proportion of variance in PSG parameter change attributable to predictor. Bold denotes P < .05, univariate linear regression.

Discussion

In this sample of children at high-risk for persistent OSA, we report that the pattern of adenotonsillar obstruction observed during DISE, but not tonsil size scored during awake clinical assessment, was significantly associated with the magnitude of improvement after AT. Children with increasing degrees of adenotonsillar obstruction showed greater improvement with respect to oAHI and several other PSG parameters after AT. We believe that the ability to account for dynamic pharyngeal collapse during DISE better reflects the physiologic process of obstruction during natural sleep than awake clinic assessment of tonsil size and therefore is better correlated with outcome of AT. This finding may seem intuitive but has not been clearly demonstrated to date. Our finding that there was agreement between DISE scores and awake tonsil grading for small (1+) and kissing (4+) tonsils but wide variation for intermediate tonsils (2+ or 3+) may further reflect the dynamic lateral pharyngeal collapse observed during DISE that could not be appreciated during awake clinic examination. Alternatively, the accuracy of findings during the awake clinic examination may have been negatively affected by poor patient cooperation during the examination. Both these factors likely contribute to our finding that DISE scores were significantly associated with changes in multiple PSG parameters, while awake tonsil scores were not.

Previous studies have found that subjective assessment of tonsil and adenoid size during awake clinical examination is not reliable in predicting the severity of OSA or the outcome of AT.^4-7 Although DISE is most often used in cases of residual OSA after AT,¹⁵ it is a low-risk procedure that has been shown to predict the outcome of upper airway procedures in adults^16,17 and has been used in some children for pre-AT evaluation and identification of obstruction.^8,9,11 Boudewyns et al reported a significant correlation between the awake tonsil score and the degree of tonsillar obstruction observed during DISE. However, they reported no correlation between tonsil score or DISE tonsil rating and AHI.⁹ Miller et al noted a similar correlation between the awake tonsil score and DISE tonsil rating and a moderate correlation between DISE tonsil rating and AHI in children without any significant comorbidities.¹⁰ Ulualp et al reported greater lateral oropharyngeal (tonsillar) obstruction among patients with moderate and severe OSA as compared with those with mild OSA.¹⁸ In contrast to these studies of baseline correlations, Alsufyani et al¹¹ attempted to examine pre-AT DISE findings and outcomes of AT, noting that tonsil size was negatively correlated with treatment failure (ie, larger tonsils during DISE correlated with success). However, the primary outcome in this study was a positive screen on a validated sleep questionnaire; the authors did not examine PSG outcomes.

This study reports a substantial improvement in PSG parameters following AT, consistent with previous studies.^2,3,19,20 However, with a criterion of AHI ≥2 to diagnose OSA, 75% of the study population continued to have residual OSA. This is a higher prevalence of residual OSA than what was reported in most previous studies.^2,3,19,20 This is likely due to the inclusion criteria in this study that specifically selected for patients at greater risk for residual OSA after AT. Post-AT OSA has been associated with several risk factors, including severe preoperative OSA, obesity, neuromuscular disorders, Down syndrome, and craniofacial anomalies.^3,21-24 As many as 88% of obese children have residual OSA after AT,²⁵ and at least 30% to 50% of children with Down syndrome have persistent OSA after AT.^22,26-28 DISE has identified multilevel obstruction in 85.4% of children with Down syndrome²⁸ and 57% to 71% of nonsyndromic children.^9,18

Our results suggest that in patients at risk for AT failure, performing DISE prior to or at the time of AT may better predict the outcome of AT than relying on clinical examination or comorbidities alone. While one could choose to schedule routine follow-up or even routine post-AT PSG in all patients with risk factors for persistent OSA, this would be an inefficient use resources and may not be feasible at most centers. Therefore, to help more efficiently identify patients most likely to fail AT who would most benefit from closer postoperative follow-up or an alternative treatment plan, we recommend consideration of routine pre-AT DISE in any patient where there is high risk for AT failure or uncertainty about the optimal surgical intervention. In this study, this included patients with severe baseline OSA, neuromuscular disorders, Down syndrome, obesity, or clinically small tonsils. However, this study was not designed or intended to determine the optimal indications for DISE, and specific subgroup analysis was not possible due to the small sample size. It is worth noting that routinely performing DISE prior to AT does add a small but substantial amount of time to each procedure, particularly given the initial loading dose of dexmedetomidine and the subsequent slower recovery. Operating room and anesthesia efficiency, reimbursement for DISE, and productivity expectations all vary widely by setting and institution, and it may be impractical for some otolaryngologists to routinely perform DISE for all patients who are considered high risk for AT failure. While we believe that there is prognostic value in routinely performing DISE in this population, larger multi-institutional studies are needed to better elucidate which patients would most benefit from this evaluation.

There are significant strengths to this study, including the systematically collected DISE assessments via a previously validated rating scale, with ratings documented at the time of AT so that the outcomes of surgery did not influence the DISE ratings, and the inclusion of patients with pre- and postoperative PSG data. However, there are important limitations as well. First, this is a retrospective study with a large number of excluded patients because of missing pre- or post-AT PSG. It is possible that patients with complete symptom resolution would be less likely to return for a postoperative PSG, therefore selecting patients with a greater likelihood of AT failure and creating a possible selection bias. However, comparison of baseline characteristics between patients with pre- and post-AT PSG and those without did not reveal any significant differences in demographics or comorbidities. Examination of the differences in awake tonsil size stratified by PSG testing suggests that patients with only pre-AT PSG were more likely to have small (1+) tonsils than any other subgroup, though this difference did not reach statistical significance (P = .10). However, the pattern of AT obstruction observed during DISE did not differ significantly between patients with and without pre- and post-AT PSG, suggesting that the relationship between pre-AT DISE findings and PSG outcomes is likely generalizable to patients who did not complete pre- and post-AT PSG.

Second, this study includes a heterogeneous population selected on the basis of risk factors for increased risk of AT failure, which may limit the application of these results to the broader population of children with OSA or any specific subgroup. It is possible that the lack of statistically significant differences across subgroups of PSG testing may have been due to a relatively small sample size, limiting statistical power. As noted, we did not have a sufficient sample size to allow robust subgroup analysis based on the different inclusion criteria; however, we think that the inclusion of multiple risk factors associated with AT failure adds to the potential generalizability of our findings.

Finally, this study focused on the assessment of tonsil and adenoid obstruction by DISE and comparison with clinic examination with respect to AT outcome. While we did investigate other anatomic levels as part of the SERS ratings, we did not observe any significant associations (data not shown). This may reflect the limited sample size and the relatively small impact on the change in PSG parameters after AT of obstruction at other anatomic sites. Further prospective studies with larger sample sizes are needed to better determine whether the relationships observed in this study are generalizable to all patients undergoing AT or specific subgroups.

Conclusion

In our study, preoperative DISE findings, but not clinical assessment of tonsil size, were significantly associated with the outcome of AT. Less severe adenotonsillar obstruction was associated with less PSG improvement. Preliminary data suggest that in this population at high risk for persistent OSA after AT, assessment of the pattern of dynamic collapse and airway obstruction with DISE may better predict the outcome of AT than tonsil size assessment in clinic and potentially help guide expectations and management of residual OSA.

Author Contributions

Derek J. Lam, study design, data collection, analysis and interpretation, manuscript editing and approval; Natalie A. Krane, data collection and interpretation, manuscript preparation, editing, and approval; Ron B. Mitchell, study design, data interpretation, manuscript editing and approval.

Disclosures

Competing interests: None.

Sponsorships: None.

Funding source: NHLBI K23 HL127132. This funding contributed to study coordination, data collection and storage, and biostatistical consulting costs.

Footnotes

Acknowledgements

We acknowledge Eleni O’Neill, MPH, for her assistance with data collection and database management and Steven Shea, PhD, for his assistance with manuscript editing.

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

This article was presented at the American Society of Pediatric Otolaryngology Annual Meeting; May 18-21, 2017; Austin, Texas.

References

Tauman

Gulliver

Krishna

, et al. Persistence of obstructive sleep apnea syndrome in children after adenotonsillectomy. J Pediatr. 2006;149:803-808.

Friedman

Wilson

Lin

Chang

. Updated systematic review of tonsillectomy and adenoidectomy for treatment of pediatric obstructive sleep apnea/hypopnea syndrome. Otolaryngol Head Neck Surg. 2009;140:800-808.

Bhattacharjee

Kheirandish-Gozal

Spruyt

, et al. Adenotonsillectomy outcomes in treatment of obstructive sleep apnea in children: a multicenter retrospective study. Am J Respir Crit Care Med. 2010;182:676-683.

Howard

Brietzke

. Pediatric tonsil size: objective vs subjective measurements correlated to overnight polysomnogram. Otolaryngol Head Neck Surg. 2009;140:675-681.

Nolan

Brietzke

. Systematic review of pediatric tonsil size and polysomnogram-measured obstructive sleep apnea severity. Otolaryngol Head Neck Surg. 2011;144:844-850.

Hwang

Guilleminault

Park

Kim

Hong

. Usefulness of adenotonsillar size for prediction of severity of obstructive sleep apnea and flow limitation. Otolaryngol Head Neck Surg. 2013;149:326-334.

Tang

Benke

Cohen

Ishman

. Influence of tonsillar size on OSA improvement in children undergoing adenotonsillectomy. Otolaryngol Head Neck Surg. 2015;153:281-285.

Truong

Woo

Koltai

. Sleep endoscopy as a diagnostic tool in pediatric obstructive sleep apnea. Int J Pediatr Otorhinolaryngol. 2012;76:722-727.

Boudewyns

Verhulst

Maris

Saldien

Van de Heyning

. Drug-induced sedation endoscopy in pediatric obstructive sleep apnea syndrome. Sleep Med. 2014;15:1526-1531.

10.

Miller

Purcell

Dahl

, et al. Clinically small tonsils are typically not obstructive in children during drug-induced sleep endoscopy. Laryngoscope. 2017;127:1943-1949.

11.

Alsufyani

Isaac

Witmans

Major

El-Hakim

. Predictors of failure of DISE-directed adenotonsillectomy in children with sleep disordered breathing. J Otolaryngol Head Neck Surg. 2017;46:37.

12.

Friedman

Tanyeri

La Rosa

, et al. Clinical predictors of obstructive sleep apnea. Laryngoscope. 1999;109:1901-1907.

13.

Lam

Weaver

Macarthur

, et al. Assessment of pediatric obstructive sleep apnea using a drug-induced Sleep Endoscopy Rating Scale. Laryngoscope. 2016;126:1492-1498.

14.

Berry

Budhiraja

Gottlieb

, et al. Rules for scoring respiratory events in sleep: update of the 2007 AASM manual for the scoring of sleep and associated events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med. 2012;8:597-619.

15.

Friedman

Parikh

Ishman

, et al. The current state of pediatric drug-induced sleep endoscopy. Laryngoscope. 2017;127:266-272.

16.

Koutsourelakis

Safiruddin

Ravesloot

Zakynthinos

de Vries

. Surgery for obstructive sleep apnea: sleep endoscopy determinants of outcome. Laryngoscope. 2012;122:2587-2591.

17.

Soares

Sinawe

Folbe

, et al. Lateral oropharyngeal wall and supraglottic airway collapse associated with failure in sleep apnea surgery. Laryngoscope. 2012;122:473-479.

18.

Ulualp

Szmuk

. Drug-induced sleep endoscopy for upper airway evaluation in children with obstructive sleep apnea. Laryngoscope. 2013;123:292-297.

19.

Mitchell

. Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and postoperative polysomnography. Laryngoscope. 2007;117:1844-1854.

20.

Marcus

Moore

Rosen

, et al. A randomized trial of adenotonsillectomy for childhood sleep apnea. N Engl J Med. 2013;368:2366-2376.

21.

Wiet

Bower

Seibert

Griebel

. Surgical correction of obstructive sleep apnea in the complicated pediatric patient documented by polysomnography. Int J Pediatr Otorhinolaryngol. 1997;41:133-143.

22.

Myatt

Beckenham

. The use of diagnostic sleep nasendoscopy in the management of children with complex upper airway obstruction. Clin Otolaryngol Allied Sci. 2000;25:200-208.

23.

Liu

Zhang

, et al. Outcome of adenotonsillectomy for obstructive sleep apnea syndrome in children. Ann Otol Rhinol Laryngol. 2010;119:506-513.

24.

Schechter

. Technical report: diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics. 2002;109:e69.

25.

Costa

Mitchell

. Adenotonsillectomy for obstructive sleep apnea in obese children: a meta-analysis. Otolaryngol Head Neck Surg. 2009;140:455-460.

26.

Fung

Witmans

Ghosh

Cave

El-Hakim

. Upper airway findings in children with Down syndrome on sleep nasopharyngoscopy: case-control study. J Otolaryngol Head Neck Surg. 2012;41:138-144.

27.

Shete

Stocks

Sebelik

Schoumacher

. Effects of adeno-tonsillectomy on polysomnography patterns in Down syndrome children with obstructive sleep apnea: a comparative study with children without Down syndrome. Int J Pediatr Otorhinolaryngol. 2010;74:241-244.

28.

Maris

Verhulst

Saldien

Van de Heyning

Wojciechowski

Boudewyns

. Drug-induced sedation endoscopy in surgically naive children with Down syndrome and obstructive sleep apnea. Sleep Med. 2016;24:63-70.