The Laryngoscope C 2015 The American Laryngological, V

Rhinological and Otological Society, Inc.

Systematic Review

Systematic Review of Site of Obstruction Identification and Non-CPAP Treatment Options for Children with Persistent Pediatric Obstructive Sleep Apnea P. Vairavan Manickam, MD; Sally R. Shott, MD; Emily F. Boss, MD, MPH; Aliza P. Cohen, MA; Jareen K. Meinzen-Derr, PhD; Raouf S. Amin, MD; Stacey L. Ishman, MD, MPH Objectives/Hypothesis: Although adenotonsillectomy is accepted as a first-line therapy for pediatric obstructive sleep apnea (OSA), there is currently no consensus regarding optimal methods for identifying the sites of obstruction or treatment of children with persistent disease after surgery. With this in mind, our aim was to systematically review the Englishlanguage literature pertaining to these issues. Study Design: Systematic review. Methods: We searched all indexed years of Pubmed, Cochrane CENTRAL, DynaMed, UpToDate, CINAHL, and Scopus for English-language articles containing original human data, with  7 participants, all < 18 years old. Data regarding study design, demographics, clinical characteristics/outcomes, level of evidence, and risk of bias were obtained. Articles were independently reviewed by two investigators. Results: Of 758 identified abstracts, 24 articles (combined population 5 960) were ultimately included. Seventeen (71%) described methods to identify site(s) of obstruction: drug-induced sleep endoscopy (11/24), cine magnetic resonance imaging (MRI) (3/24), and alternative imaging (3/24). Treatment options included lingual tonsillectomy (n 5 6), with success rates of 57% to 88% (Cohen’s effect size d 5 1.38), as well as supraglottoplasty (n 5 4), with success rates of 58% to 72% (d 5 0.64). Additional treatments included medications and surgery (e.g., partial midline glossectomy and tongue suspension). Conclusions: Drug-induced sleep endoscopy and cine MRI are the most commonly reported tools to identify sites of obstruction for children with persistent OSA; however, these techniques have not yet been clearly linked to outcomes. Evidence for treatment is extremely limited and focuses primarily on lingual tonsillectomy and supraglottoplasty. Also, reports regarding appropriate patient selection and outcomes in obese or otherwise healthy children are scant. Key Words: Diagnosis, treatment, persistent obstructive sleep apnea, pediatric, drug-induced sleep endoscopy, cine MRI, lingual tonsillectomy, supraglottoplasty, sleep state-dependent laryngomalacia. Laryngoscope, 126:491–500, 2016

From the Division of Pediatric Otolaryngology–Head and Neck Surgery (P.V.M., S.R.S., A.P.C., S.L.I.), the Division of Biostatistics and Epidemiology (J.K.M-D.), the Division of Pulmonary Medicine (R.S.A., S.L.I.), Cincinnati Children’s Hospital Medical Center; the University of Cincinnati College of Medicine (P.V.M., S.R.S., R.S.A., S.L.I.), Cincinnati, Ohio, and the Department of Otolaryngology–Head and Neck Surgery, Johns Hopkins University School of Medicine (E.F.B.), Baltimore, Maryland, U.S.A. Editor’s Note: This Manuscript was accepted for publication on May 19, 2015. Presented as a poster presentation at the Triological Society Combined Sections Meeting, Coronado Island, California, U.S.A., January 24, 2015. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Stacey L. Ishman, MD, MPH, Surgical Director, Upper Airway Center, Divisions of Pediatric Otolaryngology– Head and Neck Surgery and Pulmonary Medicine, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, MLC 2018, Cincinnati, OH 45229. E-mail: [email protected] Additional Supporting Information may be found in the online version of this article. DOI: 10.1002/lary.25459

Laryngoscope 126: February 2016

INTRODUCTION Obstructive sleep apnea (OSA) occurs in 2% to 4% of children in the United States, and guidelines from both the American Academy of Otolaryngology–Head and Neck Surgery and the American Academy of Pediatrics recommend adenotonsillectomy (T&A) as the firstline treatment.1,2 Although T&A has historically been considered curative, the most recent (2009) metaanalysis indicates a 34% to 40% likelihood of persistent OSA following T&A.3–7 Because OSA is known to be associated with physical, psychological, neurocognitive, and behavioral sequelae,8 early identification of children who are likely to have persistent disease can have a significant clinical impact. Children identified as high-risk for persistent OSA include those with craniofacial and mandibular anomalies (i.e., Down syndrome [DS], Pierre Robin sequence, and Treacher

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

491

TABLE I. MESH Terms Used for Comprehensive Literature Search on the Topic of Diagnosis and Management of Pediatric Persistent Obstructive Sleep Apnea. Database

MESH Term

Dynamed

Obstructive sleep apnea in children

UpToDate PubMed

Evaluation of suspected obstructive sleep apnea in children Management of suspected obstructive sleep apnea in children ((“diagnosis”[Subheading] OR “diagnosis”[All Fields] OR “diagnosis”[MeSH Terms]) OR (“therapy”[Subheading] OR “therapy”[All Fields] OR “therapeutics”[MeSH Terms] OR “therapeutics”[All Fields]) OR (“disease management”[MeSH Terms] OR (“disease”[All Fields] AND “management”[All Fields]) OR “disease management”[All Fields])) AND (((recurrent[All Fields] OR recurring[All Fields]) AND (“sleep apnea”[All Fields] OR “sleep apnea syndromes”[MeSH Terms] OR (“sleep”[All Fields] AND “apnea”[All Fields] AND “syndromes”[All Fields]) OR “sleep apnea syndromes”[All Fields] OR (“sleep”[All Fields] AND “apnea”[All Fields]) OR “sleep apnea”[All Fields])) AND (“infant”[MeSH Terms] OR “child”[MeSH Terms] OR “adolescent”[MeSH Terms]))—limited to English

CINAHL

(MM "Sleep apnea syndromes1") AND (recurrent OR recurring)

Scopus

(TITLE-ABS-KEY(((recurring OR recurring) "sleep apnea")) AND ((child OR infant OR adolescent)) AND (management OR therapy OR diagnosis OR prevention)

MeSH 5 Medical Subject Headings.

Collins syndrome), obesity, severe OSA prior to surgery, and neurologic impairment.3,6,9,10 Once the definitive diagnosis is established with polysomnography (PSG), the site or sites of obstruction are typically identified by performing a thorough physical examination. This includes careful inspection for possible residual tonsil tissue, adenoid hypertrophy, lingual tonsil hypertrophy, or nasal obstruction. In addition, awake flexible laryngoscopy may be performed or lateral neck plain films may be taken; however, neither allows for an evaluation of the airway during sleep or sedation or for easy identification of multiple sites of obstruction. In view of these limitations, diagnostic modalities such as cine magnetic resonance imaging (MRI) and sleep endoscopy have been introduced. Once these sites have been identified, treatment options can be determined. Depending on the severity of the OSA, these options may include monitored observation,11 continuous positive airway pressure (CPAP), antiinflammatory medications,12 oral appliances, orthodontic treatment (e.g., rapid maxillary expansion), or surgery.13,14 Given the lack of consensus regarding optimal methods for identifying sites of obstruction, as well as the most effective options other than CPAP for treating persistent pediatric OSA, we conducted a review to summarize the existing literature pertaining to these issues.

We searched the PubMed, Cochrane CENTRAL, DynaMed, UpToDate, CINAHL, and Scopus databases on January 22, 2014, and updated the search on December 17, 2014. The search strategy was created in conjunction with a medical librarian using Medical Subject Headings intended for PubMed, and then tailored for the other databases (Table I). Two or more independent investigators reviewed all identified titles, abstracts, and full-text articles to determine if they met inclusion criteria (Fig. 1). These investigators included four attendings and one medical student. In the event of

MATERIALS AND METHODS Our study methodology was based on the Preferred Reporting Items for Systemic Reviews and Meta-Analysis checklist and statement recommendations. We reviewed articles regarding the treatment of persistent OSA, which we defined as residual disease after tonsillectomy with or without adenoidectomy, as determined by overnight PSG. Articles were included if they: 1) reported on children 1 to 18 years of age who had undergone identification of site of obstruction or treatment for persistent OSA confirmed on PSG; 2) contained an abstract; 3) were published in English; 4) included a minimum of seven patients; and 5) were not classified as review articles. We excluded studies that used bony framework surgery or CPAP as the only treatment modality.

Laryngoscope 126: February 2016

492

Fig. 1. Flow diagram of literature search based on the Preferred Reporting Items for Systemic Reviews and Meta-Analysis checklist and statement recommendations.42

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

TABLE II. Characteristics of Included Studies: Cine MRI to Identify Site of Obstruction for Children With Persistent OSA Following Adenotonsillectomy. Children With Persistent OSA

Children With Identified Site of Obstruction

Author

Year

LOE

N

Mean Age (years)

Shott

2004

4

15

10.3*

15/15 (100%)

14/15 (93%)

Cine MRI may be useful in identifying multiple levels of airway obstruction, particularly in children with DS. 93% of patients in this study had multiple levels of airway obstruction, most commonly glossoptosis and hypopharyngeal collapse. Findings may be used to tailor surgical management.

Donnelly

2004

4

27

9.9*

27/27 (100%)

Not reported

Fricke

2006

3

89

9.71 (5.63)

52/89 (58%)

17/52 (33%) with OSA

Children with DS had propensity to have persistent OSA due to multiple causes, most commonly macroglossia and enlarged adenoid tonsils as identified by cine MRI. Limited data regarding use of cine MRI in surgical planning to improve AHI. LTH was present in 33% of persistent OSA patients, particularly those with DS; 0% of control patients (no history of OSA) had LTH. Cine MRI sleep studies may help identify multiple levels of obstruction not readily apparent on physical exam or radiographic studies.

Comments

*No standard deviation reported. AHI 5 apnea–hypopnea index; DS 5 Down syndrome; LOE 5 level of evidence; LTH 5 lingual tonsillar hypertrophy; MRI 5 magnetic resonance imaging; N 5 total number of patients; OSA 5 obstructive sleep apnea.

disagreement regarding study inclusion, reviewers reached a consensus on inclusion or exclusion. Information obtained from each article pertaining to the identification of site of obstruction included authors, year of publication, diagnostic technique, number of patients, mean age, and results. For studies that discussed treatment options (e.g., lingual tonsillectomy and supraglottoplasty), outcome measures of effectiveness were documented. These measures include mean apneahypopnea index (AHI) after surgery and the percentage of subjects who were successfully treated by surgery. The level of evidence for each study was determined according to the guidelines outlined by the Center for EvidenceBased Medicine (http://www.cebm.net/index). The NewcastleOttawa Scale for assessment of quality of nonrandomized studies in meta-analysis was also used to evaluate each article. This scale assigns ratings based on three categories: 1) selection of cases and controls; 2) comparability of controls on basis of design/analysis; and 3) methods of ascertaining exposure. In view of the limited literature, diverse topic and study designs in identified studies, and significant heterogeneity of data, a quantitative meta-analysis was not performed. Articles were analyzed using qualitative synthesis methods as follows: 1) they were grouped into categories by content area (identification of site of obstruction or treatment options); 2) common themes were identified; and 3) subgroup synthesis was carried out. A number of articles focused on more than one content area and were therefore included in more than one of the described content areas. Because this study reviewed previously published studies, it was exempt from evaluation by our institutional review board.

Statistical Analysis and Effect Size Calculation We reported separate Cohen’s d effect sizes for all studies selected for inclusion that elucidated the prevalence of persis-

Laryngoscope 126: February 2016

tent OSA via pre- and postoperative PSG in the content areas of lingual tonsillectomy, supraglottoplasty, and drug-induced sleep endoscopy (DISE). Cohen’s d effect sizes were calculated for each of the content areas separately by dividing the mean difference between pre- and postmeasures (within-group difference) by their pooled standard deviation (SD). The correlations between pre- and postmeasures are needed for the Cohen’s d calculation. Because this correlation was unknown for the reported studies, we entered a global estimation of r 5 0.7 into the formula.15 Because the sample size varied across studies, we computed a weighted mean across all studies for which the inverse of the estimated SD for each study served as a weight.15 Confidence intervals were calculated based on the standard error of the overall mean effect size.

RESULTS We identified 758 abstracts. Of these, 133 underwent full-text review and 24 were included in the final analysis (Fig. 1). As shown in Tables II to IV, studies of the methods used to identify sites of obstruction included plain films, sleep fluoroscopy, computed tomography, cine MRI, awake laryngoscopy, and DISE. No studies were identified that discussed oral appliances or orthodontic treatment for the management of persistent OSA.

Diagnostic Techniques Three studies16–18 used cine MRI to identify the site of obstruction (Table II). These combined studies included 131 patients (sample size range 15–89) with a pooled mean age of 9.8 years (range 9.7–10.3 years). All

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

493

studies included children with DS. The site of obstruction was identified in 33% to 93% of children. Eleven studies19–29 comprising a combined total of 375 patients (sample size range 8–22) with a pooled mean age of 7.2 years (range 2.2–11 years) reported on DISE (Table III); however, only five of these studies22–25,27 specifically reported the percentage of patients in whom DISE identified the site of obstruction. Combined findings of these five studies showed that DISE was successful in identifying this site in 100% of children (n 5 162, sample size range 8–82). Cohen’s effect size value (d 5 2.14; 95% confidence interval [CI], 2.12– 2.17) suggested a large effect size in favor of this diagnostic examination. Four studies24,25,27,28 discussed the utility of DISE in identifying multiple levels of obstruction. Most common reported sites of obstruction were the tongue base, adenoids secondary to regrowth, inferior turbinates,24 velum, and lateral oropharyngeal walls.25 One study28 proposed a scoring system to standardize the reporting of DISE outcomes. Three studies30–32 (Table IV) described other modalities to discern the site of obstruction. One study30 compared sleep fluoroscopy to direct laryngoscopy and bronchoscopy to identify upper airway obstruction in 50 patients. Authors of this study reported that sleep fluoroscopy identified sites of obstruction not recognized on direct laryngoscopy and bronchoscopy in 27 (54%) children and altered the course of treatment in 26 (52%). A study using plain film evaluation31 found that lingual tonsillar hypertrophy was more common in children with DS than in non-DS controls and could be identified on plain films obtained for evaluation of cervical spine abnormalities. The third study evaluated the usefulness of MRI in identifying volumetric changes of the upper airway following T&A in 27 children.32 Authors of this study found that obese children with OSA had significant residual adenoid tissue and enlargement of the soft palate and tongue volume.

Treatment Studies of treatment modalities after T&A included lingual tonsillectomy, supraglottoplasty, partial midline glossectomy, and tongue suspension with or without lingual tonsillectomy. Given that outcomes reported in these studies were heterogeneous, not all parameters discussed below were available in each study. For studyspecific data, refer to Tables V to VII. Four studies19,21,22,33 with a combined sample size of 77 patients (range 8 to 24) reported on supraglottoplasty to treat persistent pediatric OSA (Table V). The mean reported age was 5.7 years.22,33 The pooled mean AHI improved from 12.1 events per hour before surgery to 4.4 events per hour after surgery.19,21,22,33 Only one study21 reported on OSA resolution, finding that 58% of patients had a postoperative AHI < 5 and 16% had a postoperative AHI < 1. Cohen’s effect size value (d 5 0.84; 95% CI, 0.79 20.89) suggested a medium effect of treatment on the AHI and respiratory distress index (RDI) as compared to nontreatment. Laryngoscope 126: February 2016

494

We identified six studies that reported on the results of lingual tonsillectomy,19,20,29,33–35 three of which used this procedure in combination with other surgery19,29,33 (Tables VI and VII). The combined sample size of studies that reported on the use of lingual tonsillectomy alone19,20,33,34 was 141 patients (range 16–68 patients). The reported pooled mean age19,20,29,33 was 9.7 years (range 7–11 years). The pooled mean AHI was 13.9 events per hour before surgery and 8.0 events per hour after surgery.19,29,33 When treatment success was defined as a postoperative AHI < 5, reported success rates were 57% and 61%.20,33 When success was defined as a postoperative AHI < 1, success rates were 17% and 22%.20,33 Cohen’s effect size value (d 5 1.39; 95% CI, 1.33–1.43) suggested a large effect of treatment on the AHI and RDI as compared to nontreatment. Five studies29,35–38 investigated other therapies for the treatment of persistent OSA (Table VII). Two of these studies36,37 (n 5 43) carried out at a single institution employed subperiosteal release of the floor of mouth musculature as a primary treatment for children with Pierre Robin sequence. Authors found significant improvements in oxygen saturation and a decrease in the AHI from 46.5 to 17.4 events per hour, reporting that 84% of patients had correction of respiratory obstruction. Two studies authored by Wootten et al.29,35 reported on patients who underwent more than one procedure for persistent OSA. In the first,29 30 of 31 patients underwent tongue suspension, along with radiofrequency ablation of the base of tongue, and 10 patients underwent additional concurrent procedures. These authors reported resolution of OSA in 61% of patients. The second study35 primarily reported on the results of multilevel surgery, which included various combinations of lingual tonsillectomy, posterior midline glossectomy, revision adenoidectomy, inferior turbinate submucosal resection, uvulopalatopharyngoplasty, and supraglottoplasty. Authors found that the mean AHI decreased from 7.0 to 3.6 events per hour. Kheirandish et al.38 described the use of combined oral antileukotriene and intranasal steroid therapy for patients with residual OSA. They reported that in 22 children (mean age 6.3 years) who underwent this therapeutic approach, there was a decrease in the mean obstructive AHI from 3.9 to 0.3 events per hour as compared to controls, who had no improvement in AHI.

DISCUSSION Although T&A is the primary treatment for pediatric OSA, complete resolution of OSA occurs in only 60% to 66% of affected children.7 In view of the fact that multiple studies of adults with OSA indicate that the majority have multilevel obstruction,39–41 multilevel surgery has become the standard of care for those who undergo surgical treatment. To appropriately target treatment for these patients, accurate identification of the site(s) of obstruction is imperative. Experience with children who have persistent disease suggests that similar to their adult counterparts, they too may have multilevel obstruction; however, optimal methods of identifying

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

Laryngoscope 126: February 2016

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

495

2012

2013

2013

Truong

Fishman

Ulualp

2012

Digoy

2012

2012

Chan & Koltai

Fung

2009

Lin

2012

1999

Myatt

Durr

Year

Author

4

4

2b

4

2b

4

2b

2b

2b

LOE

Not reported

7.4†

82

28

100%; stratified into mild, moderate, severe OSA

100%; not defined per AHI

7.65†

6 (3.7)

Not reported 100%; not defined per AHI

Not reported

Not reported

69%; defined by AHI > 1/hr

Not reported Not reported

7.6 (4.14)

23 Control

39 surgically na€ıve 41 persistent OSA

7.09 (4.37)

7.8 (3.3)

100%; defined by AHI > 2/hr

100%; not defined per AHI

11†

Not reported

Not reported

Percentage of Children With Persistent OSA

Not reported

23 DS

13

36

22

26

8

N

Mean Age (years)

28 (27)

Not reported

Not reported (no intervention done)

Not reported (no intervention done)

7.79 (9.9)

15.7 (13.3)

20.7†

8.0 (8.0)

Not reported (no intervention done)

Not reported (no intervention done)

Not reported (no intervention done)

4.1 (5.0)

5.4†

8.1*,†

4.6 (4.5)

Postop AHI

13.8 (15.9)

Not reported

Not reported

7.9 (7.3)

13.3 (12.9)

15.4†

14.7*,†

48 (15.5)

Preop AHI

Not reported

Not reported Not reported

12/23 (52%)

23/23 (100%)

13/13 (100%)

36/36 (100%)

Not reported

Not reported

8/8 (100%)

Children With Identified Site of Obstruction

Majority of subjects had multilevel obstruction; most common sites identified were velum and oropharynx/lateral walls. In children with single-site

Compared to awake endoscopy, sleep endoscopy identified more cases in which the nasopharynx, lateral pharyngeal walls, tongue base, and supraglottis caused obstruction. Both underestimated collapse when compared to PSG.

39 surgically na€ıve children with OSA and 41 children with persistent OSA. Sleep endoscopy was effective in identifying sites of obstruction responsible for OSA in surgically na€ıve as well as persistent OSA patients.

23 patients with DS and 23 controls; DS children exhibited similar prevalence of tonsillar obstruction, significantly less adenoidal obstruction, and significantly more pharyngeal and lingual collapse.

Multilevel disease was prevalent in patient with persistent OSA and was readily identified by sleep endoscopy; 85% of patients had comorbidity.

Sleep endoscopy was effective in diagnosing SDL. Significant improvement in AHI was noted in patients with SDL following laser supraglottoplasty.

Occult laryngomalacia was readily detected by sleep endoscopy and was amenable to treatment by supraglottoplasty.

Sleep endoscopy enabled improved visibility and decreased surgical uncertainty in lingual tonsillectomy; 61% of patients had postop AHI < 5; 17% had postop AHI < 1.

Classified children into 4 levels of obstruction for surgical planning. Significant reduction in AHI for patients with initially severe airway disease. Rigid laryngobronchoscopy recommended to avoid missing distal obstruction.

Comments

TABLE III. Characteristics of Studies: Drug-Induced Sleep Endoscopy to Identify Site of Obstruction for Children With Persistent OSA Following Adenotonsillectomy.

496

*Respiratory disturbance index reported; not AHI. † No standard deviation reported. AHI 5 apnea–hypopnea index; DS 5 Down syndrome; LOE 5 level of evidence; N 5 total number of patients; OSA 5 obstructive sleep apnea; Preop 5 preoperative; Postop 5 postoperative; PSG 5 polysomnography; SDL 5 state-dependent laryngomalacia.

2014 Wootten

2b

11

8.1 (3.5)

100%; defined by AHI > 1/hr

Not reported

7.0 (5.8)

3.6 (1.8)

5 locations in upper aerodigestive tract were evaluated on a newly designed 4point scale for obstruction. This proposed scoring system allows standardization of reporting sleep endoscopy outcomes. Aggregate scoring in multilevel obstruction correlated with worse PSG parameters. 31 patients underwent sleep endoscopy; 11 had complete pre- and postoperative PSG data. 26 patients underwent concurrent intervention, and subjective improvement was reported in 92%. Not reported (no intervention done) 24.8† Not reported Not reported 2.2† 23 2014 Chan

4

N LOE Year Author

airway obstruction, the most common site was oropharynx/lateral walls.

Comments Postop AHI Preop AHI Children With Identified Site of Obstruction Mean Age (years)

Percentage of Children With Persistent OSA

TABLE III. (Continued)

Laryngoscope 126: February 2016

sites of obstruction as well as optimal treatment approaches remain unclear. We reviewed the existing literature regarding the identification of the site of obstruction and non-CPAP treatment approaches for children with persistent pediatric OSA after T&A. Our results showed that cine MRI and DISE were the two most commonly used diagnostic techniques and that lingual tonsillectomy and supraglottoplasty were the most commonly used treatment approaches. Although cine MRI was reported in three studies,6–18 all of these studies were carried out at the same institution—suggesting that this technique may not be widely used. Each of the studies included children with DS and reported that multiple levels of obstruction were identified in most of these children. The most commonly identified sites of obstruction were the base of tongue—with both glossoptosis and hypopharyngeal collapse—and the adenoids. However, given the paucity of literature on the use of cine MRI in children who do not have DS, these findings are difficult to extrapolate to other pediatric populations. In addition, authors of two of the three studies16,17 reported an increased likelihood of lingual tonsil hypertrophy in children with DS as compared to non-DS controls. They also cited the advantage of cine MRI over nasopharyngeal endoscopy because it enables simultaneous assessment of multiple levels of obstruction. In contrast to cine MRI studies, reports of DISE originated from multiple institutions, suggesting that this technique is more widely used than cine MRI. These reports also noted that the tongue base and adenoids were common sites of obstruction, primarily due to lingual tonsillar hypertrophy, as were the inferior turbinates19,20,24 and supraglottis.21,26 Studies in this diagnostic category had heterogeneous objectives that ranged from comparison of the technique to awake endoscopy,26 interrater reliability of scoring systems,28 and utility of DISE to accurately direct surgical plans.19–24 In six studies,19–24 children underwent surgical interventions after DISE. These interventions included uvulopalatopharyngoplasty, supraglottoplasty, coblation, lingual tonsillectomy, midline posterior glossectomy, revision adenoidectomy, and inferior turbinate submucosal reduction. None of these studies included information regarding specific phenotypes that might be useful in predicting surgical outcomes. A challenge associated with both DISE and cine MRI is the need for optimization of the anesthetic to induce sleep without inducing airway obstruction. As well, there is no agreed upon method of recording the results of either DISE or cine MRI and no clear evidence of outcomes for procedures that are selected based upon obstruction seen during these evaluations. Similarly, information pertaining to the use of sleep fluoroscopy30 to evaluate the airway and plain films31 to assess the lingual tonsils is sparse, as is data assessing their utility. Once the site of obstruction is identified, treatment of OSA is warranted. In children with persistent OSA following T&A, adenoid and tonsillar

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

TABLE IV. Characteristics of Other Studies That Identify Site of Obstruction for Children With Persistent OSA Following Adenotonsillectomy. Mean Age (Years)

Diagnostic Methodology

Children With Persistent OSA

Children With Identified Site of Obstruction

Author

Year

LOE

N

Gibson

1996

2b

50

Not reported

SF

13/50 (26%)

Not reported

SF may be a sensitive technique to tailor surgical therapy. Useful as an adjunct to direct laryngoscopy and bronchoscopy in identifying upper airway obstruction in children with hypopharyngeal collapse and multilevel disease. Disadvantages include radiation and the need for sedation.

Sedaghat

2012

2b

89 DS

5.2*

105 control

5.4*

C-Spine radiographs C-Spine radiographs

Not reported Not reported

5/105 (4.8%) 0/105 (0%)

Lingual tonsillar enlargement and hypertrophy are more common in children with DS than in age- and gendermatched controls. Evidence is limited regarding use of C-spine radiographs as a screening tool to identify LTH.

27

13 (2.3)

MRI

27/27 (100%)

Not reported

All children were obese. Persistent OSA in obese children may be due to residual adenoid tissue, increased soft palate volume, or increased tongue volumes. It is uncertain whether this is due to normal postop growth or progression of obesity.

Nandalike

2013

2b

Comments

*No standard deviation reported. DS 5 Down syndrome; LOE 5 level of evidence; LTH 5 lingual tonsillar hypertrophy; MRI 5 magnetic resonance imaging; N 5 total number of patients; OSA 5 obstructive sleep apnea; postop 5 postoperative; SF 5sleep fluoroscopy

regrowth are typically evaluated; however, we did not identify articles discussing the need for revision adenoidectomy or completion tonsillectomy. Regardless of the technique used to identify sites of persistent obstruction, surgical intervention is directed to the specific site or sites identified. Lingual tonsillectomy was the surgical method reported most often for children after T&A, with six

studies19,20,29,33–35 representing the experience at three institutions. Resolution of OSA ranged from 57% to 88% across these series, although some children were noted to have additional sites of obstruction (e.g., the supraglottis) that were also addressed surgically. Truong et al.19 noted that children with tongue base collapse from hypotonia or craniofacial anomalies did not fare as well as those who did not have collapse.

TABLE V. Characteristics of Supraglottoplasty Studies for Treatment of Children With Persistent OSA Following Adenotonsillectomy.

Author

Year

LOE

N

Mean Age (years)

Complete Resolution of OSA

O2 Saturation Nadir Preop/ Postop

Chan et al.

2012

4

24

7.3 (0.8)

14/24 (58%)

Chan & Koltai

2012

4

9

Not reported

Truong et al.

2012

4

8

Digoy et al.

2012

4

36

AHI/RDI Preop

AHI/RDI Postop

88.0/88.8 (mean)

14.9 (2.8)

4.9 (1.1)

Not reported

Not reported

10.4*

2.9*

Not reported

Not reported

Not reported

9.7 (8.3)

5.7 (5.3)

4.7 (median)

26/36 (72%)

83/86.5

13.3 (12.9)

4.1 (5.0)

Comments

58% of children had a postop AHI < 5, whereas 18% had a postop AHI < 1 in all children undergoing SGP. 50% of SGP patients had medical comorbidities. Higher BMI identified as a risk factor for poorer outcomes following SGP. 22 patients underwent SGP; 9 underwent SGP alone. Comparison of preop and postop AHI for concurrent vs. isolated SGP showed no significant difference. 4 children underwent concurrent SGP. Sleep endoscopy was useful in identifying occult laryngomalacia, which may be amenable to SGP. 72% has postop AHI < 5; 31% had postop AHI < 1. 11% did not have T&A prior to surgery due to having very small-to-average-sized tonsils. SGP can reduce AHI in patients with statedependent laryngomalacia. Transient dysphagia is a primary complication.

*No standard deviation reported. OSA = obstructive sleep apnea; AHI 5 apnea-hypopnea index; LOE 5 level of evidence; N 5 total number of patients; O2 5 oxygen; Postop 5 postoperative; Preop 5 preoperative; RDI 5 respiratory disturbance index; SGP 5 supraglottoplasty; T&A 5 adenotonsillectomy.

Laryngoscope 126: February 2016

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

497

TABLE VI. Characteristics of Lingual Tonsillectomy Studies Evaluating Treatment of Children With Persistent OSA Following Adenotonsillectomy. Mean Age (years)

Complete Resolution of OSA

O2 Saturation Nadir Preop/ Postop

AHI/RDI Preop

AHI/RDI Postop

84.0/91.0 (nadir)

10.5†

3.2†

39/68 (57%)

88.8/90.8 (mean)

11.8 (0.9)

5.7 (0.8)

7†

Not reported

Not reported

18.3 (15.3)

9.7 (11.4)

26

11†

16/26 (61%)

89/90

14.7*†

8.1*†

61% of patients had postop AHI < 5; 17% had postop AHI < 1. Inflammation of pharyngeal mucosa or muscle hypotonia may lead to treatment failure.

9

8.57 (4.1)

6/9 (66%)

83/84 (nadir)

8.51 (7.0)

4.07 (1.9)

22 children underwent LT; only 9 had preop and postop PSG data. 1 had LT alone; concurrent procedures included midline posterior glossectomy, inferior turbinate reduction, revision adenoidectomy, and supraglottoplasty. 66% of patients had postop AHI < 5; 0% had postop AHI < 1.

Author

Year

LOE

N

Abdel-Aziz

2011

4

16

Not reported

14/16 (88%)

Chan et al.

2012

4

68

11 (0.5)

Truong

2012

4

31

Lin

2009

4

Wootten

2014

2b

Comments

14/16(88%) children identified on CT or MRI had complete resolution of OSA (AHI cutoff not reported). 3 children developed postop edema. 8/68 children had concurrent supraglottoplasty. 57% had an AHI < 5 after surgery, whereas 22% had an AHI < 1 event/hour. Obese patients had postop AHI twice as high as nonobese patients. 4/31 patients had concurrent supraglottoplasty. Patients with comorbidities and craniofacial anomaly did not benefit as much from lingual tonsillectomy as those without.

*RDI reported, not AHI. † No standard deviation reported. AHI 5 apnea-hypopnea index; LOE 5 level of evidence; LT 5 lingual tonsillectomy; N 5 total number of patients; O2 5 oxygen; OSA 5 obstructive sleep apnea; Postop 5 postoperative; Preop 5 preoperative; PSG 5 polysomnography; RDI 5 respiratory disturbance index.

Chan et al.33 noted that obesity further exacerbated lingual tonsillar hypertrophy and contributed to persistent sleep apnea even after removal of the lingual tonsillar tissue. Supraglottoplasty for laryngomalacia was the next most common procedure reported in the literature, with four studies19,21,22,33 representing the experience of two institutions. Indications included laryngomalacia while awake, as well as sleep state-dependent laryngomalacia also known as occult laryngomalacia, which were seen only during sleep.21 Resolution of OSA was reported in 58% to 72% of selected patients. A subset of patients was also treated with concurrent procedures to address multiple sites of obstruction. A single study35 evaluated tongue suspension with radiofrequency to the base of tongue, which resulted in resolution of OSA in 61% of patients. A second study29 evaluated DISE-directed intervention, which usually resulted in multilevel surgery, including a combination of lingual tonsillectomy, posterior midline glossectomy, revision adenoidectomy, inferior turbinate reduction, uvulopalatopharyngoplasty, and supraglottoplasty. This trial reported that only 23% of children had OSA resolution; however, the pooling of procedures makes it hard to generalize these results. Subperiosteal release of the floor of mouth was also reported in two studies36,37 from a single institution. The first36 reported an improvement in the mean oxygen saturation in the six children who Laryngoscope 126: February 2016

498

were evaluated with pulse oximetry. The second37 used PSG to determine a resolution rate of 84%; however, the criterion for resolution was not clearly defined. Lastly, a single study38 reported on the efficacy of montelukast medical therapy to resolve residual mild OSA and noted a decrease in the AHI of 3.6 events/hour; a resolution rate was not reported. Our study had several shortcomings. We identified a limited number of studies, including only one randomized controlled trial, which focused on identification of the site(s) of obstruction or treatment of persistent pediatric OSA. Moreover, there was significant heterogeneity in diagnostic modalities and outcomes reported, especially for articles pertaining to diagnosis of site of obstruction. Additionally, there was little standardization of interpretation of DISE studies, and correlation with outcomes was infrequently reported. Despite these limitations, our review provides a synthesis and evaluation of the literature regarding treatment and site of obstruction identification.

CONCLUSION Although T&A is highly effective in decreasing the severity of pediatric OSA, many children continue to suffer from persistent OSA after surgery. Persistent disease is particularly common in children with medical comorbidities such as DS, craniofacial abnormalities, and obesity. These children typically have residual obstruction

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

Laryngoscope 126: February 2016

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment

499

2b

4

4

4

2a

LOE

26

31

31

9.3 (3.4)

0.1*

11.5*

0.1*

6.5 (1.8)

14 control

12

6.3 (1.3)

22 treated

N

Supraglottoplasty (2)

Uvulopalatoplasty (2)

Midline posterior glossectomy (16) Revision adenoidectomy (11) Inferior turbinate reduction (7)

Lingual tonsillectomy (22)

DISE-directed surgery

Subperiosteal release of the floor of mouth

Tongue suspension (termed genioglossal advancement) and radiofrequency ablation of the base of tongue

Subperiosteal release of the floor of mouth

Oral montelukast and intranasal budesonide for 12 weeks

Treatment Modality

86.1/89.1

19/31 (61%)

6/26 (23.1%)

Not reported

Not reported

90.8/94.7 (mean)

Not reported

26/31 (84%)* Not defined

87.4/84.0

87.3/92.5

O2 Saturation Nadir Preop/ Postop

Not reported

Not reported

Complete Resolution of OSA

7.0 (5.8)

46.5*

14.0 (10.1)

Not reported

3.6 (1.4)

3.9 (1.2)

AHI/RDI Preop

3.6 (1.8)

17.4*

6.5 (6.4)

Not reported

4.7 (1.5)

0.3 (0.3)

AHI/RDI Postop

28/31 patients underwent cine MRI to identify obstruction; all had base of tongue collapse. Genioglossal advancement and radiofrequency ablation may be useful for patients with retroglossal and base-of-tongue obstruction. All children had PRS. Subperiosteal release of floor of mouth is an effective primary surgery for children with airway obstruction secondary to PRS. 23/26 underwent multilevel surgery. 4/26 objectively or subjectively (2 worsened on postop PSG) DISE-directed surgery. No data supplied on the outcomes of individual procedures.

All children had PRS. Subperiosteal release of the floor of mouth is an effective treatment for upper airway obstruction in children with PRS.

In children with mild persistent OSA, combination oral antileukotriene and intranasal steroid therapy is beneficial. There is limited evidence regarding the duration of combined therapy as well as potential for rebound disease.

Comments

*No standard deviation reported. AHI 5 apnea hypopnea index; DISE 5 drug-induced sleep endoscopy; LOE 5 level of evidence; N 5 total number of patients; O2 5 oxygen; OSA 5 obstructive sleep apnea; Postop 5 postoperative; Preop5 preoperative; PRS 5 Pierre Robin sequence; RDI 5 respiratory disturbance index; MRI 5 magnetic resonance imaging.

2014

2010

Wootten

Wootten

1996

CaouetteLaberge

2012

2006

Kheirandish

CaouetteLaberge

Year

Author

Mean Age (Years)

TABLE VII. Characteristics of Other Studies That Discuss Treatment of Children With Persistent OSA Following Adenotonsillectomy.

at multiple sites in the upper airway, and sleep endoscopy and cine MRI are reported to be useful methods of both evaluating dynamic airway collapse and identifying the site(s) of obstruction. The information gleaned from these tools is useful in tailoring subsequent surgical intervention; however, information pertaining to appropriate patient selection is scant. Lingual tonsillectomy and supraglottoplasty are the most commonly reported treatment modalities, and articles suggest that evaluation for lingual tonsillar hypertrophy and laryngomalacia, during wake or sedated evaluation/sleep, should be considered. The generalizability of these reports is, however, limited by the fact that many of them have focused on children with significant comorbidities (e.g., DS). Thus, there clearly remains a need for future studies that focus on commonly seen sites of obstruction and treatment outcomes across a much more diverse patient population; these studies should also include obese as well as otherwise healthy children.

BIBLIOGRAPHY 1. Marcus CL, Brooks LJ, Draper KA, et al. Diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics 2012;130:576–584. 2. Baugh RF, Archer SM, Mitchell RB, et al. Clinical practice guideline: tonsillectomy in children. Otolaryngol Head Neck Surg 2011;144(suppl 1): S1–S30. 3. 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 Resp Crit Care Med 2010;182: 676–683. 4. Fricke BL, Donnelly LF, Shott SR, et al. Comparison of lingual tonsil size as depicted on MR imaging between children with obstructive sleep apnea despite previous tonsillectomy and adenoidectomy and normal controls. Pediatr Radiol 2006;36:518–523. 5. Mitchell RB. Adenotonsillectomy for obstructive sleep apnea in children: outcome evaluated by pre- and postoperative polysomnography. Laryngoscope 2007;117:1844–1854. 6. O’Brien, LM, Sitha S, Baur LA, Waters KA. Obesity increases the risk for persisting obstructive sleep apnea after treatment in children. Int J Pediatr Otorhinolaryngol 2006;70:1555–1560. 7. Friedman M, Wilson M, Lin HC, Chang HW. Updated systematic review of tonsillectomy and adenoidectomy for treatment of pediatric obstructive sleep apnea/hypopnea syndrome. Otolaryngol Head Neck Surg 2009;140: 800–808. 8. Mitchell RB, Kelly J. Behavior, neurocognition and quality-of-life in children with sleep-disordered breathing. Int J Pediatr Otorhinolaryngol 2006;70:395–406. 9. Thevasagayam M, Rodger K, Cave D, Witmans M, El-Hakim H. Prevalence of laryngomalacia in children presenting with sleep-disordered breathing. Laryngoscope 2010;120:1662–1666. 10. Strome M. Obstructive sleep apnea in Down syndrome children: a surgical approach. Laryngoscope 1986;96:1340–1342. 11. Marcus CL, Moore RH, Rosen CL, et al. A randomized trial of adenotonsillectomy for childhood sleep apnea. N Engl J Med 2013;368:2366–2376. 12. Fauroux, B. What’s new in paediatric sleep? Paediatr Respir Rev 2007;8: 85–89. 13. Pirelli P, Saponara M, Guilleminault C. Rapid maxillary expansion in children with obstructive sleep apnea syndrome. Sleep 2004;27:761–766. 14. Woodson BT. Non-pressure therapies for obstructive sleep apnea: surgery and oral appliances. Respir Care 2010;55:1314–1321. 15. Lipsey MW, Wilson DB. Practical Meta-Analysis. 1st ed. Thousand Oaks, CA: Sage Publication; 2000. 16. Shott SR, Donnelly LF. Cine magnetic resonance imaging: evaluation of persistent airway obstruction after tonsil and adenoidectomy in children with Down syndrome. Laryngoscope 2004;114:1724–1729. 17. Donnelly LF, Shott SR, LaRose CR, Chini BA, Amin RS. Causes of persistent obstructive sleep apnea despite previous tonsillectomy and adenoidectomy in children with Down syndrome as depicted on static and dynamic cine MRI. AJR Am J Roentgenol 2004;183:175–181.

Laryngoscope 126: February 2016

500

18. Fricke BL, Donnelly LF, Shott SR. Comparison of lingual tonsil size as depicted on MR imaging between children with obstructive sleep apnea despite previous tonsillectomy and adenoidectomy and normal controls. Pediatr Radiol 2006;36:518–523. 19. Truong MT, Woo VG, Koltai PJ. Sleep endoscopy as a diagnostic tool in pediatric obstructive sleep apnea. Int J Pediatr Otorhinolaryngol 2012; 76:722–727. 20. Lin AC, Koltai PJ. Persistent pediatric obstructive sleep apnea and lingual tonsillectomy. Arch Otolaryngol Head Neck Surg 2009;141:81–85. 21. Chan DK, Truong MT, Koltai PJ. Supraglottoplasty for occult laryngomalacia to improve obstructive sleep apnea syndrome. Arch Otolaryngol Head Neck Surg 2012;138:50–54. 22. Digoy GP, Shukry M, Stoner JA. Sleep apnea in children with laryngomalacia: diagnosis via sedated endoscopy and objective outcomes after supraglottoplasty. Arch Otolaryngol Head Neck Surg 2012;147:544–550. 23. Myatt HM, Beckenham EJ. The use of diagnostic sleep nasendoscopy in the management of children with complex upper airway obstruction. Clin Otolaryngol 2000;25:200–208. 24. Durr ML, Meyer AK, Kezirian EJ, Rosbe KW. Drug-induced sleep endoscopy in persistent pediatric sleep-disordered breathing after adenotonsillectomy. Arch Otolaryngol Head Neck Surg 2012;138:638–643. 25. Fung E, Witmans M, Ghosh M, Cave D, El-Hakim H. Upper airway findings in children with Down syndrome on sleep nasopharyngoscopy: casecontrol study. J Otolaryngol Head Neck Surg 2012;41:138–144. 26. Fishman G, Zemel M, DeRowe A, Sadot E, Sivan Y, Koltai PJ. Fiber-optic sleep endoscopy in children with persistent obstructive sleep apnea: inter-observer correlation and comparison with awake endoscopy. Int J Pediatr Otorhinolaryngol 2013;77:752–755. 27. Ulualp S, Szmuk P. Drug-induced sleep endoscopy for upper airway evaluation in children with obstructive sleep apnea. Laryngoscope 2013;123: 292–297. 28. Chan DK, Liming BJ, Horn DL, Parikh SR. A new scoring system for upper airway pediatric sleep endoscopy. JAMA Otolaryngol Head Neck Surg 2012;140:595–602. 29. Wootten CT, Chinnadurai S, Goudy SL. Beyond adenotonsillectomy: outcomes of sleep endoscopy-directed treatments in pediatric obstructive sleep apnea. Int J Pediatr Otorhinolaryngol 2014;78:1158–1162. 30. Gibson SE, Myer CM III, Strife JL, O’Connor DM. Sleep fluoroscopy for localization of upper airway obstruction in children. Ann Otol Rhinol Laryngol 1996;105:678–683. 31. Sedaghat, AR, Flax-Goldenberg RB, Gayler BW, Capone GT, Ishman SL. A case-control comparison of lingual tonsillar size in children with and without Down syndrome. Laryngoscope 2012;122:1165–1169. 32. Nandalike K, Shifteh K, Sin S, et al. Adenotonsillectomy in obese children with obstructive sleep apnea syndrome: magnetic resonance imaging findings and considerations. Sleep 2013;36:841–847. 33. Chan DK, Jan TA, Koltai PJ. Effect of obesity and medical comorbidities on outcomes after adjunct surgery for obstructive sleep apnea in cases of adenotonsillectomy failure. Arch Otolaryngol Head Neck Surg 2012; 138:891– 896. 34. Abdel–Aziz M, Ibrahim N, Ahmed A, Hamamsy ME, Abdel-Khalik MI, ElHoshy H. Lingual tonsils hypertrophy; a cause of obstructive sleep apnea in children after adenotonsillectomy: operative problems and management. Int J Pediatr Otorhinolaryngol 2011;75:1127– 1131. 35. Wootten CT, Shott SR. Evolving therapies to treat retroglossal and base of tongue obstruction in pediatric obstructive sleep apnea. Arch Otolaryngol Head Neck Surg 2010;136:983–987. 36. Caouette–Laberge L, Plamondon C, Larocque Y. Subperiosteal release of the floor of the mouth in Pierre Robin sequence: experience with 12 cases. Cleft Palate Craniofac J 1996;33:468– 472. 37. Caouette–Laberge L, Borsuk DE, Bortoluzzi PA. Subperiosteal release of the floor of the mouth to correct airway obstruction in Pierre Robin sequence: review of 31 cases. Cleft Palate Craniofac J 2012;49:14–20. 38. Kheirandish L, Goldbart AD, Gozal D. Intranasal steroids and oral leukotriene modifier therapy in residual sleep-disordered breathing after tonsillectomy and adenoidectomy in children. Pediatrics 2006;117:61–66. 39. Steinhart H, Kuhn-Lohmann J, Gewalt K, et al. Upper airway collapsibility in habitual snorers and sleep apneics: evaluation with drug-induced sleep endoscopy. Acta Otolaryngol 2000;120:990–994. 40. den Herder C, van Tinteren H, de Vries N. Sleep endoscopy versus modified Mallampati score in sleep apnea and snoring. Laryngoscope 2005; 115:735–739. 41. Ravesloot MJ, de Vries N. One hundred consecutive patients undergoing drug-induced sleep endoscopy: results and evaluation. Laryngoscope 2011;121:2710–2716. 42. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol 2009;62:e1–e34.

Manickam et al.: Persistent Pediatric OSA Evaluation and Treatment